The New Oxford Handbook of Economic Geography 9780198755609, 0198755600

Citation preview

The New Oxford Handbook of

ECONOMIC GEOGRAPHY

The New Oxford Handbook of

ECONOMIC GEOGRAPHY Edited by

GORDON L. CLARK, M A RYA N N P. F E L DM A N , M E R IC S . G E RT L E R and DA R I U S Z WÓJ C I K W I T H T H E A S SI STA N C E O F A N G E L I KA  KA I SE R

1

3 Great Clarendon Street, Oxford, ox2 6dp, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Oxford University Press 2018 The moral rights of the authors‌have been asserted First Edition published in 2018 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2017958294 ISBN 978–​0–​19–​875560–​9 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

For Peter and Shirley, Gordon, Joanna, Isabel and Miles, and Ana

Preface

The New Oxford Handbook of Economic Geography (NOHEG) brings to the fore a vibrant field of research and teaching at the interface between a host of disciplines, including geography, economics, the management sciences, and urban and regional planning. In commissioning essays we sought to engage leading scholars who have made profound and long-​lasting contributions to economic geography along with new voices, new approaches, and new topics of significance to the first half of the twenty-​first century. This sounds ambitious. And it is. But we have framed NOHEG as a compendium of essays which reach back to key concepts and ideas and forward to emerging issues and theoretical perspectives that together underwrite the field of economic geography. Quite obviously, NOHEG builds upon the success of the original Oxford Handbook of Economic Geography (OHEG; 2002). In many respects, that was a very different project than NOHEG. When we were developing the logic and building blocks of OHEG, the field of economic geography was less a shared project across social science disciplines and more a separate thread in the various disciplines that played host to ‘economic geographers’. In this respect, OHEG sought to bring together scholars who were engaged with related issues, even if their theories and analytical approaches were quite different. As such, the compilation was deliberately framed as a dialogue between economists, geographers, and urban and regional planners comparing and contrasting their approaches to common issues. By contrast, NOHEG takes this dialogue as given and seeks to represent the remarkable develop­ment of the field within and across disciplines. Once again, our intention has been to give life to this dialogue without advocating one specific way of being an economic geographer or, for that matter, restricting the focus of the Handbook to a set of issues that are the core of economic geography. Throughout this volume, pluralism reigns supreme. We leave it to the readers to make their own judgements about the salience of issues, the virtues of competing theoretical approaches, and the claims and counterclaims made by contributors about how to conceptualize twenty-​first-​century globalization. In part, our pluralism reflects the editorial team, of whom three were editors of the original volume. The newly added fourth editor brings his own programme of financial geography and valuable experience to the project. All benefit the NOHEG through their expertise in a variety of ways: our first editor has had a longstanding career in the economics discipline, the second has successfully transitioned between economics and geography and public policy and business schools, and our third editor (before becoming university president) is engaged in issues of innovation at the interface between economics, geography, and political science. Inevitably, our separate and common experience frames the project. The OHEG began with a manifesto. Simply and directly, the volume opened with a statement regarding the core principles or foundations upon which the volume was based. We term these principles ‘significant points of departure’, emphasizing difference,

viii   Preface differentiation, and the heterogeneity of the economic landscape. These principles have stood the test of time. And, as you would expect, they underwrite NOHEG. These principles challenge commonplace expectations of convergence in economic prospects and development across regions, nations, and the globe. At the same time in this volume these principles deserve and receive deeper analysis than in the first volume. So much has changed in the twenty years between conceiving OHEG and realizing NOHEG. And yet, as many readers of the first volume have observed, these principles remain contested and contestable. We are pleased that this is the case and hope that NOHEG carries through on the challenge represented by difference, differentiation, and heterogeneity. Our Handbook is dedicated to the memory of Susan Christopherson, who died on 14 December 2016. She was a valued colleague and dear friend whose work was motivated by a desire to understand people, places, politics, institutions, and economic processes. Her contribution to this book has all this and much more. On behalf of the community whose work is embodied by this Handbook, we salute her contribution to academic life and her commitment to friendship. The Editors, December 2016

Acknowledgements

This book was made possible by the support of numerous organizations. In particular, we would like to acknowledge the support of our home institutions, notably (for Gordon Clark and Dariusz Wójick) the Smith School of Enterprise and the Environment, the School of Geography and the Environment, and the Saïd Business School at Oxford University, and, further afield, the Department of Banking and Finance in the Monash Business School at Monash University, the University of Toronto (Meric Gertler), and the University of North Carolina (Maryann Feldman). Without the administrative and organizational support of our home institutions, and without the support and encouragement of our colleagues in those institutions, this book would not have been possible. At Oxford, we were fortunate to have a team that has carried the Handbook through its various stages. We would especially like to thank Alice Chautard, Seth Collins, Angelika Kaiser, Irem Kok, and Sarah McGill for their reading of submissions, their attention to the form and structure of the book and its various sections, and their help in meeting the various deadlines that a book of this size must meet. Also important in this regard has been the enthusiasm and encouragement found in our OUP editors, notably Dominic Byatt and David Musson, Olivia Wells, and the New York-​based Handbook staff led by Sarah Kain. We are very grateful for their engagement with this project, and its previous incarnation. Along the way, we had an opportunity to present both the book and the role of our contributors to the wider academic community through the Fourth Global Conference on Economic Geography (2015) in Oxford. This event was led by Dariusz Wójcik, along with the staff of the Smith School of Enterprise and the Environment and the School of Geography and the Environment. Most importantly, Patrizia Ferrari’s conference management made this possible. At the conference we held a number of meetings with contributors, allowing for a dialogue about their contributions and the significance of the book. We are very grateful to those contributors who participated for their enthusiasm and engagement in this initiative. Finally, any long-​running project such as this incurs many debts—​specifically, the patience of our families and friends. Peter and Shirley, Gordon M. Allen, Joanna, Isabel and Miles, and Ana have, in their different ways, sustained our academic careers and demonstrated their continued support throughout this particular project.

Contents

List of Figures  List of Tables  List of Editors  List of Abbreviations  List of Contributors 

Introduction: Economic Geography in the Twenty-​first Century  Gordon L. Clark, Maryann P. Feldman, Meric S. Gertler, and Dariusz Wójcik

xvii xix xxi xxiii xxxi 1

PA RT I   G RO U N DE D I N P L AC E 1. Global Prospects: The Asian Century?  Michael Berry and Benno Engels

19

2. Inequality in Advanced Economies  Danny Dorling

39

3. Income Inequality and Growing Disparity: Spatial Patterns of Inequality and the Case of the USA  Amy K. Glasmeier

63

4. The Emerging Transformation of China’s Economic Geography  Kam Wing Chan

78

5. Economic Growth and Poverty Reduction in Contemporary India  Stuart Corbridge

97

6. Crisis and Austerity in Action: Greece  Maria Tsampra

113

xii   Contents

PA RT I I   C ON C E P T UA L F OU N DAT ION S 7. Economic Growth and Economic Development: Geographical Dimensions, Definition, and Disparities  Maryann P. Feldman and Michael Storper

143

8. Heterodoxy as Orthodoxy: Prolegomenon for a Geographical Political Economy  Eric Sheppard

159

9. Relational Research Design in Economic Geography  Harald Bathelt and Johannes Glückler

179

10. Behaviour in Context  Gordon L. Clark

196

11. Evolutionary Economic Geography  Ron Boschma and Koen Frenken

213

12. Institutions, Geography, and Economic Life  Meric S. Gertler

230

PA RT I I I   I N N OVAT ION 13. Economic Ecosystems  Philip E. Auerswald and Lokesh Dani

245

14. How Geography Shapes—​and Is Shaped by—​the Internet  Chris Forman, Avi Goldfarb, and Shane Greenstein

269

15. Schumpeterian Customers? How Active Users Co-​create Innovations  286 Gernot Grabher and Oliver Ibert 16. The Geography of the Creative Industries: Theoretical Stocktaking and Empirical Illustration  Mark Lorenzen 17. Firms in Context: Internal and External Drivers of Success  Mercedes Delgado

305 324

Contents   xiii

PA RT I V   T H E F I R M 18. The Logic of Agglomeration  Gilles Duranton and William R. Kerr

347

19. Network Geographies and Geographical Networks: Co-​dependence and Co-​evolution of Multinational Enterprises and Space  366 Simona Iammarino and Philip McCann 20. The Logic of Production Networks Henry Wai-​chung Yeung 21. Global Sourcing of Business Processes: History, Effects, and Future Trends  Stephan Manning, Marcus M. Larsen, and Chacko G. Kannothra

382

407

22. Towards New Economic Geographies of Retail Globalization  Neil M. Coe and Neil Wrigley

427

23. Corporate Social Responsibility and Standards  Alex Hughes

448

PA RT V   WOR K  24. Pluralizing Labour Geography  Jamie Peck

465

25. Precarious Work and Winner-​Take-​All Economies  Kendra Strauss

485

26. Talent, Skills, and Urban Economies  Richard Florida and Charlotta Mellander

499

27. Immigration and the Politics of Skill  Natasha Iskander and Nichola Lowe

519

PA RT V I   F I NA N C E  28. Finance and Financial Systems: Evolving Geographies of Crisis and Instability  Gary A. Dymski

539

xiv   Contents

29. The Global Financial Networks  Dariusz Wójcik

557

30. Information Flows, Global Finance, and New Digital Spaces  Matthew Zook

575

31. ‘Organic Finance’: The Incentives in Our Investment Products  Ashby Monk and Rajiv Sharma

591

32. Financialization of Everyday Life  Karen P.Y. Lai

611

33. Infrastructure and Finance  Phillip O’Neill

628

PA RT V I I   R E S OU RC E S A N D T H E E N V I RON M E N T 34. The Financialization Thesis Revisited: Commodities as an Asset Class  Sarah McGill

645

35. Vulnerable Regions in a Changing Climate  Robin Leichenko

665

36. Carbon Markets: Resource Governance and Sustainable Valuation  Janelle Knox-​Hayes

683

37. Long-​run Resource Scarcity  Dieter Helm

703

38. Reconceptualizing Resource Peripheries  Caitlin A. McElroy

715

39. Outside Regional Paths: Constructing an Economic Geography of Energy Transitions  Susan Christopherson†

732

PA RT V I I I   ST R AT E G I E S F OR DE V E L OP M E N T  40. Green Growth  Cameron Hepburn, Alexander Pfeiffer, and Alexander Teytelboym † Deceased

749

Contents   xv

41. Pursuing Equitable Economic Growth in the Global South  Andrés Rodríguez-​Pose and Callum Wilkie

770

42. Just Growth: Strategies for Growth with Equity  Karen Chapple

792

43. Policy Through Practice: Local Communities, Self-​Organization, and Policy  Jennifer Clark 44. Innovation Highways and the Geography of Inclusive Growth  Anita M. McGahan and Janice Gross Stein

810 826

45. Shocking Aspects of Regional Development: Towards an Economic Geography of Resilience  Ron Martin

839

Author Index  Subject Index 

865 893

List of Figures

0.1 Frequency of Geographical Bigrams Occurring in Books Included in the English-​language Corpus of Google Books  4.1 De Facto Urban Population and Urban Hukou Population, 1955–​2013

10 85

5.1 India’s Long-​term Growth, 1950–​2009 

100

5.2 Headcount Index of Poverty Using the National Poverty Line (Percentage) 

103

5.3 India: Real Patterns of Consumption Growth 

105

6.1 Employment (15–​64 years) Rates (%) EU27, 2008, 2010, and 2012

117

6.2 Unemployment (15–​64 years) Rates (%) EU27, 2008, 2010, and 2013

118

6.3 Long-​term Unemployment Rates, 2008 and 2014 

118

6.4 Youth Unemployment (15–​24 years) Rates (Percentage Labour Force) EU27, 2008, 2012, and 2013 

119

6.5 At Risk of Poverty or Social Exclusion (Percentage of Total Population) 

121

6.6 Gross Domestic Product and Employment Rates in Greece, 1996–​2011 

122

6.7 Greek and European Union Small-​and Medium-​sized Enterprise performance, 2008–​14 (estimates for 2011 onwards) 

128

10.1 Interaction Between the Environment (Resources) and the Nature of the Decision Problem 

206

13.1 Sewall Wright’s Fitness Landscapes 

249

13.2 Ecosystem Boundaries—​US/Mexico Border

254

13.3 Succession and Reorganization of Ecosystems 

257

13.4 Phase Characteristics of the Entrepreneurial Ecosystem 

258

17.1 Portfolio of Fifty-​one Traded Clusters and their Connections 

327

17.2 Strong IT and Analytical Instruments Clusters in the U.S., 2011

329

18.1 Agglomeration and City Formation 

350

18.2 Internal Structure of Clusters 

355

18.3 Tech Sourcing in Silicon Valley Core 

357

18.4 Tech Sourcing Around Silicon Valley 

358

20.1 Strategic Coupling and Global–​Local Economic Integration Through Production Networks 

388

21.1 Ambivalent Effect of Service Commoditization on Geographical Cluster Growth

414

xviii   List of Figures 21.2 Global Delivery Model 

416

22.1 Levels of Cross-​border Online Purchases 

442

26.1 The Growth of the Creative Economy Share 

502

26.2 Creative Class Shares in US Metropolitan Regions 

504

26.3 Social, Analytical; and Physical Skills in US Metropolitan Regions

507

29.1 The Structure of Finance 

559

29.2 The Map of Finance 

560

29.3 Finance in the Global Economy 

562

29.4 Fordism and Flexible Accumulation 

568

29.5 The Share of Financial and Business Services in Total Employment 

569

30.1 Bitcoin to US Dollar Exchange Rates, 2013–​15 

583

31.1 Existing Opaque, Overly Packaged Form of Investment Management 

600

31.2 Organic Finance: Greater Investor Responsibility on Fees and Efficiency of Investment Access Points 

600

34.1 Notional Amount of Outstanding Over-​the-​Counter Commodity Derivatives, December 1998–​June 2010 (US$ trillion)

651

34.2 Financial Investment in Commodities, Assets Under Management, by Product, 2005–​11 (US$ billion)

652

36.1

688

Marshall’s Supply-​and-​Demand Curve 

36.2 Typology of the Spatial and Temporal Dynamics of Value 

691

36.3 Economies of Production and Internalization Through the Circulation of Value 

693

37.1 US Field Production of Crude Oil (1000 Barrels per Day), 1920–​2014 

706

40.1 Commodity Prices Over the Last 100 Years 

752

40.2 Forecast Exhaustion Dates of Minerals Extend into the Future 

753

40.3 Selected Planetary Boundaries Already Breached 

755

40.4 Progress in Solar versus Coal and Nuclear

763

41.1 Gini Coefficients Among Selected Developing Countries 

773

41.2 Differences in Interpersonal and Territorial Income Inequalities Among Selected Countries (Second Theil Index) 

774

45.1 Differing City Responses to the Great Recession in the USA, 2007–​09: New York, Chicago, Atlanta, and Phoenix Compared 

842

45.2 Stylized Reactions of a Regional Economy to a Shock 

847

45.3 Regional Economic Resilience as Process 

849

45.4 Some Potential Policy Foci for ‘Building’ Regional Adaptive Resilience 

856

List of Tables

1.1 Comparative Growth Rates in Real Gross Domestic Product, Selected Countries  21 1.2 Comparative Change in Per Capita Gross Domestic Product at Purchasing Power Parity (Current International Dollars), Selected Countries: 2000–​12 

22

1.3 Value Added by Sector 

23

2.1 Summary of Income Inequalities in the Five Most Populous European Union Countries (2012) 

44

2.2 Summary of UK Household Income Distribution 2012 

45

2.3 Summary of Income Inequalities in Germany, France, Italy, Spain, and the UK in 2012 

48

2.4 Share of Income Received by Best-​off 1 Per Cent Taxpayers in Rich Countries

49

2.5 The Radio of Incomes of the Best-​off 10 Per Cent of Households Versus the Worst-​off 10 Per Cent Among the World’s Richest Countries

50

2.6 Share in Top Incomes of the 1 Per Cent and Gini Measure of Inequality, Fifteen Affluent Countries Ranked by the Take of the 1 Per Cent

51

2.7 Projections for Rich Countries’ Income Inequality, Percentage Take of Top 1 Per Cent

52

2.8 The Numbers of Bankers Paid Over €1,000,000 in 2012 (Highest Numbers in the European Union) 

53

2.9 Income Inequality in the USA, 2008 and 1970–​2008 (at Real 2008 Rates) 

54

2.10 Income Received from the State by Households in 2012 

56

4.1 Urban Hukou Population, Urban Population, and Gross Domestic Product Shares, 1949−2014 (Percentage of National Total) 

84

4.2 Distribution of Cities, 1982–​2010 

87

4.3 Spatial Gini, 1957–​2010 

88

4.4 Number of Megacities and Urban Population of Large Countries, 2014

89

5.1 Ranking of India’s Poorest States by Gross State Domestic Product Per Capita and Human Development Indicators 

106

6.1 Top Employment Sectors in Greece, Percentage 2000 and 2008 

123

6.2 Top Employment Sectors in the Regions of Greece, 2008 

124

6.3 Top Employment Sectors in Greece, 2009 and 2013 

125

6.4 Top-​six Employment-​loss Sectors (NACE rev2.0) Greece and EU27, 2008Q1–​2011Q1 

125

xx   List of Tables 6.5 Top-​six Employment-​gain Sectors (NACE rev2.0) Greece and EU27, 2008Q1–​2011Q1 

126

6.6 Occupational Structure of Employment (000s) in Trade, Greece, 2008–​13 

130

6.7 Greece’s Path-​dependent Growth Patterns from the 1980s to 2008–​09 Crisis 

132

15.1

289

Situating Co-​creation 

15.2 A Typology of Co-​creation Formats and Practices 

290

20.1 Strategic Coupling, Global Production Networks, and Local Development Trajectories  390 22.1 Leading Transnational Retailers, Ranked by International Revenue in 2013 

430

22.2 Differences in Level of Globalization by Retail Sector, 2010–​13 

432

22.3 Level of Globalization of Top 250 Global Retailers by Region/​Country, 2013 

433

22.4 Top Fifteen e-​Retailers, Ranked by Online Sales, 2013 

441

25.1 Full-​time, part-​time, and temporary employment in the European Union (percentages), 2003, 2007, and 2014

489

25.2 Survey Results (Selected), UK Employees, 2013 

494

List of Editors

Gordon L.  Clark ([email protected]) is the Director of the Smith School of Enterprise and the Environment with cross-​appointments in the Saïd Business School and the School of Geography and the Environment at Oxford University. He holds a Professorial Fellowship at St Edmund Hall, Oxford. He is also Sir Louis Matheson Distinguished Visiting Professor at Monash University’s Faculty of Business and Economics (Melbourne) and a Visiting Professor at Stanford University. Previous academic appointments have been at Harvard’s Kennedy School of Government, Harvard Law School (Senior Research Associate), the University of Chicago, Carnegie Mellon’s Heinz School, and Monash University. Other honours include being Andrew Mellon Fellow at the US National Academy of Sciences and Visiting Scholar Deutscher Akademischer Austausch Dienst at the University of Marburg. Maryann P. Feldman ([email protected]) is the S.K. Heninger Distinguished Professor of Public Policy and Finance at the University of North Carolina, Chapel Hill and the Kenan Flagler Business School. In 2013, she was awarded with the prestigious Global Entrepreneurship Research Award from the Swedish Entrepreneurship Forum and Research Institute of Industrial Economics. She is a member of the Innovation Forum at the National Academies of Science. Her research interests focus on the areas of innovation, the commercialization of academic research, and the factors that promote technological change and economic growth. Meric S. Gertler ([email protected]) is President of the University of Toronto, Professor of Geography and Planning and the Goldring Chair in Canadian Studies. He was the founding co-​director of the Program on Globalization and Regional Innovation Systems (PROGRIS) at the Munk School of Global Affairs. His research focuses on the geography of innovative activity, the economies of city regions, and economic restructuring in North America and Europe. He is the author, co-​author, and co-​editor of more than ninety scholarly articles and chapters, and nine books, including Manufacturing Culture: The Institutional Geography of Industrial Practice. He has served as an advisor to local, regional, and national governments in Canada, the USA, and Europe, and international agencies such as the Organisation for Economic Co-​operation and Development. He is the founding associate editor of the Journal of Economic Geography. Dariusz Wójcik ([email protected]) is a Professor of Economic Geography in the School of Geography and the Environment at Oxford University, a Fellow of St Peter’s College, and a Visiting Professor at Beijing Normal University. His research focuses on finance, corporate governance, and economic globalization. His current project, funded by the European Research Council, investigates how financial and business services have been

xxii   List of Editors affected by the global financial crisis, and how they change in response to new financial regulation, the rise of the Global South, and the digital revolution. The project also focuses on the impacts of finance on regional development. Dariusz is a member of the editorial board of Economic Geography, the Journal of Economic Geography, Environment and Planning A: Economy and Space and GeoJournal and leads the Global Network on Financial Geography (FinGeo).

List of Abbreviations

Prelims NOHEG OHEG

New Oxford Handbook of Economic Geography Oxford Handbook of Economic Geography

Introduction GDP IMF

gross domestic product International Monetary Fund

Chapter 1 FDI GDP GFC IMF SOEs

foreign direct investment gross domestic product global financial crisis International Monetary Fund state-​owned enterprises

Chapter 2 EU European Union EU-​SILC European Union Statistics on Income and Living Conditions IMF International Monetary Fund

xxiv   list of abbreviations

Chapter 3 CMSA IMF OECD SBTC UNDP

Consolidated Metropolitan Statistical Area International Monetary Fund Organisation for Economic Co-​operation and Development skill-​biased technological change United Nations Development Programme

Chapter 4 FDI GDP

foreign direct investment gross domestic product

Chapter 5 AJR BJP GDP PPP TFP

Acemoglu, Johnson, and Robinson Bharatiya Janata Party gross domestic product purchasing power parity total factor productivity

Chapter 6 EC European Commission ECB European Central Bank EEC European Economic Community EMU European Monetary Union EU European Union FDI foreign direct investment GDP gross domestic product GVA gross value added HRADF Hellenic Republic Asset Development Fund IMF International Monetary Fund SME small-​and medium-​sized enterprise

List of abbreviations     xxv

Chapter 7 EU European Union

Chapter 11 EEG RIS

evolutionary economic geography regional innovation system

Chapter 12 CME EEG LEP LME

coordinated market economy evolutionary economic geography local enterprise partnership liberal market economy

Chapter 13 MSA

metropolitan statistical area

Chapter 14 ISP IT

Internet service provider information technology

Chapter 15 DCA dichloroacetic acid GPS Global Positioning System

xxvi   list of abbreviations

Chapter 16 IP MNE

intellectual property multinational enterprise

Chapter 17 B2C BCD DSC IT STEM USCMP

business-​to-​consumer Benchmark Cluster Definitions Dual Specialization Correlation information technology Science, Technology, Engineering, and Mathematics U.S. Cluster Mapping Project

Chapter 19 FDI foreign direct investment GCN global city network HFDI horizontal FDI IDP investment development pathway KCM Knowledge Capital Model MNE multinational enterprise NEG new economic geography OLI Ownership–​Location–​Internalization PLC product life cycle VFDI vertical FDI

Chapter 20 GPN GVC OECD UNCTAD WTO

global production network global value chain Organisation for Economic Co-​operation and Development United Nations Conference on Trade and Development World Trade Organization

List of abbreviations     xxvii

Chapter 21 BPO CMMI GDM GE ICT IS IT KSC

business process outsourcing Capability Maturity Model Integration global delivery model General Electric information and communication technology impact sourcing information technology knowledge services cluster

Chapter 22 FDI OECD TNC

foreign direct investment Organisation for Economic Co-​operation and Development transnational corporation

Chapter 23 CSR GPN GVC ILO NGO UN

corporate social responsibility global production network global value chain International Labour Organization non-​governmental organization United Nations

Chapter 24 EU European Union INE involuntary non-​standard employment SER standard employment relationship

Chapter 28 DSGE IMF

dynamic stochastic general equilibrium International Monetary Fund

xxviii   list of abbreviations

Chapter 29 CEO chief executive officer EU European Union FABS financial and business services FDI foreign direct investment GFN global financial networks GPN global production networks IT information technology OECD Organisation for Economic Co-​operation and Development TNC transnational corporation

Chapter 30 AP Associated Press EU European Union HFT high-​frequency trading

Chapter 31 LTI

long-​term investor

Chapter 33 EU European Union SUNFED Special United Nations Fund for Economic Development UN United Nations

Chapter 34 CFTC ETP GSCI OTC

Commodity Futures Trading Commission exchange-​traded product Goldman Sachs Commodity Index over the counter

List of abbreviations     xxix

Chapter 35 ILS

insurance-​linked securitization

Chapter 36 CO2 EU ETS

carbon dioxide European Union Emissions Trading System

Chapter 37 GDP

gross domestic product

Chapter 38 GDP MNC NGO

gross domestic product multinational corporation non-​governmental organization

Chapter 39 SEC

Securities and Exchange Commission

Chapter 40 GDP OECD tCO2 USGS

gross domestic product Organisation for Economic Co-​operation and Development total carbon dioxide United States Geological Survey

xxx   list of abbreviations

Chapter 41 LED OECD

local economic development Organisation for Economic Co-​operation and Development

Chapter 42 CBA

community benefits agreement

Chapter 44 GOI IIG NIS NSI WTO

geography of innovation innovation for inclusive growth national innovation system national systems of innovation World Trade Organization

Chapter 45 GDP NEG

gross domestic product new economic geography

List of Contributors

Philip E. Auerswald  ([email protected]) is an associate professor at the Schar School of Policy and Government, George Mason University. He leads the Global Entrepreneurship Research Network, an initiative of the Kauffman Foundation, and is the co-​founder and co-​editor of Innovations, a quarterly journal published by MIT Press about entrepreneurial solutions to global challenges. He is most recently the author of The Code Economy: A Forty-​Thousand Year History, published by Oxford University Press in 2017. He has served as a consultant to foundations, corporations, and national governments, including the World Bank, Gates Foundation, the National Academy of Sciences, and the National Institute of Standards and Technology. From 2010 to 2013 he was an advisor to the Clinton Global Initiative, focusing on job creation and market-​ based solutions. Harald Bathelt  ([email protected]) is the Canada Research Chair in ‘Innovation and Governance’ at the Department of Political Science, University of Toronto, Canada. He also holds an appointment as Professor in the University of Toronto’s Department of Geography and Planning. His research and teaching interests are in the areas of industrial and economic geography, political economy, and methodology, specifically in the analysis of knowledge generation and innovation processes over distance, industrial clustering, and the socio-​economic impacts of regional and industrial change. For the past six years, Professor Bathelt was a Visiting Professor at the East China Normal University in Shanghai, China, and was previously a Visiting Professor at the University of Heidelberg, Germany and HEC Montréal, Canada. Since 2012, he has been editor of the Journal of Economic Geography. Michael Berry  ([email protected]) is Emeritus Professor at Urban Studies and Public Policy, Centre for Urban Research, RMIT University, Melbourne, Australia. His research has focused on urban development processes, urban social theory, economics and public policy, and housing markets and policy. His focus has been on alternative financing approaches for affordable housing, and the impact of housing markets on the macro-​economy for which he has been awarded numerous significant grants, most notably from the Australian Housing and Urban Research Institute, which he helped found. He has been on the editorial board of a number of journals, including Housing,Theory and Society and Urban Policy and Research. He has visited a number of universities in Europe, Asia, and North America, most recently as Helen Kam visiting fellow at Girton College, Cambridge. Ron Boschma ([email protected]) is Professor in Regional Economics at the Urban and Regional Research Centre Utrecht (URU) at Utrecht University, the Netherlands. He is also Professor II in Innovation Studies at UiS Business School and Stavanger Centre for

xxxii   List of Contributors Innovation Research, University of Stavanger, Norway. In 2013, Professor Boschma received an Honorary Doctorate in the Natural Sciences (‘Doktor der Naturwissenschaften ehrenhalber’) from Marburg University (Fachbereich Geographie) in Germany. His research interests include evolutionary economic geography; spatial evolution of industries; geography of innovation; geography of knowledge networks; agglomeration externalities and regional growth; and regional diversification. He is editor of the Cambridge Journal of Region, Economy and Society, and associate editor of Industrial and Corporate Change, Papers in Regional Science, and Regional Studies. Kam Wing Chan ([email protected]) is Professor of Geography at the University of Washington. His main research focuses on China’s cities, migrant labour, and the household registration system. He is the author of Cities with Invisible Walls:  Reinterpreting Urbanization in Post-​1949 China (Oxford University Press, 1994). He is Associate Editor of Eurasian Geography and Economics. In the last two decades he has worked as a consultant for the World Bank, Asian Development Bank, United Nations, and McKinsey & Co. on several major policy projects on China. His recent commentaries and interviews have appeared in The Wall Street Journal, The New York Times, The Economist, South China Morning Post, BBC, CBC Radio, Caixin, Ming Pao, and China Daily. Karen Chapple  ([email protected]) is a Professor of City and Regional Planning at the University of California, Berkeley. Chapple specializes in housing, community, and economic development, as well as regional planning. She has most recently published on job creation on industrial land (in Economic Development Quarterly), regional governance in rural Peru (in the Journal of Rural Studies), and accessory dwelling units as a smart-​growth strategy (in the Journal of Urbanism). Her recent book (Routledge, 2014) is entitled Planning Sustainable Cities and Regions: Towards More Equitable Development. In 2015 she launched the Urban Displacement Project, a research portal examining patterns of residential, commercial, and industrial displacement, as well as policy and planning solutions. Susan Christopherson  was a Professor and Chair of the Department of City and Regional Planning in the School of Art, Architecture and Planning at Cornell University. She published a series of key articles and a prize-​winning book examining how market governance regimes influence regional economic development and firm strategies. She was a recognized expert in the field of media studies with a record of research and publication on the media entertainment industries. Professor Christopherson published more than 100 articles and book chapters on topics illuminating the spatial dimensions of economy and society, and served on numerous editorial boards. Most recently, she was editor-​in-​chief of the Regional Studies Association/​Taylor Francis Book Series on Cities and Regions. Gordon L. Clark  ([email protected]) DSc (Oxon) FBA is the Director of the Smith School of Enterprise and the Environment with cross-​appointments at the Saïd Business School and the School of Geography and the Environment at Oxford University. He holds a Professorial Fellowship at St Edmund Hall, Oxford. He is also Sir Louis Matheson Distinguished Visiting Professor at Monash University’s Faculty of Business and Economics (Melbourne) and a Visiting Professor at Stanford University. Previous academic appointments have been at Harvard’s Kennedy School of Government, Harvard Law School (Senior Research Associate), the University of Chicago, Carnegie Mellon’s Heinz School, and

List of Contributors    xxxiii Monash University. Other honours include being Andrew Mellon Fellow at the US National Academy of Sciences and Visiting Scholar Deutscher Akademischer Austausch Dienst at the University of Marburg. Jennifer Clark  ([email protected]) is Director of Georgia Tech’s Center for Urban Innovation and an Associate Professor in the School of Public Policy. Dr Clark publishes work on the development and discussion of urban and regional policies and their effect on cities and their economic resilience. Her book, Remaking Regional Economies (with Susan Christopherson), won the Best Book Award from the Regional Studies Association in 2009. She published Working Regions in 2013 and the Handbook of Manufacturing Industries in the World Economy in 2015. She has worked on innovation policy projects with a broad range of organizations, including the OECD and the Canadian, UK, and US governments. She earned her PhD from Cornell University, an MPlan from the University of Minnesota, and a BA from Wesleyan University in Connecticut. She is the current chair of the Economic Geography Specialty Group (EGSG) of the Association of American Geographers and an editor of Regional Studies. Neil M. Coe  ([email protected]) is Professor of Economic Geography at the National University of Singapore, where he is also co-​director of the Global Production Networks Research Centre. His research interests are in the areas of global production networks and local economic development; the geographies of local and transnational labour markets; the geographies of innovation; and institutional and network approaches to economic development. He has published over seventy-five articles and book chapters on these topics, and is a co-​author or co-​editor of six books. He is currently an editor of the Journal of Economic Geography, and on the editorial board of European Urban and Regional Studies. Stuart Corbridge  ([email protected]) is the Vice Chancellor and Warden of Durham University. Among his previous appointments were a Lectureship in South Asian Geography at Cambridge University, a Professorship of International Studies at the University of Miami, and a Chair in Geography at the London School of Economics. In 2013, he was appointed as the first Deputy Director and Provost of LSE. He has consulted with the UK’s Department for International Development on issues of public service delivery, joint forest management, and participation and empowerment in Jharkhand, India. His major publication (with Glyn Williams, Manoj Srivastava, and Rene Veron) remains Seeing the State: Governance and Governmentality in India. He was also a Managing Editor for the Journal of Development Studies (2005–​10) and Economy and Society (2008–​12). Lokesh Dani  ([email protected]) is a doctoral candidate in public policy at the Schar School of Policy and Government, George Mason University. He is a research assist­ ant for the Center for Regional Analysis where his interests focus on measuring entrepreneurial ecosystems, as well as evaluating the skill composition of regional labour forces. He has an MS degree from New York University with a concentration in International Business, Economics and Development, and a BA from Bucknell University in English. Prior to his studies in economics he worked in biomedical research at the Cardiovascular Research Foundation in New York City. Mercedes Delgado  ([email protected]) is a Senior Lecturer at the MIT Sloan School of Management, and the Research Director and Research Scientist of the MIT Innovation

xxxiv   List of Contributors Initiative Lab for Innovation Science and Policy. She also serves as Senior Associate at the Institute for Strategy and Competitiveness at Harvard Business School. Delgado’s research focuses on the relationship between the regional business environment and the perform­ ance of firms, regions, and countries. She examines the role of regional clusters in job creation, innovation, entrepreneurship, and resilience. In recent work she explores the interaction between the spatial organization of firms, their location choices through the value chain, and firm performance. Delgado has published articles in top economic, policy, and strategy journals. She recently served as a lead researcher on the U.S. Cluster Mapping Project. Danny Dorling ([email protected]) is the Halford Mackinder Professor of Geography, School of Geography and the Environment, University of Oxford. He was previously Professor for the Public Understanding of Social Science, University of Sheffield. Additionally he has been appointed Adjunct Professor, Department of Geography, University of Canterbury, New Zealand. He has served on leading boards and committees, including the Public Health England Mortality Surveillance Steering Group, the Economics and Health Special Interest Group, Faculty of Public Health, and Advisor UK Government Office for Science, Foresight Team. His work concerning issues of housing, health, employment, education, and poverty has resulted in more than a dozen books and several hundred journal papers. Gilles Duranton ([email protected]) is the Dean’s Chair in Real Estate Professor and the Chair of the Real Estate Department at Wharton, University of Pennsylvania. He has published several key articles on urban and transportation issues, and his recent work examines urban growth and the effects of transportation infrastructure on urban development and the evaluation of local policies. His theoretical research concerns the distribution of city sizes, the skill composition, and sectorial patterns of activities in the cities. He is also co-​editor of the Journal of Urban Economics and the Handbook for Regional and Urban Economics, as well as an editorial board member of several other journals. Gary A. Dymski  ([email protected]) is Professor of Applied Economics at the Leeds University Business School, University of Leeds. Professor Dymski has published numerous books, articles, chapters, and studies on banking, financial fragility, urban development, credit-​market discrimination, the Latin American and Asian financial crises, exploitation, housing finance, the subprime lending crisis, financial regulation, the Eurozone crisis, and economic policy. Between 2003 and 2009, Gary served as founding executive director of the University of California Center Sacramento, the University of California’s academic public-​policy programme in California’s state capitol. He is currently co-​leader of the Leeds University interdisciplinary CITIES research initiative. Benno Engels  ([email protected]) is a Senior Lecturer at the School of Global, Urban and Social Studies at RMIT University, Melbourne, Australia. He teaches in the Urban Planning programme in the fields of urban economics, social planning, and urban planning history. He does research in Marxist political economy, Australian social public policy, and urban planning history. Maryann P. Feldman  ([email protected]) is the S.K. Heninger Distinguished Professor of Public Policy and Finance at the University of North Carolina, Chapel Hill and

List of Contributors    xxxv the Kenan Flagler Business School. In 2013, she was awarded with the prestigious Global Entrepreneurship Research Award from the Swedish Entrepreneurship Forum and Research Institute of Industrial Economics. She is a member of the Innovation Forum at the National Academies of Science. Her research interests focus on the areas of innovation, the commercialization of academic research, and the factors that promote technological change and economic growth. Richard Florida  ([email protected]) is University Professor and Director of Cities at the Martin Prosperity Institute at the Rotman School of Management, University of Toronto, and Distinguished Fellow at New York University Schack Institute of Real Estate. He is also the co-​founder and editor-​at-​large of CityLab, and a senior editor at The Atlantic. His research focuses on the role of creativity, innovation, and talent in urban economic development and economic geography. His books include The Rise of the Creative Class, The Flight of the Creative Class, Who’s Your City?, The Great Reset, and The New Urban Crisis, and numerous journal articles on creativity, innovation, and urban and economic geography. Chris Forman  ([email protected]) is the Peter and Stephanie Nolan Professor of Applied Economics and Management at the Charles H. Dyson School of Applied Economics and Management at Cornell University. His research interests include the geography of IT use, electronic commerce, diffusion of IT innovations, and IT strategy. He has published widely on issues related to innovation in enterprise IT, including the business process innov­ ation that accompanies enterprise IT investment within firms, as well as the strategies of enterprise IT suppliers. He currently serves as a Department Editor at Management Science, and previously served as Senior Editor at Information Systems Research. Koen Frenken  ([email protected]) is a Full Professor of Innovation Studies at Copernicus Institute of Sustainable Development, Utrecht University. His research interests include sharing economy, innovation studies, economic geography, and the evolution of science and technology. His theoretical interests are evolutionary economics, complexity theory, and network science. He has published numerous articles in the Journal of Economic Geography and Regional Studies, and is an author of several books on innovation, complexity theory, and evolutionary economic geography. He is also an associate editor at Industrial and Corporate Change and a member of the editorial boards of Research Policy, Environmental Innovation and Societal Transitions, and Revue d’Economie Industrielle. Meric S. Gertler  ([email protected]) is President of the University of Toronto, Professor of Geography and Planning and the Goldring Chair in Canadian Studies. He was the founding co- director of the Program on Globalization and Regional Innovation Systems (PROGRIS) at the Munk School of Global Affairs. His research focuses on the geography of innovative activity, the economies of city regions, and economic restructuring in North America and Europe. He is the author, co- author, and co- editor of more than ninety scholarly articles and chapters, and nine books, including Manufacturing Culture: The Institutional Geography of Industrial Practice. He has served as an advisor to local, regional, and national governments in Canada, the USA, and Europe, and international agencies such as the Organisation for Economic Co-operation and Development. He is the founding associate editor of the Journal of Economic Geography.

xxxvi   List of Contributors Amy K. Glasmeier ([email protected]) is Professor of Economic Geography and Regional Planning and runs the laboratory on Regional Innovation and Spatial Analysis, the Department of Urban Studies and Planning, MIT. Her research focuses on the spatial interactions among economic actors and organizational structures in the provision of economic opportunity for communities and individuals. From 2012 to 2016 she was a faculty co-​investigator on the Post Traumatic Stress Innovations project, where she studied access to mental health care and other support programmes for members of the Marines and Navy and their families. She is the author of many articles and books on topics including regional and industrial development, technology and innovation, poverty and inequality, and global economic challenges. With Dr Michael Goodchild and Dr Glen McDonald, the National Academy of Sciences recently published her co-​authored report ‘Fostering Transformative Research in the Geographical Sciences’. She is a founding editor of the Cambridge Journal of Regions, Economy and Society. Johannes Glückler (glueckler@uni-​heidelberg.de) is Professor of Economic and Social Geography and Fellow of the Marsilius Center of Advanced Studies at Heidelberg University. His research and teaching focuses on economic geography, organization studies, network theory, and the service industries. He is co-​founder of the MSc programme ‘Governance of Risks and Resources’ at the Heidelberg Center for Latin America in Santiago de Chile, where he regularly teaches methods of social research. He has also been Visiting Professor at the University of Salamanca, Spain, since 2009. Apart from basic research and teaching, he has provided consulting services to federal ministries, chambers of commerce, industry associations, corporations, and civil society organizations on questions of regional development, innovation, and organizational networks. Avi Goldfarb  ([email protected]) is Ellison Professor of Marketing at Rotman School of Management, University of Toronto. He serves as Senior Editor at Marketing Science. His research focuses on understanding the opportunities and challenges of the digital economy. He has published a number of articles in areas of economics, marketing, statistics, computing, and law, and is co-​editor of two books, including Economic Analysis of the Digital Economy. Gernot Grabher  (gernot.grabher@hcu-​hamburg.de) is Professor of Urban and Regional Economics at the HafenCity University Hamburg. He is conducting research that explores how cities can learn from rare events; how new social practices in the sharing economy transform urban life; and how social network sites reshape socializing, creativity, and know­ ledge production. In his recent project, funded by the German Research Foundation, he examines how social network sites transform practices of interpersonal networking. He has published numerous articles in the leading academic journals and edited eight books, most recently the volume Self-​Induced Shocks: Mega-​Projects and Urban Development (2015). He is co-​editor of the Regions and Cities series of the Regional Studies Association, and was co-​ editor of Economic Geography. Currently, he serves on the editorial boards of Cambridge Journal of Regions, Economy and Society, Economic Geography, European Planning Studies, Regional Studies, Progress in Human Geography, and Environment and Planning A. Shane Greenstein  ([email protected]) is the MBA Class of 1957 Professor of Business Administration and co-​chair of the HBS Digital Initiative at Harvard Business School. He is also co-​director of the Program on the Economics of Digitization at the National Bureau

List of Contributors    xxxvii of Economic Research. His research spans issues of strategy, regulation, history, marketing, information systems, and organization design. He has published more than fifty journal articles on economics of enterprise IT, technological competition in computing, and the commercialization of the Internet infrastructure, and has written or edited nine books. His recent book, How the Internet Became Commercial: Innovation, Privatization, and the Birth of a New Network, traces the evolution of the Internet from government ownership to privatization to the commercial Internet, showing how interplay between government and private industry transformed the Internet. He has served as the President of the International Organization Society, and has been a member of the editorial board of Telecommunications Policy, Research Policy, and other prominent academic journals. Dieter Helm  ([email protected]) is Professor of Economic Policy at the University of Oxford, a Fellow of New College, and a Professorial Research Fellow of the Smith School of Enterprise and the Environment. His main research interests are utilities, infrastructure, regulation, and the environment, with a focus on the energy, water, communications, and transport sectors, primarily in the UK and Europe. His most recent books include The Carbon Crunch: Revised and Updated, Natural Capital: Valuing the Planet, and Burn Out: The Endgame for Fossil Fuels. He is an associate editor of the Oxford Review of Economic Policy. During 2011, Helm assisted the European Commission in preparing the Energy Roadmap 2050, serving both as a special advisor to the European Commissioner for Energy and as Chairman of the Ad Hoc Advisory Group on the Roadmap. He is Chair of the Natural Capital Committee. Cameron Hepburn  ([email protected]) is Professor of Environmental Economics at the University of Oxford, based at the Smith School of Enterprise and the Environment and the Institute for New Economic Thinking at the Oxford Martin School, and is also Professorial Research Fellow at the Grantham Research Institute at the London School of Economics and a Fellow of New College, Oxford. He is an economist with expertise in energy, resources, economic growth theory, behavioural economics, and environmental economics. His work has been referred to in publications such as The Economist and The Financial Times, and he has been interviewed on television and radio in various countries. He has provided advice on energy, environmental, and climate change policy to the UK Government DECC Secretary of State Economics Advisory Group. He is also managing editor for Oxford Review of Economic Policy. Alex Hughes ([email protected]) is a Professor of Economic Geography in the School of Geography, Politics and Sociology at Newcastle University. Since 2014 she has been the Research Director for the School. She was Chair of the Economic Geography Research Group of the Royal Geographical Society (2012–​15). Currently, she is a member of the Editorial Board of the RGS-​IBG Book Series published by Wiley-​Blackwell. She is also a member of the ESRC Peer Review College. Her wide-​ranging research interests in economic geography include cultural political economy; postcolonial economies; global value chains and production networks; knowledge and economy; rethinking economy; transnational retailers and corporate responsibility; retailer–​supplier relationships in the UK and USA; ethical trade and labour in Kenya, South Africa, and Pakistan; regulation and governance; corporate social responsibility and sustainability; the audit economy; learning networks; ethical public procurement; and ethical consumption in the Global South.

xxxviii   List of Contributors Simona Iammarino ([email protected]) is Professor of Economic Geography and Head (2014–​17) of the Department of Geography and Environment at the London School of Economics and Political Science. She is also an affiliate of the Spatial Economics Research Centre. Previous tenured positions include: Science and Technology Policy Research, University of Sussex; University of Rome, ‘La Sapienza’; and Italian National Institute of Statistics. Her main research interests lie in the following areas: multinational corporations; location and innovation strategies; local economic development; geography of innovation and technological change; regional systems of innovation; and regional and local economic development and policy. She has published extensively in top-​refereed journals in the field, and has extensive experience in both externally funded international research projects and consultancy for various government agencies and international organizations such as the European Commission, the Organisation for Economic Co-​ operation and Development, and the United Nations Economic Commission for Latin America and the Caribbean. Oliver Ibert ([email protected]) is Professor of Economic Geography at the Department of Geographical Sciences at the Freie Universität Berlin, and head of the research department ‘Dynamics of Economic Spaces’ at the Leibniz Institute for Research on Society and Space. His research interests include economic geography of knowledge practices, temporary organizations in economic settings and planning administrations, customer-​driven innovation processes, economic geography of virtual online communities, planning theory, and governance. He has published widely on these topics in journals such as Economic Geography, Journal of Economic Geography, Environment and Planning A, Research Policy, Geoforum or Regional Studies. Natasha Iskander  ([email protected]) is an Associate Professor of Public Policy at Wagner Graduate School of Public Service, New York University. She conducts research on labour migration and economic development, on labour mobilization and its relationship to workforce development, and on processes of institutional innovation and organizational learning. Her most recent book, Creative State: Forty Years of Migration and Development Policy in Morocco and Mexico (Cornell University Press, 2010), examines how the governments of Mexico and Morocco elaborated policies to build a link between labour emigration and local economic development. She is currently conducting research on processes of skill development among migrants in Qatar’s construction industry. She also worked as consultant on migration and skills transfer policy for the governments of Mexico and Morocco, and for international organizations such as the World Bank and the Inter-American Development Bank. Chacko G. Kannothra ([email protected]) is a doctoral candidate at the Organizations and Social Change Program at the College of Management, University of Massachusetts Boston. His research interests include social entrepreneurship and hybrid business models, global outsourcing, and inclusive employment and development. His dissertation focuses on entrepreneurial foundations, growth strategies, and managerial challenges of hybrid organizations in global outsourcing, so-​called impact-​sourcing service providers. Before entering academia, he worked with Accenture as a Business Analyst for their Global Compensation and Benefits Administration programme, as well as a Consultant for the Ministry of Science and Technology, Government of India.

List of Contributors    xxxix William R. Kerr  ([email protected]) is Professor of Entrepreneurial Management at Harvard Business School. His research focuses on how companies and economies explore new opportunities and generate growth. He has published articles, book chapters, and case studies on the subjects of entrepreneurship, innovation and growth, agglomeration forces and cluster structures, and global online labour markets. His recent project, Entrepreneurship Reading: Launching Global Ventures, offers insights into how young global ventures design their business models and some of the resource constraints and opportunity costs that they face. He is the editor of the Journal of Economic Geography and a member of the editorial board for the Journal of Urban Economics. Janelle Knox-​Hayes ([email protected]) is the Lister Brothers Associate Professor of Economic Geography and Planning in the Department of Urban Studies and Planning, MIT. Her research focuses on the ways in which social and environmental systems are governed under changing temporal and spatial scales as a consequence of globalization. Her latest project examines how social values shape sustainable development in the Arctic. She is an author, co-​author, and co-​editor of fifteen journal articles, and two books, including Saving for Retirement and The Culture of Markets: The Political Economy of Climate Governance. She is also book-​review editor of the Journal of Economic Geography, and editor for the Cambridge Journal of Regions, Economy and Society, as well as the Journal of Urban Planning. Karen P.Y. Lai  ([email protected]) is an Assistant Professor of Geography at the National University of Singapore. Her research interests include geographies of money and finance, markets, varieties of capitalism, service sectors, global city networks, and international financial centres. Her recent project examines everyday financialization through the know­ ledge networks of financial advisors and consumers. She is currently researching the global financial networks of investment banks in mergers and acquisitions, and initial public offerings. She has authored or co-​authored a book, several journal articles, and book chapters on financialization, knowledge networks, and economic development in East and South East Asia. She is also on the Standing Committee of the Global Production Networks Centre at the National University of Singapore, and editorial board member of Geography Compass. Marcus M. Larsen  ([email protected]) is Associate Professor at the Department of Strategic Management and Globalization, Copenhagen Business School, and Adjunct Associate Professor at BI Norwegian Business School. He received a PhD degree from Copenhagen Business School on the topic of Offshoring and Globalization. His dissertation won the Peter J. Buckley and Mark Casson Dissertation Award and the Barry M. Richman Best Dissertation Award in 2014. His research focuses on the organizational design of offshoring and outsourcing and emerging economy multinationals. He has published widely on hidden costs of offshoring, strategies, and innovation in the wind-​turbine industry. He is also a co-​editor of two books on strategic design and innovation in offshoring, and on strategies in emerging markets. Robin Leichenko  ([email protected]) is Professor and Department Chair at the Department of Geography at Rutgers University. She is also the co-​director of the Rutgers Climate Institute. Her research interests span the fields of economic geography and human dimensions of global environmental change. She studies how and why processes of global economic and environmental change differentially affect cities, regions, and sectors, and the implications of these processes for questions of vulnerability, equity, and sustainability.

xl   List of Contributors She has published more than forty-​five journal articles, and is the co-​author of two books, namely Environmental Change and Globalization: Double Exposures and Housing and Economic Development in Indian Country: Challenge and Opportunity. She is past chair of the Economy Geography Speciality Group at the Association of American Geographers. Between 2012 and 2013, she also served as a committee member at the Panel to Review the US National Climate Assessment at the National Academy of Sciences/​National Research Council. Mark Lorenzen ([email protected]) is Professor of Innovation, Entrepreneurship and Industrial Dynamics at the Department of Innovation and Organizational Economics at the Copenhagen Business School and director of the Danish Research Unit for Industrial Dynamics. His research is in the field of industrial dynamics, with a special focus on the relationships between innovation and the economic organization of the market in networks, projects, and clusters, currently within the creative industries. He has published in journals such as Journal of Economic Geography, Organization Studies, and Economic Geography, and he is editor-​in-​chief emeritus of Industry and Innovation, series editor of the Routledge Studies in Industrial Dynamics, and co-​editor of The Oxford Handbook of Creative Industries. Nichola Lowe  ([email protected]) is an Associate Professor at the Department of City and Regional Planning, University of North Carolina at Chapel Hill. Her research focuses on the institutional arrangements that lead to more inclusive forms of economic development and, specifically, the role that practitioners can play in aligning growth and equity goals. She is undertaking research on local economic development and skills transference among immigrant construction workers and job strategies for displaced workers and less educated individuals in the North American context. She has worked as a consultant for the International Labour Organization, Inter-​American Development Bank, Bank of the Northeast Brazil, and Ontario Ministry of Economic Development and Trade. Stephan Manning  ([email protected]) is Associate Professor of Management and co-​founder of the Organizations and Social Change Research Group at the College of Management, University of Massachusetts Boston. His research mainly covers three areas: sustainability standards, global services sourcing, and project-​based organizing. He has done field research in various countries, including China, Germany, Guatemala, Kenya, Romania, South Africa, and the USA. His research has been published in numerous academic journals, such as Strategic Management Journal, Journal of International Business Studies, Journal of Management Studies, Organization Studies, and Research Policy. He serves as Senior Editor of Management and Organization Review. He is also founding co-​editor and author of the Organizations and Social Change Blog, and has written for The Conversation, The Broker, and other blog platforms. Ron Martin ([email protected]) is Professor of Economic Geography, University of Cambridge, UK, and holds a Professorial Fellowship at St Catharine’s College. He is also a Research Associate of the Centre for Business Research, Judge Business School, and leads the Centre for Geographical Economic Research, at Cambridge. He has been awarded the British Academy’s ‘Thank-​Offering to Britain’ Senior Research Fellowship, 1997–​8; a Leverhulme Major Research Fellowship, 2007–​10; and a Leverhulme Emeritus Fellowship for 2015–​17. He is an Academician of the Academy of Social Sciences (2001), a Fellow of the British Academy (2005), and President of the Regional Studies Association. In 2016 he was

List of Contributors    xli awarded the Royal Geographical Society’s Victoria Medal for outstanding contributions to the study of regional economic development. His research interests span: work, financial systems, regional economic development, evolutionary economic geography, and public policy. He currently leads a major Economic and Social Research Council project on city economic evolutions. Philip McCann  ([email protected]) holds the Endowed Chair in Economic Geography at the University of Groningen in the Netherlands. He studied at and gained his PhD (1993) from the University of Cambridge; prior to moving to Groningen he worked at the University of Pennsylvania in the USA (1993–​5), the University of Reading (UK) (1995–​2005), and the University of Waikato in New Zealand (2005–​9). He has also been a Visiting Professor in the USA, Japan, Thailand, Spain, and Italy. Professor McCann’s research covers a wide range of economic geography topics and his research has received awards in various countries. He has also acted as an advisor and consultant to the European Commission, the Organisation for Economic Co-​operation and Development, the European Investment Bank, and government departments and research institutions in several different countries. Caitlin A. McElroy  ([email protected]) is a Departmental Research Lecturer at the School of Geography and the Environment and the Smith School of Enterprise and the Environment, University of Oxford. The focus of her research has revolved around the institutional dynamics of private-​sector investment in development. This has included researching the interactions between the mining industry, environment, and development. This has led to a range of projects and partnerships that address critical issues, such as improving the sustainability and development of resource-​driven economies and the creation of tools to assist corporations in the management of their environmental and social risks and opportunities. Further, it has led to her current work to build a research programme titled ‘Sharing Resource Prosperity’, which continues to research environmental and social sustainability related to the extractive sector. Also, she is involved in researching environmental impact investing in the USA and UK. She has published several papers on the governance of extractive industries’ investments for development, and is co-​editing a book titled Food, Energy, and Water Sustainability: Emergent Governance Strategies. Anita M. McGahan ([email protected]) is a Professor of Strategic Management at the Rotman School of Management at the University of Toronto. She has also served as the Associate Dean of Research at the Rotman School. Her research is focused on industry change, sustainable competitive advantage, and the establishment of new fields. An area of particular interest to her is global health and the diffusion of knowledge across international boundaries. She has authored and co-​authored more than fifty-​five articles in refereed journals and three books, including How Industries Evolve: Principles for Achieving and Sustaining Superior Performance. She is a member of the editorial boards of several journals, such as Academy of Management Discoveries, European Management Review, and Strategic Organization. Sarah McGill  ([email protected]) is a Research Associate at the Smith School of Enterprise and the Environment, School of Geography and the Environment, University of Oxford. Her current research uses behavioural and institutional approaches to examine cross-​country differences in social security provision and the

xlii   List of Contributors barriers that prevent households from adequate long-​term financial planning. She obtained her DPhil at the University of Oxford in 2015. Her previous research focused on commodity markets, mineral resource governance, and the relationship between the financial sector and the ‘real economy’; she has also published several papers and reports on sustainable investing. She is a member of the editorial board of the Journal of Sustainable Finance and Investment. Charlotta Mellander ([email protected]) is Professor of Economics at the Jönköping International Business School. She is also affiliated with the Martin Prosperity Institute at the Rotman School of Management, University of Toronto. She studies location patterns of creative individuals and firms to determine how they shape regional development and has published numerous journal articles on this topic. She has delivered more than 200 invited, external speeches, both nationally and internationally, including the European Union and the United Nations, and companies like IBM. Ashby Monk  ([email protected]) is the Executive and Research Director of the Stanford Global Projects Center. He is also a Senior Research Associate at the University of Oxford and a Senior Advisor to the Chief Investment Officer of the University of California. His current research focus is on the design and governance of institutional investors, with particular specialization on pension and sovereign wealth funds. He has published several articles and op-​eds on institutional co-​investing, pension funds, and ethical investment. His research and writing has been featured in The Economist, The New York Times, The Wall Street Journal, The Financial Times, Institutional Investor, Reuters, Forbes, and on National Public Radio, among a variety of other media. Phillip O’Neill  ([email protected]) is Professor of Economic Geography and Director, Centre for Western Sydney, at Western Sydney University. Before joining the University of Western Sydney as Foundation Director of the Urban Research Centre in 2006, he was Director of the Centre for Urban and Regional Studies at the University of Newcastle, NSW. He was recognized as a Fellow of the Institute of Australian Geographers in 2008. From 2008 to 2013 Phillip was editor-​in-chief of Geographical Research and he currently sits on the editorial boards of Environment and Planning A and the Journal of Economic Geography. His research interests include urban infrastructure, suburban labour markets, and state theory and practice. Phillip is widely published in economic geography, as well as in the mainstream media where he is a prominent columnist and commentator. Jamie Peck ([email protected]) is Canada Research Chair in Urban and Regional Political Economy, Distinguished University Scholar, and Professor of Geography at the University of British Columbia. With long-term research interests in urban restructuring, geographical political economy, labour studies, the politics of policy formation and mobility, and economic geography, his recent books include Offshore: Exploring the Worlds of Global Outsourcing, Fast Policy: Experimental Statecraft at the Thresholds of Neoliberalism (with Nik Theodore), Constructions of Neoliberal Reason, and The WileyBlackwell Companion to Economic Geography (with Trevor Barnes and Eric Sheppard). A Fellow of the Royal Society of Canada, and previously the holder of Guggenheim and Harkness fellowships, Peck is the Editor-in-Chief of the Environment and Planning series of journals.

List of Contributors    xliii Alexander Pfeiffer  ([email protected]) is studying for a DPhil in Geography and the Environment at the University of Oxford with a focus on stranded carbon assets and their effects on financial markets. He is also a research assistant for the Institute for New Economic Thinking (INET) at the Oxford Martin School. His main research interests are stranded carbon assets and the carbon bubble, energy policy, effects of climate policies on financial markets, carbon taxes, and committed emissions. Andrés Rodríguez-​Pose (a.rodriguez-​ [email protected]) is Professor of Economic Geography at the London School of Economics, where he was previously Head of the Department of Geography and Environment. His research interests include regional growth and inequality, fiscal and political decentralization, regional innovation, and development policies and strategies. In his areas of expertise, he has regularly acted as a consultant to several Directorates of the European Commission, the European Investment Bank, the World Bank, the Cities Alliance, the Organisation for Economic Co-​operation and Development, the International Labour Organization, and the Food and Agriculture Organization. He has also published more than 150 papers in peer-​reviewed journals, is an editor of Economic Geography, and sits on the editorial board of twenty-​nine other scholarly journals. He is also the author of several books such as Local and Regional Development (with Andy Pike and John Tomaney), Innovation and Regional Growth in the European Union (with Riccardo Crescenzi), and Technology and Industrial Parks in Emerging Countries (with Daniel Hardy). Rajiv Sharma  ([email protected]) is a Research Director at Stanford University’s Global Projects Center and an honorary research associate at the Oxford University Smith School of Enterprise and the Environment. He received his Doctorate from Oxford University in the field of Pensions, Sovereign Wealth Funds and Infrastructure Investment. Prior to his position at Stanford, Rajiv worked as an economist for the Organisation for Economic Cooperation and Development (OECD) in Paris. Promoting long-term investment by institutional investors is a focus of his current work at Stanford and through this, he has worked with a number of global institutional investors and governments. His latest book is entitled Reframing Finance, which looks at how more institutional investor capital can be channelled into long-term projects like infrastructure, real estate, private equity and agriculture through collaboration and co-investment. He has also worked for venture capital private equity firm Oxford Capital Partners and London-based Infrastructure/Private Equity Advisory firm, Campbell Lutyens. Eric Sheppard ([email protected]) is Alexander von Humboldt Chair and Professor of Geography at the University of California, Los Angeles. His research interests lie in geographical political economy, uneven geographies of globalization, neo-​liberalism, urbanization in the Global South, urban sustainability and environmental justice, and critical geographical information systems. He has published more than 100 journal articles and twelve books. He serves on the editorial boards of Economic Geography, The Journal of Economic Geography, and Human Geography. Between 2013 and 2014 he was President of the Association for American Geographers. Janice Gross Stein  ([email protected]) is the Belzberg Professor of Conflict Management in the Department of Political Science and the founding Director of the Munk School of Global Affairs at the University of Toronto. She is a Fellow of the Royal Society of Canada

xliv   List of Contributors and a member of the Order of Canada and the Order of Ontario. Her main research interests include intelligence, international security, negotiation processes, and behavioural explanations of decision-​making. She has published more than 100 articles and several books, including Digital Diplomacy and a forthcoming article on loss aversion and the end stage of negotiation. Michael Storper ([email protected]) is Professor of Economic Geography at the Department of Geography and the Environment, London School of Economics. He is also affiliated with the Centre de Sociologie des Organisations at Sciences-​Po in Paris, and the Department of Urban Planning in the School of Public Affairs at the University of California Los Angeles. His research interests include globalization and regional economic development processes, and the effects of new communication technologies on face-​to-​face contact and delocalization. His latest books are The Rise and Fall of Urban Economies (2015) and Keys to the City: How Economics, Institutions, Social Interaction, and Politics Shape Development (2013). He is a Corresponding Fellow of the British Academy, and has received the Regional Studies Association’s Sir Peter Hall Prize for overall contribution to the field. Kendra Strauss ([email protected]) is an Associate Professor of Labour Studies in the Department of Sociology and Anthropology, Simon Fraser University. A  geographer by training, her work focuses on occupational pensions; precarious work, migration and unfree labour; and on theorizing the relationships between production and social reproduction in contemporary capitalist economies. She is also interested in how categories of social difference like gender, race, and class shape the wage relation and unpaid work. Her recent research has looked at non-​standard work in the UK, employment agencies, and unfree labour, and the evolution of legal approaches to forced labour and trafficking. Her latest book, co-​edited with Dr Katie Meehan, is entitled Precarious Worlds: Contested Geographies of Social Reproduction (UGA Press, 2015). Alexander Teytelboym  ([email protected]) is the Otto Poon Research Fellow at the Institute for New Economic Thinking at the University of Oxford. Following completion of his DPhil in Economics from the University of Oxford, he worked for one year as a post-​doc at the Lab for Information and Decisions Systems at MIT. His research interests focus on networks, matching markets, market design, and the environmental economics. He has published several policy papers on climate change, transport, refugee resettlement, and transition economies. Maria Tsampra  ([email protected]) is an Assistant Professor at the School of Business Administration, University of Patras, Greece. Her expertise broadly lies in economic geography, regional, local, and urban development, and, more specifically, in issues of regional diversity and spatial inequalities, regional entrepreneurship and competitiveness, local labour markets, economic restructuring, and integration in south-​ eastern European countries. She has authored and co-​authored several book chapters and journal articles on regional employment and production patterns, spatial division of labour, small-​and medium-​sized enterprise innovativeness, and so on. Callum Wilkie ([email protected]) is a researcher at the Local Economic Development Programme at London School of Economics. His areas of research and expertise include regional growth and development; equitable and inclusive economic growth; regional innovation and innovation policy; and territorial development policies and strategies.

List of Contributors    xlv Dariusz Wójcik ([email protected]) is a Professor of Economic Geography in the School of Geography and the Environment at Oxford University, a Fellow of St Peter’s College, and a Visiting Professor at Beijing Normal University. His research focuses on finance, corporate governance, and economic globalization. His current project, funded by the European Research Council, investigates how financial and business services have been affected by the global financial crisis, and how they change in response to new financial regulation, the rise of the Global South, and the digital revolution. The project also focuses on the impacts of finance on regional development. Dariusz is a member of the editorial board of Economic Geography, the Journal of Economic Geography, Environment and Planning A: Economy and Space and GeoJournal and leads the Global Network on Financial Geography (FinGeo). Neil Wrigley  ([email protected]) is Professor of Human Geography, University of Southampton, and was previously Head, School of City and Regional Planning, University of Wales, Cardiff. Founding Editor of the Journal of Economic Geography (2001–​14), he has held ESRC, Leverhulme, and Erskine Research Fellowships, and was Senior Research Fellow, St Peter’s College, Oxford. His contributions to economic geography research have a distinctive focus on retail and consumption and are reflected in his 160 journal papers and agenda-​ setting books. They have provided pioneering interpretations of the role retailers play as key organisers of the global economy, of the regulatory challenges posed by emerging competitive landscapes of retail power, and insight into the disruptive consequences of online retail and the digital economy. Recognition of his work includes the Royal Geographical Society’s Murchison Award (2008), ESRC’s Outstanding Impact in Business Prize (2014), election as Academician of the Academy of Social Sciences (2003), and the distinguished Fellowship of the British Academy (2012). Henry Wai-​chung Yeung  ([email protected]) is Professor of Economic Geography at the National University of Singapore. Previously, he has held visiting professor positions at the University of Sydney, University of Manchester, University of Hong Kong, and University of Washington at Seattle. His research interests broadly cover theories and the geography of transnational corporations, global production networks and global value chains, East Asian firms, and developmental states in the global economy. Author of five monographs, his latest books are Strategic Coupling: East Asian Industrial Transformation in the New Global Economy (Cornell Studies in Political Economy Series, Cornell University Press, Ithaca, May 2016) and Global Production Networks: Theorizing Economic Development in an Interconnected World (with Neil Coe, Oxford University Press, April 2015). He has also published over 90 journal papers and 45 book chapters, and edited and co-​edited 7 books on Asian business, globalization, and emerging economies. Since 2001, he has served as an editor at Environment and Planning A, Economic Geography, and Review of International Political Economy (2004–​13). Matthew Zook  ([email protected]) is Professor of Economic Geography and the director of The DOLLY Project (Data on Local Life and You) at the University of Kentucky. His research interests are at the intersection of economic, digital, and urban geographies. In recent years, he has studied the micro-​geographies of high-​frequency trading, the production of geosocial media, and how code, space, and place interact to create the augmented realities of daily-​lived geographies. He is also interested in the multiple ways in which flows of

xlvi   List of Contributors material goods in the global economy are shaped by immaterial flows of information, and the ways in which the intertwining of material and virtual flows is providing news ways to study economic geography. He has written dozens of journal articles and book chapters on digital geographies, economic geography, and globalization. He is a member of the editorial board at Big Data & Society: Critical Interdisciplinary Inquiries and the AAG’s new journal, GeoHumanities.

Introdu c t i on Economic Geography in the Twenty-​f irst Century Gordon L. Clark, Maryann P. Feldman, Meric S. Gertler, and Dariusz Wójcik The Remit of Economic Geography Geography can be defined as the why and so what of where. In economic geography the where, why, and so what questions are focused on understanding economy. The latter needs to be defined broadly as the totality of processes through which individuals, households, and societies make a living and sustain themselves. As such, it involves the processes of production, consumption, distribution, and exchange, but not only those in the formal economy. Informal economy, both ‘grey’, such as household production, and ‘black’, such as illegal drug trade, are as much of interest to economic geographers as constituents of national income (Gibson-​Graham, 2006). The value of unpaid childcare alone, one of the informal economy’s many components, was estimated at £343 billion in the UK in 2010, an equivalent of 23 per cent of gross domestic product (GDP), in which it is not included. The broad definition of economy in economic geography also recognizes that economy is embedded in society and nature, relying on social and natural resources and processes. As such, human–​environment relations are central to economic geography, just as they are to geography in general. Where is about the spatiality of economic processes and involves the key concepts of location, place, and territory (Coe et al., 2013). Location can be understood simply as the geographical coordinates of a person or object. Places are parts of space with a specific meaning. They are socially constructed and can refer to body parts, buildings, streets, workplaces, shops, parks, neighbourhoods, cities, and an infinite number of other phenomena, which can be defined differently by different people. Territory is also a part of space, but this time defined by a specific system of power that exercises (even if only partial) control over that part of space. This can be a nation state, with control exercised by national government, but it can also be a subnational state, province, or city under the control of a regional or municipal government. Thus, as we move from location through place to territory, we shift focus from physical, through social, to political space. Locations, places, and territories are crucial to understanding economy. If you are born in the USA, you can expect to earn 100 times more and live thirty years longer than if you are born in Zambia. A Bolivian man with nine years of

2   Clark et al. schooling earns, on average, three times less than his counterpart in the USA. The best predictor of income is not what and whom you know but where you work (World Bank, 2009). What helps economic geographers relate locations, places, and territories to each other are the concepts of distance, proximity, diversity, and scale. Distance is defined in terms of physical space, as physical distance between locations. Proximity is a much broader concept, helping us relate people, objects, places, and territories in social and political space, in addition to physical space. As such, we can distinguish many types of proximity, including social, organizational, institutional, and cognitive (Boschma, 2005). For example, people may be distant physically, but very close in terms of their language and culture, organizations they belong to, rules they follow, or ways in which they think, and vice versa, very close physi­ cally but far apart with regard to other types of proximity. Both distance and proximity are ultimately about understanding the difference between people and objects in their locations, places, and territories. This emphasis speaks to the preoccupation of economic geography with diversity of economic life, and caution regarding the search for universal laws that shape this life. Equally important to economic geography is commitment to a multiscalar inquiry, starting with the human body and an individual as an economic agent, through the scale of workplace and home, local (e.g. neighbourhood), urban, regional, national, to the supranational and global scale (Castree et al., 2004). Why is about explaining the spatiality of economic processes and the diversity of economic life they create. Here, geographers study a mesmerizing variety of forces ranging from economic universals like demand and supply, through behaviour of economic actors and organizations, formal and informal institutions setting the rules of economic activities, to ethnicity, sex, and other cultural factors. The centre stage is no longer occupied by a utility maximizer—​homo economicus—​but rather a satisficer, searching for satisfactory rather than optimal solutions, whose decisions are affected by her or his cognition on one side and environment on the other (Simon, 1956). What is more, in their decisions people are subject to all kinds of systematic anomalies and biases, including myopia and loss aversion, and are affected by the way problems facing them are framed (Kahneman, 2011). Complacency is mixed with overreaction to environmental stimuli, and herd behaviour with individualism in complex ways, making the economic landscape even more difficult to explain and predict. Economic geography considers economic landscape as heterogenous, the product of history and processes that can be distinctive to a place or territory. This does not mean that economies are incomparable across space. It does, however, mean that they cannot be reduced to a set of universal variables like GDP growth rate or Gini coefficient. This also highlights the need for using a diverse set of quantitative and qualitative research methods, as well as knowledge gained through experience, rather than applying a distant, and purportedly objective, view on economic life. In the spirit of steering clear of determinism, some economic geographers talk about factors facilitating rather than determinants of spatial economic structures (Massey, 1995). The so what question concerns the consequences of the spatial variation and organization of economic processes for economy, society, and nature at large. One way to think about these consequences is to focus on four key dimensions of economic development: growth, equity, stability, and sustainability. With its focus on uneven development across space, economic geography has contributed significantly to debates on growth and equity. Of particular relevance here is the geographers’ interest in the processes of differentiation, whereby difference, including inequality, is produced by economic processes that sustain long-​term

Introduction   3 spatial variation (Clark et al., 2000). This contribution was crucial in counteracting hyper-​ globalist views, prominent in the 1990s, emphasizing homogenizing forces of globalization, envisaging a global society, and predicting the end of geography in economy, politics, and culture. While the battle fought by economic geography against a simplistic view of globalization continues, as we argue in the following section, the twenty-​first century poses new challenges to the discipline, particularly in the areas of equity, stability, and sustainability. To be sure, beyond informing scholars inside and outside of geography, the crucial aspect of the so what question involves the impact of economic geography on public debate and policy. Even at the very basic level, the quintessentially geographical curiosity about where a product or a meal comes from can contribute to a better informed, more reflexive, and responsible consumption. In this sense, economic geography is arguably more than a discipline. It is a perspective, a mindset that can help responsible behaviour and citizenship. The where, why, and so what questions have permeated the history of economic geogra­ phy, although with changing emphasis. The roots of economic geography can be found in commercial geography of the nineteenth century, which was preoccupied with the origin of different products and documenting trade patterns around the world. Environmental geography in the first three decades of the twentieth century shifted focus from where to why, explaining economic, as well as social, patterns across space with features of natu­ ral environment, predominantly climate, often in an excessively deterministic manner. Dissatisfied with environmental determinism, for the next quarter century or so, starting from the late 1920s, in a wave of research referred to as aerial differentiation, economic geography returned to focus on the question where, emphasizing the mutual interdependence of place and economy, but without privileging the natural environment as a causal factor. In the 1950s, the discipline evolved again to the question why, employing quantitative locational analysis, allied with economic and regional science, modelling the location of economic activity as a function of demand, supply, and distance as key factors (Berry et al., 1987). The last three decades of the twentieth century have witnessed further developments in economic geography that modified the way economic geographers ask the questions where and why. One was the rise of behavioural approaches in the 1970s, investigating locational decisions of economic actors who satisfice and cope with uncertainty. A more radical turn was a political economy approach, from the 1970s shifting emphasis from location to territory, and from ‘purely’ economic to social and political processes, in which economies are embedded. Finally, a cultural economy approach, flourishing since the 1990s, has complemented the political economy approach by focusing on place and cultural embeddedness of economic life and lived economic experiences. All waves and approaches in economic geography have engaged with the so what question, but in different ways and with different emphasis. In commercial geography it was about predicting the patterns of commerce and identifying commercial opportunities. Environmental geography ventured into predicting the capacity of different environments to carry human activity, including a famous forecast made in 1940 by Griffith Taylor, when the population of Australia was seven million, that the natural environment would limit its population to a range of 20–​30 million (exactly where it is now) (Taylor, 1940). Locational analysis contributed to corporate decision-​making, regional and urban economic planning, and forecasting economic growth at the subnational level. Political and cultural economy approaches have

4   Clark et al. brought new perspectives to the table, including big normative questions about inequality, the future of capitalism, and its alternatives. A shift in emphasis in terms of questions has been accompanied with movement on the continuum between nomothetic (theory-​driven) and idiographic (descriptive) approaches. Commercial geography and the study of aerial differentiation were on the idiographic end of the scale, with locational analysis marking the height of the nomothetic tradition. Social and natural sciences influencing economic geography have been changing as well. While environmental determinism pursued grounding in natural sciences, locational analysis did so in economics, political economy approaches in heterodox political economy, while cultural economy approaches find inspiration primarily in social theory and economic sociology. Ultimately, all of the waves and shifts in the history of economic geography contribute to the palimpsest of the discipline, in which the new does not necessarily supersede and replace the old, but adds a new layer and angle to the collective intellectual project. They reflect the dynamism of an open and interdisciplinary field, well suited to explaining the trends and addressing the challenges facing the world in the twenty-​first century. It is these trends and challenges, and how economic geography has responded to them, to which we now turn.

The World Economy Since 2000 More than fifteen years have passed since the publication of the first Oxford Handbook of Economic Geography. This is a long time for an academic discipline with not much over 100 years of history, particularly during a period of momentous change and challenge in the world economy. We have seen an unprecedented growth of many economies, particularly in Asia. The share of India in global nominal GDP increased from 1 per cent in 2000 to 3 per cent in 2015; China’s from 3 per cent to 15 per cent. The term emerging markets has been planted in the popular imagination, as have more recent terms like BRICs (Brazil, Russia, India, and China), and the Global South (Africa, Central and Latin America, and Asia), all recognizing the contribution of hitherto underdeveloped countries to the world economy. A related phenomenon is the growing significance of south-​to-​south economic relations, including Chinese trade and investment in Africa and Latin America. Combined with the effects of the recent financial crises on the US and European economies, these trends mark the emergence of new geo-​economics, which collides with old US-​dominated geopolitics, with the Global South severely under-​represented in global economic governance structures such as the World Bank and the International Monetary Fund (IMF). There are potential signs of hope, as in December 2015 the US Congress finally approved the reform of the IMF, which now offers more voting rights to Brazil, Russia, India, and China (IMF, 2016). To be sure, growth in the Global South has been very uneven, both between countries, excluding large parts of Africa in particular, and within countries. It has been accompanied by a huge wave of urbanization, with the percentage of world population living in cities rising from 48 per cent in 2000 to 55 per cent in 2015. This represents 1.1 billion more urban dwellers, with 90 per cent of this growth in emerging and developing economies of the Global South. Megacities have grown particularly fast. At the end of 2015 there were twenty-​ nine cities with populations over ten million, up from seventeen in 2000 (United Nations,

Introduction   5 2016). The Pearl River Delta which hosts two of them, Shenzhen and Guangzhou, in addition to the eight-​million-​strong Hong Kong, was developing into the world’s largest megacity region (overtaking Tokyo), with a high-​speed railway to link Guangzhou with Hong Kong, a 40-​km-​long bridge to link Hong Kong with Macau and Zhuhai, and a subway linking Guangzhou with Shenzhen (The Economist, 2016). Fast growth in many emerging and developing countries has left many behind, making societies more unequal, with the biggest contrasts following the urban–​rural divide. While Shanghai may enjoy the average GDP per capita level (in purchasing power parity terms) comparable with Israel, in Gansu province it is comparable with Angola. Inequalities of income and wealth have also continued to increase in advanced economies. This ascent, as documented by Thomas Piketty (2013), is now over forty years old. In the 2010s it has made many economies (with the USA and UK in the lead) as unequal as they were in the 1930s. What is worse, only high growth of income combined with much higher redistribution of income and wealth can reverse this trend in a peaceful manner, and neither looks likely as of 2016. While driven by a complex combination of globalization, technology, and ideology, growing inequality undeniably erodes cohesion and health of societies and presents a fundamental challenge to sustainability. Technology, including the rise of the digital economy, has contributed to growth, urbanization, and inequality. It has facilitated the development of long chains and complex networks of investment, production, and trade, which have enveloped emerging and developing economies. At the same time, through economies of scale and agglomeration, it has privileged cities, particularly large ones, as nodes in these networks, which concentrate innovation, financial, and coordinating activities. To add to spatial inequalities, the gap has grown between those who have the skills to keep up with new technology, and those who do not. Educational systems fight an uphill struggle to address this growing skills gap, as the latter is inextricably linked with growing income and wealth inequality. In 1998–​ 99 when the first Oxford Handbook of Economic Geography was edited, Dropbox and Skype did not exist, editors would more often use a university library than a web browser, and probably few geographers dreamt of big data. Since then the amount of information stored digitally around the world has increased thousands of times. Having reshaped media and services (the world of information bytes), digital technologies, armed with open-​source design and three-​dimensional printing, are now affecting manufacturing (the world of atoms). For some, we are on the brink of a technological transformation, whereby production is brought much closer, also in terms of physical distance, to consumption, thus reducing the need for very long production chains (Anderson, 2012). Others, however, are sceptical, arguing that technology has ‘run out of steam’, and the new technologies are unable to generate sustained economic growth at high rates experienced since the 1950s (Gordon, 2016). Instead, an annual growth rate of 1 per cent or lower may be a new normal. Crucially, we have not managed to lessen the pressure of the world economy on the environment. Global carbon dioxide emissions have risen by nearly 40 per cent since 2000, with China replacing the USA as the largest polluter (PBL Netherlands Environmental Assessment Agency, 2015). There is headway, however, in both technology and policy. The capacity of solar and wind power has multiplied. Hybrid and electric cars are increasingly common on the streets, not only in advanced economies. The profile of the Intergovernmental Panel on Climate Change has grown with every new assessment report,

6   Clark et al. boosted by the Nobel Prize awarded to the organization in 2007. The Stern Review, published in 2006, highlighted the devastating effects of climate change on the world economy, calling for major investment of 1 to 2 per cent of global GDP per year to address the worst effects (Stern, 2007). The Paris agreement of 2015 within the framework of the United Nations Framework Convention on Climate Change also marked major progress in this area, with commitment to ambitious goals in terms of greenhouse gas emissions, climate change adaption efforts, and financial contributions to achieve both aims. In addition to these shifts, the world economy has experienced a major shock. This began with the subprime crisis in the US in 2007–​08 and spilled over to Europe in 2009, exposing the structural weaknesses of the Eurozone. As the most severe and far-​reaching economic crisis since the 1930s, it has been referred to as the global financial crisis or the Great Recession. While the financial sector in the most affected countries has been saved from the consequences of its own reckless behaviour, and a full-​blown depression has been prevented, the resulting explosion of public debt is unprecedented, putting a huge burden on the public sector and threatening long-​term growth prospects. To add irony to the situation, wealthy individuals of advanced economies own an absolute majority of public debt of their countries. Europe as a whole has not been in so much debt, in relation to the size of its economy, since the aftermath of World War II. In terms of the wealth accumulated (extremely unevenly so) by its citizens, however, it is richer than ever (Piketty, 2013). In sum, we live not only in very interesting, but also in highly uncertain times. Globalization and integration, arguably the liveliest topics in economic geography in the 1990s, which started with the collapse of communism and ended with the introduction of the Euro, have continued hand in hand with financialization until 2007, but have slowed down since, and can no longer be taken for granted. New financial regulations may be beneficial in curtailing the excesses of international and national financial markets, but the implicit assumption that the purpose of a financial system is to serve the needs of the national ‘real’ economy, rather than a global ‘real’ economy, as made by many proponents of new regulation, might be problematic. Elsewhere, in the ‘real’ economy, economic nationalism and protectionism are also in ascendency, including the growing popularity of national champion policies. Of particular importance to the future of globalization are obviously the relationships between the USA and China. Whether a balance of power conducive to international economic integration can be established remains one of the big question marks. A pessimist may interpret these challenges and uncertainties by turning to Karl Polanyi’s idea of double movement (Polanyi, 1957). As the excesses of marketization, including the forces of neoliberalization, financialization, and globalization, have led to a crisis of capitalism, the fundamental problems they have created, with inequality in the lead, have not been resolved. In fact, inequality has, in many advanced economies, been aggravated by austerity policies. However, a countermovement is mounting. Put very simply, people in mature economies appear tired of elites hoarding the benefits of globalization rather than sharing them more evenly. While the Great Recession has discredited the old economic model, the economic and political establishment has done little to change it and revert the trend of rising inequality. As a result of rising inequality many people turn to populist politicians proposing to erect walls, literal and not, in a promise to protect their countries from foreign influence, and fuelling xenophobia. The slogan ‘take back control’ used successfully by the Leave campaign in

Introduction   7 the UK’s European Union referendum (2016) is a reflection of this trend, as is the election of Donald Trump as the US President. Analogies with the early twentieth century are ominous, when a wave of marketization of the late nineteenth and early twentieth century was followed by international conflict and economic crisis. Polanyi reminds us that tensions between marketization and the countermovement do not need to resolve themselves peacefully. For that reason one could also argue that stemming the tide of rising inequality is a precondition for addressing the collective problem of climate change. While we may disagree with a Polanyian warning for our times, there is no doubt economic geography remains crucial, maybe more important than ever, to understanding the global economic change. Urbanization led by megacities reminds us about the significance of location, place, and proximity in today’s economy. Digital and other technologies may overcome the friction of distance for some economic relations and activities, such as communication, while at the same time enhancing the role of proximity for others, such as corporate control. Despite globalization and integration, territoriality remains key to understanding the geoeconomics and geopolitics of the twenty-​first century. The spectacular growth of economies ranging from Singapore to China, India, and Angola testify to the principles of diversity and heterogeneity, where there is not one economic model to follow. Growing inequality speaks to the underlying processes of differentiation. Scale plays a fundamental role, too. We cannot interpret the secessionist movement in Catalonia without considering Catalonia as a city region that does not only see itself as different from the rest of Spain, but also seeks independence to function more flexibly in a world where growth is to a large and increasing degree driven by successful city regions. The where, why, and so what questions of economic geography are relevant to understanding economy wherever you are. The last editorial meeting for this handbook took place in San Francisco in March 2016, where these questions resonated particularly loudly. In 2000, when the first handbook went to print, the dot.com boom, of which the San Francisco Bay Area was the centre, came to an end. The Nasdaq Index reached its peak of 5048 in March only to lose half of it by the end of the year, wiping out trillions of US dollars in market capitalization. The area was a symbol of irrational exuberance, congestion of economic activity, and unaffordability, with an average rent exceeding $3000, and thousands of homeless people. Geography students around the world were writing essays about San Francisco as a prime example of diseconomies of agglomeration. Surely, many in 2000 must have thought that this agglomeration cannot increase any more, and would not have banked on the future of this city region. Fast forward sixteen years. The San Francisco Bay Area economy is larger and more dominant in the world of innovation and technology than ever before. It hosts the command centres of the two largest corporations in the world by market value (Apple and Google), and two more in the global top ten: Facebook and Wells Fargo—​the world’s largest bank by market capitalization. In 2000, it had two companies in the global top ten: Cisco and Intel. It is by far the leading global centre of financial technology (fintech), which reinforces its position as a financial centre. An average rent is now over $4000—​possibly the highest of all cities in the world. The issue of homelessness has been aggravated by the subprime mortgage crisis and the following wave of repossessions (Schafran, 2013). At the same time, it is a place of mesmerizing diversity and creativity. Its corporations are at the centre of debates on global cyber security as well as clean technology innovation. As such, the San Francisco Bay Area and its history is a monument to the extremely uneven and spiky economic landscape—​a powerful

8   Clark et al. demonstration of how the issues of growth, inequality, stability, and sustainability all meet in one place with implications for the whole interdependent world.

Economic Geography Since 2000 San Francisco is also a good place for reflection on how economic geography as a discipline has responded to the changes and challenges of the twenty-​first century. As the cradle of the Flower Power movement against the Vietnam War, and the sexual revolution of the 1960s and the 1970s, the city provided considerable inspiration for debates on gender, capitalism, and the rise of political and cultural economy approaches. It was also here, in Berkeley, where Allan Pred led the behavioural turn. To start with, it is worth highlighting that all schools of economic geography that have developed since the nineteenth century continue to be relevant. Documenting and predicting patterns of production and trade is as important to businesses and governments today as it was in the heyday of commercial geography. It may even be more relevant given the intensity of cross-​border investment flows spanning the globe. One of the recent developments in consulting is referred to as financial cartography, mapping financial flows and networks to analyse investment opportunities and risks. As John Authers wrote in the Financial Times, ‘Investors have much to learn from academics—​and certainly not just from economists—​ (…) now it is the turn of geographers’ (2015, p. 28). Many would consign environmental determinism to history, but considering that our civilization has in some respects already breached environmental limits to sustainable development, more engagement of economic geography with environmental issues, including resource management and climate change, is warranted. Regional economic geography, a la mode of aerial differentiation research, also has a place in the world where many hitherto understudied economies become the stage of key economic processes. Take Myanmar, for example, where a large population and resources have attracted a wave of Chinese investment, shaping Myanmar’s geoeconomic and geopolitical position. Locational analysis continues to offer key analytical tools for studying economic development at a regional and urban level, growing only more important the further we move away from the world economy organized by industrial nation states to a complex international network economy. By cultivating quantitative skills, and spatial econometrics, locational analysis together with geographical information systems also plays a major part in responding to the challenge and opportunity of big data. Political economy approaches remain vital for studying inequality, uneven development, capitalism, and for cultivating relationships between economic geography and heterodox economics. Cultural economy offers invaluable insights into the rise of cultural and creative industries, not to mention social media and networks. In addition to the application of existing approaches, new perspectives have been developed or gained prominence since 2000. The global financial crisis has renewed interest in the economic geography of finance (Clark and Wójcik, 2007). The US subprime crisis had a fundamental spatiality, arising at the intersection between predatory financial practices and speculative investment in urban land and housing. Competitive deregulation driven by competition among nation states and financial centres contributed to the credit boom in the

Introduction   9 US and elsewhere. The consequences of the crisis have been extremely uneven both across and within countries. In Europe, the gap between the wealthier north and the poorer south has grown. In the UK, London has boomed, while smaller cities and towns, particularly in the north, have suffered the most. Geographers have been busy documenting and interpreting these developments (Aalbers, 2012; Christopherson et al., 2015). One could even notice a shift from the economic geography of finance, focused on the spatiality of finance, to a much broader financial geography, a geography infused with attention to and understanding of finance, drawing on influences from political economy and cultural economy, as well as financial sociology, history, anthropology, and economics. Financial geography is an example of the way in which economic geography is responding to the crisis of confidence in mainstream economics. Arguably, financial economics, with its blind faith in market efficiency and expansion of financial markets, has been compromised most of all, and so it is not surprising to see other social sciences stepping in to respond to demands for new perspectives. To be sure, in the wake of the crisis, economists themselves search for new ideas. For example, the Institute for New Economic Thinking, founded in 2009, aims among other goals at revising the university curriculum for economics to make it more empirically grounded and methodologically diverse, more reflexive and critical, and less canonical, with less emphasis on calibrating established one-​size-​fits-​all models, and more attuned to the diversity of economic life in the real world (INET, 2016). One cannot help thinking that this looks like a manifesto to convert economics into economic geography. While the willingness of the economics profession as a whole to reform itself is a controversial topic (Mirowski, 2013), geographers should see it as an opportunity for greater collaboration and competition with economists. Take, as an example, Thomas Piketty’s magisterial work on inequality (Piketty, 2013). On the one hand, the book represents a model of historically informed, and empirically rich research on a topic hitherto notably neglected by economists. On the other hand, the work remains rooted in methodological nationalism, with no attention to the urban and regional dynamics of inequality, and relatively little attention to its international dimensions, beyond a comparison between nation states. As in economics, in economic geography we have seen an increasing engagement with ideas from biology and engineering. One group of studies is focused on the notion of resilience—​the ability of a regional economy to withstand, absorb, or overcome an external economic shock (Martin and Sunley, 2015). Another area of research builds on evolutionary economics to develop evolutionary economic geography, interested in the processes involved in the evolution of the economic landscape, and the role of regions and cities as arenas of economic evolution. Theses writings are rife with such concepts as learning, path dependence, selection, and novelty (Boschma and Frenken, 2006). A related but younger strand of research in evolutionary economic geography draws inspiration from complexity economics, treating the economic landscape as a non-​deterministic, non-​linear system, where spatial and other macrostructures emerge spontaneously out of the microscale behaviours and interactions of system components, such as firms or individuals (Martin and Sunley, 2007). A focus on agency in shaping economic landscape, prominent in evolutionary economic geography, is the centre of relational economic geography, which focuses on relations among economic actors as key processes shaping economic landscape (Bathelt and Glückler, 2003). A related practice approach builds more directly on cultural economic geography to study practices understood as stabilized, routinized, or improvized social actions that constitute

10   Clark et al. and reproduce economic space, and through and within which socioeconomic actors and communities embed knowledge, organize production activities, and interpret and derive meaning from the world (Jones and Murphy, 2010). At the heart of this approach is an interest in the everyday economic life. Both the relational and practice approach pay a lot of attention to social and economic networks, but not quite as much as groups of studies that make networks their explicit focus, with the Global Production Networks approach in the lead. The latter is concerned with processes of strategic coupling between actors representing particular regions, such as government agencies, local businesses or business associations, and transnational corporations (Coe et al., 2004). In sum, whether focused on the micro, meso, or macro, the new turns and approaches in economic geography demonstrate that the discipline has remained open and dynamic in a pursuit of explaining the spatiality of economic processes and their impacts on growth, uneven development, stability, and sustainability. While the progress made by economic geography since 2000 escapes easy quantification, to illustrate it in a historical context, in Figure 0.1 we use the Google Books Ngram Viewer, which captures the frequency of a bigram occurring in books included in the English-​ language corpus of Google Books. Results since 1900, smoothed over ten years, indicate that the popularity of the term ‘economic geography’ had grown steadily since the turn of the century and peaked in the 1960s, possibly owing to the influence of locational analysis and regional science. It had declined in the 1970s and the 1980s, which saw the rise of social and then cultural geography. Since the trough of 1988, the popularity of the term has grown, probably revitalized by the wave of interest in globalization in the 1990s. This upward trajectory continued into the 2000s, and by 2008, on which the data end, economic geography was more popular than political, cultural, or social geography, and even surpassed human geography. We should view this diagram with a spoonful rather than a pinch of salt, but broad trends indicated therein do make some sense, and they definitely paint economic geography as 0.0000260% 0.0000240% 0.0000220% 0.0000200% 0.0000180% 0.0000160% 0.0000140% 0.0000120% 0.0000100% 0.0000080% 0.0000060% 0.0000040% 0.0000020% 0.0000000% 1900

1910

1920

1930

1940

Economic geography Cultural geography

1950

1960

1970

Human geography Social geography

1980

1990

2000

Political geography

Figure  0.1   Frequency of Geographical Bigrams Occurring in Books Included in the English-​language Corpus of Google Books. Source: Authors, based on Google Books Ngram Viewer.

Introduction   11 a discipline on the rise. There is little doubt that this upward trend has continued beyond 2008. To start with, in 2008 Paul Krugman was awarded the Nobel Memorial Prize in Economic Sciences for his contribution to economic geography and international trade theory. Later in the same year, the World Bank published its 2009 World Development Report entitled ‘Reshaping Economic Geography’, reflecting not only the influence of Krugman, but also related work in urban economics led by Gilles Duranton, Ed Glaeser, and J. Vernon Henderson, among others (World Bank, 2009). In addition, the crisis of confidence in mainstream economics in the wake of the Great Recession has also promoted interest in economic geography as an alternative way of viewing the economy. What has contributed to the rise of economic geography since 2000, in addition to individual effort, ingenuity, and timely responses to opportunities created by political, economic, and other changes in the discipline’s external environment, were major organizational developments within it. In terms of academic journals, the discipline now has two pillars or flagships publishing cutting-​edge research in the field. In 2001 the Journal of Economic Geography joined the Economic Geography journal, published since 1925. The specific contribution of the new journal, developed under the leadership of Neil Wrigley, is reinvigorating the intersection between economic geography and economics, helping the journal to become one of the top outlets in both geography and economics. On a different front, Jamie Peck led the organization of the Summer Institute in Economic Geography, a biannual gathering of graduate students and early-​career academics together with leading scholars who present and discuss theoretical and methodological innovations in the discipline. By 2016, over 300 young scholars from nearly 50 countries attended these meetings. Finally, in 2000 the Global Conference on Economic Geography was born in Singapore, through the initiative of Henry Yeung with the aim of facilitating a global dialogue among economic geographers, with a specific focus on the Global South. The following Global Conferences took place in Beijing, Seoul, and the most recent one in 2015 in Oxford, hosting over 700 delegates from over 50 countries. This successful participation has changed the timing from every four years to a triennial cycle, with the next conference to be held in Cologne in 2018. Geographers have also been at the forefront of many new interdisciplinary initiatives, such as the Centre for Innovation, Research and Competence in the Learning Economy, established in Lund in 2004. The Regional Studies Association launched a Chinese division in 2014, and established new journals led by economic geographers, including Spatial Economic Analysis in 2006, and Area Development and Policy in 2015. There are many other developments that reflect a discipline that has become larger, more globalized and diverse, and more influential that could also be mentioned.

Our Handbook Given the success economic geography has enjoyed in the last fifteen years, it is not surprising that it has generated many state-​of-​the-​art textbooks and handbooks. The most recent range from introductory texts for undergraduate students (Wood and Roberts, 2011; Coe et al., 2013), through more advanced texts focused on the political economy approach (Leyshon et al., 2011; Barnes et al., 2012) and key concepts (Aoyama et al., 2010) to more specialized

12   Clark et al. texts, such as The Handbook of Evolutionary Economic Geography (Boschma and Martin, 2012), not to forget the seventh edition of Peter Dicken’s Global Shift (2015). Building on the success of the first edition, this volume offers a radically revised, updated, and broader approach to economic geography. The structure of the first handbook was based primarily on pairing perspectives on different topics from geographers and economists, in order to highlight the distinctive theoretical contribution of the former, as well as encourage dialogue between geographers and economists. Now that economic geography has matured, and dialogue with economists has been reinvigorated, we can take a bolder approach to tackling the questions of why and so what of where. As argued earlier, the first fifteen years of the twenty-​first century have thrown into sharp relief the challenges of growth, equity, stability, and sustainability facing the world economy. In addition, they have exposed the inadequacies of mainstream economics in providing answers to these challenges. In this context the objective of the handbook is to provide a forward-​looking perspective to understanding the various building blocks, relationships, and trajectories in the world economy. Reviewing the state-​of-​the-​art of economic geography, setting agendas, and with illustrations and empirical evidence from around the world, the book aims to serve as an important reference for advanced-​level undergraduate students, graduate students, researchers, strategists, and policy-​makers. The volume opens with Part I, ‘Grounded in Place’, which explores some of the key developments in the world economy of the twenty-​first century. It starts by investigating the phenomenal growth of Asian economies (Chapter 1), with particular focus on China (Chapter 4) and India (Chapter 5). Key challenges to future development in both countries, including political structures and social exclusion, are discussed. Part I also highlights the unprecedented rise of inequality within affluent countries with high levels of income inequality (Chapter 2), including regional disparities (Chapter 3), aggravated by the financial meltdown of the US subprime and Eurozone crises. While in many countries provincial areas suffered most from the crisis and the subsequent austerity policies, as Chapter 6 shows, in Greece losses have been concentrated in metropolitan Athens. As a whole, Part I introduces economic geography as a discipline sensitive to the experiences of particular economies in specific places. The issue-​driven and empirically focused Part I  sets the stage for the presentation of ‘Conceptual Foundations’ of economic geography in Part II. Building on Part I, Chapter 7 reflects on the distinction between economic growth and development, offers a definition of the latter focused on shared prosperity and human fulfilment, and calls for policies focused on building capacities aimed at reducing the highly unequal social and geographical distributions that result from current frameworks obsessed with growth. Chapter 8 goes further, demonstrating how thinking geographically disrupts core propositions about capitalism in mainstream economic theory, and proposes geographical political economy as a way forward. This perspective is complemented by relational, behavioural, and evolutionary perspectives, explained and argued for passionately in Chapters 9, 10, and 11, respectively. For instance, Chapter 10 states the case that the behavioural revolution is a tool to understand common problems in economic geography, such as co-​location, clusters of innovation, diffusion of innovation, and home bias. This is extended in Chapter 12, which concludes Part II by focusing on the virtues and intricacies of comparative approaches in economic geography. In Part III, ‘Innovation’, we zoom in on innovation as a driver of growth and development. As Chapter 13 demonstrates, our understanding of innovation is broad, encompassing

Introduction   13 different types of actors in entrepreneurial ecosystems. In this vein, Chapter 15 highlights the role of consumers in co-​creating innovations. Chapter 14 explains the role of the Internet as the key innovation of late twentieth and early twenty-​first century, its diffusion, and its impacts on local economies. Chapter 16 takes stock of the evolving theoretical and empirical research on the creative industries, while Chapter 17 distinguishes between factors shaping innovation that are internal and external to firms, including the crucial question about the impact of co-​location on innovation. The ‘Firm’—​as a key unit of analysis in economic geographical research—​takes centre stage in Part IV. Chapter  18 picks up where Chapter  17 left off, examining the logics of agglomeration, including city and cluster formation, and highlighting opportunities for economic geography created by new types of data. The following chapters delve into the geographical nature of multinational enterprise (Chapter 19), global value chains and production networks (Chapter  20), global services sourcing (Chapter  21), and retail globalization (Chapter  22). All of these chapters describe opportunities, threats, and limits to corporate globalization, as well as their implications, and are complemented by Chapter 23, which charts the development of corporate social responsibility. As chapters on firms demonstrate, the new spatial architecture of firms enabled by globalization is key to the new spatial division of labour. This theme is developed further in Part V, devoted to ‘Work’. Chapter 24 opens this part with an overview of labour geography as shaped by restructuring, regulation, reorganization, and reproduction, followed by Chapter 25, which delves into precariousness in work in relation to income and labour market polarization. Chapter 26 explores the recent shift in urban and regional research towards talent, human capital, and skills. Part V concludes with Chapter 27, which extends the theme of skill, by exploring the politics of skill and the contribution of immigrant workers to collective learning in host countries and regions. The all-​permeating significance of ‘Finance’ to understanding economic geographies of the twenty-​first century is the focus of Part VI. On the empirical front, readers will engage in a discussion of the diffusion of subprime lending (Chapter 28); the expansion of offshore finance and investment banking (Chapter 29); the rise of digital finance, including high-​ frequency trading and bitcoin (Chapter 30); growing interconnectedness and complexity in investment networks, which paradoxically make finance more disconnected in terms of purpose and control (Chapter 31); financialization of everyday life (Chapter 32); and growing presence of private capital and diverse non-​state organizational structures in the provision of infrastructure (Chapter 33). To make sense of the changing world of finance, contributors to this part propose new conceptual frameworks for financial geography, including a spatial analysis of financial instability built on Hyman Minsky’s approach (Chapter 28); the concept of Global Financial Networks, focusing on financial and business services as agents of financialization (Chapter 29); ‘organic finance’ to move away from the overly processed and engineered products (Chapter 31); and a functional model of infrastructure analysis focused on capital, organizational, and regulatory structures (Chapter 33). Chapter 34 on the financialization of commodity markets takes us into Part VII on ‘Resources and the Environment’. Following this important segue, Chapter 35 outlines an agenda for economic geographers to take a more central role in the study of climate change and in broader, interdisciplinary conversations about the meaning and implications of the Anthropocene. Chapter 36 zooms in on the development and shortcomings of carbon markets, proposing a novel framework for analysing the spatial and temporal dynamics of value.

14   Clark et al. Resource governance is the focus of Chapters 37 and 38, which consider long-​term resource scarcity and resource periphery, respectively, with Chapter 37 discussing the idea of aggregate natural capital, operationalized through a balance sheet and risk register. Chapter 38, in turn, draws attention to resource-​driven economies in the developing world, changing trends, and the way resource peripheries are enmeshed in the global economy. Part VII ends with emphasis on energy transitions, including the recent boom in shale oil and gas extraction, and how these can be conceptualized through the lens of evolutionary economic geography and other complementary perspectives (Chapter 39). Addressing the dangers of inequality, instability, and environmental crisis head on, the volume concludes with ‘Strategies for Development’. Following on from Part VII, Part VIII starts with green growth (Chapter 40), defined as a long-​run increase in GDP alongside the enhancement, or at least protection, of natural capital. The following two chapters shift emphasis to equity. Chapter 41 develops a conceptualization of ‘equitable economic growth’, and provides insights into the operationalization of this concept in the Global South, focusing specifically on the suitability of localized equitable economic growth strategies. Chapter 42 explores the idea of ‘just growth’, whereby a floor for equity is created by providing a safety net and/​or building new capabilities, accompanied by strong local participation and a national welfare state. The theme of community development is extended in Chapter 43, which focuses on ‘third-​sector intermediaries’ and how they contribute to the evolving practices of self-​organizing within local communities. Chapter  44 demonstrates how collaboration among communities can contribute to more inclusive development through the operation of ‘innovation highways’. The volume ends with Chapter 45, which examines the relationship between economic shocks and resilience and the process of uneven regional development. The ‘new’ in the title of the Handbook was central to its constitution. Out of sixty-​five contributors only nine (including three editors) are the same as in the first handbook. A third of all contributors are women, and a third of all contributors started their academic careers after 2000. Authors of this volume work in ten different countries, based in a wide range of departments, including sociology, politics, management, economics, urban planning, business, and public policy schools, in addition to geography. The diversity of topics and styles reflects the diversity of its contributors. Within a broad remit of reviewing state-​of-​the-​art literature, mapping new economic developments, and setting agendas, we gave the authors freedom to choose their style and emphasis. This perspective presented in the handbook is at the same time grounded in theory and in the experiences of particular places. We thought it important to keep a balance between nomothetic and idiographic approaches, between theory and empirics. Some may say that mapping takes too long, and before you finish it, it is out of date, so it is better to focus on new conceptual developments and apply them to selected empirical illustrations, rather than trying to map the world economy in any systematic way. In response, we would like to quote an anecdote used by President John F. Kennedy: The great French Marshal Lyautey once asked his gardener to plant a tree. The gardener objected that the tree was slow-​growing and would not reach maturity for 100 years. The marshal replied: In that case, there is no time to lose, plant it this afternoon. (Forbes, 2016)

Introduction   15 Answering the question where may feel tedious, time-​consuming, and perhaps at times unsatisfactory. But if we do not map the world economy, in a broad sense of the term ‘mapping’, in order to explain the map and its implications, who else will, and what will be our distinctive contribution? Where is the centre of economic geography.

References Aalbers, M. (2012). Subprime Cities: The Political Economy of Mortgage Markets (Oxford: Wiley Blackwell). Anderson, C. (2012). Makers: The New Industrial Revolution. (London: Random House). Aoyama, Y., Murphy, J.T., and Hanson, S. (2010). Key Concepts in Economic Geography (London: SAGE). Authers, J. (2015). ‘Regulatory reform has increased New York’s slice of asset pie’. Financial Times, 20 August 2015, p. 28. Barnes, T., Peck, J., and Sheppard, E. (eds) (2012). The Wiley-​Blackwell Companion to Economic Geography (London: Wiley-​Blackwell). Bathelt, H. and Glückler, J. (2003). ‘Toward a relational economic geography’. Journal of Economic Geography 3: 117–​144. Berry, B.J.L, Conkling, E.C., and Ray, D.M. (1987). Economic Geography (Englewood Cliffs, NJ: Prentice Hall). Boschma, R. (2005). ‘Proximity and innovation. A  critical assessment’. Regional Studies 39: 61–​74. Boschma, R. and Frenken, K. (2006). ‘Why is economic geography not an evolutionary science? Towards an evolutionary economic geography’. Journal of Economic Geography 6: 273–​302. Boschma, R. and Martin, R. (eds) (2012). The Handbook of Evolutionary Economic Geography (Amsterdam and London: Edward Elgar). Castree, N., Coe, N.M., Ward, K., and Samers, M. (2004). Spaces of Work: Global Capitalism and Geographies of Labour (London: SAGE). Christopherson, S., Clark, G.L., and Whiteman, J. (2015). ‘Introduction: the Euro crisis and the future of Europe’. Journal of Economic Geography 15: 843–​853. Clark, G.L. and Wójcik, D. (2007). The Geography of Finance: Corporate Governance in the Global Marketplace (Oxford: Oxford University Press). Clark, G.L., Feldman, M.P., and Gertler, M.S. (eds) (2000). The Oxford Handbook of Economic Geography (Oxford: Oxford University Press). Coe, N.M., Kelly, P.F., and Yeung, H.W.C. (2013). Economic Geography:  A  Contemporary Introduction (London: Wiley). Coe, N.M., Hess, M., Yeung, H.W.C., Dicken, P., and Henderson, J. (2004). ‘Globalizing regional development:  a global production networks perspective’. Transactions of the Institute of British Geographers 29: 468–​484. Dicken, P. (2015). Global Shift:  Mapping the Changing Contours of the World Economy (London: SAGE). Forbes (2016). ‘Forbes quotes’ http://​www.forbes.com/​quotes/​5238/​ (last accessed 16 June 2016).

16   Clark et al. Gibson-​ Graham, J.K. (2006). A Postcapitalist Politics (Minneapolis, MN:  University of Minnesota Press). Gordon, R.J. (2016). The Rise and Fall of American Growth (Princeton, NJ: Princeton University Press). IMF (2016). ‘IMF Managing Director Christine Lagarde welcomes U.S. congressional approval of the 2010 quota and governance reforms’, Press Release 15/​573, December 18, 2015 https://​ www.imf.org/​external/​np/​sec/​pr/​2015/​pr15573.htm (last accessed 16 June 2016). INET (2016). ‘The mainstream economics curriculum needs an overhaul’ http://​www.inet. ox.ac.uk/​news/​economics-​curriculum-​overhaul (last accessed 16 June 2016). Jones, A. and Murphy, J.T. (2010). ‘Theorizing practice in economic geography: foundations, challenges and possibilities’. Progress in Human Geography 35: 366–​392. Kahneman, D. (2011). Thinking Fast and Slow (London: Allen Lane). Leyshon, A., Lee, R., McDowell, L., and Sunley, P. (eds) (2011). The SAGE Handbook of Economic Geography (London: SAGE). Martin, R. and Sunley, P. (2007). ‘Complexity thinking and evolutionary economic geography’. Journal of Economic Geography 7: 573–​601. Martin, R. and Sunley, P. (2015). ‘On the notion of regional economic resilience: conceptualisation and explanation’. Journal of Economic Geography 15: 1–​42. Massey, D. (1995). Spatial Divisions of Labour: Social Structures and the Geography of Production (New York: Routledge). Mirowski, P. (2013). Never Let a Serious Crisis Go to Waste: How Neoliberalism Survived the Financial Meltdown (London: Verso). PBL Netherlands Environmental Assessment Agency (2015). Trends in Global CO2 Emissions (The Hague: PBL). Piketty, T. (2013). Capital in the Twenty-​first Century (Cambridge, MA: Harvard University Press). Polanyi, K. (1957). The Great Transformation: The Political and Economic Origins of Our Time (Boston, MA: Beacon Press). Schafran, A. (2013). ‘Origins of an urban crisis: the restructuring of the San Francisco Bay Area and the geography of foreclosure’. International Journal of Urban and Regional Research 37/​2: 663–​688. Simon, H.A. (1956). ‘Rational choice and the structure of the environment’. Psychological Review 63: 129–​138. Stern, N (2007). The Economics of Climate Change: The Stern Review (Cambridge: Cambridge University Press). Taylor, G. (1940). Australia: A Study of War Environments and Their Effect on British Settlement (London: Methuen and Co.). The Economist (2016). ‘Over troubled water: cross-​border transport links are overshadowed by political fears’ http://​www.economist.com/​news/​china/​21692935-​cross-​border-​transport-​ links-​are-​overshadowed-​political-​fears-​over-​troubled-​water (last accessed 1 March 2017). United Nations (2016) ‘Population Division data’ https://​esa.un.org/​unpd/​wup/​CD-​ROM/​ (accessed 16 June 2016). Wood, A. and Roberts, S. (2011). Economic Geography:  Places, Networks and Flows (London: Routledge). World Bank (2009). World Development Report 2009:  Reshaping Economic Geography (Washington, DC: World Bank).

Pa rt  I

G ROU N DE D I N  P L AC E

Chapter 1

Gl obal Prospe c ts :  T h e Asian Cen t u ry? Michael Berry and Benno Engels Introduction The Asian region now accounts for more than half the world’s population and more than a third of total output. The various economies that make up this culturally and politically diverse region are coming to express a ‘new economic geography’ of the twenty-​first century in which a ‘catch up’ with the advanced Western nations appears to be a predominant feature (Nayyar, 2013). However, like earlier globalization regimes, the current era is conditioned by complex patterns of uneven development and interdependencies that lock particular countries and regions into the variegated whole. Geography and geographers can make a substantial contribution to understanding these developments. This chapter provides an initial step in this direction. Asia is host to the second and third largest economies in the world, China and Japan, respectively. In 2010, Asia had 3.3 million millionaires; only slightly behind North America’s 3.4 million. Fifty per cent of the world’s container traffic originates from here, and this proportion is growing as the Chinese economy progressively becomes the industrial factory and engine room of the world. We are possibly standing on the cusp of an ‘Asian Century’, which will determine humanity’s future and the economic viability of global capitalism. However, whatever ‘the Asian Century’ means, it cannot be allowed to refer to a homogenous super-​ region. Instead, the geographer’s traditional perspective in looking for the drivers and outcomes of uneven spatial development is as relevant and important as ever if we are to get a grip on development trends and future projections for this region in the world’s economic geography in the twenty-​first century. The chapter therefore seeks to untangle and extract what is, in fact, happening through select parts of Asia into the twenty-​first century. The discussion commences with a broad overview of those national economies in Asia that have been among what Sharma (2012) has called ‘the breakout nations’, drawing on the available comparative economic, environmental, and social data. We then move to a consideration of the nature of the complex and changing interrelations between economies in the broad region. This is followed by a discussion

20   Berry and Engels of the key development issues identified and the drivers at work. Is the ‘convergence thesis’ a useful framework for understanding what is happening? In looking to the future, what might happen in the leading Asian economies, both established and emerging, over the next decade and beyond? Finally, what challenges do the study of a dynamic region like Asia raise for the development of economic geography as a discipline? It is argued, in brief, that the very different patterns of economic development experienced through the Asian region are driven by a combination of forces—​economic, cultural, and political—​unleashed leading up to and following the 2008 global financial crisis (GFC) and the efforts of national governments to deal with the emerging outcomes.

Growth and Transformation in the Asian Region Much has happened over the past decade, notably the impacts and repercussions of the GFC triggered in 2008. In order to trace the uneven geographical patterns and drivers of change leading to and since the GFC, we have chosen to focus on eleven Asian countries:  India, China, Indonesia, Bangladesh, Vietnam, the Philippines, Pakistan, South Korea, Japan, Malaysia, and Thailand. This group we call the N–​(8 + 3) to distinguish it from another group of eleven emerging countries identified by Goldman Sachs (see Wilson and Stupnytska, 2007). Our grouping includes the fast-​growing BRIC (Brazil, Russia, India, and China) economies in Asia, six identified as newly emerging, one established developed economy (Japan), and two South East Asian countries (Thailand and Malaysia) ignored by Goldman Sachs. By covering this group of eleven, we can explore the diversity of development trajectories and interrelations unfolding across a region as large and complex as Asia, while still testing the implied thesis that it is from within this region that the dynamic economies of the twenty-​first century will come. We also test the corollary that the world is witnessing a convergence by which the economic lead of the West is rapidly being lost, inevitably to result in a ‘return to the future’ in which Asia leads the world, as it did in the centuries prior to the Industrial Revolution. China and India are the largest Asian economies and among those that have recently grown most quickly. Their economic and geopolitical shadows loom large over the entire region and they are increasing their economic and political engagement with the rest of the world. Japan was the first Asian country to ‘break out’ into sustained economic development and to interact intensively with its regional neighbours; it has experienced significant stagnation, economically and demographically, since the early 1990s, in contrast to its large western neighbour. The remaining eight countries in the selected group were differentially affected by the GFC and its aftermath, and have formed somewhat different interactions with each other and with the hegemonic ‘big three’, particularly China. Taken together, these country cases allow us to explore the complex patterns of development occurring within the dynamic Asian region. The following sketch outlines the main contours emerging. By 2012, the broad Asian region accounted for 36 per cent of global real gross domestic product (GDP), an increase of eight percentage points between 2000 and 2012 (Asian

Global Prospects: The Asian Century?    21 Development Bank, 2013). China, Japan, and India accounted for 70 per cent of the region’s GDP. However, 2012 also saw economic growth moderate across the larger economies in the region; for example, China’s growth fell, but to a still high 7.8 per cent. Table 1.1 compares the average growth of the eleven economies over the past five years and their 2012 results. There is variability in recent economic performance, at the aggregate level, within the Asian region. The stellar growth of India and China during the first decade of the twenty-​first century has clearly slowed, while growth of smaller economies like Thailand and Malaysia has picked up. Japan has begun to grow again after two decades of economic decline and stagnation. Korea, one of the ‘Asian tigers’ of the late twentieth century, has also slowed. Bangladesh, Indonesia, and the Philippines are growing strongly, as is Pakistan, in spite of a deteriorating internal security environment. Vietnam continues to grow strongly, although at a slower pace than earlier in the decade (see Table 1.1). At the regional level, the Asian economies overtook Europe and North America in its share of global GDP (at purchasing power parity) in the first decade of the 2000s (Maddison, 2008). The associated increase in average living standards, measured by real GDP per capita, has been impressive over the past two decades, especially during the new century, as Table 1.2 suggests. These data suggest large differences between countries with respect to current and recent living standards (see Table 1.2). Even though China had overtaken Japan with respect to the size of its economy by 2005, its per capita income is currently barely one-​quarter of Japan’s. As Japan has marked time over the past five years, the other ten countries have moved closer, especially China and India but so too the lesser-​developed nations like Bangladesh and Vietnam. Living standards in South Korea continue to converge on Japan.

Table 1.1 Comparative Growth Rates in Real Gross Domestic Product (GDP), Selected Countries Country

Real GDP growth: 2012 (%)

China

7.8

9.3

India

5.0

7.2

Japan

2.2

–​0.1

Bangladesh

6.1

6.2

Pakistan

4.5

3.2

South Korea

2.0

2.9

Indonesia

6.1

5.9

The Philippines

7.1

4.7

Vietnam

5.1

5.8

Malaysia

5.3

4.2

Thailand

6.5

3.0

Source: Asian Development Bank (2013).

Average real GDP growth: 2008–​12 (%)

22   Berry and Engels Table 1.2 Comparative Change in Per Capita Gross Domestic Product at Purchasing Power Parity (Current International Dollars), Selected Countries: 2000–​12 Country China India Japan Bangladesh Pakistan South Korea

2000 ($)

2012 ($)

2357

9210

1536

3950

25,914

35,204

862

1917

1690

2788

17,197

30,722

Indonesia

2407

4949

The Philippines

2410

4454

Vietnam

1426

3998

Malaysia

9393

17,084

Thailand

5086

10,757

Source: Asian Development Bank (2013).

Nayyar (2013, p. 136) has traced the different rates of convergence of per capita incomes in selected Asian countries on the developed world.1 While South Korea (and Taiwan) understandably are standout performers, with per capita incomes reaching around 80 per cent of levels in the industrialized West, Malaysia (40%), Thailand (33%), and China (26%) were also significant movers. Indonesia almost doubled its per capita income in relation to the developed economies from a low 10 per cent. India, however, barely made ground on this measure. The Asian Development Bank (2013) identifies a number of key factors in the process of structural transformation underlying rapid economic growth or development in countries that do not rely mainly on exploiting natural resources like oil. Typically, development proceeds by switching resources over time away from agriculture towards the manufacturing and service sectors. This is achieved through productivity improvements in agriculture and increasing imports that release labour from the sector.2 Capital invested in manufacturing and related industries attracts labour where productivity levels are (initially) higher and export opportunities greater, especially where the government imposes protective policy measures. Eventually, industrialization moderates as resources are increasingly drawn into the diverse service sector, both producer-​and consumer oriented. Table 1.3 outlines this pattern of development for the selected countries. The two largest economies diverge somewhat. China’s transformation is more dramatic in relation to both the proportion of employment and shifts in the shares of value added by sector (see Table 1.3). The service sector in India has grown more rapidly and agriculture has retained a more significant and traditional role than in China. The latter’s growth

Global Prospects: The Asian Century?    23 continues to be driven by manufacturing and related industries, based on continued growth in merchandise exports. China’s exports recovered quickly in the aftermath of the GFC of 2008–​09, helping to cushion the impact of the GFC on countries—​for example Australia and Canada—​from which they import raw materials. On a smaller scale, Indonesia and Vietnam have followed China’s path, while Bangladesh, the Philippines, Malaysia, and Thailand have, like India, witnessed rapid growth in services. The established developed economies in South Korea and Japan are heavily service sector focused; Japan now generates almost three-​quarters of its value added from service-​sector industries, which employ four out of five workers in the economy. Of the group, only Pakistan has shown modest structural change during the past two decades. The apparent decline in the significance of agriculture masks the fact that the shift to manufacturing and services has depended, in part, on the increase in agricultural productivity due to major technological changes over the past forty years and the shift from traditional to high-​yield and higher-​value agricultural products. Across all sectors, rapid development has entailed diversification of products and the creation of new linkages and complementarities in production, financing, and distribution. Development has occurred most in economies that have diversified and built strong export platforms, utilizing new production methods and processes that underpin a switch to higher value, more complex products supported by local supply chains, associated services, and complementary infrastructures (Nayyar, 2013). Later stages of development result in the spatial division of supply chains across regions and national borders (see section ‘Trajectories of Development’).

Table 1.3 Value Added by Sector Country

Agriculture as % of GDP

Industry as % of GDP

1990

2012

China

27.1

India

29.3

Services as % of GDP

1990

2012

10.1

41.3

17.4

26.9

2012

45.3

31.5

44.6

25.8

43.8

56.9

2.4

1.2

37.9

26.1

59.8

72.7a

Bangladesh

30.2

17.7

21.5

28.5

48.3

53.8

Pakistan

26.0

24.4

25.2

22.0

48.8

53.6

8.7

2.6

39.9

39.1

51.5

58.2

19.4

14.4

39.1

46.9

41.5

38.6

The Philippines 21.9

11.8

34.5

31.1

43.6

57.1

Vietnam

38.7

19.7

22.7

38.6

38.6

41.7

Malaysia

15.0

10.2

41.5

41.2

43.5

48.6

Thailand

10.0

11.4a

37.2

38.2a

52.8

50.3a

Japan

South Korea Indonesia

a

1990

a

GDP, gross domestic product. a2011, latest year available. Source: Asian Development Bank (2013)

24   Berry and Engels In most cases, these structural changes have been deliberately encouraged by the policies of the national governments concerned, often against the technical advice and political pressures of Western governments and international agencies like the International Monetary Fund (IMF) that the former control. This resistance was most obvious in the cases of South Korea and Taiwan, when their governments adopted import replacement policies designed to boost domestic industrialization and exports prior to gradually and selectively liberalizing their commercial and financial sectors (Berry, 1989). China, India, and, to some extent, Indonesia have followed this trajectory, incurring the opposition of G7 governments and international agencies. Changes in social structure that challenged conservative attitudes and economic practices also figured in the creation of entrepreneurial activity; expansion of the opportunities for and economic role of women is one factor in the success of the more dynamic Asian economies. ‘(i)t is impossible to become a modern economy with social structures that do not favour change’ (Asian Development Bank, 2013, p. 5). Most significantly, development has meant urbanization. Economic growth within most of the N–​(8 + 3) has been associated with rapid urbanization. This pattern—​familiar in the light of nineteenth-​century growth trajectories in the West—​has been driven by large-​scale internal migration flows from rural to urban areas, especially in China. Twelve of the world’s largest metropolitan regions are located in Asia, including eight of the ten most densely populated cities; Tokyo, Delhi, Shanghai, Mumbai, and Beijing figure in the eight largest. Rapid urban growth is also associated with a range of new and intractable problems for governments in the region, particularly with respect to the provision of adequate housing, health, transport, and environmental standards, and the increase in urban poverty. Aspirational values that emulate Western consumerism have stimulated local entrepreneurial activity, no longer suppressed by traditional mores and government policy; the rapid growth of a materialistic middle class in the three large countries is a stark manifestation of this sociocultural process. Deng Xiaoping’s radical turn towards ‘Socialism with Chinese characteristics’ was legitimated by his purported claim that ‘to be rich is glorious’. Very high levels of domestic savings and investment have underpinned China’s growth. Since 2005, gross domestic capital formation has exceeded 40 per cent of GDP, reaching almost 50 per cent by 2012; domestic savings tracked at similarly high rates. India and Indonesia also exhibited rates in excess of a third of GDP (Asian Development Bank, 2013). Increasingly, future growth in these countries will depend on the expansion of domestic consumption by their growing middle classes. Although China has dominated Asian development over the past twenty years this overview also suggests that the whole region has been transformed over this period. Nayyar (2013) has coined the term ‘catch up’ to capture this phenomenon, and he identifies the turning point as 1950. In the period 1820–​1950, he argues, the industrializing Western nations diverged from the rest of the world, including Asia. Since 1950, the gap has been rapidly closing. Thus, Asia’s share of GDP rose from around 15 per cent of world GDP in 1950 to 38 per cent by 2008, roughly the situation holding in 1900 (Nayyar, 2013, pp. 15, 50). At the beginning of the nineteenth century Asia accounted for more than half of global GDP; the recent trend suggests that history may be well on the way to repeating itself. It is this projection that lies at the heart of claims that the world is on the eve of ‘the Asian Century’.

Global Prospects: The Asian Century?    25

Trajectories of Development This overall picture of a region undergoing dynamic change masks complex spatial and temporal development paths. As Yusuf and Nabeshima (2010) argue, Asian industrialization came in waves, starting with that of Japan from the mid-​1950s, driven in successive decades by rapid technological progress and expanding exports to the region and beyond, to meet expanding consumer demand in the West. From the late 1960s, export-​driven industrialization took off in Hong Kong, South Korea, Taiwan, and Singapore, the four Asian tigers (Amsden, 1989; Berry, 1989). A decade later, Malaysia, Thailand, the Philippines, and Indonesia began their transformations, creating what has been termed the second wave of ‘fast followers’. These late starters benefitted from both the import demands of US and Japanese firms and consumers and increasing flows of foreign direct investment (FDI) as large multinational firms domiciled in the advanced economies ‘disarticulated’ production processes and reintegrated production and supply chains on a global basis. This new international division of labour depended on tapping into low-​cost labour markets, low-​tax regimes, and compliant domestic government regimes, largely within the ‘Pax Americana’ created after World War II by US foreign policy interests during the heyday of the Cold War. Inevitably, emerging domestic capitalists drew on technology transfer through FDI and human capital development through investment in education and training, financed by foreign aid and international agencies, in order to support the strengthening forces of industrial transformation. It was not until the 1980s, following the decisions taken at the Communist Party’s central committee in 1978, that China arrived on the scene (Nayyar, 2013). The ‘end of the iron rice bowl’—​a policy that guaranteed a minimum subsistence regardless of productive contribution—​saw staged moves to introduce market incentives to stimulate domestic economic growth. Starting first with micromarkets, as peasants and small entrepreneurs were allowed to trade modest surpluses and reinvest the proceeds, the government moved to reform large state-​owned enterprises (SOEs), and reoriented their focus to world markets by creating dual-​pricing regimes. These reforms were financed by plentiful credit provided by the state-​owned banks, boosted by large-​scale investment in production facilities and supported by massive infrastructure investments in ports, roads, and power (Ding and Knight, 2008). More controversially, the Chinese currency was manipulated to entrench China’s export supremacy. The proportion of exports to GDP in China rose from 6 to 40 per cent from 1985 to the eve of the GFC in 2007, accounting for almost 8 per cent of global exports (Maddison, 2009; Yusuf and Nabeshima, 2010). Spatially targeted policies creating special economic zones attracted FDI on the Eastern seaboard and to the South, which increasingly drove growth, attracting large-​scale immigration of workers from rural areas and smaller towns. By the mid-​1990s, China was closing fast on the Asian tigers. India began its growth surge even later than China, associated with lower but still significant and rising investment and savings rates. Exports in relation to GDP converged on Germany’s performance (around 20 per cent) in the first decade of the new century. Indian growth was strongly concentrated in information and communications technologies, particularly with respect to software development and complementary services, both low and high value. However, the contribution of manufacturing to Indian development has

26   Berry and Engels been modest, compared with China; the share of value added in this sector actually fell in the two decades after 1990 (see Table 1.3). Conversely, value added in services increased significantly over that period, underscoring the comparative and competitive advantages generated in the information and communications technology sector supported by the location of back office functions from the advanced economies, the high levels of technical education acquired by Indian nationals, at home and abroad, and the investment priorities of domestic entrepreneurs. India, like China, has gradually emerged from an initial import replacement phase of development to one increasingly reliant on (selectively) opening its economy to FDI, large-​scale investment in energy production and a more outward-​looking foreign policy. Both governments seek to mobilize the nationalistic sentiments of their peoples to exert a larger footprint on world affairs, economically, diplomatically, and politically. On a much smaller scale, Vietnam and Bangladesh are following India in this fourth wave of development. The rapid growth of the two large Asian economies created increasing opportunities for smaller economies in the region. The momentum developed in the two leading economies, their large domestic markets, and stimulus policies imposed by their central governments allowed them to survive the fallout from the GFC in 2008–​09 and continued to boost demand throughout the region and beyond. As growth flagged in the advanced economies of Europe and North America in the period since 2008, Asia continued to prosper. Both India and China have bucked the trend of fast-​growing economies regressing to mean growth over time (Yusuf and Nabeshima, 2010). Most of the other N–​(8 + 3) economies have likewise confounded the conventional model, pulled along by the dominant duo. The future will tell whether all continue to do so. The policies of their governments and business elites will be of critical importance to their development prospects. Uneven development through the region will unfold in part as a result of how the brake of corruption and cronyism is relaxed and key institutional supports that developed through the centuries in Western countries, such as the rule of law, the regulation of market failures, robust public finance, and banking institutions, supplant older deeply entrenched cultural and political traditions still in force. As noted in the section ‘Corruption’, many of the N–​(8 + 3) countries are among the most corrupt in the world. They may—​failing institutional change—​yet embrace the stagnating fate of a number of South American and African nations during the post-​World War II period, whose development was blighted by the structural social and political forces that locked compliant local comprador elites into relations of dependency with the neocolonial policies of developed countries. The scale of China’s growth in output and exports over the past thirty years has been unprecedented. Even more impressive has been the extent to which its export mix has transformed from being dependent on agriculture and resources to high value-​added finished products and services (Yusuf and Nabeshima, 2010). During the first half of the 1980s, 60 per cent of Chinese exports comprised primary products. By 1990, this proportion had fallen by half, with the share of textiles, clothing, and footwear rivalling it. In the following fifteen years, China maintained its dominance of global exports in this latter category, while also emerging as the leading exporter of electronics and many high-​tech products, including transport equipment. By the beginning of the current decade, China’s technology intensive export sector had overtaken the lead of Japan and South Korea. The high proportion of high-​tech exports from the Philippines, however, reflects the small size of its manufactured export sector and its

Global Prospects: The Asian Century?    27 strong concentration on the assembly and re-​export of high-​tech products for Japanese and Korean companies. The nature, scale, and pace of China’s development have indelibly imposed economic change throughout Asia. The impacts have been transferred through emerging patterns of international trade and FDI, creating new and more complex production networks across borders. In this latter sense, China followed Japan’s trajectory in relocating lower-​skilled manufacturing processes within the region. For example, the recent decade has seen Chinese companies shifting light manufacturing production to Vietnam to benefit from lower local wage rates, mirroring the same process that earlier marked capital flows from the West to China. The scale of intraregional trade in finished products also expanded as China focused on the production and export of electrical, transport, telecommunications, and office equipment, with countries like Malaysia, Indonesia, and Vietnam supplying in return key primary and intermediate products. Australia also prospered during this period as a major supplier of coal and iron ore. Japan and the Asian tigers exported high-​value components and completed products to China in an increasingly symbiotic trade relationship. Intraregional trade patterns have thus intensified over the past three decades, driven by the increasing dominance of China and, to a lesser extent, India. Yusuf and Nabeshima (2010) provide a detailed breakdown of the patterns emerging in the broader region. Intraregional trade (exports plus imports) has become more concentrated in the East Asian region (including China and Japan); that is, by 2006, 60 per cent of international trade by the countries of East Asia was between themselves. However, while China’s share of regional trade has increased to 20 per cent, Japan’s share has declined to 11 per cent, reflecting the surge in Chinese growth since 1980. Nevertheless, Japan, like other countries in the region, increasingly focused on regional trading, while China diversified its trade somewhat outside the region, most notably via booming exports to the USA and Europe, and global sourcing to meet their escalating energy requirements. This suggests that China is increasingly taking over not only as the dominant industrial pole for the region, but also as the locus of assembly for re-​export of products to the world, in particular to the USA. The resulting trade imbalance between China and the USA has formed an increasing focus of tension between their governments, while the consequential escalating Chinese investment in US Treasury bonds has rendered both economies increasingly dependent on the other. Until the GFC, the consumption boom in the USA underpinned the Chinese miracle along with robust growth in other export-​oriented Asian economies. A critical feature of the emergence of Asian development has been the increasing economic integration of nations within the region. The role of intraregional trade has been noted earlier in this section. Cavoli (2012) has looked at further indicators of both real and financial integration within East Asia.3 He finds that, on a number of measures, the original ASEAN countries—​Malaysia, Singapore, Indonesia, Thailand, and the Philippines—​ are relatively well integrated with each other and with China, Japan, and South Korea. Real integration is observed with respect to factors like business-​cycle synchronization and relative purchasing power parity. Financial integration can be measured by interest rate parity and asset price correlations. Integration of the smaller, much less developed ASEAN nations, including Vietnam, is much lower on most or all of these measures. However, the smaller ASEAN countries are somewhat more integrated with each other than with their larger neighbours. Interestingly, integration of the five larger ASEAN countries is more developed with respect to financial factors, while the three large economies—​China,

28   Berry and Engels Japan, and South Korea—​are more closely integrated with each other in real terms (Cavoli, 2012, p. 652). On this analysis, there seems to be three broad blocs or subregional clusters emerging within Asia: the three large northern economies; the middle-​sized ASEAN founders; and the smaller South East Asian member nations of Vietnam, Myanmar, Laos, Cambodia, and Brunei. South Asia would form a fourth subregion that is only partly integrated with the broad Asian region. However, crosscutting non-​economic factors also further differentiate the regions and subregions, creating tensions that must be managed by their governments. Thus, the nations of Indochina must deal politically with the burden of historical cultural and territorial conflicts built up over centuries, both among themselves and through the interventions of Western powers. Similarly, tensions that go beyond the economic national interest still bedevil relations between Japan and its one-​time conquered colony, South Korea. New tensions arise from the increasing dominance geopolitically of China in North Asia (see sections ‘Environmental Degradation’ and ‘Looking Forward’). Important as the lingering impacts of the GFC have been in the region, a case can be made that it was the crisis that occurred in Asia a decade earlier—​the 1997–​98 financial crisis—​that marked a distinct break in the region-​wide growth spurt and subsequently saw the smaller economies of South East Asia, in particular, fall behind China and India. The sharp halt to growth in Indonesia, Thailand, and the Philippines, reinforced by austerity policies imposed by international agencies like the IMF, was overcome in relatively quick order, but the lost ground was never made up (Woo, 2007). Intraregional trade has moved up the value chain, from primary to medium-​and high-​ tech products in most East Asian countries. However, in recent times, China (and to a lesser extent Vietnam) has become less dependent on sourcing within the region, reflecting the robust growth of backward linkages and integration within China’s diversifying economy. Increasingly, Chinese businesses are sourcing sophisticated components and production equipment in areas like electronics from Japan and South Korea and relying on domestic sources for other inputs. China’s dominant position in the production and export of clothing and footwear is also becoming less dependent on other Asian suppliers as production networks focus more tightly on domestic supply chains. This trend has potentially negative implications for some of China’s other trading partners. Chinese direct foreign investment is also diversifying, aimed at securing natural resources and food security in African countries and Australia; Indian multinational companies are moving in similar directions. Central to understanding the economic success of China has been the spatial structure of its domestic economy and, in particular, the emergence of distinctly different regional economies. It is important to grasp that this has not only emerged ‘naturally’ through market processes—​although the normal force of agglomeration economies of scale and scope have been apparent—​but also as the result of deliberate government policy, with both unintended and intended effects. The coastal eastern strip, traditionally more open to the world from the nineteenth century, with the megacities Beijing and Tianjin to the north and Shanghai and Hong Kong to the south, cluster manufacturing, finance, and government activities that drive much of the country’s success. These urban corridors, variously termed ‘megacities’ or ‘extended metropolitan regions’, include Beijing–​Tianjin–​ Tangshan–​ Qinhuangdao in the north east; Shanghai–​ Nanjing–​ Suzhou–​ Changzhou–​ Zhenjiang–​Nantong–​Yangzhou–​Wuxi in the south; and Guangzhou–​Shenzhen–​Hong

Global Prospects: The Asian Century?    29 Kong–​Macao–​Zhuhai in the Pearl River Delta, the latter with a regional population of 150 million. They have acted as magnets for workers drawn from rural areas and direct investment attracted from abroad. The ports of Tianjin, Shanghai, and Hong Kong join China to the world. Shenzhen is now one of the world’s leading financial hubs. Shanghai has become one of the global command-​and-​control cities, joining New York, London, and Tokyo as hubs directing financial flows and intelligence throughout an increasingly borderless world. Hong Kong maintains its traditional role as a financial and commercial clearinghouse for the mainland. Beijing retains its role as the seat of national government, straining to govern a country growing and transforming at breakneck speed and showing signs of administrative stress, a developing case of ‘under-​reach’ in the face of increasing uneven development. Beyond the megacities, China’s population is rapidly urbanizing in ‘middle-​sized cities’ that are increasingly being drawn into the orbit of the former: ‘(t)hey act as “bridges” between rural areas and large urban centres’ (ESCAP, 2010, p. 8). The ESCAP report notes that 60 per cent of Asia’s population lives in urban centres of less than one million people. In addition to the scale of urbanization in Asia, population densities range from 10,000 to 20,000 per square kilometre, double the rates in Latin American cities, triple those of Europe, and ten times greater than in the USA. Debates rage about what forces underlie the large changes transforming Asia’s leading cities and economies. Orthodox economics points to the influence of liberalizing economic policies and the role of comparative advantage and ‘free trade’. Heterodox accounts focus on dynamic efficiency, technological innovation, increasing returns, path dependence, import substitution policies, and competitive advantage (Rodrik, 1992; Nayyar, 1997; Amsden, 2001). Not surprisingly, the ‘late industrializers’ have pursued policies and institutional milieux to create conditions conducive to economic ‘take-​off ’ and ‘catch-​up’. As stressed earlier, the visible hand of government, rather than simply the invisible hand of the market, has sparked growth in the Asian tigers, China, India, and most of Asia’s smaller followers (Amsden, 1989; Stiglitz, 1989; Wade, 1990; Chang, 2007). In the case of Asia, Nayyar (2013, pp. 124–​5) comments: ‘(t)he role of the state in evolving trade and industrial policies, developing institutions and making strategic interventions, whether as a catalyst or as a leader, was central to this process’. Successful interventions ranged from import substitution policies and ‘guided’ lending by financial institutions to export promotion and facilitation. Growth and transformation in the successful economies of the region was driven not by textbook concerns with allocative efficiency in a static economy, but by the construction of institutions and practices conducive to both emulation and innovation within an increasingly integrated global economy. Strong, often authoritarian governments and powerful local elites mobilized nationalistic, ethnic, or religious coalitions to push through programmes of economic development. In versions of managed capitalism countries like China, Malaysia, Indonesia, and Thailand mobilized the substantial volumes of domestic and international capital financing the basic infrastructure necessary to support the rapid development occurring.

Key Challenges to Continuing Development As the second decade of the twenty-​first century continues to unfold, the economies of the region are experiencing increasing economic, social, and political stresses arising from their

30   Berry and Engels impressive recent growth trajectories. We briefly note the four main areas of concern in the following subsections.

Environmental Degradation Associated with large-​scale urban industrial development have arisen major problems of pollution, traffic congestion, waste generation, and habitat loss. Environmental degradation costs the Chinese economy between 5.5 and 10.2 per cent of GDP, according to data supplied by the United Nations (n.d., p.  25). Two-​thirds of this cost relate to land degradation—​ including deforestation, desertification, erosion, and downstream pollution to watercourses—​and one-​third specifically to urban air pollution. Indonesia, the Philippines, and Thailand have also paid significantly in environmental terms for their recent growth spurts. More than three-​quarters of South East Asia’s original forests have been lost and this trend continues (United Nations, n.d., p. 26). Many of these costs cross national borders, creating political problems for the governments concerned. For example, poor land management practices leading to large-​scale burning in order to clear land for commercial crops in Indonesia impact negatively on communities in neighbouring Malaysia and Singapore. Development is increasingly leading to shortages of clean water for agricultural and urban use. Conflicts between nations over shares of common water sources are likely to increase and intensify. For example, the long-​running dispute between India and Bangladesh over water from the Ganges shows no sign of resolution. Over-​fishing is threatening global fish stocks and raising the prospect of accelerating species extinction. The logic underlying this danger is the familiar ‘tragedy of the commons’. India, China, and Indonesia between them face 479 threatened fish species, 355 threatened mammal species, and 1205 threatened plant species (World Bank, n.d.). Nowhere is this simmering conflict more prevalent and dangerous than in the South and East China Seas where China is coming into conflict with Japan, Vietnam, and the Philippines over control and exploitation of fishing and mineral resources. All N–​(8 + 3) nations except South Korea have substantially increased their reliance on fossil fuels over the past thirty years. A business-​as-​usual scenario forecasts that Asia will account for 45 per cent of global CO2 emissions by 2030 and 60 per cent by 2100. Transport-​ related emissions are expected to increase by 57 per cent by 2030, with China and India accounting for more than half. Of the ten countries in the world most vulnerable to climate change, six are in Asia and the Pacific (UNEP, 2012, p. 4). The future promise by China (in 2015) to introduce a national cap-​and-​trade carbon emission system may substantially change this trend. Waste generation has escalated with urbanization, creating major health problems through communicable diseases and water pollution. Toxic chemicals generated by industrialization pose particularly intractable problems, especially in countries like Bangladesh that have established toxic waste dumps and recycling facilities for the region. Adverse health effects are, in turn, exacerbated by urban air pollution. Nine of the fifteen cities with the highest particulate levels and six of the fifteen cities with the highest sulphur dioxide levels are in East Asia (United Nations, n.d., p. 28).

Global Prospects: The Asian Century?    31 Hyper-​urbanization in Asia is thus a double-​edged sword. Mega-​urban regions and urban corridors provide the dynamism for national development, while also concentrating and accelerating environmental decline.

Inequality Along with rapid economic and urban growth, many of the N–​(8 + 3) group experienced increasing economic inequality (Asian Development Bank, 2012). Unsurprisingly, in most of these countries urban inequality is more pronounced than inequality in rural areas. Like the advanced Western economies, inequality in Asia is being strongly concentrated at the top end of the income distribution, although not (yet) to the extreme extent of the former. By 2010, the share of total income received by the top income percentile (‘the top 1 per cent’) ranged between 11 and 13 per cent for China, India, and Indonesia (Piketty, 2013, p. 327). Rising inequality can have negative effects in any society. In Asia, increasing economic inequality threatens to undercut the institutional basis on which recent rises in per capita income—​and the growth of a substantial middle class—​rest. Pronounced inequality reduces opportunities for low-​income people to access education and other basic services that would otherwise increase productivity. Likewise, people without wealth find it hard to secure credit on favourable terms to finance entrepreneurial activities. Although debates on the link between inequality and growth persist, recent studies come down in favour of a moderate negative relationship over the long term (see Quiggin, 2010; Berg and Ostry, 2011a, 2011b; Stiglitz, 2012). Local perceptions of increasing inequality may also stoke political conflict, especially when intersecting with ethnic divisions. Although some 700 million people globally have been lifted out of extreme poverty over the past twenty years, mostly in Asia (Nayyar, 2013, pp. 161–​9), this significant achievement would have been even more impressive but for the rise in inequality within most Asian countries. Slowing economic growth since 2008 is likely to have hit the poor and lowest income households hardest. The drivers of increasing inequality are several and interrelated.4 Globalization through trade and financial integration has placed a premium on skilled labour, causing wages between skilled and unskilled workers to diverge across Asia. Technological change has benefitted capital over labour, resulting in a declining labour share of national income (a general phenomenon in the advanced economies as well) (Piketty, 2013). Existing political elites, especially in key urban centres, are well placed to influence directly and indirectly economic opportunities and outcomes. The location of industrial activities and their associated infrastructural supports also helps to account for the widening average income gaps between urban and rural incomes, particularly in China (Lin, 2005). Increasing inequality in China poses particular problems for the Chinese government, one of the very few regimes overtly committed to creating ‘a harmonious society’. Substantial inequalities within and between China’s regions threaten harmony and the government’s legitimacy. Galbraith (2012) presents detailed analysis showing inequality increasing through the first decade of the new century within Chinese provinces, but from 2001 onward it declines between provinces, although from a high base. At the national level most of the rise in inequality is accounted for by developments in one province (Guangdong) and two

32   Berry and Engels cities, Beijing and Shanghai (Galbraith, 2012, p. 236). Galbraith argues that high incomes in the banking, finance, and information technology sectors concentrated in those locations (especially the capital, Beijing) were reinforced by the large profits derived from China’s manufacturing export boom that fed into rising real estate speculation and construction that, in turn, intensified wealth inequalities and underpinned the emergence of the super-​ elite, the top 1 per cent. Tellingly, the export and property booms coincide closely from 2003 onward.5

Financial Instability The financial crisis that hit Asian economies in the late 1990s was a forcible reminder of the darker side of globalization. Many of the N–​(8 + 3) nations suffered sharp recessions, with China and Malaysia partly protected by government-​imposed capital and currency controls. This and the larger global crisis a decade later underscored the increasing vulnerability of Asia to membership of an increasingly integrated world economy, rendered more extreme in countries with non-​transparent and poorly developed financial regulatory systems. It is not clear that these economies will be able to withstand future financial shocks, from wherever they emanate. The Chinese economy is particularly vulnerable to future financial breakdown, given the opaque, uneven, and highly government-​constrained nature of its banking system. Like Japan in the 1980s and 1990s, the Chinese banks appear to have been used by government to keep lending to both private and state-​owned enterprises, regardless of credit worthiness and repayment capacity and performance. The central government has been loath to allow SOEs, in particular, to go bankrupt; as the earlier example of Japan demonstrates, maintaining insolvent banks on government life support simply puts off the day of reckoning and prolongs the crisis. Alongside the official system, unregulated private lenders have sprung up to service smaller entrepreneurs unable to access formal sources of finance. This informal financial sector is ripe for contagion and corruption.

Corruption Government corruption is widespread throughout the Asian region, although significant differences as to scale, scope, and impact exist. The annual reports of Transparency International (2005, 2013) provide perceived corruption rankings for the N–​(8 + 3) countries. These countries can be roughly divided into three groups: relatively low corruption, moderate corruption, and high corruption. In 2005, Japan, Malaysia, and South Korea fell into the low category; Thailand, China, and India into the moderate category; and the remaining six countries in the high-​corruption group, with Bangladesh sharing the title of the world’s most corrupt country. These relative positions were relatively stable from 2005 to 2013; however, three of our eleven nations did change categories. The Philippines moved from the high to moderate category, while Thailand moved in the reverse direction, and India threatened to follow suite. Malaysia went from the low-​to moderate-​corruption group.

Global Prospects: The Asian Century?    33 High and/​or uncertain degrees of official corruption increase the costs and risks of doing business for overseas investors and trading interests, especially in countries with authoritarian regimes and where government presence in the economy is high, as in China, Vietnam, and Malaysia. This interpretation broadly fits with a separate ranking of the ease of doing business published regularly by the World Bank (2013). In general, perceived corruption correlates well with difficulty of doing business. A country perceived to be rife with corruption is also likely to be difficult to conduct business operations in. The serious nature of this problem is reflected in the ongoing anti-​corruption campaign of Chinese President Xi Jinping, which has reached from local and regional scales to the highest organs of government, party, and the military.

Looking Forward A key question in the context of this chapter is: Will the rise of China continue and, if so, what are the implications for the wider Asian region? We have already pointed out a number of challenges facing China’s leaders, notably dealing with the micro-​and macro-​environmental costs of growth, the inexorable increase in economic inequality, the fragility of the financial system, and the pervasive drag of political corruption. However, China’s major concern is likely to be managing its relations with the USA, especially following Donald Trump’s election to the US Presidency and his promise to introduce protectionist policies explicity aimed at China. On the economic front, the two countries are locked together as the world’s leading creditor and debtor nations. There appear to be clear signs of a reversal favouring US manufacturing, as wage differentials with emerging economies tighten, new technologies come online, and the low-​cost energy revolution reduces US reliance on foreign oil (Sharma, 2012, pp. 261–​70). The USA has become the world’s biggest producer of natural gas and is moving towards becoming a major energy exporter. US recovery since 2008 has seen manufacturing account for three-​quarters of its gain in its growing global export market share, concentrated in energy, automobiles, and airplanes (Sharma, 2012, p. 267). China’s increasing global share of manufacturing has arisen as a result of declining shares in Europe and other advanced nations like Australia, while the USA has maintained and may be on the verge of increasing its share. In short, caught between a regenerated US industrial sector and competition from low-​wage neighbours, regardless of any hostile trade and investment policies by the Trump administration, China may not be able to rely on manufacturing as the engine of future growth as it has in the recent past. Continuing economic growth may have to come, as argued above, from the spur of growing middle-class consumption, and a shift up the manufacturing value chain, and, in particular, to the advanced service sector. Rodrik (2013) has demonstrated empirically that ‘unconditional convergence’ is sector specific. Manufacturing lends itself to catch-​up across countries, while other sectors do not; in agricultural and service-​sector industries, the degree of convergence is ‘conditional’ on institutional factors like demographic, social and political institutions, and histories.6 The futures of the other N–​(8 + 3) members will be influenced by how the symbiotic relationship between China and the USA plays itself out in the context of country-​specific

34   Berry and Engels institutional conditions. India’s growth may be constrained by continuing low productivity in agriculture, bottlenecks in the provision of productive infrastructure, the limited spread of educational opportunities, the drag of corruption and administrative inefficiencies, and the dead weight of traditional values and cultural practices (Asian Development Bank, 2013; Nayyar, 2013, p. 180). Indonesia is facing increasing internal social conflict in the aftermath of a divisive 2014 presidential election in which the two candidates, Prabowo Subianto and Joko Widodo, were widely perceived to represent links back to the Suharto and Soekarno regimes, respectively, after a decade of relatively stable and benign rule by President Yudhuyono. Thailand is split between supporters and opponents of the Thaksin Shinawatra dynasty, returning regularly to periods of military rule and having to deal with urban congestion and a growing refugee crisis on its border. Japan shows little evidence that it is close to breaking out of its decade-​long economic malaise. The Philippines is well placed to benefit from the geographical disarticulation of value chains in the region but is too dependent on China with whom they, like Japan, are in constant territorial tension over islands and resources in the China Sea. Such disputes also increasingly conflict with US strategic aims in the Western Pacific. Vietnam also has an uneasy relationship with China and is struggling to move up the value chain in a highly competitive world. Malaysia’s stellar economic performance late in the last century has slowed, and the ambitious plan to develop the ‘information super highway’ as an engine of growth is stuttering. Only South Korea appears to be well positioned in East Asia to take on the competitive might of China, as it trades on world-​class scientific and technical infrastructures, stable political conditions (North Korea apart), and close links with both Chinese and American investors and trading partners. However, hovering over the fate of the global capitalist economy, including Asia, is the lingering aftermath of the GFC. ‘The end of normal’ (Galbraith, 2015)  may see growth collapse to well below past trend rates, a step change down from postwar norms. Yanis Varoufakis (2013) has coined the metaphor of ‘the Global Minotaur’ to refer to the process by which global macroeconomic stability was precariously maintained in the period of neoliberalism following the collapse of the Bretton Woods Agreement during which US consumers drove demand for the world’s—​especially China’s—​exports and Wall Street banks financed this process by the recycling of trade surpluses. This system collapsed in the GFC and nothing has arisen to recycle the surpluses in a situation characterized by a deep global liquidity trap.7 The twenty-​first century may not belong to (all) Asia. Catch-​up may turn into fight-​back and breakout into break-​back, as the West makes a political–​economic comeback on the coat tails of a resurgent USA and a recovering Europe, unless the pessimists are right and the new normal undercuts growth and living standards worldwide, including throughout Asia. The decision of the UK to leave the European Union and the centrifugal forces threatening the future of that major regional bloc add a further dimension of uncertainty. More likely, some Asian countries will continue to chase down the smaller G7 nations, although differences in growth rates will need to remain significant for decades in order for average living standards to also sharply converge. What is certain is that these forces will play themselves out unevenly in both space and time in complex ways that are simply impossible to grasp in full at present.

Global Prospects: The Asian Century?    35

Challenges for a New Economic Geography The discussion in this chapter suggests that whatever the precise trajectory of change in the global economy, and the position of Asia, a key feature will continue to be geographical unevenness, conditioned by the different systems of ‘variegated capitalism’ (Lane and Wood, 2012; Dixon, 2014). In this approach, institutions and political agency matter. In particular, the state (in its various forms) has and will continue to have a major impact on the nature and pace of economic development across and within dynamic regions like Asia. ‘Old’ theories of uneven development, including modern development theory, the new international division of labour, dependency theory, and the various theories of imperialism advanced to explain the rise of the Asian tigers and the fate of lagging countries, will no longer do. Standard economic trade and development analyses also fail. None of these established approaches foresaw the seismic changes discussed in this chapter. Back in the 1980s and 1990s, Western governments and elites were busy congratulating themselves on ‘the end of history’ and the triumph of Anglo-​American capitalism. Europeans were fixated on the prospect of a 300-million-​strong integrated domestic market, ignoring the potential of a market three times that size in China and double again throughout Asia. The West and many of its scholars were blindsided by the resulting developments, as they were by the scale, reach, and aftermath of the global financial crisis of capitalism. Although a few prophets suggested that China’s immense potential could eventually change the face of global capitalism, this was assumed to be a distant prospect. It has been the pace, as well as nature, of the changes discussed that have left attempts to explain them behind. A plausible analysis of how the Asian region has developed in the recent past and might develop in future offers geographers a rich field for understanding the new contexts in which the discipline’s traditional focus on uneven development is relevant to the twenty-​first century. An adequate theory of uneven development will need to explain how the current crisis and stagnation-​prone phase of globalization and the financialization of capitalism—​and the associated failed political project of neoliberalism—​has altered the spatial flows of commodities, capital, and labour across and within nation states, both within the dynamic Asian region and beyond it, how the increasing economic linkages within Asia confront geopolitical tensions created by the growing power of China and its relations to its people, its neighbours, and the faltering reach of the USA. The various contributions to this volume offer aspects of this account, as do those contained in Peck and Yeung (2010); the recent work of Dixon (2014) has also been noted. However, no one has yet produced a satisfactory analysis of the totality of political economic and cultural forces that are driving developments across the space-​time of real-​existing capitalism in the uncertain times we now face.

Notes 1. Nayyar identifies the ‘Next-​14’ breakout nations, eight of which are in the Asian region, most overlapping our group. 2. This follows the well-​known theory proposed by Simon Kuznets (1966).

36   Berry and Engels 3. Rani (2007) also points to the growing trade and financial integration of East Asia following the regional financial crisis of 1997. Increasing financial integration globally since then, tying Asia into world financial flows, has substantially increased the systemic future risks of financial contagion and crisis overall (Dixon, 2014), including within seemingly ‘variegated’ regimes and regions of capitalism. 4. The Asian Development Bank (2012, pp. 62–​73) provides a summary of the main arguments and evidence around what is driving increasing inequality in contemporary Asia. See also Kanbur et al. (2014). 5. Galbraith stresses that correlation is not causation and independent evidence of the latter is not available. 6. See Dixon (2014) for an analysis of the patterned integration of financial systems in the current phase of globalization. 7. However, attempts are being made by the two competing hegemonic powers to reconfigure global economic arrangements to support their respective national interests. China has launched the Asian Infrastructure Investment Bank as a foil to the US-​dominated World Bank and Japanese-​led Asian Development Bank. The USA attempted to push through the Trans Pacific Partnership (TPP) by more closely integrating the economies of twelve Asian and Pacific nations on terms favourable to its leading industries and major corporations. It remains to be seen, firstly, if these new institutional architectures can be withstand the opposition of centrifugal domestic political forces within the various nations, including the USA under a Trump presidency, and, secondly, if together or separately, they can resolve the entrenched problems arising from the structural asymmetries of world trade and investment, and the increasing vulnerability of global capitalism to financial crises and contagion.

References Amsden, A. (1989). Asia’s Next Giant: South Korea and Late Industrialization (New York: Oxford University Press). Amsden, A. (2001). The Rise of the Rest:  Challenges to the West from Late Industrializing Economies (New York: Oxford University Press). Asian Development Bank. (2012). Asian Development Outlook 2012:  Confronting Rising Inequality in Asia (Manila: Asian Development Bank). Asian Development Bank. (2013). Key Indicators for Asia and the Pacific, Part I (Manila: Asian Development Bank). Berg, A. and Ostry, J. (2011a). ‘Equality and efficiency: Is there a tradeoff between the two or do the two go hand in hand?’ Finance and Development 48: 12−15. Berg, A. and Ostry, J. (2011b). ‘Inequality and unsustainable growth: two sides of the same coin?’ IMF Staff Discussion Note, SDN 11/​08 (Washington, DC: International Monetary Fund). Berry, M. (1989). ‘Industrialization, Deindustrialization and Uneven Development: The Case of the Pacific Rim’ in M. Gottdiener and N. Komninos (eds) Capitalist Development and Crisis Theory: Accumulation, Regulation and Spatial Restructuring, pp. 174–​216 (London: Macmillan). Cavoli, T. (2012). ‘Exploring dimensions of regional economic integration in East Asia: More than the sum of its parts?’ Journal of Asian Economics 23: 643–​653.

Global Prospects: The Asian Century?    37 Chang, H.-​J. (2007). Institutional Change and Economic Development (London:  Anthem Press). Ding, S. and Knight, J. (2008). ‘Why has China grown so fast? The role of structural change’ Economic Series Working Paper 415 (Oxford: University of Oxford Department of Economics). Dixon, A. (2014). The New Geography of Capitalism: Firms, Finance and Society (Oxford: Oxford University Press). ESCAP (2010). The State of Asian Cities 2010/​11 (Fukuoka: UN-​Habitat). Galbraith, J. (2012). Inequality and Instability: A Study of the World Economy Just Before the Great Crisis (New York: Oxford University Press). Galbraith, J. (2015). The End of Normal:  The Great Crisis and the Future of Growth (New York: Simon and Schuster). Kanbur, R., Rhee, C., and Zhuang, J. (2014). Inequality in Asia and the Pacific: Trends, Drivers and Policy Implications (Manila: Asian Development Bank). Kuznets, S. (1966). Modern Economic Growth: Rate, Structure and Spread (New Haven, CT: Yale University Press). Lane, C. and Wood, G. (2012). Capitalist Diversity and Diversity Within Capitalism (London and New York: Routledge). Lin, S. (2005). ‘International trade, location and wage inequality in China’ in R. Kanbur and A.J. Venables (eds) Spatial Inequality and Development, pp. 260–​291 (Oxford: Oxford University Press). Maddison, A. (2008). ‘The West and the Rest in the world economy, 1000–​2030’. World Economics 9: 75–​99. Maddison, A. (2009). ‘World population, GDP and per capital GDP, 1-​2006 AD’. www.ggdc. net/​maddison/​Historical_​Statistics/​horizontal-​file_​03-​2009.xls (last accessed 3 May 2017). Nayyar, D. (1997). ‘Themes in trade and industrialization’ in D. Nayyar (ed.) Trade and Industrialization (Delhi: Oxford University Press). Nayyar, D. (2013). Catch Up:  Developing Countries in the World Economy (Oxford:  Oxford University Press). Peck, J. and Yeung, H. (2010). Remaking the Global Economy:  Economic–​ Geographical Perspectives (London: SAGE Publications). Piketty, T. (2013). Capital in the Twenty-​first Century (Cambridge, MA: MIT Press). Quiggin, J. (2010). Zombie Economics:  How Dead Ideas Still Walk Amongst Us (Princeton, NJ: Princeton University Press). Rani, P. (2007). ‘Economic integration and synchronization of business cycles in East Asia’. Journal of Asian Economics 18: 711–​725. Rodrik, D. (1992). ‘Closing the productivity gap: does trade liberalization really help?’ in G. K. Helleiner (ed.) Trade Policy, Liberalization and Development (Oxford: Clarendon Press). Rodrik, D. (2013). ‘Unconditional convergence in manufacturing’ Research Paper, School of Social Science, Institute of Advanced Studies (Princeton, NJ: Princeton University). Sharma, R. (2012). Breakout Nations: In Pursuit of the Next Economic Miracles (London: Penguin Books). Stiglitz, J. (1989). ‘On the economic role of the state’ in A. Heertje (ed.) The Economic Role of the State (Oxford: Blackwell). Stiglitz, J. (2012). The Price of Inequality: How Today’s Divided Society Endangers Our Future (New York: W.W. Norton & Co.). Transparency International (2005, 2013). ‘Corruption Perceptions Index’. www.transparency. org/​research/​cpi/​overview (last accessed 3 May 2017).

38   Berry and Engels UNEP (2012). ‘Global Environment Outlook 5: Summary for Asia and the Pacific Region on the eve of Rio+20’. United Nations Environment Program https://​www.nies.go.jp/​whatsnew/​ 2012/​20120607/​GEO-​AsiaPacific.pdf (last accessed 3 May 2017). United Nations. (n.d.). ‘Environmental Issues in Asia-​Pacific Region, United Nations, New York’ http://​unpan1.un.org/​intradoc/​groups/​public/​documents/​apcity/​unpan010365.pdf (last accessed 3 May 2017). Varoufakis, Y. (2013). The Global Minotaur: America, the True Origins of the financial Crisis and The Future of The World Economy (2nd edition) (London: Zed Books). Wade, R. (1990). Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization (Princeton, NJ: Princeton University Press). Wilson, D. and Stupnytska, A. (2007). ‘The N-​11: More than an acronym’ Global Economics Paper No. 153 (New York: Goldman Sachs). Woo, M.Jung-​en. (2007). ‘Neoliberalism and institutional reform in East Asia: A comparative study’ Discussion Paper, Food and Agriculture Organization of the United Nations (Geneva: United Nations). World Bank (n.d.). ‘World Bank Open Data’ http://​data.worldbank.org (last accessed 3 May 2017). World Bank (2013). Ease of Doing Business Index http://​data.worldbank.org/​indicator/​ IC.BUS.EASE.XQ (last accessed 3 May 2017). Yusuf, S. and Nabeshima, K. (2010). Changing the Industrial Geography in Asia: The Impact of China and India (Washington DC: The World Bank).

Chapter 2

In equalit y in A dva nc e d Ec onom i e s Danny Dorling ‘The 1 Per Cent and the Rest’: Not Just a Slogan A growing body of evidence points to high and rising income inequality as one of the most important global issues in the decade spanning 2010–​19. One recently published index alludes to the far-​reaching implications of growing economic inequality, describing it as by far the greatest threat to social cohesion in many nations (Stiles et al., 2015). Paradoxically, academic geographers tend to have become less concerned with such concrete measures of ‘everyday life’ in recent years, while economists have measured these trends but have tended to neglect their wider social implications. This chapter makes the claim that the discipline of economic geography must engage more in this area, not only to remain relevant, but also because studies of geographical variations in inequality offer tremendous insight into the wide range and significant negative externalities generated by tolerating high-income inequality. Firstly, it is argued that income inequality is best assessed with respect to a particular demographic. Specifically, the best-​off 1 per cent of households provides a crucial lens through which to evaluate inequality. At root, this is because the extent to which the top 1 per cent of earners disproportionally accumulate income appears to drive overall levels of inequality within nations. It is also important to recognize that one of the favoured measures of inequality, the Gini coefficient, is very insensitive to large increases in the take of the top 1 per cent and so is a poor measure of inequality when the size of that take has far wider implications. There are now substantial differences in income inequality between affluent nations. This was not the case in the 1970s, when most affluent nations were more equitable than they are today, or in the 1920s, when all were substantially more inequitable with very similar inequality profiles to each other. This chapter contributes a clear methodology for estimating the incomes of earners at the very top of income distributions. These include the ‘super rich’ who are not captured by many surveys and hence often render calculations of income

40   Dorling inequality as underestimates if they are left unaccounted for. Moreover, a clear distinction between mean and median ratio measures of equivalized household income is shown to be crucial for assessing the true extent of income inequality. Such technical points highlight how academic study can benefit policy debates over inequality. Measuring inequality is far from simple. Accounting for its effect is even more difficult. An illustration of current patterns is paired with a discussion of the potential implications of high and rising inequality. A growing literature points to a significant, negative relationship between income inequality and a host of social indicators. Indeed, there is now an accumulation of evidence for economic inequality’s widespread occupational, demographic, and spatial impacts. In this sense, the discipline of geography as a whole must recognize the nature of high rates of economic inequality as a negative externality: the proportion of total income accruing to the richest few affects the utility of all citizens directly, in a way left unpriced by the market mechanism. In other words, there are non-​negligible social costs to tolerating high levels of inequality. This finding lends currency to the argument that economic geographers must take account of divergent trends in income inequality in affluent nations if they are to do work that has far-reaching implications and relevance. Crucially, the American and British income distributions are found to be distinctive. In particular, in recent years income inequality has risen significantly in both the UK, and a little earlier (and higher) in the USA. Given the seemingly wide-​ranging impacts of inequality, from the environment to the property market, these trends have important ramifications. In short, these two nations should not be treated by economic geographers as models from which to base general analysis of affluent countries. The UK and USA are economic outliers and not representative of rich nations in general. Very recent trends suggest that inequalities in the UK are now rising even faster than they have been in the USA and if this were to continue then the UK will soon become the affluent nation with the 1 per cent who take the most of all.

The Inequality Paradox Growing income inequality is one of the major macroeconomic events of our time (Cowen, 2012). In 2014, the bank Credit Suisse reported that the richest 1 per cent of people in the world had increased their share of global wealth from 41 per cent to 48 per cent in a mere twelve months. Among that 1 per cent, the wealth of a minute number of multibillionaires is rising so rapidly that new inequality statistics swiftly become out-​dated (Moreno, 2014). The Nobel Prize-​winning economist Robert Shiller has suggested that the renewed greed of the top 1 per cent within the USA has wrecked more destruction on livelihoods of normal households than the 2008 global financial crisis achieved. He stated, ‘the most important problem we are facing now … is rising inequality’ (Wilkins, 2013). Given the intensity of current concerns over the global situation, this chapter seeks to revitalize thinking on income inequality among economic geographers. World-​ leading economists now accept that levels of inequality in the most unequal of affluent countries are unsustainable (Atkinson, 2015) and a large number of these economists can now be listed who are beginning to suggest many remedies (listed in the introduction to Dorling, 2016).

Inequality in Advanced Economies    41 Geographers have occasionally turned their attention towards inequality and poverty, especially since the 1970s (Harvey, 1973; Philo, 1995; Ross et  al., 2005; DeVerteuil, 2009; Wyly, 2009; Li and Wei, 2010; Hay, 2013; Hennig and Dorling, 2014); however, this remains a neglected subject in the field of economic geography. As economic inequalities have widened in recent years, many geographers have specialized in alternative areas: notably, more theoretical considerations of more ephemeral aspects of life. This might have partly come about because David Harvey’s Marxist theories were most prominent within the discipline in the 1970s, a time when inequalities were at a historic minimum within both the UK and USA. Harvey and other radical geographers then did not mention the growing equality at that time as being unusual or an achievement. Hence, when inequalities subsequently rose, such geographers made little mention of the rises because they had not celebrated the falls. The academic study of inequality remains demarcated by the significant work of sociologists, historians, anthropologists, economists, and epidemiologists (Nowatzki, 2012; Sassen, 2014; Sayer, 2014). Thomas Piketty’s best-​selling Capital in the Twenty-​First Century transported the subject of economic inequality to the top of social science agendas in 2014 (Piketty, 2014). Concurrently, a number of the world’s most acclaimed science journals have covered the issue (for a recent review, see Pickett and Wilkinson, 2015). In 2014, world leaders, including the presidents of the USA, China, and the World Bank, ranked economic inequality among the top three most important global issues (Dorling, 2014). In 2015 the Pope made inequality the central issue of his teaching, ‘Laudato Si’: On care for our common home’, and in January 2016 Oxfam explained that the accumulation of wealth by the very richest in the world was not just rising, but accelerating (Dorling and Lee, 2016). Current cross-​national trends in income distributions necessitate a focus on the consequences of inequality. The discipline of geography has a significant contribution to make as a framework through which to understand and analyse these heterogeneous trends. It is imperative to recognize the spatial—​the geographic—​dimension of inequality. In other words, just as ‘poverty is inherently geographic’, geographical factors are crucially interwoven within economic inequalities (Glasmeier, 2002, p. 156). This chapter explores both the drivers and consequences of rising income inequality, embracing the importance of exposing economies and economics to geographical scrutiny (Smith et al., 2010).

Why the 1 Per Cent Matter The economic concept of inequality is operationalized through a host of indicators, from the (insensitive) Gini index to the (more useful) Palma ratio (see Atkinson, 1970). The slogan ‘We are the 99%’, appropriated by the academic and anarchist David Graeber, shifted public attention towards measuring incomes and wealth at the very top of the income distribution (Runciman, 2013). In this chapter, we define the 1 per cent from the cut-​off in annual income level that divides the best-​off one in one hundred people among a particular group from the rest of the population. This is done for groups of people when those people are clustered into households and when only the overall household income of each household is known. This group was holding 50 per cent of all global wealth at the point the academic paper in which some of these estimates were first made was published (Dorling, 2015a). By the time you read this, the share of the best-​off 1 per cent by wealth will be more

42   Dorling than half of all the wealth in the world. But the best-​off 1 per cent by income are usually not quite so rich. The top 1 per cent is one of the best-​defined categories in social science. For any population, given a definition of income—​this chapter utilizes surveys that include all income received by households before tax and additional deductions and additions –​the make-​up and size of the 1 per cent group can be calculated exactly.1 To be among the best-​off 1 per cent of all households in the USA in 2013 required a gross household income, before tax, of at least $394,000 a year (Currid-​Halkett, 2013). In the same year, in the UK, living in a household with one adult earning an annual income of at least £160,000 qualified that household entry into the best-​off 1 per cent club (Cribb et al., 2013).2 The top 1 per cent take about 20 per cent of all income in the USA, 15 per cent in the UK, 10 per cent in Germany, and nearer 5 per cent in Japan (and not much more than that in the Netherlands and across all of Scandinavia). Once the ‘take’ of the next best-​off 1 per cent and the 1 per cent below that are considered it becomes clear just how little is left when just a few take so much. Measuring inequality via the incomes of the top 1 per cent is also preferable to other possible measures given the paucity of the data available on wealth, and also partly because this income inequality measure touches on the realm of the very wealthy. Although the wealth of super-​rich individuals is published by several sources, the wealth of the 1 per cent most affluent as a complete category across a range of countries is not well recorded, we rely on crude estimates to make the statements given above concerning half the world’s wealth. Thus, by default, income remains social scientists’ best-​enumerated and preferred measure for cross-​national analysis. To evaluate the most affluent 1 per cent we must define them through measures of their annual income; we simply cannot achieve a reliable comparable cross-​country study with current measures of wealth. Although we can talk about the global wealth holdings of the very richest with some confidence, this is because we do not have to assign any particular wealthy individual to one particular country to do that. To make cross-​national comparisons this chapter uses several measure of income inequality. The twenty/​twenty ratio compares the top 20 per cent with the bottom 20 per cent and is used in one of the most successful social science publications of the last decade (Wilkinson and Pickett, 2010). However, I prefer to use the share of the top 1 per cent. The best-​off 1 per cent in any one society are—​by definition—​the very highest paid twentieth of the best-​ off 20 per cent. They constitute 5 per cent of that group by population and can receive each year as much as 50 per cent of the income of that group. Thus, the twenty/​twenty ratio largely reflects the share of income held by the 1 per cent. The ninety/​ten ratio focuses in on the tails of income distributions, and can also be used to further underscore the importance of the 1 per cent for assessing inequality. There is a key theoretical defence for a focus on the top 1 per cent measure. How much the best-​off 1 per cent take is the factor that emerges as the crucial driver behind nations’ overall levels of income inequality. In other words, when assessing the impacts of inequality, the share taken by the top 1 per cent seems to be the key independent variable, decisively affecting the skewness of countries’ income distributions. It is possible that in future this measure will not be so sensitive and income will be better shared out in societies. At the moment, the top 1 per cent tend to take so much more than the next 1 per cent that the take of the top 1 per cent remains the single simplest measure of income inequality that tells us the most about how different affluent countries differ from each other.

Inequality in Advanced Economies    43 The reasoning underpinning the argument that what the 1 per cent take is of such great importance is simple. A proportional relationship is apparent between the top 1 per cent’s share of national income and that of the other households near the top of the income distribution. In general, the greater the percentage of total income obtained by the top 1 per cent, the greater the percentage acquired by the 1 per cent of households just below them. The take of the 1 per cent ‘trickles down’ through imitation and expectation, bolstering incomes within the most very affluent households very near the top as well as the top 1 per cent themselves. Crucially, however, this ‘trickle-​down’ effect is necessarily strictly limited: the trickle does not trickle far, probably to only a few among the most affluent 20 per cent of households and not at all outside of that group, from whom potential income not received trickles up. The more the 1 per cent takes as a share of all income, the less there is for the remaining 80 per cent of the population. This is largely because when the 1 per cent take more, the other 19 per cent also take more. This is why low-​in-​work-​incomes and low benefit levels tend to be common in those countries where the 1 per cent takes the most. There will always be a 1 per cent best-​off group in any country or area under consideration. It is the extent to which this category is set apart from average households that matters. It is also often found that the 1 per cent take tends to be high when social mobility is low. This measure establishes the depth of the divide between the have-​mosts and have-​nots. Parents in very unequal affluent countries try very hard to ensure their children do not fall far. The data analysis set out in the next section finds support for the case for this theory of inequality: that when there are greater economic inequalities between the ‘1 per cent’ and ‘the rest’, overall inequality is more acute in many other aspects of life, including having low social mobility. When the best-​off 1 per cent take less, overall inequality almost always falls and all other indices of inequality tend to be lower.

Uncovering Recent Trends in Income Inequality in Affluent Nations Contemporary advances in data collection provide rich insights into current trends in economic inequality. Cross-​country levels of economic inequality report rates of inequality peaking in some of the large poor countries of the world (including Brazil and South Africa). Yet recently, the more affluent nations of the world have become increasingly set apart from others by stark contrasts between them in their levels of economic inequality. This chapter focuses, in particular, on this contemporary transformation in income inequality across the richest nations. Analysis of new data contributes empirical support towards the two key themes outlined above: the impact of the share of the top 1 per cent and the importance of geography’s engagement with inequality. To produce a fair set of statistics on economic inequalities, a particular form of triangulation is needed and many data sources have to be combined. Here, estimates of the incomes of the 0.01 per cent, 0.001 per cent, and so on, are compared, up to the very smallest groups of extremely rich households with details published on the numbers of highly paid financiers (the majority of the most highly rewarded in both the UK and USA) in each affluent country.

44   Dorling These data are compared with a variety of other sources for robustness, including the numerous newspaper ‘rich-​lists’, and with official survey data on median household incomes. This is especially important for studies of the USA and UK, given that within these two nations so few households now fare well. Once this triangulation is carried out, sample surveys of national populations can be used to estimate the income distribution of the top 1 per cent. Using data from the European Union Statistics on Income and Living Conditions (EU-​ SILC) survey, new estimates have recently been made for the household income distributions of all large European countries for the year 2012. The EU-​SILC replaced its predecessor in 2004, significantly expanding the number of countries sampled. By assigning each household a cross-​sectional weight, these samples have been adjusted by the author to produce national estimates that are representative of each country’s population. This method counters sample selection bias. Table 2.1 provides a summary of income inequality in the five countries in the European Union (EU) that have the largest populations: Germany, Italy, Spain, France, and the UK. Only these countries are sufficiently populous that discussion of the best off one in ten million households is relevant. Focusing on the 1 per cent, the table suggests that the UK is an outlier. The threshold to be a member of the best-​off 1 per cent is 6.3 times the median UK household income, in contrast to a comparative value of between 4.2 and 4.9 in the other four countries. Table 2.2 reaffirms the conclusion that the UK is a case apart through a detailed analysis of the UK income distribution in 2012. The data cover these households’ gross annual income, including investment income. This table elucidates a number of significant points, crucial for students of economic inequality to understand. While Table 2.1 shows money recorded in euros, for Table 2.2 the sums have been converted back into pound sterling. The distinction between mean and median has fundamental implications in inequality analyses. The median annual income in 2012 for UK households was £30,267. After tax (including tax credits) and once benefits and pensions are excluded, the figure is reduced to £13,481 for the actual median incomes of all UK households from post-​tax earnings alone. In contrast, the arithmetic mean annual household income was significantly higher, standing at £43,909. This mean is weighted with respect to individual households, such that the sample population distribution is representative of the UK population. The stark divergence arises because the UK’s income distribution was—​and continues to be—​so skewed in favour of the most affluent households.

Table 2.1 Summary of Income Inequalities in the Five Most Populous European Union Countries (2012) Country

Sample (n) As share of all households (%)

1% cut-​off (€)

Median(€)

Ratio of 1% to median

Germany

13,512

0.03

154,000

36,400

4.2

France

11,360

0.04

189,000

39,000

4.8

Italy

19,399

0.08

164,000

33,400

4.9

Spain

13,109

0.08

105,000

22,700

4.6

UK

8058

0.03

227,000

36,300

6.3

Inequality in Advanced Economies    45 Table 2.2 Summary of UK Household Income Distribution 2012 na

Income (£s)

8058

Gross income

Post-​tax

Post-​benefit

Post-​pension

Variable

HY010

HY020

HY022

HY023

Median

30,267

25,458

22,469

13,481

43,909

33,121

29,809

23,808

104,386,279

56,138,025

56,138,025

56,138,025

1/​million—​  …

32,988,872

17,741,126

17,741,126

17,741,126

1/​100,000—​  …

10,425,371

5,606,673

5,606,673

5,606,673

1/​10,000— ​…

3,294,698

1,771,859

1,771,859

1,771,859

Weighted mean b

1/​10million

8

1/​1000— ​…

1,041,213

559,955

557,909

554,839

63

1/​100—​0.1%

305,395

180,487

179,204

168,586

678

The top 9%

116,942

80,326

78,828

74,533

7309

Bottom 90%

32,863

26,332

22,817

16,688

782

Bottom 10%

7671

6835

4358

194

8058

R Geo-​meanc

3.2

2.8

2.9

3.2

d

EOnePercent

420,648

240,639

239,290

229,553

E1%:median

14

9

11

17

136

82

128

2857

0.1%:10%

a‘n’ refers to number of households in the sample. b1/​10 million is the best off one in 10 million

households of which there are just three in the country. c‘R Geo-​mean’ refers to the geometric mean of the inequality ratios calculated between different categories in the income distribution (i.e. 3.2 is the mean of 3.4, 2.6, and 3.6: respectively the ratios of the 0.1%’s mean income to the 1%’s, the 9%’s to the 1%, and the 90%’s to the 10%’s). d‘EOnePercent’ refers to the mean annual income of the top 1%, including estimates for the super rich. Source: Calculations by author using European Union Statistics on Income and Living Conditions (EU-​SILC)-​weighted household sample.

The distinction between the mean and median may appear painstakingly obvious, yet surprisingly often policymakers conflate mean and median income statistics, producing invalid conclusions on inequality. For instance, a recent report by the UK Department for Work and Pensions claimed that inequality among UK households had fallen (Shale et al., 2015). Yet, the report measured the ratio of the median income of the top quintile to the bottom quintile. This is therefore the ninety–​ten percentile ratio. Crucially, given the definition of median, any statistic that compares medians excludes the top 1 per cent, in contrast to the ninety/​ten ratio of decile means. Given the aforementioned importance of the top 1 per cent (and those percentiles just beneath them, which are also excluded when these medians are compared), it is of necessity to use the latter measure in inequality analysis, which includes the extremes of the income distribution. So skewed is the UK income distribution that even if the top 10 per cent are excluded, the weighted mean in the UK is still higher than the sample median.

46   Dorling It is valuable to recognize that, in general, some of the greatest economic inequalities reside within the top 1 per cent rather than within the remaining 99 per cent of the population. The higher the share of national income share that the best-​off 1 per cent take, the greater income inequalities within the 1 per cent tend to be. In the UK, those households in the 0.001–​0.1 per cent bracket earn or otherwise receieve, on average, £1 million a year—​ £1,041,000 in mean annual income. The remainder of the 1 per cent have a weighted mean annual income of £305,000. This is 3.4 times less than the 0.1 per cent receives on average. The rest of the top 10 per cent have a weighted mean annual income of £117,000. This is 2.6 times less than that of the top 1 per cent. The remaining 90 per cent have a weighted average income of £32,863 (3.6 times less than the top 10%). In the UK the mean annual income of the 1 per cent is £370,000 a year, or over 2.3 times as much money as the threshold required to belong to that group. This mean for the income of the top 1 per cent in the USA will be of an even greater magnitude. In more equitable countries, such as the Netherlands and Sweden, the 1 per cent both take far less and have less inequality within their group (Dorling, 2014). In any society in which the best-​off 1 per cent take a low proportion of national income, overall income inequality within that society always turns out to be less. This is not only because the 99 per cent will have more, but also because there will almost certainly be less inequality within that 99 per cent, and within the 1 per cent when the two are not so far apart. We still need to pose the question of how to estimate the most extreme incomes, the incomes of the ‘super rich’. Because extremely well-​off individuals are unlikely to be included in the EU-​SILC sample owing to their rarity, the extent of inequality will be underrepresented without estimations such as that just presented. In short, the degree of inequality in the sample itself must be extrapolated in order to estimate the top incomes of those in the 0.01 per cent and above. The assumption made here is that the ratios of inequality remain approximately constant as one proceeds up the income scale. Once we are talking about incomes at the very top we are dealing with very small numbers of families and so need to triangulate between sources. The geometric mean of the three ratios calculated earlier (3.4, 2.6, and 3.6) is 3.2. Applying this ratio upwards, the average income of the best-​off 0.01 per cent (less the 0.001%) is £3.29 million a year. Employing the same method, the average income of the best-​off 0.001 per cent (less the 0.0001%) is estimated to be £10.43 million a year. The most affluent 0.0001 per cent (less the 0.00001%) secures £32.99 million a year. Finally, the best of 0.00001 per cent of households, equating to three families, accumulates an average annual income of £104.39 million a year. These calculations are far from fanciful in light of the fact that approximately three of the UK’s richest families saw The Sunday Times’ estimate of their wealth rise by at least a tenth of a billion pounds (i.e. by at least £100,000,000) between publications of two recent versions of the super-​rich list, implying income gains of at least those amounts within a year (The Sunday Times, 2014). For example, between 2013 and 2014 in the UK the wealth of the Mittal family rose by £250 million. When The Sunday Times’ rich list of 2015 was published in spring of that year it confirmed that such increases in wealth at the top were continuing, although often for a different set of very wealthy families. There is great precarity at the top of the income scale just as there is at the bottom in the most unequal of affluent countries. All the figures required to estimate income inequality in the UK and used in the paragraph above are also in Table 2.2. To draw up this table it is assumed that the very rich pay income tax at the same rate as the best-​off 0.1 per cent (who pay 46% of their income in income tax).

Inequality in Advanced Economies    47 It is also presumed that the very rich receive negligible benefits and pensions. However, it is worth noting that even the richest 0.1 per cent in the UK in 2012 received just over £2000 a year in benefits and about £3000 a year, on average, in pension payments. Although roughly half of these benefits came from child benefit payments, which have been phased out for high earners from 2012 and are no longer paid to them by government. The data on the UK household income distribution in Table 2.2 can be used to calculate the mean annual income of the top 1 per cent in the UK (labelled ‘EOnePercent’), which includes the super rich, using the methodology just described. The result in 2012 was that when the very richest were also included, the best-​off 1 per cent by income received £420,648 (or €504,778) a year, which is fourteen times the median UK income. Put another way, in simple terms, the top 1 per cent in the UK enjoy fifty-​five times more income than the average income of households within the bottom 10 per cent. The top 1 per cent thus receive 550 times as much money a year as the bottom 10 per cent. This is extreme among affluent countries. It is not normal. Table 2.2 illustrates that although these inequalities are muted after taxes are paid, they are aggravated once benefits and pensions are subtracted (benefits reduce income inequality, but only slightly). Without benefits or pensions, and after having paid tax, the bottom 10 per cent of the population would have, on average, 2857 times less a year to live off than the best off 0.1 per cent. This equates to living off £194 per annum, or £0.53 per day. Without benefits, people in the UK would starve. Exactly the same methods were employed by the author to generate the corresponding estimates and calculations for Germany, France, Italy, and Spain. Table 2.3 below displays the results. In terms of household income distribution, the four other countries in Table 2.3 exhibit far more similarities with one another than they do with the UK. On average, the best-​ off 1 per cent in the UK receive a gross annual income almost twice that of Germany and almost exactly four times as great as that enjoyed by their Spanish counterparts. In contrast, the median income of UK households is lower than that in both Germany and France. This, again, underscores the important distinction between mean and median income. France is the second most inequitable country studied here, yet it can still boast far great equality than the UK. The top 1 per cent in France receives ten times the French medium income, or thirty-​five times the average incomes of the worse-​off 10 per cent. As aforementioned, the corresponding UK figures are fourteen and fifty-​five. In short, the UK is the outlier within Europe. It is resoundingly the most economically unequal country within Europe (see independent research from Fernández-​Macías and VacasSoriano, 2015; Sellers, 2015). Additional data sources allow for a wider perspective on global inequalities. For this chapter a comparison of the top 1 per cent’s income share in twenty-​five countries across the world has also been constructed. The twenty-​five countries compared in Table 2.4 below were selected on the basis that they had the highest gross domestic product per capita in the world in 2009 out of all nation states with a total population exceeding two million. Hence, the table ranks the twenty-​five richest countries in the world in 2009 excluding Luxembourg, Iceland, and San Marino (populations of less than two million). According to the International Monetary Fund (IMF), as of April 2015, this selection of twenty-​five countries would now include Kuwait, the United Arab Emirates, Hong Kong, and Qatar. These four countries would replace Slovenia, Greece, Portugal, and Spain in the club of rich

48   Dorling Table 2.3 Summary of Income Inequalities in Germany, France, Italy, Spain, and the UK in 2012a Germany

France

Italy

Spain

UK

1%

257,067

375,063

270,099

126,285

504,778

9%

103,060

110,521

101,336

73,826

140,331

90%

33,466

35,776

31,362

23,558

39,436

Bottom 10%

7656

10,695

6280

3916

9206

Median

36,425

38,999

33,389

22,700

36,321

Mean

41,785

45,514

39,862

29,089

52,691

7

10

8

6

14

34

35

43

32

55

1%/​median 1%/​bottom 10%

aFigures in Euros (€); the final two rows show ratios. This income estimate includes the super

rich living in each country. Source: Calculations by author using European Union Statistics on Income and Living Conditions (EU-​SILC)-​weighted household sample.

countries. However, owing to data availability this chapter continues to compare the group of twenty-​five countries selected using 2009 data. To support the argument that the top 1 per cent drives additional measures of inequality, comparison is made with the ninety–​ten inequality measure: the ratio of the mean income share of the richest 10 per cent in a country to that of the poorest 10 per cent. For example, as Table 2.5 highlights, in 2010 the top 10 per cent in the USA received 20.32 times more than the bottom 10 per cent. In Denmark the poorest tenth are almost four times better off in comparison (ratio 5.16). One important finding is that there is a very close correlation between the income take of the best-​off 1 per cent and the overall measures of income dispersion in each country, such as the Gini coefficient of income inequality, as measured by the Luxembourg Income Study. This is apparent even although the Gini coefficient is a less sensitive measure of income inequality. These relationships are summarized in Table 2.6, which can be compared with Tables 2.4 and 2.5. So, there is wide variation between affluent countries, but how are the ratios changing over time? In addition to these estimates, The New York Times’ Income Distribution Database provides time series data, allowing projections to be generated for income inequality in Canada, Finland, Germany, Greece, Ireland, Israel, Italy, the Netherlands, Spain, the UK, and the USA. Owing to data limitations it was not possible to use the database to calculate predictions for the remaining fourteen countries used in this chapter. Projections (presented in Table 2.7) were estimated by taking simple linear averages between data points from 2004–​05 and 2010. If the same annual trends in income inequality, as measured by the ninety/​ten ratio, continued then the UK would become the most unequal rich country in the world in the year 2026. In other words, should the average trend between 2004 and 2010 be replicated across the eleven countries, the UK would overtake both the USA and Israel to a point where the richest 10 per cent receive, on average, 22.4 times more than the poorest 10 per cent per year by that date. As of the UK budget of July 2015 that continuation of current

Inequality in Advanced Economies    49 Table 2.4  Share of Income Received by Best-​off 1 Per Cent of Taxpayers in Rich Countriesa Country

Year

Top 1% income share (% of all national income)

USA

2013

17.54

Germany

2008

13.89

Singapore

2012

13.57

UK

2012

12.70

South Korea

2012

12.23

Canada

2010

12.22

Switzerland

2010

10.63

Ireland

2009

10.50

Portugal

2005

9.77

Japan

2010

9.51

Italy

2009

9.38

Australia

2010

9.17

France

2012

8.94

Spain

2012

8.20

New Zealand

2011

8.13

Norway

2011

7.80

Finland

2009

7.46

Sweden

2013

7.24

Denmark

2010

6.41

Netherlands

2012

6.33

aAll country data from Alvaredo et al. (2015). Data unavailable for

Israel, Greece, Austria, Belgium, and Slovenia. Other sources give much lower figures for both Germany and Japan. Source: Data from the New York Times’ (NYT) Income Distribution Database, assembled by Janet Gornick (LIS),Thierry Kruten (LIS), Branko Milanovic (LIS), David Leonhardt (NYT), and Kevin Quealy (NYT), www.lisdatacenter.org/​resources/​other-​databases/​—​linear projection of most recent trends forward in time to 2026 (see New York Times, 2014).

trends appeared possible; but few commentators at the time reported this. By February 2016 it was looking even more likely that the UK could become the most inequitable of all affluent nations as the presidential debate in the USA began to focus on income inequalities with in the USA and as the Singaporean government continued to implement policies to prevent wealthy individuals from overseas investing in the Singapore’s property market through the implementation of capital controls. Trends in Israel remain very uncertain. Unless inequities in Israel grow much higher in the coming years, for instance owing to wealthy investors moving to live in Israel, then it is increasingly likely that the UK will become a more

50   Dorling Table 2.5  The Ratio of Incomes of the Best-​off 10 Per Cent of Households Versus the Worst-​off 10 Per Cent Among the World’s Richest Countries Country

Year

90:10

USA

2010

20.32

Singapore

2012

18.48

Israel

2010

17.42

UK

2010

17.35

Canada

2010

14.54

Spain

2010

13.62

Greece

2010

12.71

Italy

2010

11.26

Ireland

2010

11.08

Germany

2010

10.35

Portugal

2012

10.10

South Korea

2013

10.08

Finland

2010

9.20

Australia

2012

8.80

Netherlands

2010

8.60

New Zealand

2012

8.21

France

2012

7.40

Japan

2009

7.28

Austria

2012

7.00

Switzerland

2012

6.70

Sweden

2012

6.30

Norway

2012

6.20

Belgium

2012

5.85

Slovenia

2012

5.45

Denmark

2012

5.16

Sources: 2010 data from the New York Times’ (NYT) Income Distribution Database, assembled by Janet Gornick (LIS),Thierry Kruten (LIS), Branko Milanovic (LIS), David Leonhardt (NYT), and Kevin Quealy (NYT), www.lisdatacenter.org/​resources/​other-​ databases/​(see New York Times, 2014). (The 90:10 ratio calculated is the fraction of mean equalized household income (2005 purchasing power parity $) of the top to the bottom decile, where the equivalence scale is income divided by the square root of the number of household members). 2012 and Korean data from the OECD Income Distribution Database (see OECD, 2013). Singapore data from UNU-​WIDER (2014). Japanese data on equivalized household disposable income shares from Ballas, D. et al. (2013). Note: the OECD database obtained a value of 10.7 for Japan and the UN-​WIDER database obtained a value of 7.9. Here the Ballas et al. (2013) estimate is used.

Inequality in Advanced Economies    51 Table 2.6  S hare in Top Incomes of the 1 Per Cent and Gini Measure of Inequality, Fifteen Affluent Countries Ranked by the Take of the 1 Per Centa Country

Top 1% income share (in earlier dates)

Gini inequality measure

USA

17.7

0.37

UK

15.5

0.34

Canada

13.8

0.32

Germany

10.9

0.29

Ireland

10.3

0.31

Italy

9.4

0.34

France

8.9

0.28

Spain

8.6

0.32

Australia

8.6

0.31

Norway

7.9

0.26

Finland

7.9

0.26

Switzerland

7.8

0.27

Sweden

6.7

0.24

Netherlands

5.4

0.27

Denmark

4.3

0.23

aNote that the data here for the 1% are drawn

from a year or so earlier than the data shown in Table 2.4. The rich can suddenly decide to declare more of their income. Source: Dorling (2013, Figure 1.1, and online tables).

unequal country than Israel by 2024, and the USA by 2026. However, political events are very unpredictable. All these statistics on rising economic inequality can become overwhelming. By 2014, fatigue was setting in among reporters:  over four years’ worth of news since the Great Recession had repeated the same message:  while most people in affluent countries (and especially the most unequal of affluent countries) were becoming worse off, a few were gaining at an accelerating pace. The average wealth of billionaires was spiralling upwards, as was the prosperity of the large majority of the upper echelons among those recorded in national income distributions. Unsurprisingly, in 2015 it emerged that defence of extremes of inequality was one of the key battlegrounds for political parties in the UK’s general election (Dorling, 2015b). Indeed,

Year

Canada

Finland

Germany

Greece

Ireland

Israel

Italy

Netherlands

Spain

UK

USA

2008

14.5

9.3

10.3

12.8

11.7

16.84

11.6

8.6

13

13.7

20.2

2009

14.5

9.3

10.3

12.7

11.4

17.12

11.45

8.6

13.3

14.2

20.3

2010

14.5

9.2

10.4

12.7

11.1

17.4

11.3

8.6

13.6

14.7

20.4

2011

14.5

9.1

10.5

12.7

10.8

17.68

11.15

8.6

13.9

15.2

20.5

2012

14.5

9.1

10.5

12.6

10.5

17.96

11

8.6

14.2

15.7

20.6

2013

14.5

9.0

10.6

12.6

10.2

18.24

10.85

8.6

14.5

16.1

20.7

2014

14.5

8.9

10.7

12.6

9.8

18.52

10.7

8.5

14.8

16.6

20.8

2015

14.5

8.9

10.7

12.5

9.5

18.8

10.55

8.5

15.1

17.1

20.9

2016

14.5

8.8

10.8

12.5

9.2

19.08

10.4

8.5

15.4

17.6

21.0

2017

14.5

8.7

10.9

12.5

8.9

19.36

10.25

8.5

15.7

18.1

21.1

2018

14.5

8.7

10.9

12.4

8.6

19.64

10.1

8.5

16

18.6

21.2

2019

14.5

8.6

11.0

12.4

8.3

19.92

9.95

8.5

16.3

19.0

21.3

2020

14.5

8.5

11.1

12.4

7.9

20.2

9.8

8.4

16.6

19.5

21.4

2021

14.5

8.4

11.2

12.4

7.6

20.48

9.65

8.4

16.9

20.0

21.5

2022

14.5

8.4

11.2

12.3

7.3

20.76

9.5

8.4

17.2

20.5

21.6

2023

14.5

8.3

11.3

12.3

7.0

21.04

9.35

8.4

17.5

20.9

21.7

2024

14.5

8.2

11.4

12.3

6.6

21.32

9.2

8.3

17.8

21.4

21.8

2025

14.5

8.2

11.4

12.2

6.3

21.6

9.05

8.3

18.1

21.9

21.9

2026

14.5

8.1

11.5

12.2

6.0

21.88

8.9

8.3

18.4

22.4

22.0

Source: data from the New York Times’ (NYT) Income Distribution Database, assembled by Janet Gornick (LIS),Thierry Kruten (LIS), Branko Milanovic (LIS), David Leonhardt (NYT), and Kevin Quealy (NYT), www.lisdatacenter.org/​resources/​other-​databases/​ (see New York Times, 2014)—​linear projection of most recent trends forward in time to 2026.

52   Dorling

Table 2.7  Projections for Rich Countries’ Income Inequality, Percentage Take of Top 1 Per Cent

Inequality in Advanced Economies    53 the high, rising, and prolonged trends in economic inequality outlined here are not merely statistics. Inequality changes fundamental relationships in society and touches on the lives of the entire population (Dorling, 2016). This is precisely why an interdisciplinary, and specifically an economic and geographic, appreciation of inequality is so important.

Inequality as a Negative Externality Economic inequality generates significant occupational consequences. In a very unequal country the 1 per cent are mostly employed within the financial services sector. In other words, as a country becomes more economically unequal a greater proportion of the 1 per cent is drawn from just this one sector. This trend is even more pronounced within the top 0.1 per cent of earners. Notably, as Table 2.8 highlights, over ten times more bankers earn very high salaries in the UK as compared with the country ranked in second place, Germany. In fact, there are fewer highly paid bankers in every other EU country, combined, than there are in the UK. Remarkably, there are now more bankers paid over £1 million in just one bank in London (Barclays) than in all of Japan, in all sectors of the economy combined (Jones, 2014). Recently these groups have been documented in great detail (Birtchnell and Caletro, 2013). The phenomenon whereby having to host an extremely large number of very rich bankers and living in a extremely unequal society appears to go hand-​in-​hand is possibly driven by strong links between the UK and USA. Indeed, in the UK it has become more common in recent decades to hear and read of US attitudes to very high pay being that ‘high rewards’ are presented and often seen as being socially acceptable. As introduced in Tables 2.5, 2.6, and 2.7, the USA leads the rich world in terms of income inequality and is likely to do so until at least 2025. Table 2.9 depicts in detail how the distribution of incomes within the USA is even more extreme than that detailed for the UK. It shows what the top 0.1 per cent receive each year in the USA and how that compares with what the distribution of incomes in the USA was in 1970. It also shows what average mean salary and other income levels in the USA would be if a twenty-​to-​one limit was imposed at the very top of that society

Table 2.8  T he Numbers of Bankers Paid Over €1,000,000 in 2012 (Highest Numbers in the European Union) Country

Number of highly paid bankers

UK

2714

Germany

212

France

177

Italy

109

Spain

100

Source: EBA (2013).

54   Dorling Table 2.9 Income Inequality in the USA, 2008 and 1970–​2008 (at Real 2008 Rates)a Income level

Number of people

Average income ($)

Overall change Annual salary 1970–​2008 (%) in 1970 ($)

Salary with 20:1 limit ($)

Top 0.1%

152,000

5.6 million

385

1.15 million

631,000

Top 0.1–​0.5%

610,000

878,139

141

364,000

199,000

Top 0.5–​1%

762,000

443,102

90

233,000

127,000

Top 1–​5%

6.0 million

211,476

59

133,000

72,000

Top 5–​10%

7.6 million

127,184

38

92,000

50,000

137.2 million

31,244

–​1

31,560

31,000

Bottom 90%

aNote the figure of –​1% in the final row means that, on average, someone in the bottom 90% of

US society saw their ‘real’ household income fall between the years 1970 and 2008 in most cases. Sources: Whoriskey (2011), which, in turn, used ‘The World Top Incomes Database’ and reports by Jon Bakija, Williams College; Adam Cole, US Department of Treasury; Bradley T. Heim, Indiana University; Carola Frydman, MIT Sloan School of Management and National Bureau of Economic Research (NBER); Raven E. Molloy, Federal Reserve Board of Governors; Thomas Piketty, Ehess, Paris; Emmanuel Saez, University of California Berkeley and NBER. Reproduced from original source (Whoriskey, 2011).

and the distribution below that remained the same for all but the bottom 90 per cent of the American people. The demographic implications of high and rising inequality are significant. Canadian data reveal that, of those within the top 1 per cent who were receiving the bulk of their income from their earnings (rather than as interest on capital or in rent), more than 80 per cent in very recent years were men. As the income share of this 1 per cent has grown, so too has the share that is taken by men (Breau, 2014). The 1 per cent is also ageing. The group’s members now predominantly reside within the fifty to sixty-​four-​year-​old age bracket (Breau, 2014). There is no reason not to suppose a similar microdemography in the UK and USA to that found in Canada. In fact, recent research has revealed a very rapidly widening income gap between men and women in the UK as women and the jobs done more by women have suffered form the majority of public spending cuts made since 2008 (Dorling, 2016). The direct spatial outcomes of growing inequalities are startling. In general, during periods of higher income equality, the top 1 per cent is spread relatively evenly across a country. But when inequalities rise, the top 1 per cent tends to congregate in particular areas, neighbourhoods where they feel more socially acceptable and can be nearer to the few places of work where the highest wages are paid. A series of very detailed maps are presented in a recent social atlas of the UK to illustrate this growing concentration (Dorling and Thomas, 2016). A substantial—​and growing—​body of research points to a range of negative economic, social, and geographical externalities associated with income inequality. The Noble Prize-​ winning economist Paul Krugman cites growing recognition of the ‘Wilkinson-​type

Inequality in Advanced Economies    55 views about the corrosive effects of inequality’ (Wilkinson, 1996, 2005; Wilkinson and Pickett, 2010; Pickett and Wilkinson, 2015). Krugman (2012) suggests that a vast number of social ills appear to be significantly exacerbated within countries where income inequality is higher. Another world-​leading economist, Stiglitz (2012), also argues that the 1 per cent erodes ‘our sense of identity in which fair play, equality of opportunity and a sense of community are so important’. Extensive empirical evidence also indicates that income (and wealth) inequality can be detrimental for environmental sustainability, including increased biodiversity loss (Mikkelson et al., 2007; Holland et al., 2009), increased consumption, and increased waste generation per capita (see Dorling, 2010a, 2010b, 2011; Dorling et al., 2007). Finally, economic prosperity seems to be significantly related to lower rates of income equality, and the IMF recently found an inverse relationship between the income share accumulating to the top 20 per cent and economic growth (Dabla-​Norris et al., 2015). The potential for such serious relationships only reinforces the need for more refined analyses in this area, in order to inform policy change before further, irrevocable harm occurs. It is almost certainly the indirect effects of growing greed, and the justification of greed as somehow reflecting desert, that does the most destructive harm to society.

Income Distributions in the UK and US Studying income inequality data highlights a stark pattern. The USA has long been a glaring outlier: grossly unequal in terms of the divergence between the top 1 per cent and the rest. Yet the trend since the mid-​1980s in UK income inequality is bringing the UK’s income distribution more in line with the USA’s and in contrast with other affluent nations. In 2012, in the UK it was revealed that, compared with the very top 1 per cent of households, the rest of the top 20 per cent were taking home less and less. Between 2007 and 2012, the real disposable income of the top fifth of households in the UK dropped by £4,200 a year, a 6.8 per cent real-terms fall for that category (ONS, 2013). The average fall for all households was £1,200 a year. This development reduced inequality within the 99 per cent but, crucially, not between the top 1 per cent and the 99 per cent. In the period 2011–​12, the median household income in the UK fell by 2.8 per cent (when taking inflation into account). However, mean incomes fell by only 1.6 per cent because the top 1 per cent did not see a fall in their incomes. Very similar findings were reported in the USA. Indeed, in the five years after President Obama had come to office 95 per cent of all income gains made had gone to the top 1 per cent (Yousuf, 2013). The narrative of the UK as a nation now set apart from other European countries with regard to income inequality has been corroborated by independent research (see Inequality Briefing, 2015; OECD, 2015). Moreover, it appears as if inequality will be further exacerbated in the UK in the near future. The Institute of Fiscal Studies recently concluded that poorer households have done worse than those in the middle and uppermiddle parts of the income distribution during the recent period of austerity in the UK (Johnson, 2015). The Institute for Fiscal Studies showed that the overall array of measures announced by the UK Chancellor in his July 2015 budget was regressive: taking more from

56   Dorling poorer households than affluent ones. Indeed, the Resolution Foundation, an independent think tank, predicts that the cuts to working-age welfare will obstruct the recovery in living standards for many families on low incomes for very many years to come (Resolution Foundation, 2015). There is a growing body of evidence with respect to the UK lending great and growing currency to the Wilkinson-​style argument that the social, political, and economic (as well as medical) implications of these blunt and wide income and wealth inequalities are substantial. Even simply looking at economic indicators illustrates the idiosyncratic situation of the UK. For one, a distinctive tax system emerges. The richest 1 per cent and their behaviour in the UK have created circumstances in which taxes have to be raised to supplement the incomes of the poorest just so they can be housed and fed. This is because the incomes of the poorest in the UK, including those with jobs, are now so low that very large sums of housing benefit have to be paid out. Ironically, that benefit mostly goes to landlords in the UK, and a disproportionate amount to landlords whose annual incomes from rent put them in the top 1 per cent. Table 2.10 shows the proportions of household income received in the form of benefits by each of the five income groups in Germany, France, Italy, Spain, and the UK. The UK must raise relatively high rates of tax as a result, not only to pay for the benefits required by the very poorest in a low-​income-​for-​most society, but also to compensate ‘the 90 per cent’ for their relatively low wages and to ensure households have the ability to afford increasingly expensive necessities (e.g. gas and electicity bills). Indeed, the UK has lower average wages than any other large European country, and also has very high average housing costs and other essential living prices (including fuel, local transport, and food). It is important to consider whether the median North American or Briton experiences a better quality of life, relative to the median citizen of Spain, Italy, France, or Germany. As reported earlier in the chapter, the median household in Britain received an income equivalent to €36,300 in 2012 (before housing costs but after tax deductions). The corresponding figure in Germany was greater, at €36,400, and even more so in France, where it

Table 2.10 Income Received from the State by Households in 2012 Germany

France

Italy

Spain

UK

0.10%

7%

0%

0%

0%

0%

1%

3%

1%

2%

2%

0%

9%

3%

3%

2%

3%

1%

90%

9%

9%

5%

9%

11%

Bottom 10%

56%

31%

10%

34%

32%

Mean

7%

7%

4%

7%

8%

Ratio Bottom 10%:mean

7.6

4.4

2.6

4.8

4.3

Source: Calculations by author using European Union Statistics on Income and Living Conditions (EU-​SILC)-​weighted household sample.

Inequality in Advanced Economies    57 was €39,000. Housing costs are lower in both Germany and France. In the USA the median household earned €36,450. Yet, the cost of healthcare is significantly higher in the USA (Dorling, 2015b, Chapter 6, footnote 74). Once this is taken into account, as a necessity vital to assessing quality of life, it is the median family in the USA that will have the lowest standard of living among these six countries, and then the UK. Such is the contribution today of inequality in affluent nations. Furthermore, beyond income and tax economic measures, social indicators point to significant deprivations in the UK and US relative to their status as two of the world’s major advanced economies. A relatively recent review of over thirty studies concluded that children in lower-​income families exhibit lower cognitive, social–​behavioural, and health outcomes partly because they are poorer (Cooper and Stewart, 2013). In other words, income has a direct impact; the association between income and social outcomes does not just arise because low income is correlated with other household and parental characteristics. These are points that need to be stressed more than they currently are in geographical research, especially given the UK government’s recent decision to abandon the current income measure of child poverty. Comparatively, a US study concluded that family income affects the likelihood of graduating from university far more than any measured mathematical talent a child might demonstrate (Dynarski, 2015). Another study found that, again relatively recently, an American child from a low-​income family that does graduate still has less chance of being in the top 20 per cent of earners than a child from a high-​income family that failed to graduate (Urahn et al., 2012). The UK is rapidly moving towards becoming very like the USA and will overtake it in the inequality stakes by 2026 if current trends continue. UK levels of inequality are now vast by usual European standards. In light of their status as anomalies, and given the evidence for the important implications of inequality, geographers must now recognize how unusual the UK and the USA are in the global order of income inequality. This makes US and UK societies extremely unrepresentative of affluent nations in general, which tend to be much more socially cohesive countries. Because of this, it is hardly valid to use what is discovered in the UK or USA by researchers as general models by which to study affluent societies more widely. Arguably a very large number of academic studies by geographers in these two countries are studies of the extreme, where a host of social and other factors are distorted by high and rising income inequality.

Mapping Inequality: Where Do We Go From Here? Indicators of inequality require further clarification and improvement. Although it is possible to define the 1 per cent with a remarkable degree of precision, there are still several ambiguities to bear in mind. To produce meaningful figures for any particular type of household can imply using different thresholds if people are single or are in a family with children. The figures presented here apply to all households, not to individuals. But for historical time series that allow us to compare the growth of income inequalities over decades, often it is only individual taxation records that are included.

58   Dorling Furthermore, the figures presented in this chapter do not take wealth into account. This is because data on wealth, especially on the wealth of the very richest, is mostly incomplete and highly unreliable. An improved definition of the 1 per cent would, perhaps, combine wealth and income statistics such that for any high particular level of wealth held, the income required to fall within the best-off 1 per cent by income would be less than for those with low or no wealth. A further improvement on our estimation of the top 1 per cent measure would be to include wider family wealth. An affluent postgraduate student, being supported by their wider family, and studying at Oxford University, may appear not to be well off in their current household, but if their wider family wealth were known they might be included in the 1 per cent best-​off households, and another household on the margin of the definition would be relegated to the 99 per cent. Better measures of wider family income may find that wealth and income inequalities are even greater than we currently suppose. This chapter has made the case that economic geographers must turn their attention to income inequality and its implications, particularly in analyses of the UK and USA. Among the twenty-​five richest nations on Earth, the UK and USA have been found to rank within the top four with respect to income inequality. Intriguingly, both these countries are also renowned as home to the majority of academic geographers worldwide. Third place (see Table 2.4) goes to Singapore, another country celebrated for its geographers. Could it be that geography, the discipline of empire, mainly still survives and prospers where great economic inequalities remain? To what degree is rigorous thinking within the discipline skewed by the contexts in which geographers are writing and researching? Could this be as true for radical and critical geographers as for the more mainstream kind of geographical studies? Maybe radical and critical geographers are not as radical or as critical as they might have been had they not mostly been brought up in and acclimatized to extreme inequality as being normal? If you were choosing to study the world, as an economic geographer, why would you choose a vantage point on the extreme edges? Surely you would choose to study one of the most important trends of contemporary social science from the vantage point of a relatively equal, less divided society; or even from one of the more equitable affluent countries of the world where people’s political views and presumptions are often very different again? According to the tables in this chapter, more normal countries among the world’s richest, which by various measures are neither extremely equal nor unequal, include Finland, France, Germany, Ireland, Italy, the Netherlands, Portugal, South Korea, and Spain; if you want to see how unequal and odd the UK and USA are, then look at them from there. And similarly, if you want to look towards some of the most equitable countries of the world such as Norway and Japan from a normal vantage point, you don’t want to be looking from the UK or USA—​at least not if you think that your viewpoint is in any way normal.

Acknowledgements Just after completing her PPE degree Natasha Stotesbury helped with research assistance for this chapter. Her help and that of the anonymous referees; Gordon Clark and Seth Collins, in editing; and Angelika Kaiser in administering the submission is gratefully acknowledged.

Inequality in Advanced Economies    59

Notes 1. A case can be made for looking at income after housing costs are taken into account. However, in general, the majority of the housing costs of the best-​off 1 per cent are non-​ essential and as such should not be equated with the inelastic consumption of household necessities by lower-​income groups. Households are the unit of the analysis used in this chapter in order to capture the fact that children and partners of rich individuals in general benefit in terms of high material living standards. 2. These figures are income received before tax and after receipt of any benefits. The Institute for Fiscal Studies has calculated the UK statistics, but they are also verified here by cross-​ checking with other sources and through extending those analyses done by that Institute to other large countries in Europe (more details are given in Dorling, 2015a).

References Alvaredo, F., Atkinson, A., Piketty, T. and Saez, E. (2015). ‘The World Top Incomes Database’. http://​topincomes.g-​mond.parisschoolofeconomics.eu/​ (last accessed 10 May 2017). Atkinson, A. (1970). ‘On the measurement of inequality’. Journal of Economic Theory 2: 244–​263. Atkinson, A. (2015). Inequality (London: Harvard University Press). Ballas, D., Dorling, D., Nakaya, T., Tunstall, H., and Hanaoka, K. (2013). ‘Income inequalities in Japan and the UK: a comparative study of two island economies’. Journal of Social Policy and Society 13: 103–​117. Belfield, C., Cribb, J., Hood, A., and Joyce, R. (2015). ‘Living standards, poverty and inequality in the UK: 2014’. Institute for Fiscal Studies http://​www.ifs.org.uk/​publications/​7274 (last accessed 1 July 2017). Birtchnell, T. and Caletro, J. (eds) (2013). Elite Mobilities (Abingdon: Taylor and Francis). Breau, S. (2014). ‘The Occupy Movement and the top 1% in Canada’. Antipode 46: 13–​33. Cooper, K. and Stewart, K. (2013). ‘Does money affect children’s outcomes?’ Joseph Rowntree Foundation http://​www.jrf.org.uk/​publications/​does-​money-​affect-​childrens-​outcomes (last accessed 10 May 2017). Cowen, T. (2012). The Great Stagnations: How America Ate All the Low Hanging Fruit of Modern History, Got Sick, and Will (Eventually) Feel Better (New York: E P Dutton & Co Inc.). Cribb, J., Hood, A., Joyce, R., and Phillips, D. (2013). ‘Living standards, poverty and inequality in the UK: 2013’. Institute for Fiscal Studies. http://​www.ifs.org.uk/​comms/​r81.pdf (last accessed 10 May 2017). Currid-​Halkett, E. (2013). ‘The 21st century silver spoon’. The New York Times, 9 November http://​opinionator.blogs.nytimes.com/​2013/​11/​09/​the-​21st-​century-​silver-​spoon (last accessed 10 May 2017). Dabla-​Norris, E., Kochhar, K., Suphaphiphat, N., Ricka, F. and Tsounta, E. (2015). ‘Causes and consequences of income inequality:  a global perspective’. International Monetary Fund http://​www.imf.org/​external/​pubs/​cat/​longres.aspx?sk=42986.0 (last accessed 10 May 2017). DeVerteuil, G. (2009). ‘Inequality’ in R. Kitchin and N. Thrift (eds) International, Encyclopedia of Human Geography, pp. 433–​445 (Amsterdam: Elsevier). Dorling, D. (2010a). Is more equal more green? (London: University of Sheffield).

60   Dorling Dorling, D. (2010b). ‘Social inequality & environmental justice’. Environmental Scientist 19: 9–​13. Dorling, D. (2011). ‘The economics of social inequality and the natural environment’, presentation given to the Royal Geographical Society November 2011 http://​sasi.group.shef.ac.uk/​ presentations (last accessed 10 May 2017). Dorling, D. (2013). Unequal Health (Bristol: Policy Press). Dorling, D. (2014). Inequality and the 1% (London: Verso). Dorling, D. (2015a). ‘Income inequality in the UK:  comparisons with five large Western European countries and the USA’. Applied Geography 61: 24–​34. Dorling, D. (2015b). Injustice: Why Does Social Inequality Still Persist (Bristol: Policy Press). Dorling, D. (2016). A Better Politics: How Government can Make us Happier (London: London Publishing Partnership). Dorling, D., Barford, A., and Wheeler, B. (2007). ‘Health impacts of an environmental disaster: a polemic’. Environmental Research Letters 2: 045007. Dorling, D. and Thomas B. (2016). People and Places: A 21st Century Census Atlas of the UK (Bristol: Policy Press). Dorling, D. and Lee C. (2016). Geography (London: Profile). Dynarski, S. (2015). ‘For the poor the graduation gap is even wider than the enrolment gap’. New  York Times. 2 June http://​www.nytimes.com/​2015/​06/​02/​upshot/​for-​the-​poor-​the-​ graduation-​gap-​is-​even-​wider-​than-​the-​enrollment-​gap.html?_​r=0&abt=0002&abg=0 (last accessed 10 May 2017). EBA (2013). ‘EBA presents data on high earners in EU banks for 2012’. European Banking Authority http://​www.eba.europa.eu/​-​/​eba-​presents-​data-​on-​high-​earners-​in-​eu-​banks-​ for-​2012 (last accessed 10 May 2017). Eurostat (2012). ‘EU Statistics on Income and Living Conditions’ http://​www.eui.eu/​Research/​ Library/​ResearchGuides/​Economics/​Statistics/​DataPortal/​EU-​SILC.aspx (last accessed 10 May 2017). Fernández-​Macías, E. and Vacas-​Soriano, C. (2015). ‘Recent developments in the distribution of wages in Europe’. Eurofound Report EF1510 http://​www.eurofound.europa.eu/​publications/​report/​2015/​working-​conditions-​labour-​market/​recent-​developments-​in-​the-​distribution-​of-​wages-​in-​europe (last accessed 10 May 2017). Glasmeier, A. (2002). ‘One nation, pulling apart: the history of poverty in the United States’. Progress in Human Geography 26: 155–​173. Harvey, D. (1973). Social Justice and the City (Oxford: Blackwell). Hay, I. (ed.) (2013). Geographies of the Super-​Rich (Cheltenham: Edward Elgar). Hennig, B.D. and Dorling, D. (2014). ‘The shape of the nation’s wealthiest’. Political Insight November 20–​21. Holland, T., Peterson, G., and Gonzalez, A. (2009). ‘A cross-​national analysis of how economic inequality predicts biodiversity loss’. Conservation Biology 23: 1304–​13013. Inequality Briefing (2015). ‘Briefing 61: regional inequality in the UK is the worst in Western Europe’ http://​inequalitybriefing.org/​brief/​briefing-​61-​regional-​inequality-​in-​the-​uk-​is-​ the-​worst-​in-​western-​europe (last accessed 10 May 2017). Johnson, P. (2015). ‘Opening remarks, summer post-​budget briefing 2015’. Institute for Fiscal Studies http://​www.ifs.org.uk/​publications/​7853 (last accessed 10 May 2017). Jones, O. (2014). The Establishment (London: Allen Lane). Krugman, P. (2012). ‘The economics of marginalization and hopelessness’. New York Times, 12 May http://​krugman.blogs.nytimes.com/​2012/​05/​12/​the-​economics-​of-​marginalization-​and​hopelessness/​ (last accessed 10 May 2017).

Inequality in Advanced Economies    61 Li, Y. and Wei, Y.H.D. (2010). ‘The spatial-​temporal hierarchy of regional inequality of China’. Applied Geography 30: 303–​316. Mikkelson, G., Gonzalez, A., and Peterson, G. (2007). Economic inequality predicts biodiversity loss. PLoS ONE 2: e444. Moreno, K. (2014). ‘The 67 people as wealthy as the world’s poorest 3.5 billion’. Forbes Magazine, 25 March http://​www.forbes.com/​sites/​forbesinsights/​2014/​03/​25/​the-​67-​people-​as-​wealthy​as-​the-​worlds-​poorest-​3-​5-​billion/​ (last accessed 10 May 2017). Nowatzki, N.R. (2012). ‘Wealth inequality and health:  a political economy perspective’. International Journal of Health Services 42: 403–​424. OECD (2013). ‘Income Distribution Database’. www.oecd.org/​social/​income-​distribution-​ database.htm (last accessed 15 June 2015). OECD (2015). ‘Income inequality data update and policies impacting income distribution:  United Kingdom’. http://​www.oecd.org/​unitedkingdom/​OECD-​Income-​Inequality-​ UK.pdf (last accessed 10 May 2017). ONS (2013). ‘The effects of taxes and benefits on household income, 2011/​12, London’. Office for National Statistics http://​www.ons.gov.uk/​ons/​dcp171778_​317365.pdf (last accessed 10 May 2017). Philo, C. (1995). Off the Map:  Social Geography of Poverty (London:  Child Poverty Action Group). Pickett, K.E. and Wilkinson, R.G. (2015). ‘Income inequality and health: a causal review’. Social Science and Medicine 128: 316–​326. Piketty, T. (2014). Capital in the Twenty-​First Century (Cambridge, MA: Harvard University Press). Resolution Foundation (2015). ‘Welcome boost on low pay but severe welfare cuts will mean big losses for many low-​income working families’. Press release, 8 July http://​www.resolutionfoundation.org/​media/​press-​releases/​welcome-​boost-​on-​low-​pay-​but-​severe-​welfare-​cuts-​ will-​mean-​big-​losses-​for-​many-​low-​income-​working-​families/​ (last accessed 10 May 2017). Ross, N., Dorling, D., Dunn, J.R., Henriksson, G., Glover, J., Lynch, J., et  al. (2005). ‘Metropolitan income inequality and working-​age mortality:  a cross-​sectional analysis using comparable data from five countries’. Journal of Urban Health: Bulletin of the New York Academy of Medicine 82: 101–​110. Runciman, D. (2013). ‘The Democracy Project: a history, a crisis, a movement by David Graeber’. https://​w ww.theguardian.com/​b ooks/​2013/​mar/​2 8/​democracy-​project-​david-​graeber-​ review (last accessed 10 May 2017). Sassen, S. (2014). Expulsions: Brutality and Complexity in the Global Economy (Cambridge, MA: Harvard University Press). Sayer, A. (2014). Why We Can’t Afford the Rich (Bristol: Policy Press) Sellers, P. (2015). ‘Britain now most unequal EU country, says official report’. TouchStone Blog, 18 May http://​touchstoneblog.org.uk/​2015/​05/​britsian-​now-​unequal-​european-​countyry-​ saysy-​new-​reportc/​ (last accessed 10 May 2017). Shale, J., Balchin, K., Rahman, J., Reeve, R., and Rohlin, M. (2015). ‘Households below average income: an analysis of the income distribution, 1994-​5—​2013/​4’. Department for Work and Pensions https://​www.gov.uk/​government/​collections/​households-​below-​average-​income-​ hbai-​-​2 (last accessed 10 May 2017). Smith, S., Pain R., Marston, S., and Jones J. (eds) (2010). The SAGE Handbook of Social Geographies (London: SAGE Publications). Stiglitz, J. (2012). The Price of Inequality: How Today’s Divided Society Endangers Our Future (New York: W. W. Norton & Company).

62   Dorling Stiles, T., Davies, T., and Cooper, A. (2015). ‘2015 Change Readiness Index’. KPMG International https://​assets.kpmg.com/​content/​dam/​kpmg/​pdf/​2016/​06/​it-​2015-​change-​readiness-​index. pdf (last accessed 10 May 2017). The New York Times (2014). ‘Income Distribution Database’. www.lisdatacenter.org/​resources/​ other-​databases/​ (last accessed 19 June 2015). The Sunday Times (2014). ‘Sunday Times Rich List 2014’. http://​en.wikipedia.org/​wiki/​Sunday_​ Times_​Rich_​List_​2014 (last accessed 10 May 2017). UNU-​WIDER (2014). ‘World Income Inequality Database (WIID3.0b)’ https://​www.wider. unu.edu/​project/​wiid-​world-​income-​inequality-​database (last accessed 10 May 2017). Urahn, S., Currier, E., Elliott, D., Wechsler, L., Wilson, D., and Colbert, D. (2012). ‘Pursuing the American Dream: economic mobility across generations’. The Pew Charitable Trusts http://​ www.pewtrusts.org/​~/​media/​legacy/​uploadedfiles/​pcs_​assets/​2012/​PursuingAmerican Dreampdf.pdf (last accessed 10 May 2017). Washington Post (2011). ‘(Not) spreading the wealth’, 18 June http://​www.washingtonpost.com/​ wp-​srv/​special/​business/​income-​inequality/​ (last accessed 10 May 2017). Whoriskey, P. (2011). ‘With executive pay, rich pull away from the rest of America’, Washington Post, 19 June https://​www.washingtonpost.com/​business/​economy/​with-​executive-​pay-​ rich-​pull-​away-​f rom-​rest-​of-​america/​2011/​06/​13/​AGKG9jaH_​story.html?utm_​term=. dc7e6572309b/​(last accessed 10 May 2017). Wilkins, B. (2013). ‘Nobel Prize winner Shiller:  inequality biggest problem facing US’. The Digital Journal http://​www.digitaljournal.com/​article/​360347 (last accessed 10 May 2017). Wilkinson, R. (1996). Unhealthy Societies (London: Routledge). Wilkinson, R. (2005). The Impact of Inequality (London: Routledge). Wilkinson, R. and Pickett, K. (2010). The Spirit Level (London: Allen Lane). Wyly, E. (2009). ‘Strategic positivism’. The Professional Geographer 61: 310–​322. Yousuf, H. (2013). ‘Obama admits 95% of income gains gone to top 1%’. CNN Money http://​ money.cnn.com/​2013/​09/​15/​news/​economy/​income-​inequality-​obama/​ (last accessed 10 May 2017).

Chapter 3

Inc om e Ine qua l i t y and Grow i ng Disparit y:  Spat ia l Pat terns of Ine qua l i t y an d the Case of t h e  U S A Amy K. Glasmeier Introduction Rising income inequality is a global concern. There has been a sizeable increase in income inequality since the 1970s in most markets, from advanced economies to developing nations. Economic disparity has garnered considerable attention since the last recession. Described by former US President Barack Obama as the defining challenge of our time, rising inequality is one of the most hotly debated issues among researchers and policymakers around the globe (Furman et al., 2015; International Monetary Fund, 2015). While a widespread concern, public discussion of income inequality in the USA and elsewhere still lacks clarity about its extent, its drivers, the resulting consequences, and what to do about it. Likewise, there is confusion over what exactly is meant by income inequality, why it matters, and how it compares to wealth inequality, which in many countries is more extreme than income inequality. Although rising income and wealth inequality are global concerns, still these terms, their implied composition, and numerical measurement lack a shared connotation. If there are universal features that help to define the current moment of disparity, they are the apparent impact of rising wealth and income inequality on economic growth and the resulting political apprehension accompanying this moment, evident as it is in developed and developing countries alike. Distinct characteristics of nations, including their economic and political histories, combine forces in generating disparities across groups and locations. Inequalities of income and wealth are characteristically bonded to and embedded in the economic base of different countries and the legacy of the institutional linkages related to prior rounds of job and

64   Glasmeier wealth-​generating capital investment. Emblematic here is the important role of manufacturing in bringing wealth and prosperity to industrial communities of developed economies. Industrial restructuring, automation and importantly, international trade policy have wreaked havoc on the economic base of working class communities around the world (Autor et al., 2016). Traversing the last thirty years, economists, political scientists and historians, and economic and political geographers have chronicled the geographical expressions of growing inequality taking root around the globe (Harvey, 2001; Martin and Morrison, 2002; Firebaugh, 2003; Beramendi, 2012; Florida and Mellander, 2013). In fact, 1959 marks the year in the USA that poverty thresholds were established. These markers are now institutionalized (though flawed they are) and they track the divergence in incomes across locations starting with end of the 1960s (Fisher, 2008, referencing Orshansky, 1958). From investigations of capital mobility at the systemic level, to the recognition of the state as the mediating influence and fiscal agent receiving and redistributing taxes to surmount the forces of inequality, to the susceptibility of trade-​sensitive sectors that, in turn, expose metropolitan areas to the effects of large declines in middle-​income jobs, the forces of inequality are inescapable and indelible. Geographical space serves as a pivot around which both explanations for and drivers of income and wealth inequality find expression. The effects accompanying geographical difference are integral to and foundational of economic inequality. Time’s geographical expression is a fulcrum around which disparity forms. Sources of income and capital have short-​and long-​term durability, each with their own consequence. The precipitating forces of capital mobility unleash downside risks peculiar to the economic base of specific countries and communities resulting in shifts in income, employment, and wealth opportunities. The fixity of objects and the differential mobility of people are lasting reminders of the power underlying the formation of geographical variation. Inter-​generational mobility of wealth is exemplary of capital’s durability through time and, secondarily, its consequential effect on income inequality. Accumulated wealth serves as a buffer against unforeseen events enabling continued income generation and wealth accumulation over time. This reciprocal process of disparity formation is fundamentally spatial and tractable; and yet it is not wholly unidirectional or irreversible—​witness the marked decline in the fortunes of the rich and the poor during the Great Recession. However, accumulated wealth’s inter-​temporal hold on the future enables the possibility of recovery after crises and serves as the foundation of continued income generation and growing inequality. In turn, the absence of stored wealth eliminates the prospect of self-​funded security in periods of unstable or declining income. Changes in public policy—​taxation, trade, and labour market regulation—​have tractable consequences on income and wealth inequality. Economic geographers have traced the diffusion of policy, showing that policies are neither immutable nor geographically fixed. Research by economic geographers record the speed with which policies replicate where suitable political conditions arise (Peck and Theodore, 2015). Historical precedent is a prudent guide to understanding the role of the state at this moment. In the USA, progressive policies hard won during past eras of political liberalism were susceptible to interest-​group influence that shaped and reshaped the direction of government practices. The 1970s were an economic watershed. The combined forces of globalization and the consequences of technological change transformed the economies of countries around the world and the local communities within them. The 1970s and 1980s (especially Ronald Reagan) saw the rollback of federal policies, including reduced tax rates on investment income privileging capital gains over earned

Income Inequality and Growing Disparity    65 income. Changes in antitrust regulations, in part a reaction to global competition, and in part reflecting changing practices in sectors like retail and the dismantling of former monopolies like AT&T, lead to employment instability. The failure of the executive branch to raise the minimum wage to keep pace with the rising cost of living compounded the effect of stagnant wage rates. Labour market regulations guaranteeing the rights of workers to organize and bargain for better wages and working conditions faced persistent challenges. Starting in the early 1990s, trade policies designed to rein in non-​tariff-​based forms of protectionism succeeded in establishing new rules governing market access. However, these so-​called ‘reforms’ failed to deliver the required designs for and sufficient funding of compensatory mechanisms able to support the scale of transition assistance required to retrain, relocate, and retire affected workers. The lack of a policy framework capable of addressing the twin effects of increasing trade liberalization and skilled-​biased technological change got lost in a divided government. The scale of labour market adjustment required in the face of such convergence failed to materialize. Workers lacking little more than a high-​school diploma had little to fall back on in the face of an increasingly globalized economic landscape. The lack of necessary government investments to offset the inevitable effects of dislocation resulting from skill-​biased technological change (SBTC) and job relocation contributed to rising income inequality in the USA. As the effects of economic displacement mounted, the geographical mobility of capital played one community off against another. Billions of dollars of subsidies changed hands in offensive and defensive bidding wars between states. On the one hand, states aggressively pursued foreign direct investment emanating from international companies seeking access to the US market. Taking up residence simultaneously provided protection against unforeseen trade policy actions, while ensuring access to the USA’s rich pool of human capital and technology. While the intensity of these policy influences is subject to debate, few would argue that the last thirty years have seen an erosion of the Depression era-​inspired social safety net conceived in the 1930s. In the current context, the well-​documented influences of these events warrant the selection of the USA as a case study of disparity formation. Tracing geographers’ role in identifying antecedents of the ‘great divergence’ remains a vital element in understanding the causes and the consequences of rising inequality and growing disparity formation.

Why the Concern About Inequality Now? Over the last five years, numerous reports have dwelled on the consequences of growing income and wealth inequality. The rise in inequality is worrying many organizations around the world from the United Nations Development Programme (UNDP) to the Organisation for Economic Co-​operation and Development (OECD), from the private World Economic Forum to the non-​profit Oxfam and Pew’s Center for Social and Economic Research, and influential organizations such as the research arm of the bank Credit Suisse and national governments around the world. Whether measured by income based on wages and earnings from work, or wealth—​the value of assets individually owned including capital and material resources—​or corporate executive compensation (a topic I mention here but do not discuss further in this chapter), rising inequality is evident everywhere (Kiatpongsan and Norton, 2014).

66   Glasmeier The literature on rising inequality is becoming vast. Research underpinning this scholarship focuses on both income and wealth disparity, compares developed and developing countries, and in several cases examines the effects of rising inequality across different spatial scales from cities to regions to countries. Many perspectives are proffered to explain this concern. Given this growing research area, in writing this chapter I have two inspirations. Firstly, reviewing the vast evidence I believe the economics-​inspired theories contending that inequality is a prerequisite for and in some cases a necessary prerequisite of early stages of development rests on simplifying assumptions and reflects historical circumstances that ignore contemporary evidence of modern experiences of development. Past acceptance of the conceptual trade-​off between short-​term consequence and the pursuit of long-​term transformation no longer stands on firm ground as the sole pathway leading to development. Recent research indicates that this perspective inaccurately reflects the empirical evidence which reveals that countries can and have pursued development paths leading to greater equality over time (United Nations Development Programme, 2013). Secondly, moral and ethical arguments can and are made to justify the pursuit of equality because it supports human dignity and enables the achievement of individual self-​actualization. Empirical evidence supports the notion that less equality and rising inequality are linked to reduced rates of economic growth, lower savings rates, and reductions in human capital investment. Inequality is also linked to lower skill levels, which in turn lead to reduced literacy levels, the ability to care for one’s self and consequently necessitates a greater reliance of an individual on public support. Health, civic engagement and levels of social trust are also diminished (OECD, 2013). Their positive form is considered a prerequisite to the formation of a middle class, which serves as a bedrock of societal development. History demonstrates that the underlying predicates of the American dream, ‘everybody deserves a change and at the same time you are expected to make it on your own’, are no longer in reach for millions of Americans. While this chapter focuses on a single case, that of the USA, it reflects themes evident in the literature on the wider experience of developed and developing countries around the world.

Causal Forces Precipitating Growing Disparity There are numerous causal forces behind the current rise in inequality, everything from the resource curse in resource-​dependent countries to the shift in national policies away from production-​induced growth in places like China. Disentangling them is beyond the scope of this chapter. Here, instead, a range of explanations for current economic circumstances is examined. At a global scale, many commentators attribute the rise in inequality in advanced economies to the effects of SBTC and the decline of particular labour market institutions (International Monetary Fund, 2015). Unconvertible evidence indicates globalization’s increasing speed contributes to rising inequality. In less developed countries, rising inequality occurs when conventions to support the growth of a middle class are absent. These forces are neither new nor are they intractable. However, their absence represents missing links required to support economic growth leading to development.

Income Inequality and Growing Disparity    67

Globalization, Trade, Outsourcing, and Foreign Direct Investment At the level of policy, the case of the USA offers convincing evidence that the features of inequality are both global and local in the making, with resulting consequences. Research by David Autor and colleagues at MIT highlights the consequence of China’s ascension to the World Trade Organization, opening up the US labour market to wage competition in labour-​intensive sectors sensitive to the influence of trade. This work comes on the heels of more than three decades of research by economic geographers identifying the end of the postwar era of relatively widely distributed economic growth. Deindustrialization was ushered in by the rise of reconstituted economies of Europe and Japan, the catch-up of early Asian industrializers rewarded with access to high-​income markets in return for a pledge to pursue liberal ideals and uphold democratic institutions (Glasmeier, 1986). The first break in Bretton Woods institutions arose as countries like South Korea, Japan, Hong Kong, and Singapore reshuffled the geography of labour-​intensive industries, progressing quickly from a low-​skilled to a higher-​skilled, but still a low-​wage, labour base. Cracks emerged in the early 1970s in the relationship between a country’s level of development and its ability to perform complex skill-​intensive tasks. Always fragile, the Bretton Woods post-​World War II policy framework consensus gave way to job losses and a long, drawnout period of adjustment that left millions out of work. As my analysis indicates, initial adjustments in wages and labour force participation rates occurred, slowly at first. For workers and communities that lost jobs, the theoretical predictions that labour market adjustment would prevail, regrettably failed to materialize. For more than a decade, job displacement and employment losses went unmet by offsetting employment increases in other industries (Acemoglu et al., 2016; Autor et al., 2016). Adjustment mechanisms to dampen the worst effects of change—​skill augmenting training, industrial upscaling, and new industry formation—​spawned only weak and spatially uneven responses; many efforts were too small in scale to have much effect. Recognized early by economic geographers, economists eventually modelled the effects of SBTC, a powerful force underpinning the current era of rising income inequality. Long after the spatial division of labour was recognized, economists Card and DiNardo (2002) summarized the problem of SBTC in writing that as developed countries lost less skilled jobs with declines in low-​value production activities, wage rates fell while the ranks of less skilled workers swelled. This, in turn, increased the global supply of less skilled workers, making it even more likely that low-​skill-​related goods would be produced in places where wage rates for such work were lowest. This caused a simultaneous shift in demand for skilled labour in high-​income countries as industries increased the share of skilled labour, even given rising or stable wages for skilled workers. Demand for skilled labour in affected industries increased globally and skill premiums formed even in developing countries (Card and DiNardo, 2002). First evident in home countries, SBTC quickly enabled employers to refine their locational strategies to target distant pools of highly skilled workers compensated at exceedingly low home country wages (Glasmeier, 1986). As companies grew more adept at designing products based around modularity, the spatial division of labour spread outward in search

68   Glasmeier of pools of highly skilled yet low-​cost labour found around the globe. Home-​country labour markets no longer constrained access to skilled labour. Firms refined their locational strategies, shopping for places boasting not only skilled workers, but also exceedingly lucrative business incentives, including low levels of regulation and pools of investment capital able to subsidize capital intensive assembly operations. And in some cases, firms returned to their home countries, employing automation to reduce their wage bill while improving the quality of their final product. Foreign direct investment in cheap labour locations within the US drew skilled workers from job-​scarce regions towards the non-​unionized US south (Bluestone and Harrison, 1984). Ever nimble, companies played states one off against another, receiving multimillion-​dollar capital subsidies at free or exceedingly low interest rates while enjoying multi-​year tax-​free status on income and property (Berman et al., 1998; Dicken, 2015).

Labour-​market Policies National policies are both impediments to and causal forces encouraging growing income inequality. Since the early 1970s, economic geographers have recognized and charted the role of policy simultaneously aiding and abetting the spatial dispersal of labour-​intensive activities, while enabling skill-​intensive activities to concentrate amid select urban areas. Middle-​ income jobs followed low-​skilled ones to lower-​cost locations, eventually evaporating in the wake of SBTC. The collapse of middle-​income employment restructured wage rates, contributing to the convergence of income and rates of gross domestic product per capita across regions. Further eroding the value of labour, labour market regulation, and stagnant minimum wages ensured that entry-​level jobs more often than not served as dead-​end positions for first-​time labour-​market entrants. While geographers have made contributions to the examination of the role of minimum wage rates and their effect on standards of living, scholars studying institutions such as unions demonstrate the effect of differential rates of regulation among states. Economists recognize the influence of this change, and their work postdates that of geographers. At the same time, more flexible labour market policies over the last thirty years have in many countries resulted from dual labour markets of good and bad jobs. Experienced workers occupy ‘good-​paying, benefit-​providing jobs’, whereas newer entrants, young people, women, and unskilled workers occupy part-​time and flexible work positions with limited economic mobility or security. Support for unions is another means whereby national policies affect the labour market and the prospect of economic mobility. Both developed and developing countries experienced the steady erosion in institutional provisions ensuring employment mobility and labour protections (Dreger et al., 2015). Labour-​market effects work through multiple policy vehicles. Educational deferments in the Vietnam War combined with the subsequent utilization of the GI Bill by thousands of returning veterans flooded the US labour market with college-​educated workers seeking jobs requiring post-​secondary education. On the eve of the microelectronics revolution and the ensuing era of miniaturization, wages fell in reaction to a substantial influx of skilled, college-​educated employees, diminishing incentives to seek higher education. A resulting gap opened in the skill profile of the US workforce that took decades to overcome (Autor, 2014).

Income Inequality and Growing Disparity    69 Perhaps selectively, referential of the macroforces unleashed by the war in Vietnam and the oil crises of the 1970s, economic geographers unquestionably led the way in interpreting globalization’s effect on the manufacturing regions of the USA and other industrial countries and the resulting labour market consequences. Job-​destroying industrial restructuring emptied out manufacturing regions formerly reservoirs of medium-​skilled, well-​paying, labour-​intensive work employing millions of (predominantly male) workers. Emphasizing sectoral practices and firm policies, economic geographers traced both their causes and consequences, including business cycles, labour relations, spatial divisions of labour, and state policies. Combining these forces, the restructured demand for male workers led to depressed rates of remuneration for college-​educated individuals through the outright destruction of jobs for unskilled and semi-​skilled workers. Stunted returns to education, especially for men further added to rising inequality by diminishing the links between higher education and income (Goldin and Katz, 2008; Goldin and Katz, 2009; Autor, 2014). Thus, the three forces—​‘technological change, deunionization, and globalization—​work in tandem’ (Autor, 2014, p. 840). The next section focuses on the USA and discusses the consequences of inequality on communities around the country by region. Tracing causal forces leading to rising inequality requires consideration of the interaction effects of spatial mobility of capital, investment, and opportunity in the form of technology transfer that has distributed employment globally.

A Focus on the USA In no other country is evidence of growing income inequality more meaningful than in the USA. The USA is a model of development coveted by many around the world. The rate of continued migration is a testimony to the belief that the USA is an exemplary nation where the pursuit of individual freedom enables anyone, through hard work, to achieve personal realization, economic security, and intergenerational economic well-​being. Admittedly, the long-​held belief that the American dream was available to all those who applied themselves has always been true for some and not others. The exclusion of groups in society represents an important and often neglected part of the American story. Today’s concern about rising inequality is heightened given the changes in the economy and the declining role of government protections that have further reduced the percentage of the population effectively able to share in the American dream (Autor, 2014).

Inequality in the USA: Income and Wealth Income In the USA, inequality of income and wealth have been rising since the late 1970s (Rank et al., 2014). Income reflects annual wages and earnings plus income from investments otherwise known as labour and taxable income. Wealth measures estimates of capital income (the value of all assets of worth owned). Despite a difference of opinion about how to calculate

70   Glasmeier and measure income or wealth inequality, on both counts the evidence indicates a few broad patterns. Looking at income (money income resulting from work and investments), from the 1950s to the 1970s shares to different income groups remained stable. This period of distributional stability gave way to a period of wage stagnation and eventually inequality began to rise. Inequality rose steadily over the next three and a half decades (Noah, 2013). From the late 1970s until 2011 wages for the median US worker increased by just 6 per cent over that period. As Paul Krugman points out, the late 1980s and through the 1990s until 2005 incomes grew fat, but 80 per cent of the total increase in income went to the top 1 per cent (Krugman, 2007). Growing differences in shares by income bracket saw greater gains in the highest quintile from the 1980s onward. This pattern was checked somewhat during the Great Recession, mainly because occupants of high-​income brackets experienced larger declines in the value of assets they held compared with asset poor individuals. Poor people, by and large, do not own assets, nor do they have much in the way of savings. Starting in 2009, the shift in shares began to favour high-​income groups. Seen today, nationwide, 3.1 per cent of income earned annually goes to the poorest 20 per cent of people, while 51.4 per cent is received by the richest 20 per cent. In a multi-​year investigation starting in 2012, the Pew Research Center began to track the consequences of the economic crisis on the American middle class (Pew Research Center: Social & Demographic Trends, 2012). In 2012 the Center reported that since 2000 the middle class has shrunk in size, measured as ‘all adults whose annual household income is two-​thirds to double the national median household income’. According to Pew’s research, in 2011 the middle-​income group comprised 51 per cent of all adults, down from 61 per cent as calculated for 1971. The corresponding figures for the upper tier reflected a significant rise in incomes. The upper tier increased to 20 per cent of all adults in 2011, up from 14 per cent in 1971. The high-​income tier now comprises 46 per cent of household income, up from 28 per cent back in 1971. The decade 2000–​ 10 saw the median income of the middle-​income tier decline by 5 per cent, while wealth of that tier decreased by 28 per cent. The reverse pattern was true for the upper-​income tier, although median wealth gains increased by 1 per cent.

Wealth Another indicator of growing disparity is observed in changes in the distribution of wealth over the 1970–​2010 period. Wealth is of particular interest because of the role it plays in underpinning the options individuals and families have to make decisions about their future. Economists remain divided about the scale of the impact of changes in wealth across income groups—​not that change has occurred, but rather the range of gains by income group. Recent research focusing on the distribution of wealth parallels the Pew Research Center’s findings of shifts in the shares received by families across wealth tiers. Data covering much of the twentieth century suggests wealth, like income, has become increasingly unequal over time. According to this new work, ‘by our estimates, the share of wealth owned by the top 1% families has regularly grown since the late 1970s and reached 42% in 2012. Most of this increase is driven by the top 0.1%, whose wealth share grew from 7% in 1978 to 22% in 2012, a level comparable to that of the early 20th century’ (Saez and Zucman, 2016, p. 520).

Income Inequality and Growing Disparity    71 The research demonstrates not only the distribution of wealth over time, but also how increases in wealth were shared over the last four decades. These findings conclude that ‘from 1986 to 2012, for example, almost half of U.S. wealth accumulation has been due to the top 0.1% alone’ (Saez and Zucman, 2016, p. 521). Explanations for this rise in wealth inequality relate to increases in income in the top-​ income brackets and the ability of upper-​income groups to save more income than others over time. Savings, in turn, make possible more capital investment, leading to additional capital gains. Saez and Zucman’s research shows that persons becoming wealthy today are younger and their labour incomes are substantially higher than previous generations of comparably aged workers. At the same time, Pew research also shows that the wealth share of the bottom 90 per cent, which rose during the twentieth century (from 20 per cent in 1920 to 34 per cent in the mid-​1980s to 23 per cent in 2012), began to decline in the 1980s. Middle-​class savings decline primarily explains this development. In the USA, middle-​class fortunes are increasingly tied to changes in factors emerging over the last thirty years, especially the decline in defined pension plans supporting retirement, the rising cost of housing, and the burden of student debt reducing the potential of homeownership for individuals in their late twenties to mid-​thirties (Li and Goodman, 2015). The disparities in income and wealth go hand in hand. According to the Federal Reserve Board’s Survey of Consumer Finances, poorer families’ net worth declined between 2010 and 2013 (Buchholz et al., 2016). In stark contrast, the nation’s wealthiest families continued to make modest gains. These trends relate to the types of investments low-​and high-​ wealth households own. For most Americans, their house is their primary asset. An increase in housing value represents their greatest means of wealth accumulation over time. Over the last seven years home prices have been stagnant, except where the limited supply of housing combined with high-​income-​earning jobs drove up housing prices. In contrast, the wealthy have access to much greater earnings, which, in turn, can fund a variety of wealth-​generating instruments, including stocks and bonds.

Spatial Pattern of Inequality in the USA Admittedly, the spatial pattern of urban development reflects unique characteristics related to, if nothing else, the site and situation of a particular city or region. Still, as the process of urbanization accelerated over the last three decades, models and standards of development accompanying this period of urban growth exhibit tendencies of enclave formation based on income and wealth. In Global South countries, the pattern is often hyper-​accentuated—​a primate city forms the core of the urban city system. Wealth and income, infrastructure, and opportunity accompany the process of urbanization. While urbanization has unique features, nonetheless its consequences represent points of concentrated and agglomerated activity. In exploring patterns of economic inequality in urban areas of the USA, I am careful to draw very broad generalizations about spatial variation, while recognizing distinctions evident at the scale of particular locations. Situating the consequence of rising income and wealth inequality geographically requires reflecting on past patterns of change to make sense of the current context. Over the last

72   Glasmeier eighty years, regional per capita income as a percentage of the national average showed signs of converging until the late 1970s. As much as anything, starting in the late 1970s, repeated recessions, major industrial restructuring and both age-​and employment-​related migration brought an end to the trend of convergence. Incomes and wealth began to concentrate in selected locations while bleeding out of others, reasserting the importance of places of economic power. Over the last fifty years, even as the population fanned out towards the south and west, the US economy slowly but steadily consolidated around and within a few key locational nodes, forming a pattern reminiscent of the economic city system of the early-​to-​ mid-​twentieth century (Pred, 1964). During the late 2000s, there was a decided trend towards inter-​and intra-​state inequality (McNichol et al., 2012). Incomes across the nation have increased since the recession, but income inequality has also increased. Incomes among the richest 20 per cent of households grew faster from 2006 to 2012 than they did among the poorest 20 per cent of households in all fifty states, without exception (Frohlich, 2015). A state-​by-​state analysis of income trends identifies thirty-​nine states where the top 1 per cent of households received 50 per cent or more of total income gains over the 2009–​12 period. In seventeen states, predominantly in the South, East, and Far West, the top 1 per cent received the entire income gains over the same interval. Comparing the poorest fifth with the top fifth of all households across states indicates incomes are extremely concentrated in the states with the highest income inequality. In New York and Connecticut—​the first and second worst states for income distribution—​more than one-​quarter of all income was held by the richest 5 per cent of state households (Sommeiller and Price, 2015).

The Metro Consequences of Rising Income and Wealth Inequality in the USA The advent of big data, more timely federal and state statistics, and ease of analysis makes possible the ability to consider these consequences from a range of views. Here I examine three perspectives:  structural features that are built-​in and undergird existing inequality such as minimum-​wage laws and city-​level fortunes related to place-​and people-​based characteristics. Two lenses—​one state-​and one metro-​scale—​offer perspectives on the origin and consequence of income and wealth inequality in the USA. The first lens represents the state variation in the minimum wage. The minimum wage was introduced in the 1930s to serve as a floor under which employers could not pay substandard wages. At its peak, 1968 minimum wages paid the equivalent in today’s dollars of $10.16 per hour. It was last increased in 2009 when the Obama administration raised the wage from $6.55 an hour (equivalent to $7.02/​ hour) to $7.25 an hour (a level first achieved in 1950). Today the minimum wage pays the equivalent of the minimum wage of the 1970s. Between 2009 and 2014 the last increase lost 8 per cent of its purchasing power in adjusted dollars. States have been most active in adjusting local minimums to account for differences in costs of living. Twenty-​nine states and more than two dozen cities and counties legislated higher minimums. Some states enforce no minimum wage (Alabama, Mississippi,

Income Inequality and Growing Disparity    73 South Carolina, Tennessee, and Louisiana). As expected, the minimum wage hits younger workers hardest. Of the three million workers earning at or below the minimum wage, almost 1.5 million are young workers between the age of sixteen and twenty-​four years, and another 22 per cent are between twenty-​five and thirty-​four years of age. An additional twenty million workers or almost 30 per cent of all hourly workers are paid at or near the minimum wage. Certain jobs in fields like food service almost exclusively pay the minimum. Employment in some of the nation’s biggest annual job generators, including sectors such as retail sales, construction, and elementary and secondary schools, almost all start as minimum-​wage jobs (Bureau of Labour Statistics, 2012; Department of Labour, 2016; Desilver, 2015). The second lens focuses on the relationship between rising inequality and location and the extent that there is bias in high-​and low-​paying jobs across the country and among cities. Results of three studies suggest there is a marked pattern of spatial bias with relatively few locations in the USA, where income inequality is extreme. Brookings researchers analyse inequality in American cities by applying a ratio comparing the top and bottom ends of the income distribution (the ratio of households earning more than 95 per cent of all other households—​and those earning more than only 20 per cent of all other households) (Berube and Holmes, 2016). Not surprisingly, using the twenty/​ninety-​five ratio the top two cities in the USA with the highest income inequality are the New York–​Newark–​Jersey City Consolidated Metropolitan Statistical Area (CMSA) and Bridgeport–​Stamford–​Norwalk, Connecticut. These CMSAs are part of a super-​­metropolitan region that encompasses the midsection of the North Atlantic sea coast. Including Pennsylvania, the metropolitan region with the highest level of demonstrated income inequality is a ­megaregion of 21,000,000 people. Differences between places of high-​income inequality and low-​income inequality are explained by unique circumstances. Of locations where inequality is highest and has risen the most over the 2007–​14 period, research attributes this growing inequality to the difference in employment stability between high-​and low-​income families. In contrast, metropolitan areas where income inequality was relatively low reflect circumstances related to demographic, race, and economic factors, including the homogeneity of occupations and small variances in wage rates by industry, low poverty rates, and high levels of employment (Davidson, 2011). In rare cases, unionization and a comparable economic base comprising jobs of a similar type accompany low-​income inequality. A second study highlights differences in metropolitan characteristics based on wages, as measured by hourly compensation versus income, measured as wages and other sources of income (dividends, interest, rents). The findings indicate that differences in wage inequality reflect place-​based characteristics such as the size of a metropolitan area, human capital levels, and levels of technology. Differences in income inequality relate to people-​ based characteristics, including differences in rates of unionization, race, and poverty. Both dependent variables appear to measure distinct qualities of inequality (Florida and Mellander, 2013). A third study examines places where people earning the highest level of income live (‘One Percenters’) compared with the bottom 99 per cent of local residents. These are locations where basic resources are at the highest level, but these are also locations where recovery from the recession was the greatest. The locations of high income and wealth, while experiencing a dip in income and wealth during the Great Recession,

74   Glasmeier recovered their losses and captured 85 per cent of the income and wealth earned post-​2009. These metropolitan areas represent exclusive locations around the USA, from the Mountains of Wyoming to the shores of Florida. The top twenty-​five metropolitan areas with the highest ratio of individual income earners in the 1 per cent group compared with the remaining 99 per cent income group represent a collection of enclave communities that are both homogeneous and exclusive. These places boast mountain resorts, coastal communities, and locations of old industrial and inherited wealth outside major metropolitan areas. The geography of locations with smaller ratios of high to low household incomes is diverse. Many of these locations are unique, ranging from old manufacturing towns to sites of higher education to communities attractive to so-​called ‘snowbirds’, retirees from cold climates relocating to more temperate places during the winter.

Conclusion The characteristics of rising inequality, falling numbers of the middle class, lower shares of income and wealth for all but the wealthiest, and increasing indebtedness by age group, are evident in many countries around the world. These trends have real consequences for places. The income-​sorting process is leading to people segregating themselves by income and earning capacity. As incomes concentrate, reductions in social diversity isolate low-​and high-​wealth communities from one another. Absent a range of incomes, poor communities face declining tax bases and endure Spartan public sector services, including reduced contributions to schools and other social activities. Even the basics can be missing as residents’ incomes in low-​wealth communities prove insufficient to support basic food stores (Berube and Holmes, 2016). Contrasting circumstances exist in high-​income communities where wealthy individuals buy all manner of special services and experiences beyond those provided by local public services and school programmes. Economic geographers were early contributors to research on the antecedents of rising income and wealth inequality. In many instances, geographers identified the shifts taking place in communities based on middle-​ income, labour-​ intensive economic activities. Economists’ contributions are more recent and reflect the interests of labour economists concerned about changes in the availability of jobs, the quality of work, and the long-​term consequences on families. This recent research offers important insights derived from major data developments and computational tools that enable studies of occupations, skill levels, the effects of international trade flows, and institutional transformations. As quoted earlier, reports written by the OECD, the International Monetary Fund, the UNDP, and numerous think tanks and special-​interest organizations indicate income and wealth inequality is highly geographically uneven. In a modest way, this chapter set out to explore the intersection of macro-​patterns and micro-​foundations of rising income and wealth inequality and the role of economic geography in illuminating those trends. Rising income and wealth inequality results from long-​term processes working themselves out over time and space. Recognizing the limitations of a single-​country case study, conditions precipitating income and wealth inequality are nonetheless surprisingly consistent across locations at multiple scales. In the USA, important centres of corporate headquarters, finance,

Income Inequality and Growing Disparity    75 government, and education reflect both the highest incomes and also exhibit the greatest inequality among income groups. Around the world, so-​called ‘world or global cities’ fall within this group. This chapter suggests rising inequality is found almost everywhere from the staunchest conservative setting where neoliberal policies operate with few constraints to even those countries where social policies actively promote fairness and opportunity. Indeed, a recent paper on the future of liberal democracy in the development of the Global South emphasizes the point that emerging democratic powers, like the BRIC countries (Brazil, Russia, India, and China), are facing serious criticisms of the persistent income disparities present in their nations, while developed northern democracies also face deterioration in income equality (Öniş, 2015). While it is a comprehensive trend that income inequality is mounting, support must be given within the USA to diminish the wealth and income gap between the rich and poor, which will hopefully propagate worldwide once one leading country is successful in doing so. The starting point has to be the recognition that the consequences of rising inequality will be to reduce social advancement, diminish the belief in civil society, and result in unhealthy and unsustainable populations. The lack of opportunities to aspire to a better future reduces a person’s sense of commitment to group values. Rising income inequality can be reduced by reversing many of the policies enacted over the last three decades that reduced protections to workers by diminishing labour market institutions, allowing minimum wages to stagnate, lessening the progressivity of personal income taxes rates, and deregulating markets, which enabled new forms of market power to emerge. Proliferating countervailing investments in effective transition assistance to offset the inevitable effects of dislocation resulting from SBTC of trade policies will lead not only to policy innovation, but will also prompt a revaluation of the importance of skill creation and capital investment in fixed assets. Another approach would be to reduce benefits that favour the rich, such as capital gains tax rates, stock options, and carried interest. Such actions would allow commensurate reductions in rates of taxation on labor income and allow for an increase in the ability of individuals to retire absent the burden of financial insecurity. All of these methods for shrinking inequality gaps in wealth and income will increase prospects of future membership in the middle class and ultimately lead to a better future for ourselves and others.

Acknowledgements This chapter benefited from extensive comments and suggestions contributed by Dr William Bonvillian, Ms Molly Nebiolo, Dr Tracey Farrigan of the Economic Research Service of the U.S. Department of Agriculture, and Dr Robin Leichenko, Head of the Department of Geography, Rutgers University.

References Acemoglu, D., Autor, D., Dorn, D., Hanson, G.H., and Price, B. (2016). ‘Import competition and the great US employment sag of the 2000s’. Journal of Labor Economics 34: S141–​S198.

76   Glasmeier Autor, D. (2014). ‘Skills, education, and the rise of earnings inequality among the other 99 percent’. Science 344: 843–​851. Autor, D.H., Dorn, D., and Hanson, G.H. (2016). ‘The China shock: learning from labor market adjustment to large changes in trade’. Annual Review of Economics 8: 205–​280. Beramendi, P. (2012). The Political Geography of Inequality:  Regions and Redistribution (New York: Cambridge University Press). Berman, E., Bound, J., and Machin, S. (1998). ‘Implications of skill-​biased technological change: international evidence’. The Quarterly Journal of Economics 113: 1245–​1279. Berube, A. and Holmes, N. (2016). ‘City and metropolitan inequality on the rise, driven by declining incomes’. Metropolitan Policy Program. Brookings Institution http://​www.brookings.edu/​research/​papers/​2016/​01/​14-​income-​inequality-​cities-​update-​berube-​holmes (last accessed September 2016). Bluestone, B. and Harrison, B. (1984). Deindustrialization of America (New York: Basic Books). Buchholz, D., Larrimore, J., and Thompson, J. (2016). Federal Reserve Surveys of the Economic Well-​Being of US Households: SCF and SHED (Washington, DC: Urban Institute. Housing Finance Policy Center). Bureau of Labor Statistics. (2012). ‘Characteristics of minimum wage workers’. http://​www.bls. gov/​opub/​reports/​minimum-​wage/​archive/​characteristics-​of-​minimum-​wage-​workers-​ 2014.pdf (last accessed September 2016). Card, D. and DiNardo, J.E. (2002). ‘Skill-​biased technological change and rising wage inequality: some problems and puzzles’. Journal of Labor Economics 20(4): 733–793. Davidson, L. (2011). ‘Utah has nation’s lowest income inequality’. Salt Lake Tribune http://​archive.sltrib.com/​story.php?ref=/​sltrib/​politics/​52790774-​90/​utah-​income-​inequality-​among. html.csp (last accessed September 2016). Department of Labor (2016). ‘Minimum wage—​ U.S. Department of Labor—​ Chart1. Washington, DC’. https://​www.dol.gov/​featured/​minimum-​wage/​chart1 (last accessed September 2016). Desilver, D. (2015). ‘Five facts about the minimum wage’. Pew Research Center. Fact Tank: News in the Numbers. July 23. http://​www.pewresearch.org/​fact-​tank/​2015/​07/​23/​5-​factsabout-​the-​minimum-​wage/​; https://​www.dol.gov/​whd/​minwage/​america.htm (last accessed September 2016). Dicken, P. (2015). Global Shift, 7th ed. (New York: Guilford Press). Dreger, C., LÓpez-​Bazo, E., Ramos, R., Royuela, V., and Surin~ach, J. (2015). ‘Wage and Income Inequality in the European Union. 2015’. Policy Department A: Economic and Scientific Policy http://​ec.europa.eu/​eurostat/​cros/​system/​files/​05-​2014-​wage_​and_​income_​ inequality_​in_​ the_​eu_​0.pdf (last accessed 2 March 2017). Firebaugh, G. (2003). The New Geography of Global Income Inequality. (Cambridge, MA: Harvard University Press). Fisher, G.M. (2008). ‘Remembering Mollie Orshansky—​the developer of the poverty thresholds’. Social Security Bulletin 68: 3. Florida, R. and Mellander, C. (2013). ‘The geography of inequality: difference and determinants of wage and income inequality across US metros’. The Royal Institute of Technology Centre of Excellence for Science and Innovation Studies (CESIS) http://​www.cesis.se (last accessed September 2016). Frohlich, T.C. (2015). ‘States with widest gaps between rich, poor’. USA Today, 10 October http://​www.usatoday.com/​story/​money/​personalfinance/​2015/​10/​10/​24-​7-​wall-​st-​states-​ rich-​poor/​73618858/​ (last accessed September 2016).

Income Inequality and Growing Disparity    77 Furman, J., Obstfeld M., and Stevenson B. (2015). The 2015 Economic Report of the President. The Council of Economic Advisers. 69th Annual Economic Report of the President (Washington DC: The White House). Glasmeier, A.K. (1986). ‘High-​tech industries and the regional division of labour’. Industrial Relations 25: 197–​211. Goldin, C., and Katz, L. (2008). The Race Between Education and Technology (Cambridge, MA: Harvard University Press). Goldin, C., and Katz, L. (2009). ‘The future of inequality’. Milken Institute Review. 3rd Quarter, 26–33. Harvey, D. (2001). Spaces of Capital (New York: Routledge). International Monetary Fund (IMF) (2015). ‘Causes and consequences of income inequality: a global perspective’. Prepared by Era Dabla-​Norris, Kalpana Kochhar, Frantisek Ricka, Nujin Suphaphiphat, and Evridiki Tsounta https://​www.imf.org/​external/​pubs/​ft/​sdn/​2015/​ sdn1513.pdf (last accessed September 2016). Kiatpongsan, S. and Norton, M.I. (2014). ‘How much (more) should CEOs make? A universal desire for more equal pay’. Perspectives on Psychological Science 9: 587. Krugman, P. (2007). The Conscience of a Liberal (New York: W.W. Norton & Company). Li, W. and Goodman, L. (2015). Americans’ Debt Styles by Age and Over Time. Research Report (Washington, DC: November Urban Institute). McNichol, E., Hall, D., Cooper D., and Palacios, V. (2012). ‘Pulling apart: a state-​by-​state analysis of income trends.’ Economic Policy Institute, 15 November 2012. http://​www.epi.org/​ publication/​pulling-​apart-​2012/​ (last accessed September 2016). Martin, R. and Morrison, P. (2002). The Geography of Labor Market Inequality (New York: Routledge). Noah, T. (2013). The Great Divergence: America’s Growing Inequality Crisis and What We Can Do About It (New York: Bloomsbury Press). OECD (2013). OECD Skills Outlook 2013: First Results from the Survey of Adult Skills. (OECD Publishing) http://dx.doi.org/10.1787/9789264204256-en. Öniş, Z. (2015). ‘Democracy in uncertain times: globalization, inequality and the prospects for democratic development in the Global South’. Koç University, Istanbul: Department of International Relations http://​citeseerx.ist.psu.edu/​viewdoc/​download?doi=10.1.1.730.6922 &rep=rep1&type=pdf (last accessed 3 September 2016). Peck, J. and Theodore, N. (2015). Fast Policy: Experimental Statecraft at the Thresholds of Neoliberalism (Minneapolis, MN: University of Minnesota Press). Pew Research Center: Social & Demographic Trends (2012). ‘The lost decade of the middle class fewer, poorer, gloomier. Chapter 1: overview’ http://​www.pewsocialtrends.org/​2012/​ 08/​22/​the-​lost-​decade-​of-​the-​middle-​class/​ (last accessed September 2016). Pred. A. (1964). ‘The intrametropolitan location of american manufacturing’. The Annals of The Association of American Geographers 54: 165–​180. Rank, M.R., Hirschl, T.A., and Foster, K.A. (2014). Chasing the American Dream: Understanding What Shapes Our Fortunes (New York: Oxford University Press). Saez, E. and Zucman, G. (2016). ‘Wealth inequality in the United States since 1913: evidence from capitalized income tax data’. The Quarterly Journal of Economics 131: 519–​578. Sommeiller, C. and Price, M. (2015). ‘The increasingly unequal states of America income inequality by state, 1917 to 2012’. Economic Policy Institute, 26 January 2015 http://​www.epi. org/​publication/​income-​inequality-​by-​state-​1917-​to-​2012/​ (last accessed September 2016). United Nations Development Programme (2013). Humanity Divided: Confronting Inequality in Developing Countries (New York: Poverty Reduction Programme).

Chapter 4

The Em e rg i ng T ransform at i on of China’s Ec onomi c Geo gra ph y Kam Wing Chan Introduction China’s super-​fast economic growth, averaging about 10 per cent for the last three decades, has won the admiration of many. Indeed, China’s share of the global gross domestic product (GDP) has leapt from about 5 per cent in 2005 to 15 per cent ten years later (Wall Street Journal, 2015). Since 1979, China has gradually risen to be the ‘world’s factory’, dominating the globe through its mass increase in exports of manufactured products. China is the world’s largest low-​and medium-​end contract manufacturer, producing for brand-​name companies like Apple, Toshiba, and Sony. This was partly made possible by the massive migration of underemployed agricultural labour to industry and service in coastal cities, mainly in east and south China. This process has been achieved via a special type of migration and urbanization, operating within a rigid dual socioeconomic system under its hukou (household registration) system, a residence-​control mechanism. This chapter focuses on the domestic scene and examines China’s economic and urban geography by examining its urbanization and city size distribution. The next section provides some essential background about China’s spatial economy, with an emphasis on its configurations and institutions. The section entitled ‘The Dual Structure and Hukou System’ examines China’s dual system and the hukou system. The section ‘Incomplete Urbanization and Under-​agglomeration of Cities’ offers a synthesis of China’s economic geography with attention on its ‘incomplete urbanization’ and under-​agglomerated city economies. The ‘Concluding Discussion’ section ends the chapter with a summary, discussion of the policy significance, and directions for future research.

The Emerging Transformation of China’s Economic Geography    79

Background: Institutional Foundation of the Economy and Major Changes Many outside analysts have long proclaimed that China is fully or largely capitalist (Lin, 1997; Huang, 2008). Many of those who think that China is already capitalist have often overlooked the tenacious institutional continuities from Mao’s era. Indeed, despite the significant change in many aspects of Chinese people’s life (e.g. expansion of personal freedoms) and the economy (e.g. the resumption of many market practices) since 1979, some substance of the pre-​reform political economy has not been fundamentally altered: China remains a Leninist one-​party state, which continues to hold sway in many major aspects of not just the polity, but also of society and the economy. The Chinese ‘market economy’ has to be qualified and at best it is very much a state-​led market economy. The Chinese governments of all levels still run a key part of the economy directly and the rest indirectly. The barrage of measures unleashed by the central government to counter a sudden stock market downturn in July in 2015 and again in January 2016 is widely seen as an example of the continuing state omnipotence in the major sectors of the economy, including the stock exchanges. Indeed, the government continues to manage many spheres of society directly (e.g. births and residence registration)1 and, inevitably, exercises a decisive role in the economy geography and many spatial processes, most obviously in migration and urbanization. In studying China’s urbanization, it is essential to foreground the ‘administrative region economy’ (Liu, 1996, 2010). China’s planned economy was set up in the 1950s to function through a hierarchy of administrative–​economic units with powers concentrated at the central level, controlled by the party. Not surprisingly, with the continuing one-​party rule, this system of unitary hierarchical administrative units has remained intact, although local governments now have more economic decision-​making powers. This multi-​tier hierarchy, made up of the central government, provincial-​level units, prefecture-​level units, county-​ level units, and towns and townships, set up not only to serve the needs of a command economy, but also, politically (and socially), the one-​party rule. This structure determines the basic configuration of China’s spatial economy and the number of administrative units (cities and towns). Cartier has argued convincingly that ‘any theorization of the urban process in China cannot claim an empirically grounded spatial analytic without incorporating the administrative divisions’ (2015, p. 22). Largely corresponding to the local governmental structure, especially in the present-​day, more urbanized China, there are four major levels of political-​cum-​economic urban units under the central government: (in order) provincial-​ level cities, prefecture-​level cities, county-​level cities, and towns.2 Despite devolution of many powers to lower-​level governments in the last three decades, the hierarchical nature of the top-​down polity remains. Upper-​level governments still control the appointments of key personnel in their subordinate units. The power remains vertically organized and determined by the top. The higher ranks not only reflect the political/​ administrative power, but also affect the distribution of fiscal resources in the formal state budgetary system and local economic development (Wong, 1997; Ma, 2005). With industrialization (which is basically in cities) forming the core of China’s development strategy since the 1950s, urban jurisdictions have enjoyed higher administrative ranks, enhancing China’s urban bias, which persists to this day.3

80   Chan The hierarchical system of urban administrative jurisdictions also means that local governments are evaluated and controlled by their supervisory units. All the units within the hierarchy are charged with administrative powers and responsibilities in accordance with their level, with the central government units at the apex, provincial-​level units at the next level, and so on. Because the governments directly participate in the economy and because economic growth (fairly narrowly defined) is the priority objective of the central government, the evaluation criteria of local officials are necessarily heavily tilted towards this set of rather parochial economic indices, along with adherence to the party line. In the pre-​reform era, the targets were the output quantity of the key products, such as steel, coal, and grain. With the economy becoming more monetarized in the last thirty years, the targets set for local governments have been revised to the amount of locally generated regional GDP, GDP growth rate, foreign direct investment (FDI), per capita GDP, budgetary revenue, and the like, but they remain narrowly and overwhelmingly ‘economic’ (Whiting, 2001; Tsui and Wang, 2004). As a result, individual local governments naturally pursue practices and policies focusing on the fulfilment of those targets on the list of evaluation, often at the expense of goals beyond those (e.g. the environment and labour protection) and other, non-​measurable, broader regional and even national interests. The emphasis on each individual unit’s economic performance easily leads local governments to favour local protectionism, often resulting in costly duplications of building infrastructure and production capacity at a significant loss of economic efficiency (Li, H., et al., 2005). The media are rife with stories of individual jurisdictions putting up administrative barriers that impede the free flow of goods and factors of production. Critics have likened the thousands of local governments to ‘feudalist fiefdoms’. Weak enforcement of environmental and labour protection by local governments in their fervent pursuit of local industrial growth has also become commonplace.4 By its very nature, the architecture of the hierarchical administrative system is not congenial to horizontal cooperation and efficiency. Answerable to upper-​tier governments and not subordinate to neighbouring jurisdictions, local governments often can only resolve interjurisdictional conflicts through the intervention of upper-​level governments. The barriers are reinforced by institutions such as the hukou system, which have restricted interjurisdictional factor mobility.

The Dual Structure and Hukou System The paramount goal of the Chinese state in the last six decades, in the name of achieving ‘national development’, has been to attain rapid industrialization, sometimes at huge human and environment costs. Another important institutional foundation that has enabled reaching this goal is the division of the society and economy into two rather disparate systems, commonly referred to in China as the ‘dual system’ (eryanzhi). This separation allows the rural and urban populations to be treated differently, leading also to a persistent dual social and economic structure with a permanent second class of peasants (Chan, 2009). The dual system was set up to enable a Stalinist-​type, ‘big push’ industrialization strategy China adopted in the 1950s (Naughton, 2007). The industrial system/​sector, almost totally located in cities, was designated as the priority sector of the economy and was nationalized (i.e. state owned). It was put under strict state management and received strong state support

The Emerging Transformation of China’s Economic Geography    81 and protection. The state provided basic welfare only for urban workers and their families to maintain social and political stability of this small priority (urban) sector (about 15–​16% of the population in 1955). It was largely maintained at that level until 1978 (refer to urban hukou population in Figure 4.1). The other subsystem was the non-​priority, agricultural/​rural sector encompassing the rest, roughly 85 per cent of the population in the mid-​1950s. Remaining outside of the state’s responsibility, it was largely treated as a ‘residual’, with its main functions being a provider of cheap raw materials (including food grain), labour, and capital for the urban–​industrial sector. The rural population and production were collectivized to serve the above functions, with the collectives operating, among other functions, as the state’s policing agent. The farm population, excluded from state-​supplied welfare, had no claim on national resources and was expected to fend for itself except during time of extreme duress or emergencies. The rural sector’s main task was to produce food grain and raw materials at state-​ dictated (low) prices to support state industry. This allowed the state to reap the monopoly profits of industry for its coffers for the purpose of implementing the ambitious industrialization programs. The hukou, or household registration, system is a necessary instrument in making this separation possible. Under the hukou system, each individual also is fixated within his/​her own small hukou administrative unit (a neighbourhood in the city, or a village) at a different position in the hierarchy.5 Permanent migration from one administrative unit to another is generally not permitted without the approval of the hukou authorities rested in the police. Furthermore, in addition to the huge rural–​urban disparities in state-​provided services, the hukou location also determines the level of benefits that one can get and they also vary from one administrative unit to another within the urban subsystem. Under China’s big push industrialization strategy, the state siphoned off resources in the rural sector for capital accumulation in industry through the well-​known process of ‘scissors prices’ (Lardy, 1983; Tang, 1984; Chan, 2009). As Yang and Cai (2003) point out, to enforce such an extraction, the state needed to exercise coercion using a ‘trinity’ of institutions simultaneously: the compulsory procurement and monopoly of sales of farm produce, the rural collective (commune) system, and the hukou system that controlled population mobility. The first tool was to generate an unequal intersectoral (economic) exchange, and the second and third served as the administrative, policing, and implementation mechanisms to ensure the success of the first. As happened in many other communist countries before, such a Stalinist industrialization strategy was notorious for creating huge disparities between the urban–​industrial and the agricultural sectors, and was bound to generate immense outflows from the countryside until mobility controls were fully in place. As China pressed forward with industrialization in the 1950s, large numbers of peasants began to flee the countryside for the cities. Even although the freedom of migration and residence was freshly enshrined in the first Constitution in 1954, the state took measures to stem the flows by setting up checkpoints and using other administrative measures at various major transportation nodes in 1955 through 1957 (Tien, 1973). It soon became obvious that a more systematic and powerful coercive mechanism would be needed to prevent or at least regulate such ‘undesirable’ rural-​to-​urban migratory flows. It was then that hukou came into full play as a central component of the command system. Hence, the hukou system was finally codified in 1958 (Chan, 2009). The decree required that all internal migration be subject to approvals from the authorities at the

82   Chan destination, just like the Soviet’s prospika system. Each person’s hukou was classified as rural and urban, and for newborns, the hukou classifications would follow that of the mother. By immobilizing the peasantry, forcing them to tend the land at mostly subsistence levels of compensation, and excluding them from access to social welfare and ability to move to cities, this approach created two very different societies. And given the immutable, hereditary nature of the hukou classifications, the peasantry became an immobile underclass. Without such a system, China would not have been able to achieve the paramount goal of the command economy—​rapid industrialization within a short time, albeit at very high cost, during the early decades of the communist rule (Tang, 1984; Chan, 1994).6 Despite this ‘achievement’, it is no secret that the Chinese leadership publicly admitted that the economy as a whole was on the brink of collapse on the eve of the reform in 1978, and that some 250 million peasants were in abject poverty. The economy was in dire need of a new direction (Lardy, 1983). Decollectivization of agriculture in the late 1970s and the relaxation of migration controls since the mid-​1980s have resulted in large volumes of ‘temporary’ migrants to cities, many of whom belonged to what was called ‘rural migrant workers’ (nongmingong). This is a group of industrial and service workers with rural hukou working in cities. However, these labourers are not legally considered as urban workers, and are therefore not eligible for the regular urban welfare and rights that are available to any urban resident. Nor are they supposed to settle in their destination and make it their permanent home. The amount of rural migrant labour, having grown from about 20−30 million in the early 1980s to about 170 million at the end of 2014, is now prodigious (National Bureau of Statistics, 2015). For university graduates, there are now rudimentary national labour markets, although the hukou impediments to settling remains. In many ways, other factors of production are not ‘mobile’ either. This is especially so with regard to land in the rural areas, where transactions are formally forbidden and conversion of farmland to non-​agricultural uses by the farmers is stringently restricted. This spatial fixity of resources, ‘owned’ and controlled to a great extent by local governments, has many implications for the organization of the spatial economic, political, and social structures. Rural de-​collectivization also quickly gave rise to a large army of surplus rural labour, the root cause of China’s poverty in the last two-​to-​three centuries. More interestingly, the hukou system, a major instrument of that command system and, more broadly, the dual system, were refashioned in the post-​Mao era to serve China’s strategy of becoming the ‘world’s factory’ (Zhang, 2014; Lim, 2017). When China’s export-​processing industry gradually roared into high gear in the 1990s, the deployment of rural labour to the cities for the export industry became a major post-​Mao strategy; ironically, this time it was through unshackling labour from the rural collectives. By the mid-​1990s, rural hukou labour had become the backbone of the export industry and, more generally, the manufacturing sector. In coastal export-​oriented cities such as Shenzhen and Dongguan, migrant labour easily accounted for the great majority (70–​80%) of the labour force (Liang, 2001). Even for a more typical urban site, like the inland city of Wuhan, workers without local hukou accounted for 43 per cent of the employment in manufacturing in 2000 (Chan, 2009). Incomes of rural migrant labour have become an important part of peasants’ incomes, accounting for about 40 per cent of their average net incomes since the late 2000s (Caijing, 2009). In cities, in addition to the lack of access to many basic social services, these migrant workers also face many formal and informal obstacles to securing jobs other than low-​skilled

The Emerging Transformation of China’s Economic Geography    83 ones (Solinger, 1999; Li, 2003; Cai, 2007). The lack of local hukou for migrant workers, combined with other unfavourable conditions, such as the plentiful supply of labour and lack of access to legal information and support, created a huge class of super-​exploitable, yet highly mobile and flexible industrial workers for China’s new economy, closely integrated into global trade networks (Lee, 1998; Alexander and Chan, 2004). The ‘China price’, mainly due to its low labour costs, was the lowest among major developing countries (Chan and Ross, 2003). Many of these workers are vulnerable, and often subject to exploitation and labour abuses (Chan, 2001). Their ‘temporary’ nature and lack of local citizenship also made them easily expendable. The 2008–​09 global financial crisis seriously hit China’s export sector, leading to unemployment of about 20 million migrant workers (Chan, 2010a). The new approach of ‘freeing’ peasant labour has served very well China’s economic growth strategy of being the world’s low-​cost industrial producer. In essence, the persisting dual system has helped defer the arrival of the critical ‘Lewis turning point’,7 so that China can continue to draw labour from rural to urban areas and export-​processing zones at rural subsistence wage rates (Chan, 2010b).

Incomplete Urbanization and Under-​agglomeration of Cities China’s national urbanization strategy is very much part of its national industrialization and sectoral strategies. An ‘incomplete urbanization’ model has been adopted. This means that industrialization is achieved without the proportionate increase of the associated costs of social expenditure or ‘costs of urbanization’ (Ofer, 1977; Chan, 1994; Sjoberg, 1999). The most stringent measures to limit the growth of population are imposed on large cities, where the social expenditure per capita is usually the highest. In the pre-​reform era, China’s strategy favoured industry at the expense of agriculture, and gave priority to investment over consumption (Lin et al., 1996). An important part of that strategy was the differential treatment of the rural and urban sectors and their populations, as explained earlier. To maintain such an imbalance, migration to urban areas had to be and was strictly controlled through a web of regulations governing residence, employment, and social services. Urbanization, measured by the de facto population in urban centres, was maintained at a low level in the 1960s and 1970s. As a consequence, while China had high industrial growth rates from 1950 to 1980, the rate of urban population growth was comparatively low, leading to ‘incomplete urbanization’, also known as ‘under-​urbanization’ (Chan, 1994). In the reform era, the ‘incomplete urbanization’ approach still persists, but in a different form. There have been higher growth rates of the (de facto) urban population, especially through net rural–​urban migration, but a significant source of the urban population growth is in the form of migration of those without the hukou status of the destination (the so-​called ‘temporary’ or ‘non-​hukou’ migrants). This population segment, mostly from the countryside, is not eligible for the benefits that are normally available to urban residents. As pointed out earlier, the ‘temporary’ population or rural migrant labour is mostly not ‘temporary’, and its size is huge. In essence, China has adopted a strategy of letting migrants from the countryside into the cities and export-​processing zones to sell their labour at very low wages, but

84   Chan without giving them urban residence status and access to social services, thereby making their real wages much lower. Evidently, this has been a critical ingredient of China’s export competitiveness in the world economy. Table 4.1 uses three sets of statistics to illustrate the process of incomplete urbanization, as well as highlighting their different forms in the pre-​reform and reform eras. While the other two statistics are self-​explanatory, the urban hukou population refers to those who are eligible for welfare and social benefits provided by the state (Chan, 2007, 2009). The Chinese economy, in output (GDP) terms, has been industrialized relatively quickly in the last sixty years: the share of the non-​agricultural sectors rose from about 54 per cent in 1955 to about 90 per cent in 2010, but the population was only half urbanized. Within the urban population, the proportions of the urban population and urban hukou population were almost the same or very close in the period 1955–​80, indicating the congruence of these two population

Table 4.1 Urban Hukou Population, Urban Population and Gross Domestic Product (GDP) Shares, 1949−2014 (Percentage of National Total) Year

Urban Hukou Urban population population (B)b (A)a

1949

17.4

10.6

1955

15.2

13.5

53.7

–​38.5

1.7

1958

18.5

16.2

65.9

–​47.4

2.3

1965

16.7

18.0

62.1

–​45.4

–​1.3

1970

15.3

17.4

64.8

–​49.5

–​2.1

1975

15.4

17.3

67.6

–​52.2

–​1.9

1978

15.8

17.9

71.8

–​56.0

–​2.1

1980

17.0

19.4

69.8

–​52.8

–​2.4

1985

20.1

23.7

71.6

–​51.5

–​3.6

1990

21.1

26.4

72.9

–​51.8

–​5.3

1995

23.8

31.7

80.0

–​56.2

–​7.9

2000

26.1

36.2

84.9

–​58.8

–​10.1

2005

32.0

43.0

87.8

–​55.8

–​11.0

2010

34.3

49.7

90.4

–​56.1

–​15.4

2014

36.5d

54.8

91.8

–​55.3

–​18.3

a

GDP of non-​ agricultural sectors (C)c

D =A–​C

E =A–​B

6.8

Refers to feinongye (‘non-​agricultural’) hukou population, eligible for state-​provided, urban-​equivalent welfare, and social benefits. b Based on de facto population in urban areas in cities and towns. c Refers to the combined GDP share of the secondary sector and tertiary sector. d Author’s estimate based on 2012 figure. Sources: compiled by the author from Chinese Statistical Yearbook and China Population Statistical Yearbook (National Bureau of Statistics, various years).

The Emerging Transformation of China’s Economic Geography    85 groups in the pre-​1980 period (Table 4.1, last column). That is, the urban population and those who received state-​provided welfare were basically the same group; other groups were kept out of the cities. However, beginning in the mid-​1980s, these two numbers started to diverge, reaching a difference of 15.4 per cent (of China’s total population) in 2010 (see also Figure 4.1). This suggests that while more people are being allowed into the cities, an increasing proportion of the migrants are not eligible for state-​provided social welfare entitlement. As a result, China has a huge and expanding supply of ‘cheap’ urban–​industrial migrants. China’s ‘incomplete urbanization’ is also manifested at the individual city level. Migration restrictions in China limit the ability of labour to move permanently from low-​productivity locations to settle in high-​productivity ones; and, more generally, they limit the ability of the population to agglomerate at different points in space. Urban production is characterized by localized external economies of scale (Henderson, 1988), exploitation of which requires the population to move and agglomerate in high-​density cities. An issue with China’s cities is whether the restrictions on mobility have prevented population from agglomerating in cities in sufficient numbers to exploit fully scale externalities relevant to the local activity of the area. Tackling this issue requires estimation of how urban productivity varies with city size and an examination of what are efficient sizes for Chinese cities. A few earlier attempts have been made to measure the Chinese city size efficiency or the like.8 The most sophisticated studies thus far available are recent research by Au and Henderson (2006), and Li, X., et al. (2005). The former uses employment and other data from the China City Statistical Yearbooks for 1996 and 1997 for 212 prefecture-​level and above cities. They estimate city production functions and how productivity changes with city employment, controlling for variables such as industrial composition, investment, FDI, market potential of the city, and access to coastal markets.9 Based on their models, they reason that 43 per cent of the cities under study are below the 95 per cent confidence interval on the optimum city employment 60

% of Population

50 40 De facto urban population 30 20 Urban Hukou population 10 0 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

Figure 4.1   De Facto Urban Population and Urban Hukou Population, 1955–​2013. Sources: compiled by author from China Statistical Yearbook and China Population Statistical Yearbook (National Bureau of Statistics, various years).

86   Chan size. In other words, those cities are significantly undersized, and there would be gains in productivity from moving from the then city size to the optimum size. Li, X., et al. (2005) analyse data of a similar set of 202 prefecture-​level cities in 1990 and 2000 through an optimization model and argue that while the pure technical efficiency of most Chinese cities is high, the scale (city size) efficiency is low. Both of the aforementioned studies were carefully implemented, with consideration given to various possible tendencies of variables under study. But there are still questions about their numerical results, because the data they used have significant limitations for assessing efficient city sizes.10 Under-​or over-​counting city population or employment significantly affects the per capita efficiency indicators used in those studies. Because of the limitations of the data they use, neither study has truly addressed the city size efficiency issue in the most desirable way. However, the data problems do not refute the main thrust of the arguments in both studies: migration restrictions prevent many cities from growing to their optimum sizes and reaping the benefits of agglomeration economies. There is a rather large body of literature showing that similar industrial structures and duplication of infrastructures are found in many neighbouring cities in China (e.g. Liu, 1996; Shu and Zhou, 2003; Bai et al., 2004). Those studies are consistent with the arguments of both Au and Henderson (2005) and Li et al. (2005). Indeed, the jurisdictional system described earlier also allows individual jurisdictions to erect barriers that distort interregional or intercity flows of labour and goods in contravention of comparative advantage and economies of agglomeration, resulting in low efficiency. Alternatively, there are now reasonable data to demonstrate that China’s urban system displays a rather low agglomeration of population. Based on Chan’s work (2015) on China’s city population and data,11 Table 4.2 presents a statistical overview of China’s urban system in 1982–​2010. In 2010, about 40 per cent of the 666 million urban population was in cities over one million in size; another 30 per cent in towns, the smallest urban centre group; the group in between accounted for the remaining 30 per cent. A similarly even distribution or lack of concentration of the size distribution was maintained throughout the last three decades. The spatial Gini coefficient, which is a standard measure of the aggregate degree of geographical concentration, confirms this point (Table 4.3). For the world’s 1657 metropolitan areas with populations of over 200,000 in 2000, the spatial Gini was 0.56. For the same set of Chinese cities in the same size category, the Gini was only 0.33 in 2000 and 0.36 in 2010, way below the world’s, and compares with 0.52−0.65 for these large countries, namely, Brazil, Japan, Indonesia, UK, Mexico, Nigeria, France, India, Germany, the USA, and Spain. China’s relatively flat city-​size distribution is similar to those in former Soviet bloc countries (e.g. Russia’s Gini was 0.45 and Ukraine’s 0.40), reflecting a similar (previous) socialist system and development strategy. Based on historical data assembled by Chan (2015), Table 4.3 also shows that China’s low spatial concentration began to appear only after 1957, after the implementation of a socialist urbanization strategy. Another manifestation of China’s low agglomeration is that China has fewer very large cities, relative to its huge urban population, the world’s large (Chan et al., 2008; Henderson, 2009). A comparison of the world’s megacities (those with more than ten million people) and the size of the urban population in all the large countries (those with more than 100 million population) in Table 4.4 shows that China has the fewest megacities per 100 million urban population. China’s ratio is only about half of the average of all the

The Emerging Transformation of China’s Economic Geography    87 Table 4.2 Distribution of Cities, 1982–​2010 No. of cities City sizea

1982

1990

2000

2010

Urban population (millions)a

% of urban population of the nation

1982

1990

2010

1982

1990

2000

2010

2000

Provincial-​and prefecture-​level (PPL) citiesb > 5 million

3

3

7

14

17.1

21.4

58.6 127.3

8.0

7.2

12.8

19.1

1–​5 million

33

38

54

67

55.9

71.9

107.4 136.7

26.0

24.2

23.4

20.5

0.5–​0.99 million

31

46

75

99

20.9

31.8

52.1

70.5

9.7

10.7

11.3

10.6

0.2–​0.49 million

33

66

108

91

12.5

23.1

38.7

34.0

5.8

7.8

8.4

5.1

6

11

18

16

0.7

1.7

3.1

2.6

0.3

0.6

0.7

0.4

Total

106

164

262

287

107.1 149.9 259.9 371.1

49.8

50.5

56.6

55.7

PPL cities

115

188

263

287

120.0 187.2 259.9 371.1

55.9

63.1

56.6

55.8

County-​level cities

129

279

400

367

29.3

85.7 102.0

13.6

8.1

18.7

15.3

All cities

244

467

663

654

149.3

211.3 345.6 473.1

69.5

71.2

75.3

71.1

2,664 12,084 19,883 19,410

65.5

30.5

28.8

24.7

28.9

< 0.2 million

All towns National urban

24.1

85.3

113.5 192.5

214.8 296.6 459.1 665.6

100

100

100

100

Total may not add up due to rounding. aThe city size and urban population refer to the de facto population in the ‘urban areas’ (as defined by China’s National Bureau of Statistics, see Chan, 2007, 2015) in the beginning year of the period. bData for the PPL cities are approximates based on cities in the dataset used in Chan (2015). Source: Chan (2015).

other large countries. The population of China’s largest city, Shanghai, also accounts for a far lower percentage of the nation’s urban population and total population (only 3% and 1.6%, respectively) in 2014 than the averages (13% and 22%, respectively) of all the other large countries. The low spatial urban agglomeration is strongly indicative of some serious obstacles in the interjurisdictional mobility of factor of production in the economy. This problem is obvious, but it is not simply caused by the low mobility of labour to move permanently from low productivity locations to settle in higher one because of the hukou restrictions, and so on (Au and Henderson, 2006). The administrative region economy and the architecture of local governments within it, combined with the cadre evaluation system, also incentivize local governments to pursue protectionist measures to guard often parochial and short-​term interests within a small jurisdiction but at the expense of the broader regional or national, or longer-​term interests. Local economic protectionism is symptomatic of the Chinese administrative system, contributing to significant suboptimization and low levels of spatial agglomeration. Under that system, there are few incentives for local government to increase

88   Chan Table 4.3 Spatial Gini, 1957–​2010 1957

1982

1990

2000

2010

Cities, 200,000+

0.483

0.423

0.399

0.334

0.364

All cities

0.609

0.478

0.394

0.387

0.403

Population size

Number of cities Cities, 200,000+

382

455

523

All cities 173 244 631 Population of largest city (Shanghai) as a % of

80

172

666

654

Urban population

6.1

2.9

2.7

2.9

3.0

National population totals

0.9

0.6

0.7

1.1

1.5

Source: Chan (2015).

the hukou population but a lot of incentives to expand urban land under the local government’s jurisdiction (Wang, 2012; Yew, 2012; Tian, 2015). China’s low urban concentration, an outcome of various government policies to control the population growth of large cities, is far more attributable to the systematic features outlined earlier than the political stability considerations, as suggested by Wallace (2014). In summary, from an economic spatial perspective, the story of Chinese city size distribution is that there are too many cities, and many of them are too small to take advantage of agglomeration economies and to develop a higher level of functional specialization among cities. Large cities in China have been rapidly expanding the service sector, especially producer services, in the last ten years or so. Technological change in urban production, as is ongoing in China at a high level, increases efficient city sizes, as do improvements in organization of land markets and improved land use patterns (Black and Henderson, 2003). As bigger cities become more business service-​oriented, their efficient sizes should increase, as business services experience higher degrees of scale externalities than do manufacturing activities. It appears that there is ample room for big Chinese cities to benefit from the scale externalities in this respect. Moreover, China’s bigger cities are still heavily engaged in manufacturing. As they become more specialized in higher-​tech and high value-​added sectors, they will also enjoy greater benefits from agglomeration. For smaller urban centres, including many well-​developed towns, there is also insufficient agglomeration, partly because of the dispersed (rural) industrialization policy and the incentive system inherent in the jurisdictional system (Li and Li, 2005), and partly because of the significant policy biases against them in infrastructural investment, fiscal resources, and access to capital, as implied in China’s administrative hierarchy. Many of them cannot develop to their full capacity; nor can most of them agglomerate a large-​enough population to foster healthy growth of the service sector. The smaller urban centres are too many in number and too dispersed.

Nation

Total population (millions)

Urban percentage

Urban No. of population megacities (millions)

No. of Largest city megacities per 100 million urban population

Population of Population of the largest city largest city as (millions) % of nation’s population

China

Population of largest city as % of urban population

1394

54

758

5

0.66

Shanghai

23

1.6

3.0

Indonesia

253

53

134

1

0.75

Jakarta

30

11.9

22.4

USA

323

81

263

2

0.76

New York

19

5.9

7.2

Russia

142

74

105

1

0.95

Moscow

16

11.3

15.2

Mexico

124

79

98

1

1.02

Mexico City

21

16.9

21.4

Brazil

202

85

172

2

1.16

Sao Paulo

21

10.4

12.2

Pakistan

185

38

71

1

1.41

Karachi

16

8.6

22.5

India

1267

32

410

6

1.46

Delhi

24

1.9

5.9

Japan

127

93

118

2

1.69

Tokyo

38

29.9

32.2

Bangladesh

159

33

53

1

1.89

Dhaka

17

10.7

32.1

Philippines

100

45

45

1

2.25

Manila

22

22.0

49.4

12.9

22.1

Unweighted mean (excluding China)

1.33

Population size of the cities refers to the metropolitan area. Data on the total and urban populations are from United Nations (2014). The number of China’s megacities is estimated by the author (see Chan, 2015). Others are based on Kotkin (2014), with adjustment of Karachi’s based on United Nations (2014) and Forstall et al. (2009).

The Emerging Transformation of China’s Economic Geography    89

Table 4.4 Number of Megacities and Urban Population of Large Countries, 2014

90   Chan

Concluding Discussion Future urbanization and urban growth will continue to pose serious challenges to the Chinese policymakers. Even assuming a modest urban growth rate of 2.5 per cent per year, the large Chinese urban population base will mean that in the next ten years, more than 200 million people will be added to urban centres of various sizes in the country. This will generate an enormous demand for urban jobs and urban social infrastructural services. So far, China has been able to get away with not paying the ‘full bill’ for industrialization by means of its ‘incomplete urbanization’ approach based on a dual society, that is, institutionally excluding migrants from urban benefits and programmes. Evidently, as experience elsewhere has shown, this cannot be sustainable: unassimilated migrants (often undereducated, especially young people) are much more costly in social and political terms in the longer run. China’s existing model runs a high risk of breeding a huge urban underclass (Solinger, 2006). Hundreds, if not thousands, of protests lodged by peasants and migrants over an array of issues in the last decade also raise the alarm that the rights awareness of the rural population and migrants is on the rise. Those can no longer be easily ignored. Crucial to the assimilation and acceptance of migrants as equals are the hukou reforms and the capacity to provide reasonable employment for them. On the employment side, given that a large proportion of urban population growth in the coming one to two decades will be from the countryside, it is necessary to continue focusing on a job-​oriented development strategy in order to generate positions suited to the skill levels of migrant labourers and provide training for them so that they can match demand as the economy evolves. This point has been made more obvious by the vulnerability of migrant labour in the global recession in 2008 and 2009, when some 20 million migrant labourers lost their jobs without any unemployment compensation protection (some even without full payment of the wages they had earned) (Chan, 2010c). In normal times, if migrants can find reasonable jobs and can compete equally in the urban labour market, they will also have the wherewithal to finance the expansion of some of the urban social services, hence reducing the political resistance of the existing urban local population to accepting more migrants in the cities (Cai and Chan, 2000; Cai et al., 2005). As migrant labour has been the main workhorse of the Chinese export-​processing industry, and migrants are an indispensable part of China’s industrialization and the economy of large cities, the importance of maintaining a stable migrant labour force and converting the experienced and skilled ‘temporary’ migrant workers into ‘permanent’ citizens is clear. Some cities have made small progress in easing restrictions on mobility for some of the more sought-​after types of labour (mainly college-​educated and professionals). Similar steps towards gradual reforms could be taken to help keep skilled migrant workers settled in cities. This would be a largely win–​win situation for both parties (Chan, 2014). At the more fundamental level, China needs to abandon its ‘incomplete urbanization’ or ‘industrialization on the cheap’ approach and consider truly ‘uprooting’ the rural population who are willing to resettle in urban centres, and giving them equal access to urban benefits, and so on. China’s development of a national high-​speed rail system in the last few years has helped to increase agglomeration economies of cities. However, according to the analysis by Bosker et al. (2015), there remain large potential of productivity gains locked by obstacles created

The Emerging Transformation of China’s Economic Geography    91 by institutions such as the hukou system. China’s urbanization policy all along has been to control the population size of the large cities; this policy has been accentuated by the latest hukou reform plans (Chan, 2014). According to the estimate made by Yukon Huang, a former World Bank chief in China, the economic loss due to the current urbanization (limiting growth of big cities) policy amounts to about one percentage point of China’s GDP (Hamlin, 2014). Combes et al. (2015) have also shown the potential of significant gains by free migration and greater agglomeration. The policy of controlling urbanization in large cities and channeling migrants to smaller cities, counter-​intuitively, actually contributes to increasing inequality. The policy is also counter to the goal of rebalancing the economy to increase the share of consumption in the GDP (Wan and Cai, 2012; Chen and Lin, 2013; Lu and Wan, 2014). Reforming the hukou system will foster true geographical mobility of labour and reduce the lack of spatial agglomeration in its urban system. From an economic efficiency perspective, there are too many urban centres in China, and most of them are too small in population size to benefit from agglomeration. Under the current administrative jurisdictional system and hukou system, the size of the permanent population of each city is relatively rigid, and any possible changes are limited. With freer migration and an economy where the main duties of governments are limited largely to providing social goods, more competitive cities will experience accelerated growth, and less competitive ones will face depopulation. This will help to create a system of fewer cities but ones which have greater variations in size, which will better fit China’s increasingly diverse regional and local conditions and residents’ preferences. Seen from this angle, the latest policy of controlling the population growth of China’s megacities is counter-​productive, a step in the wrong direction. Ultimately, the agent for achieving agglomeration effects in the economy should be the market, not the government—​in other words, companies and urban and rural residents (workers and consumers), rather than government policies, should be the driver of urbanization. That is because officials do not possess all the relevant information; they do not know which companies will do better in large or small cities and thus where they should be located. Neither do officials know if a certain migrant is better suited in a large city or small city for the simple reason that the specific conditions of each individual are different. In the end, such decisions should be left to companies and workers themselves based on market conditions, not determined by a one-​size-​fits-​all policy to control the population of very large cities and encourage migrant workers to settle in small cities and towns (Meng, 2015). Reforms of the local government systems are equally fundamental. This will involve changing major elements of the legacy of the command economy: the state-​run economy, the top-​down systems of hierarchical administrative jurisdictions and evaluations, and the associated incentive system for local bureaucrats. Local governments and officials should be incentivized to do a good job of taking care of the provision of public goods (environment, social services for local population, etc.), instead of directly running the economy. There must also be input from local citizens in assessing local officials. This will help correct several major distortions in the local urban economies identified in this chapter and help China create a more desirable geographical configuration of the government and the economy. The present analysis also clearly demonstrates a greater need for more research, as there are many gaps and some confusion in the urbanization literature. One important aspect is that the current Chinese urban definition and application, while they are useful for studies at the national level, are quite problematic at the individual city level, especially as metropolises

92   Chan develop and sprawl to encompass low-​density suburbs. This has prevented real progress in studying Chinese urban systems in any meaningful way, as has been pointed out before (Chan, 2007). More work is urgently needed to design a conceptually sound and operationally workable system of city population statistics in the special context of China. There is also a clear need to analyse more closely the interrelationship among urbanization/​urban growth, the changes in the system of local governments, and local development strategies. I believe that this is where one can write the most interesting story of China’s real emerging political, economic, and urban geography.

Notes 1. The government today still controls many aspects of people’s life, sometimes in a highly intrusive manner. For example, in implementing the family-​planning policy, some cities in Guangdong, until recently, would grant registration of a newborn’s local hukou only upon the provision of proof of sterilization of the mother (Southern.com, 2015). 2. Since the early 1980s, several other special status cities have also been established. See Chan (1997). 3. Given the top-​down configuration of power, local jurisdictions always have incentives to move up the administrative ladder: to be upgraded to a higher administrative rank. Rapid economic growth in some locales in the reform era has allowed many units to seek urban designations and ‘upgrades’, resulting in significant increases in the number of especially prefecture-​level and county-​level cities in the 1980s and 1990s (Chung and Lam, 2003). 4. See an examination of environment and labour aspects in Dongguan, one of the cities that has recorded the highest GDP growth rates in China in the last twenty years (Yeung, 2001). 5. China’s ‘internal borders’ created by the hukou system have been compared with the national borders within the European Union (Pasquali, 2015). 6. In addition to its general deleterious effects on agriculture, the forced collectivization and industrialization programmes of the late 1950s contributed significantly to a famine that resulted in at least 20 million deaths in the early 1960s. 7. This is named after economist Arthur Lewis, a Nobel laureate. According to Lewis, developing countries’ industrial wages begin to rise quickly at that point when the supply of surplus labour from the rural areas tapers off. Wang (2005) has stated that the hukou system enabled China to bypass the Lewis turning point. The author of this chapter has argued that the hukou system has allowed China to extend the period before reaching the turning point (Chan, 2010b). 8. Some earlier studies include Perkins (1990) and Chang and Kim (1994). Many of them suffer from serious data problems, as pointed out by Chan (2007). 9. In their model, output is value-​added per worker in the non-​agricultural sector of the city proper. Determinants include the capital stock to labour ratio, share of accumulated FDI in capital stock, distance to the coast, education, and scale measures. 10. A check of Au and Henderson’s (2006) sources with the available 2000 census data reveals that the city employment data contained in the China City Statistical Yearbooks undercount the true city employment, most probably by excluding a significant number of workers engaged in informal employment, especially the employment of rural migrant labour (Cai and Chan, 2009). Li and Li (2005) used census population data to produce

The Emerging Transformation of China’s Economic Geography    93 the best set of statistics possible then, but the census city district-​based population data greatly over-​count the true ‘city’ population in many instances (see Chan, 2007). 11. The definitions of ‘urban population’ and ‘city population’ in China are quite complicated. Here I refer to the de facto population in the ‘urban areas’ defined by China’s National Bureau of Statistics. For details, see Chan (2007, 2015).

References Alexander, P. and Chan, A. (2004). ‘Does China have an apartheid pass system?’ Journal of Ethnic and Migration Studies 30: 609–​629. Au, C.-​C. and Henderson, J.V. (2006). ‘How migration restrictions limit agglomeration and productivity in China’. Journal of Development Economics 80: 350–​388. Bai, C., Du, Y.,Tao, Z., and Tong, S.Y. (2004). ‘Local protectionism and regional specialization: evidence from China’s industries’. Journal of International Economics 63: 397–​417. Black, D. and Henderson, J.V. (2003). ‘Urban evolution in the USA’. Journal of Economic Geography 3: 343–​372. Bosker, M., Deichmann, U., and Roberts, M. (2015). ‘Hukou and highways:  the impact of China’s spatial development policies on urbanization and regional inequality’. World Bank Policy Research Working Paper 7350, June 2015. Cai, F. (2007). Zhongguo liudong renkou wenti (Question on China’s Floating Population) (Beijing: Shehui Kexue Wenxian Chuban She). Cai, F. and Chan, K.W. (2000). ‘The political economy of urban protectionist employment policies in China’. Working Paper Series No. 2, Institute of Population Studies, Chinese Academy of Social Sciences. Cai, F. and Chan, K.W. (2009). ‘The global economic crisis and unemployment in China’. Eurasian Geography and Economics 50: 513–​531. Cai, F., Du, Y., and Wang, M. (2005). Zhongguo laodongli shichang zhuanxing yu fayu (Transition and Development of China’s Labor Markets) (Beijing: Shangwu Chubanshe). Caijing (2009). ‘Nongmingong shiye diaocha’ (‘Survey of unemployment of rural migrant labor’). http://​www.caijing.com.cn/​ (last accessed 1 July 2017). Cartier, C. (2015). ‘Territorial urbanization and the party-​state in China’. Territory, Politics, Governance 3: 294–​320. Chan, A. (2001). China’s Workers under Assault:  The Exploitation of Labor in a Globalizing Economy (Armonk, NY: M. E. Sharpe). Chan, A. and Ross, R.J.S. (2003). ‘Racing to the bottom: industrial trade without a social clause’. Third World Quarterly 24: 1011–​1028. Chan, K.W. (1994). Cities with Invisible Walls: Reinterpreting Urbanization in Post-​1949 China (Hong Kong: Oxford University Press). Chan, K.W. (1997). ‘Urbanization and urban infrastructure services in the PRC’ in C. Wong (ed.) Financing Local Government in the People’s Republic of China, pp. 83–​125 (New York: Oxford University Press). Chan, K.W. (2007). ‘Misconceptions and complexities in the study of China’s cities: definitions, statistics, and implications’. Eurasian Geography and Economics 48: 383–​412. Chan, K.W. (2009). ‘The Chinese Hukou system at 50’. Eurasian Geography and Economics 50: 197–​221.

94   Chan Chan, K.W. (2010a). ‘Fundamentals of China’s urbanization and policy’. The China Review 10: 63–​94. Chan, K.W. (2010b). ‘A China paradox: migrant labor shortage amidst rural labor supply abundance’. Eurasian Geography and Economics 51: 513–​530. Chan, K.W. (2010c). ‘The global financial crisis and migrant workers in China: there is no future as a labourer; returning to the village has no meaning,’ International Journal of Urban and Regional Research 34: 659–​77. Chan, K.W. (2014). ‘China’s urbanization 2020:  a new blueprint and direction’. Eurasian Geography and Economics 55: 1–​9. Chan, K.W. (2015). ‘China’s city system, changes, and policy: an analysis of the census data, 1982–​2010’, Asian Development Bank, unpublished policy paper. Chan, K.W., Henderson, V., and Tsui, K.Y. (2008). ‘Spatial dimensions of Chinese economic development’ in L. Brandt and T. Rawski (eds) China’s Great Transformation, pp. 776–​828 (Cambridge: Cambridge University Press). Chang, S.-​D. and Kim, W.B. (1994). ‘The economic performance and regional systems of China’s cities’. Review of Urban and Regional Development Studies 6: 58–​77. Chen, B. and Lin, Y. (2013). ‘Fazhan zhanlue, chengshihua yu zhongguo chengxiang shouru chaju’ (‘Development strategy, urbanization and the rural–​urban income disparity in China’). Zhongguo Shehui Kexue (Social Science in China) 4: 81–​102. Chung, J.H. and Lam, T.-​C. (2003). ‘China’s “city system” in flux: explaining post-​Mao administrative changes’. The China Quarterly 180: 945–​964. Combes, P.-​P., Démurger, S., and Li, S. (2005). ‘Migration externalities in Chinese cities’. European Economic Review 76: 152–​167. Forstall, R.L., Greene, R.P., and Pick, J.B. (2009). ‘The world’s largest cities? Why so little consensus?’ Tijdschrift voor Economische en Sociale Geografie, 100: 277–​297. Hamlin, K. (2014). ‘The $2 trillion megacity dividend China’s leaders oppose: cities,’ Bloomberg Business, 19 October http://​www.bloomberg.com/​news/​articles/​2014-​10-​19/​the-​2-​trillion-​ megacity-​dividend-​china-​s-​leaders-​oppose-​cities (last accessed 21 October 2014). Henderson, J.V. (1988). Urban Development:  Theory, Fact, and Illusion (Oxford:  Oxford University Press). Henderson, J.V. (2009). ‘Urbanization in China: policy issues and options’. China Economic Research and Advisory Programme http://​www.econ.brown.edu/​Faculty/​henderson/​ Final%20Report%20format1109summary.doc (last accessed 6 March 2017). Huang, Y. (2008). Capitalism with Chinese Characteristics:  Entrepreneurship and the State (Cambridge: Cambridge University Press). Kotkin, J. (2014). Problem of MegaCities (Orange, CA: Chapman University Press). Lardy, N. (1983). Agriculture in China’s Modern Economic Development (New York: Cambridge). Lee, C.K. (1998). Gender and the South China Miracle (Berkeley, CA: University of California Press). Li, Q. (2003). ‘Yingxiang woguo chengxiang liudong renkou de tuili yu lali yinsu fenxi’ (‘Factors affecting the push and pull of migration in China’). Zhongguo Shehui Kexue 1: 125–​136. Li, S., Hou, Y., Liu, Y., and Chen, B. (2005). ‘Zhongguo guonei defang baohu de tiaocha baogao’ (‘An Investigation Report of Local Protection in China’) in H. Ma and M. Wang (eds) Zhongguo fazhan yanjiu 2005 (China Development Studies 2005), pp. 93–​105 (Beijing: Zhongguo Fazhan Chubanshe). Li, X. and Li, Y. (2005). ‘Nongcun jiti suoyouzhi yu fensanshi nongcun chengshihua kongjian’ (‘Rural collective system and dispersed urbanized space in the countryside’). Chengshi guihua (City Planning Review) 29: 39–​42.

The Emerging Transformation of China’s Economic Geography    95 Li, X., Xu, X., and Chen, H. (2005). ‘20 shiji 90 niandai zhongguo chengshi xiaolü de shikong bianhua’ (‘Spatial and temporal change of urban efficiency in China in the 1990s’). Dili xuebao 60: 615–​625. Liang, Z. (2001). ‘The age of migration in China’. Population and Development Review 27: 499–​524. Lim, K.F. (2017). ‘On the shifting spatial logics of socioeconomic regulation in post-​1949 China’. Territory, Politics, Governance 5: 65–​91. Lin, G.C.S. (1997). Red Capitalism in South China: Growth and Development of the Pearl River Delta (Vancouver: UBC Press). Liu, J. (1996). Zhongguo xingzheng quhua de lilun he shijian (Theory and Practice of the Planning of Administrative Regions in China) (Shanghai: Huadong shifan daxue chubanshe). Liu, J. (2010). ‘Regional cooperation in China’s administrative region economy’ in A. Yeh and J. Xu (eds) China’s Pan-​Pearl River Delta, pp. 63–​77 (Hong Kong: Hong Kong University Press). Lin, J.Y., Cai, F., and Li, Z. (1996). The China Miracle:  Development Strategy and Economic Reform (Hong Kong: The Chinese University Press). Lu, M. and Wan, G. (2014). ‘Urbanization and urban systems in the People’s Republic of China’. Journal of Economic Surveys 28: 671–​685. Ma, L.J. (2005). ‘China’s changing urban administrative system: spatial restructuring and local economic development’. Political Geography 24: 477–​97. Meng, X. (2015). Harnessing China’s Untapped Labor Supply (Chicago, IL: Paulson Institute). National Bureau of Statistics (NBS) (various years). Zhongguo tongji nianjian (China Statistical Yearbook) (Beijing, China: Zhongguo tongji chubansh). National Bureau of Statistics (2015). ‘2014 nian quanguo nongmingong jiance diaocha baogao (‘A monitoring and investigative report of China’s rural migrant labor in 2014’) http://​www. stats.gov.cn/​tjsj/​zxfb/​201504/​t20150429_​797821.html (last accessed 30 April 2015). Naughton, B. (2007). The Chinese Economy: Transitions and Growth (Cambridge, MA: MIT Press). Ofer, G. (1977). ‘Economizing on Urbanization in Socialist Countries: Historical Necessity or Socialist Strategy?’ in A.A. Brown and E. Neuberger (eds) Internal Migration: A Comparative Perspective, pp. 277–​303 (New York: Academic Press). Pasquali, P. (2015). ‘Borders of migration: a comparative legal perspective between EU and China’. China-​EU Law Journal DOI: 10.1007/​s12689-​014-​0046-​8. Perkins, D.H. (1990) ‘The Influence of Economic Reforms on China’s Urbanization’ in R.Y.-​W. Kwok and G. Blank (eds) Chinese Urban Reform: What Model Now?, pp. 78–​106 (New York: M. E. Sharpe). Shu, Q. and Zhou, K. (2003). Cong fengbi zouxiang kaifang (From Closed Doors to Openness) (Shanghai: Huadong shifan daxue chubanshe). Sjoberg, O. (1999). ‘Shortage, priority and urban growth:  towards a theory of urbanisation under central planning’. Urban Studies 36: 2217–​2236. Solinger, D. (1999). Contesting Citizenship in Urban China (Berkeley, CA:  University of California). Solinger, D. (2006). ‘The Creation of a New Underclass in China and Its Implications’. Environment & Urbanization 18: 177–​193. Southern.com (2015). ‘Dongguan liangbumen lianhe fawen chongshen ruhu buyu shanghuan jieza guagou’ (‘Dongguan’s two departments issued a joint statement that registering hukou does not need installation of IUD and ligation’) http://​dg.southcn.com/​content/​2015-​07/​16/​ content_​128580883.htm (last accessed 18 July 2015). Tang, A.M. (1984). ‘An analytical and empirical investigation of agriculture in mainland China, 1952–​1980’. Economic Series No.4 (Taipei: Chung-​Hua Institute for Economic Research).

96   Chan Tian, F. (2015). ‘Evolution and Reform of China’s Hukou System’ in K. Göymen and R. Lewis (eds) Public Policymaking in a Globalized World, pp. 185–​201 (Istanbul: Istanbul Policy Center at Sabanci University). Tien, H.Y. (1973). China’s Population Struggle (Columbus, OH: Ohio State University Press). Tsui, K.Y. and Wang, Y. (2004). ‘Between separate stoves and a single menu: fiscal decentralization in China’. The China Quarterly 177: 71–​90. United Nations (2014). World Urbanization Prospects:  The 2014 Revision (Washington, DC: United Nations). Wallace, J. (2014). Cities and Stability: Urbanization, Redistribution, & Regime Survival in China (Oxford: Oxford University Press). Wall Street Journal (2015). ‘Unraveling the China puzzle’. 8 July http://​www.wsj.com/​articles/​ unraveling-​the-​china-​puzzle-​1436397000 (last accessed 8 July 2015). Wan, G. and Cai, F. (2012). ‘Zhongguo jingji mianlin de tiaozhan yu chengshihua’ (‘Challenges and Urbanization Faced by China’s Economy’) in G. Wan and F. Cai (eds) Zhongguo de chengshihua daolu yu fazhan zhanlue:  lilun tantao he shizheng fenxi (Urbanization and Development in China:  Challenges, Prospects and Policies), pp. 3–​14. (Beijing:  Economic Science Press). Wang, F. (2005). Organizing Through Division and Exclusion: China’s Hukou System (Stanford, CA: Stanford University Press). Wang, X. (2012). ‘Zhongguo chengshihua lujing yu chengshi guimo de jingjixue fenxi’ (‘Economic Analysis of the Road of China’s Urbanization and City Size’) in G. Wan and F. Cai (eds) Zhongguo de chengshihua daolu yu fazhan zhanlue: lilun tantao he shizheng fenxi (Urbanization and Development in China: Challenges, Prospects and Policies), pp. 17–​35. (Beijing: Economic Science Press). Whiting, S. (2001). Power and Wealth in Rural China: The Political Economy of Institutional Change (Cambridge: Cambridge University Press). Wong, C. (ed.) (1997). Financing Local Government in the People’s Republic of China (New York: Oxford University Press). Yang, D. and Cai, F. (2003). ‘The Political Economy of China’s Rural–​urban Divide’ in N. Hope, D. Yang, and M. Yang (eds) How Far Across the River? Chinese Policy Reform at the Millennium, pp. 389–​416 (Stanford, CA: Stanford University Press). Yeung, G. (2001). Foreign Investment and Socio-​economic Development in China: The Case of Dongguan (London: Palgrave). Yew, C.P. (2012). ‘Pseudo-​Urbanization? Competitive government behavior and urban sprawl in China’. Journal of Contemporary China 21: 281–​298. Zhang, J. (2014). ‘Global Economic Crisis and the “Spatial Fix” of China’s World Factory: The Great “Long March” Inland’ in Y. Atasoy (ed.) Global Economic Crisis and the Politics of Diversity, pp. 132–​154 (London and New York: Palgrave Macmillan).

Chapter 5

E c onomic Grow t h a nd P ov ert y Redu c t i on i n C ontem p ora ry I ndia Stuart Corbridge Introduction There are very many ‘rules’ in social science to which India stands as an interesting and salutary exception. Democracies generally do not survive long at low average levels of per capita income (Przeworski et al., 1996). Except in India, save for the ‘Emergency Period’: 1975–​77. Countries with extremely high levels of ethnic fractionalization do not tend to hold together (Fearon and Laitin, 2003). Except in India since 1947. Rich urban men are more likely to vote in major elections than poor, rural women. Except recently in India (Banerjee, 2014). And countries with high levels of corruption and poor governance regimes do not tend to support long-​term economic growth (Spence, 2011). Except in India, which has sustained annual rises now in gross domestic product (GDP) every year since a steep contraction in 1979–​80, and which among large developing countries has sustained per capita GDP growth rates over the same period second only to China. India has maintained these high rates of growth but without reducing rates of extreme income poverty anything like as effectively as China or other fast-​developing countries in Asia and Latin America (Winters and Yusuf, 2007; Chant and McIlwaine, 2009; Sharma, 2009). This chapter addresses two key puzzles in India’s recent economic development: the mainsprings of its growth transition(s), and the country’s failure to promote growth that is sufficiently inclusive in social and spatial terms to secure reductions in extreme income poverty on a par with its peers. The next section reviews the long-​term drivers of economic growth. It considers the competing roles of Geography (as understood by Jeffrey Sachs and colleagues) and institutions (the formal and informal rules of the game). An account is developed to explain sustained economic growth in India post-​1980 in the absence of significant improvements in the country’s Geography or institutions. This account is rooted in a relational account of Geography and spatial competition. This is very different to the fixed-​effect model of geography

98   Corbridge developed by Sachs. The model developed here also pays attention to historical path dependencies to explain why stable economic policies over a period of decades can have larger-​ than-​expected effects on economic growth. The focus is on latent or underlying institutional quality. The section ‘Poverty Reduction, Inequality, and Political–​Economic Geography’ considers why the transmission mechanism linking economic growth to poverty reduction in India has been weak. Geography is, again, central to the argument. Economic growth in the east and centre-​north of India has been undercut over the last thirty years by successive rounds of underinvestment in state capacity. It has also been harmed by systems of political calculation that made investments in security and growth seemingly unnecessary for incumbent re-​election. Happily, there have been signs recently that growth is being re-​ ignited in some parts of eastern India by dynamic political entrepreneurs. In other areas there is a continuing low-​level civil war between state forces and the Maoists. Growth and poverty reduction remain unlikely in this conflict zone. The reasons for and significance of these developments are considered in a brief conclusion.

Geography, Institutions, and the Drivers of Economic Growth Economists generally assume that the proximate drivers of growth are physical capital deepening, human capital formation, and improvements in productivity. They also recognize that these drivers are both causes and consequences of economic growth, which is to say they are largely endogenous in standard growth models. Countries with high levels of GDP per capita have resources to invest in schools and education and high-​tech city parks. Their middle classes also demand these investments. Accordingly, the deeper drivers of economic growth are generally thought to reside in some combination of greater international integration, improved institutional quality, or improvements to a country’s Bad Geography (Rodrik, 2003). The Geography thesis has been advanced with particular vigour by Jeffrey Sachs (Sachs et al., 2001; Sachs, 2005). In its simplest version the thesis holds that rich and poor countries divide geographically between temperate and tropical areas. Whether we look at GDP per capita (adjusted for purchasing power parities (PPPs)), or patents issued, Sachs insists that key regional patterns of income and innovation map out as distance decay functions from zero latitude. Fundamentally, countries close to the equator pay a price for their Bad Geographies. Tropical climate systems are extreme and give rise both to poor soil systems (notably the red soils or laterites) and damaging human and animal disease ecologies (whether malaria, schistosiamsis, or trypanosomiasis). In some parts of the tropics, too, and most notably in Africa, countries can be landlocked. In these cases, the penalty of Bad Geography is compounded. Of course, Sachs is far more a Promethean than an environmental determinist. Sachs has consistently maintained that rich countries have a moral duty and an economic interest in transferring over $150 billion annually to poor tropical countries in the form of official development assistance. Foreign aid can be used to overcome local poverty traps by funding road and infrastructure projects, malaria eradication schemes, and so on. However, Sachs’s

Economic Growth and Poverty Reduction in Contemporary India    99 critics—​and there are many—​insist that financial transfers on this scale are likely to be misused by local political and economic elites (Easterly, 2006). They also maintain that the bad geography thesis makes a number of category errors (Acemoglu and Robinson, 2012). Singapore is tropical but thriving. Switzerland is landlocked but has good neighbours. And so on. While it is attractive in a modelling sense to have a wholly exogenous variable (geography) driving economic growth, most mainstream social scientists insist that the problems of poor tropical countries have to do mainly with history, or the political settlements that were produced as a result of colonialism. Particularly in tropical Africa, where white colonialists did not settle in large numbers and where the preference was for extractive regimes serviced by local compradors, the legacies of colonial rule tend to be felt still in a relative absence of manufacturing capital, stark divisions between cities and the countryside, low levels of literacy and life expectancy, and political systems devoid of meaningful social contracts (Mamdani, 1996; Besley and Persson, 2011). The key development need of these countries is to change the formal and informal rules of the game. Current institutions provide few incentives for long-​run economic accumulation or the building of democratic systems of rule and accountability. In the words of Rodrik et al. (2004, p. 135), summing up a large body of work based on cross-​country regressions, ‘Once institutions are controlled for, integration has no direct effect on incomes, while geography has at best weak direct effects’. Bill Easterly and Robert Levine (2003) have pushed the ‘institutions matter’ line even further. They maintain that macroeconomic policies do not have a significant effect on incomes once the quality of institutions is controlled for. If true, this is a galling finding for the aid industry and the global political elite. In a country like the UK it provides the basis for a sharp retort to those who say that the UK should challenge the power of the financial industry and become ‘more like Germany’. Well, yes, but only if that is a meaningful proposition. Even in the wake of the global financial crisis the willingness of the UK’s political elites to challenge the hegemony of finance capitalism has been minimal. The contrary suggestion of Easterly and Levine is that countries have the institutions they have for a reason: the weight of history and the formation of political classes who benefit from them. In this light, institutional change is either rather slow or it occurs mainly through revolutions or war: the USA in 1776, France in 1789, Japan in 1868, China in 1949 and again in 1976, and so on. Absent such ruptures there is unlikely to be great change in the international league table of economic growth performance. But here, too, India stands out as an exception. It is significant that a focus on India is rarely developed in either the strong Bad Geography or institutionalist models of comparative economic performance. Sachs might reasonably maintain that the Gangetic plain areas of India are north of the Tropic of Cancer, although that would hardly help his broader argument: northern India tends to be poorer than southern (or tropical) India. And India simply does not fit in the model of colonial path dependencies developed in a brace of famous papers by Acemoglu, Johnson, and Robinson (AJR model) (2001, 2002). Although their research paradigm has been developed by authors who have tried to measure the effects on growth of different colonial rulers (British, French, Spanish, for instance (Lange and Mahoney, 2006; Fielding and Torres, 2008)), or particular colonial policies (education or health care (Bolt and Bezemer, 2009)), the core AJR model is developed around a binary typology of extractive regimes (e.g. the Belgian Congo) or Neo-​Europes (the white settler colonies of the UK: the USA, Canada, Australia, and New Zealand) that finds no place for a country like India. Even more significantly, the performance of the Indian economy post-​1980 has

100   Corbridge been remarkably robust—​notwithstanding any evident positive changes in the quality of the country’s geography or institutions. Consider the former. Even on a less ‘physical/​climatic’ reading of Geography than is presented by Sachs, it is far from clear that improvements in India’s transport infrastructures drove an upwards shift in the country’s economic growth. Major improvements in India’s air and road systems—​including the Golden Quadrilateral linking Delhi, Kolkata, Chennai and Mumbai—​have mainly occurred post-​2000, twenty years after the Indian economy started to grow at consistently higher rates. And then consider institutions. Economists like to proxy the quality of institutions with reference to the strength of the rule of law or property rights protection. This makes sense: a high risk of kidnap, murder, or expropriation is unlikely to encourage entrepreneurship or long-​term investment. But most serious studies of the quality of India’s institutions post-​c.1980 do not suggest that the rule of law has significantly improved over time. Nor has the quality of the country’s politicians, rather too many of whom have spent time in jail or have been charge-​sheeted with one serious offence or another. Nor too has India climbed the ‘corruption’ league tables compiled by Transparency International. Levels of corruption have now become so high in the country that major political movements have sprung up post-​2000 to articulate concerns about the persistent misuse of public office for private gain (Sharma, 2014). And yet what remains undeniably true is that the Indian economy grew after 1980 in a way that it did not previously, and least of all from c.1965–​79 (see Figure 5.1). India’s GDP grew at an average annual rate of 3.7 per cent from 1950–​51 to 1979–​80, and at over 6 per cent annually from 1980–​81 until the time of the global economic crisis in 2008–​09. Throughout this period the economy grew year on year: the last time it contracted was in 1979–​80 (by more than 5%). This is extraordinary—​and was wholly unexpected. Over the period 1980–​2003 the Indian economy grew faster than all of its rivals save for China, Botswana, South Korea, Singapore, and Oman. The economy also recorded two significant upturns. According to Hausmann et al. (2005, p. 305) there were just eighty-​three growth accelerations in the global economy from 1950 to 1982. They define a growth acceleration as an increase in per capita

9 –2 80

97

19

–1 50 19

00

9

9 –0 00 20

90

–9

9

9 19

9

–8 80 19

9

–7 70 19

–6 60 19

19

50

–5

9

8 7 6 5 4 3 2 1 0

Year GDP

Per Capita GDP

Figure 5.1   India’s Long-​term Growth, 1950–​2009. GDP, gross domestic product.

Economic Growth and Poverty Reduction in Contemporary India    101 growth of two percentage points or more, where the increase in growth has to be sustained for at least eight years and the post-acceleration growth rate has to be at least 3.5 per cent per year. In these terms, India has had two growth accelerations, the first beginning in the early 1980s—​ten years ahead of so-​called economic reform in India, of which more shortly—​and another in the early 2000s that was spurred on by India’s high-​tech boom and greater integration into the global economy. There is, thus, a puzzle. What caused this strong growth performance if not positive changes in geography or institutions? Three parts of an answer suggest themselves: consistency in economic policy; the emergence of competition States; and—​notwithstanding what was said earlier—​an underlying robustness (comparatively) in the quality of the country’s underlying institutions. Firstly, economic policy. Discounting Easterly and Levine, it seems clear that a continuing high rate of economic growth in India post-​1980 is connected to consistency of economic policymaking. There are two dimensions to this. When Indira Gandhi returned to power in India in 1980 she retired much of the socialist rhetoric she had rolled out in the 1970s. She also began to align herself with leading corporate interests. As Atul Kohli (2006) has rightly noted, this was not a pro-​market turn in Indian economic management, but it was a pro-​ business turn. The governments of Indira and Rajiv Gandhi in the 1980s eased restrictions on capacity expansion for incumbent firms, removed many of the price controls imposed in the 1960s and 1970s, and reduced corporate taxes. Bosworth and Collins (2007) have calculated that total factor productivity (TFP) in India increased sharply in the 1980s to an average rate of growth of 2.49 per cent per annum. This contrasts markedly with negative TFP growth rates in the 1970s. Relatedly, when India was forced at last to begin the liberalization of parts of its economy in 1991—​in the wake of a severe crisis of external indebtedness—​the reform process that was unleashed, and which initially involved significant relaxations in trade and financial sector regulations, built upon the earlier pro-​business tilt (Panagariya, 2008). Despite not uncommon references to ‘neoliberalism’ in India, the truth is that the formal sector employs no more than 10 per cent of the workforce (Harriss-​White, 2003; Breman, 2010) and state services have not been ruthlessly privatized (Nayak, 2010). Nor has India relaxed all of its constraints on foreign inward investment or property acquisition. Similarly, while various Left parties maintained their public opposition to ‘economic reform’ in India through the 1990s and early 2000s, and while the Hindu nationalist Bharatiya Janata Party (BJP) occasionally took pot shots at foreign icons like Coca Cola or Michael Jackson, the deeper truth is that the liberalization project begun by Manmohan Singh when he was India’s Finance Minister in the early 1990s was continued by successive coalition governments led by the BJP or third-​ party movements (the United Front government, 1996–​98). Over a thirty-​year period, governments in India have tied themselves faithfully to the mast of big business. Rhetorics of rule have changed over this period, along with flagship social programmes like the National Rural Employment Guarantee Scheme rolled out by Manmohan Singh when he returned to New Delhi in 2004 as Prime Minister. But business leaders have been able to make investment decisions in most of the country in the knowledge that a return to the dirigiste policies of the late 1960s and early 1970s (including the notorious Monopolies and Restrictive Trade Practices legislation that was enacted in 1969) was not on the cards. Secondly, competition States. Noting that the Indian economy grew overall at average rates in excess of 6 per cent per annum through the period 1980–​2009 fails to register the fact

102   Corbridge that growth rates of more than 10 per cent were recorded in States like Gujarat and Punjab at this time, while many of India’s central and eastern States, including Bihar, Jharkhand, and Chhattisgarh (the latter two from 2000, when they were carved out of Bihar and Madhya Pradesh, respectively), barely grew at more than 1 or 2 per cent per annum. Aseema Sinha (2004) has suggested that the pro-​business and pro-​market tilts of the 1980s and 1990s liberated economic growth most quickly in those States which decades earlier had large entrepreneurial, merchant, and trading communities—​such as the Marwaris of western India. Private sector growth, which had been suppressed in mid-​century, now bloomed again. In addition, the Government of India chose in the 1990s to allow States to compete for foreign inward investment. An early example would be the competition between Maharashtra and Tamil Nadu to host a Ford motor plant, a battle won in 1996 by Tamil Nadu. Indeed, the government did more than this. Prior to the reform period, States within India’s federal system of rule competed with each other mainly to petition New Delhi for greater grants in aid from the more elastic tax revenues raised at the centre. New Delhi, in turn, used this system of patronage to reward political clients in the States. This game changed markedly when States were set free to raise loans themselves and to court foreign capital, as they have been doing increasingly over the past twenty years. Here, indeed, is one key reason why economic reform has been relatively audacious in India notwithstanding the conservatism of the political class in New Delhi (Grindle, 2000; Corbridge et al., 2013). Many of India’s key economic reforms have been made in the provinces. This is where land acquisition by big business has been at its fiercest, often at the expense of poorer communities who have seen their property dispossessed. And this is where India’s allegedly strong labour laws have been most regularly ignored and flouted. Further, as Rob Jenkins (1998, 2011)  has noted, the unbinding of Prometheus in states like Gujarat—​including under then Chief Minister, Narendra Modi, now India’s Prime Minister—​has put pressure on political leaders elsewhere in India to behave more and more like chief executive officers. Chief Minister Chandrababu Naidu was the very embodiment of this new political–​economic creation in Andhra Pradesh in the mid-​2000s, but the model also spread in that decade to include the ostensibly communist political leadership of West Bengal. Rightly fearful that the urban middle classes would turn against them if they did not start to deliver growth and consumption goods, the ruling Communist Party of India (Marxist) significantly changed its policies under the leadership (from 2000 to 2011) of Chief Minister Buddhadeb Bhattacharya (Das, 2012). Finally, it is worth noting that regional competition and policy consistency in India post-​ 1980 would have been unlikely to give rise to high and sustained rates of economic growth in the absence of a fundamentally enabling institutional framework. This will seem an unlikely claim, but it is consistent with two important observations. Firstly, as Brad DeLong has shown in an important paper, ‘India’s economic growth from 1960 to 1992 lies smack in the middle of world growth rates’ (DeLong, 2003, p. 189). That is to say, on the basis of cross-​ country regressions of the average rate of growth of output per worker against three proximate determinants of growth in a standard Solow model (the share of investment in GDP, the population growth rate, and the log of output per worker), DeLong finds that India did pretty well as expected, even although it is widely accepted that the country formulated a raft of policies in the 1960s and 1970s that actively discouraged private capital accumulation. Put another way, this seems to suggest that India’s core institutional framework, compared with those of many developing country peers, was sufficiently robust to keep the economy

Economic Growth and Poverty Reduction in Contemporary India    103 afloat, despite the best efforts of the country’s leadership. Turning this around, it is then not unreasonable to maintain, with Arvind Subramanian (2007), that the legacies of an English-​ speaking middle class, of relatively strong property rights guarantees (a legacy of both the colonial and pre-​colonial periods), and of a stable political settlement at the centre of the country, allowed India to leverage a consistent run of non-​damaging (but not radical or neoliberal) economic policies post-​1980 in order to secure higher growth rates than otherwise would have been the case. Institutions matter, after all.

Poverty Reduction, Inequality, and Political–​E conomic Geography Institutions also matter when it comes to maps of poverty reduction and inequality. This is true both within India and in comparisons between India and other emerging countries. A key observation about poverty reduction in India is that it was more efficient in the 1970s and 1980s than in the period following liberalization in the 1990s. The transmission mechanism between economic growth and reductions in extreme income poverty did pick up again in the 2000s. But it will be argued here that further reductions will depend heavily on improvements on state capacity and institutional quality in those regions of India that hitherto have benefitted only weakly from sustained economic growth. Figure 5.2 conveys a good sense of key trends and issues in poverty reduction in India since independence. Notwithstanding reasonably high rates of economic growth from 1950 to 1965, the percentage of Indian men and women living below the national poverty line (which recently has been defined even more harshly than the $1.25 per day/​PPP used by the 70 Proportion of India’s population living in poverty 60 50 40 30 20 Proportion of India’s poor living in urban areas

10 0 1950

1960

1970

1980

1990

2000

2010

Figure 5.2   Headcount Index of Poverty Using the National Poverty Line (Percentage).

104   Corbridge World Bank (Himanshu, 2010)) failed to decrease from 1950 to 1970. The total number of Indians living in extreme income poverty soared in this period from about 180 million in 1950 to over 300 million by 1970. Significantly, though, the proportion of Indians living in extreme poverty declined sharply from the early 1970s until the late 1980s, and declined with some consistency (as can be seen from the shape of the curve). The main drivers of this poverty reduction necessarily have to be found in some of the key economic policies of the much derided pre-​reform years, including the so-​called Nehru–​ Mahalanobis years of the Second and Third Five-​Year Plans (1956–​66). Several economists have pointed out that some of the more benign effects of economic policy in these years—​ large investments in capital-​intensive industries in poorer regions of India, the development of Indian Institutes of Technology and Management, targeted poverty reduction programmes—​were not greatly felt until the 1970s and 1980s (McCartney, 2009). In addition, the urban-​biased policies pursued by Nehru and his chief planner, Mahalanobis, in the 1950s and early 1960s were abruptly reversed following the food crises of 1965–​67. From that point onwards the income terms of trade have rarely moved against rural India, and politicians have generally been reluctant to take on the power of organized farming lobbies (Varshney, 1995). More positively, from about this time the Indian state invested heavily in the Green Revolution, which flourished particularly in areas with established irrigation systems or where local governments invested heavily in new systems of groundwater management and rural electrification. It was the expansion of agrarian capitalism that largely pulled people above the poverty line in India in the 1970s and 1980s. Throughout this period, the bulk of India’s population, and the great mass of extreme income poverty, were to be found in the countryside. The Green Revolution, and associated developments in rural marketing and industrialization, tightened rural labour markets in key areas like Punjab, western Uttar Pradesh, and Tamil Nadu, and helped check increases in the real price of staple grains. As a consequence, as Petia Topolova (2008) has shown (Figure 5.3, column 1), proportionately the main beneficiaries of economic growth in India in the 1980s (and, indeed, in the late 1970s) were rural households in the bottom third of the income distribution. This pattern of relatively inclusive growth changed markedly in the mid-​to late 1990s, as can be seen in the right-​hand column of Figure 5.3. In the period from 1993–​94 to 2004–​05 most of the benefits of real consumption growth were captured by a small urban elite and a slightly larger urban middle class (Banerjee and Piketty, 2005). This largely explains the sharp reduction in the rate of poverty reduction in India, described by the shaded section of Figure 5.2 (in effect, the difference between what happened from c.1990 to c.2000 and what would have happened had India maintained the rate of poverty reduction recorded from c.1970 to c.1990). As Robin Burgess and Tim Besley (2003) report, India by c.2000 had an elasticity of poverty reduction with respect to income per capita of around –​0.65. This was slightly better than the figure of –​0.59 they report for South Asia as a whole, but it is significantly down on the figure of –​1.00 in East Asia and the Pacific, or the even better number for China alone (–​1.05) In plain English, the capacity of economic growth to drive income poverty reduction in India is only about two-​thirds that in China. China’s great success in pulling hundreds of millions of people out of extreme income poverty in the 1990s was centred squarely on the countryside (Eastwood and Lipton, 2000). Households with comparatively high levels of education and health care were able to access new employment opportunities in and

Economic Growth and Poverty Reduction in Contemporary India    105 4.5

All India 1983–1993/94

4.0 3.5 3.0

1.5

1.5

1.0

1.0

0.5

0.5 0

10 20 30 40 50 60 70 80 90

Urban India 1983–1993/94

4.0 3.5 3.0

Urban India 1993/94–2004/05

3.5 3.0

1.5

1.5

1.0

1.0

0.5

0.5 0

10 20 30 40 50 60 70 80 90

0.0 –0.5

0

10 20 30 40 50 60 70 80 90

4.0

4.0 Rural India 1983–1993/94

3.5 3.0

3.0 2.5

2.0

2.0

1.5

1.5

1.0

1.0

0.5

0.5 0

10 20 30 40 50 60 70 80 90

Rural India 1993/94–2004/05

3.5

2.5

0.0

10 20 30 40 50 60 70 80 90

2.0

2.0

–0.5

0

2.5

2.5

–0.5

0.0 –0.5 4.0

4.5

0.0

3.0 2.0

2.0

0.0

All India 1993/94–2004/05

3.5 2.5

2.5

–0.5

4.0

0.0 –0.5

0

10 20 30 40 50 60 70 80 90

Figure 5.3   India: Real Patterns of Consumption Growth (after Topolova 2008). outside agriculture as the old system of collective farming was undone and long tenancies were introduced in their place (giving de facto stability in property rights over a long time horizon). In contrast, India’s economic reforms privileged the urban economy and tradeable good sectors from the start. They set in motion pressures for rural-​to-​urban migration, and widening spatial inequalities, which only became pressing in China in the second stage of its economic reform process. Socially, too, it is very apparent that economic growth has not delivered benefits to the poorest quintile of urban Indians at the same rate that benefits have been delivered to the average or median household: the famous proposition of Dollar and Kraay (2002, p. 195) in their well-​known paper ‘Growth is good for the poor’. Undoubtedly,

106   Corbridge economic growth is necessary to pull poor people out of poverty, but the type and composition of growth matter just as much as headline rates (Donaldson, 2008). Maps of extreme income poverty in India also coincide for the most part with maps of a wide range of human development indicators (see Table 5.1). There are some important exceptions that need to be noted. Literacy rates in the south Indian State of Kerala are much higher than would be predicted on the basis of average household income. Disturbingly, too, the sex ratios of women to men are worst in some of India’s richest States, including Punjab and Haryana, where there are fewer than 900 women for every 1000 men (Dreze and Sen, 2013). By and large, though, the States in India with the lowest gross state domestic

Table 5.1 Ranking of India’s Poorest States by Gross State Domestic Product Per Capita and Human Development Indicators Rank by GSDP per capita

Literacy rateb HDI ranka Diff.

Rank

Infant mortalityc

Underweight childrend

Sex ratio (0–6 years)e

Diff.

Rank

Diff.

Diff.

Rank

Bihar

13

13

–​17.9

13

9

–​7.4

11

15

5

Orissa

12

10

–​1.8

9

–​13.4 11

–​7.4

12

26

3

Uttar Pradesh 11

12

–​8.0

12

–​19.1 13

–​4.7

10

–​11

8

Madhya Pradesh

10

11

–​1.3

8

–​18.5 12

–​8.1

13

5

7

Rajasthan

9

8

–​4.4

11

–​12.8 10

–​3.6

9

–​18

10

West Bengal

8

7

3.8

5

18.9

3

–​1.7

7

33

2

Andhra Pradesh

7

9

–​4.3

10

1.8

8

9.3

4

34

1

Karnataka

6

6

1.6

7

16.1

4

3.1

5

19

4

Tamil Nadu

5

2

8.1

2

19.4

2

10.3

3

15

6

Haryana

4

4

3.2

6

10.8

5

12.4

2

–​108

12

Gujarat

3

5

4.6

3

5.0

7

1.9

6

–​44

11

Punjab

2

1

4.6

4

10.5

6

18.3

1

–​129

13

Maharashtra

1

3

11.9

1

23.9

1

-​2.6

8

–​14

9

All-​India (avg.) Kerala

65.4 7

1

25.5

–​5.3

Rank

67.6 1

51.3

47 1

20.1

927 1

33

3

Notes: GSDP, gross state domestic product; Diff., difference between state-​level indicator and all-​ India average. aHDI ranking refers to the Human Development Index methodology in the United Nations Development Programme’s Human Development Report 2011. Ranking across sixteen major states, including Assam. b2001, per cent of population aged seven years and older. c1998/​9, per 1000 live births. d1998/​9, per cent of children under three years of age. e2001, girls per 1000 boys in 0–​ 6 years age group. Sources: UNDP (2001); Office of the Registrar General & Census Commissioner, India (2001); International Institute for Population Scienes (2000); Global Health Data Exchange (1999); World Bank (2006, p. 21).

Economic Growth and Poverty Reduction in Contemporary India    107 product—​Bihar, Orissa (Odisha), Jharkhand, and Chhattisgarh—​are the same States that have high levels of underweight children, high infant mortality rates, and low literacy rates. If the capabilities of people in the central/​eastern region of India are to be put on a par with their fellow citizens in the west and south the state will need to channel investment directly or indirectly to key public services and new growth platforms. At a minimum this will mean significant investments in rural electrification and rural banking (Kale, 2014), but to be more effective there will need to be a significant redistribution of assets in the countryside (agrarian reform). Inclusive growth also presupposes actions to enable members of the most seriously disadvantaged and discriminated against communities—​dalits (scheduled castes), adivasis (scheduled tribes), Muslims, and particularly women in each of these communities—​to fight more effectively to gain access to better paid opportunities in local employment markets. But herein lie two problems, which, taken together, return us to an account of the geographical dimensions of institutional quality. Firstly, eastern India is simply not well joined up with the rest of the country. Notwithstanding the imagery of India Shining that peaked around the turn of the millennium, which liked to reference the new India through photographs of the new office blocks and industrial spaces of Gurgaon, near Delhi, or Bandra Kurla in Mumbai, it remained impractical at this time to think of driving from capital cities in eastern India to Delhi, or even to relatively nearby Kolkata (Corbridge, 2011). As late as 2007 the middle classes of eastern India were forced either to fly to Delhi on one of two or three daily flights (in the case of Ranchi) or to take an overnight train journey. Moreover, within the region, while there were a few decent roads—​as from Ranchi to Jamshedpur—​very often it could take four or five hours to travel just 100 kilometres. And within districts like Bhojpur at this time, in the middle of the so-​called flaming fields of Bihar, petrol stations thinned out markedly as territories moved quickly out of government control into the hands of the Maoists (Naxalites) or various landlord armies (senas). In the wake of the low-​intensity wars that have raged in this region for decades—​and following years of investment neglect—​the state has simply withered away in some Blocks and Districts. Schools have been closed (or occasionally blown up), and dispensaries have few or no stocks of key drugs. Many Block Development Offices that were open for business two decades ago had effectively been shut down or decommissioned by around 2010. Rotting paper files and a chaotic system for the distribution of government employment or pensions had given way to ghost buildings and never-​present public employees (including teachers). Within this environment any capacity to enforce order or property rights tended to be at the point of a gun—​the state having long since given up its claim to an effective and legitimate monopoly over the means of violence. In these circumstances, capital accumulation made no sense and had little chance. In Bihar during the 1990s and early 2000s this inauspicious terrain for growth and poverty reduction was made even worse by the political calculations of the state’s longtime Chief Minister, Lalu Yadav. Mr Yadav had come to power in Bihar on the basis of a political alliance between the state’s Muslims and three of Bihar’s most powerful Backward Castes—​the Yadavs, the Kurmis, and the Koeris. As Jeffrey Witsoe (2013) has shown in his outstanding account of ‘democracy against development’, Mr Yadav offered the last three communities the dignity or honour (izzat) of not being trampled upon by the leading forward castes (Brahmins, Bhumihars, Rajputs) who ran the state during the years of the Nehru–​Gandhi dynasty. Just as importantly, Mr Yadav offered the Muslims of Bihar protection against the riots that gripped India through the 1990s, which in other states often led to killings of

108   Corbridge Muslims at the hands of organized Hindu groups while the police looked on (Corbridge et al., 2012). Mr Yadav made sure that the police force in Bihar protected local Muslims in the event that a riot flashpoint was lit. Many lives were saved as a consequence. However, what Mr Yadav did not promise was either ‘development’ or effectively functioning public services. On the contrary, Lalu Yadav often dismissed ‘development’ on the basis that its benefits would be captured by the State’s existing economic elites (the forward castes), and/​or that it was environmentally damaging and linked to destruction of Earth’s ozone layer (Corbridge et al., 2005). He also reasoned that dignity could be delivered in short order, along with restructured bus companies and police forces in which his supporters could be quickly placed. What he believed he could not guarantee within a five-​year electoral cycle was infrastructural development or significant economic growth. Indeed, his policy of favouring his political clients, and running down the all-​India services in Bihar (the Indian Administrative Service, the Indian Police Service, etc.), ensured that growth and social development could not take place under his leadership (or, indeed, under the Chief Ministership of his wife, Rabri Devi, when Mr Yadav found himself in jail). Tens of millions of people were condemned to a decade of almost zero economic growth on a per capita basis and hardly any serious reduction in extreme income poverty. This was institutional failure on a grand scale. A disconnected transport geography, coupled with a political geography that encouraged significant state failure, added up to a recipe for economic ruin at the same time that west and south India began to boom.

Conclusion Since the world economic crisis broke in 2008 rates of economic growth in India have faltered, but only from around 9 per cent in 2007 to 6.5 per cent in 2012. (There was a worsening to around 5% growth in 2013–​14.) On the plus side, as can be seen in Figure 5.2, the ability of economic growth in India to power poverty reduction resumed its pre-​reform path in the late 1990s. The tilt to the urban middle classes that happened in the early 1990s was partially corrected in later years. What happens next will depend on the interplay between the two geographical dialectics that have structured discussion in this chapter. On the one hand, it is a standard Kuzenetzian expectation that social and spatial inequalities will increase in the early and mid-​phases of sustained economic growth. Factories have to locate somewhere and some labour markets tighten more quickly than others. What might be called ‘benign inequalities’ open up in the wake of these investment and location decisions. They get corrected over time—​in standard modernization parlance—​as other locations catch up with the growth-​friendly policies of the early adopters (the Provincial Darwinism model described by Jenkins) and when the state is forced to enter a new social contract with its citizens based around better funded and more efficient public service delivery and democratic accountability. Keeping men and women in Gujarat in extreme income poverty to lower the income gap with Bihar is hardly a sensible way of organizing economic affairs. However, in India, as in many other countries, this first and largely inter-​temporal dialectic interacts with a second that tends to the reinforcement of spatial and social inequalities

Economic Growth and Poverty Reduction in Contemporary India    109 for insistently malign reasons. The caste system has been reworked in recent decades from a largely vertical hierarchy of ‘pure’ and ‘impure’ castes to a more horizontally organized system of competing ethnic groups (Chandra, 2004). Nevertheless, India’s Scheduled Communities continue to be locked out of local and national circuits of power and prosperity. Merit generally goes unrewarded and social mobility is limited. If these groups are to work their way out of extreme deprivation they will need far more than government employment schemes or generally tighter labour markets to help them. In the long term they need the support of politicians who are forced to be responsive to their weight of voting numbers. They also need the support of politicians who will build and not undermine effective state capacity and public service delivery. In Bihar, since the end of Mr Yadav’s rule, there have been signs that politicians of this calibre can emerge and win elections. Nitish Kumar, the Chief Minister of Bihar from 2005 to 2014, raised the growth rate in Bihar from close to 2 or 3 per cent per annum when he took office to closer to 10 per cent when he stepped down. Much remains to be done to bring the poorest of Bihar’s dalit communities (e.g. the Musahars) into the tent of economic growth, but for the urban middle classes especially, and for many in the countryside, the improvements in law and order that Mr Kumar secured did lay the foundations for concerted investments in physical and human capital formation. However, in neighbouring Jharkhand no such transformation has materialized. There, and in Chhattisgarh and parts of northern Orissa (Odisha), the conflict between the state and the Maoists became more violent in the mid-​2000s. Absent a ceasefire and a lasting political settlement, there is no prospect of economic growth or poverty reduction in this large swathe of central and eastern India. Significantly, too, and here there is a faint echo of the physicalist account of geography that one finds in Sachs, the hilly and forested terrain of this ‘Maoist belt’ affords opportunities for guerrilla warfare that are not present in the Gangetic plains (Shah, 2013). In this part of India institutional failings have given way to institutional collapse. The sad truth is that effective state capacity will only be rebuilt after years of bloody struggle. Prosperity remains a long way off.

References Acemoglu, D., Johnson, S., and Robinson, J. (2001). ‘The colonial origins of comparative development: an empirical investigation’. American Economic Review 91: 1369–​1401. Acemoglu, D., Johnson, S., and Robinson, J. (2002). ‘Reversals of fortune:  geography and institutions in the making of the modern world income distribution’. Quarterly Journal of Economics 117: 1231–​1294. Acemoglu, D. and Robinson, J. (2012). Why Nations Fail: The Origins of Power, Prosperity and Poverty (London: Profile Books). Banerjee, A. and Piketty, T. (2005). ‘Top Indian incomes: 1922–​2000’. World Bank Economic Review 19: 1–​20. Banerjee, M. (2014). Why India Votes? (New Delhi: Routledge). Besley, T. and Persson, T. (2011). Pillars of Prosperity: The Political Economics of Development Clusters (Princeton, NJ: Princeton University Press). Bolt, J. and Bezemer, D. (2009). ‘Understanding long-​run African growth: colonial institutions or colonial education?’ Journal of Development Studies 45: 24–​54.

110   Corbridge Bosworth, T. and Collins, S. (2007). Accounting for growth in China and India’. NBER Working Paper 12943 (Cambridge, MA: National Bureau of Economic Research). Breman, J. (2010). Outcast Labour in Asia: Circulation and Informalisation of the Workforce at the Bottom of the Economy (New Delhi: Oxford University Press). Burgess, R. and Besley, T. (2003). ‘Halving global poverty’. Journal of Economic Perspectives 17: 3–​22. Chandra, K. (2004). Why Ethnic Parties Succeed: Patronage and Ethnic Head Counts in India (Cambridge: Cambridge University Press). Chant, S. and McIlwaine, C. (2009). Geographies of Development in the 21st Century (Cheltenham: Edward Elgar). Corbridge, S. (2011). ‘The Contested Geographies of Federalism in post-​Reform India’ in S. Ruparelia, S. Reddy, J. Harriss, and Stuart Corbridge (eds) Understanding India’s New Political Economy: A Great Transformation? pp. 66–​80 (London: Routledge). Corbridge, S., Harriss, J., and Jeffrey, C. (2013). India Today:  Economy, Politics and Society (Cambridge: Polity Press). Corbridge, S., Kalra, N., and Tatsumi, K. (2012). ‘The search for order: understanding Hindu-​ Muslim violence in post-​Partition India’. Pacific Affairs 85: 287–​311. Corbridge, S., Williams, G., Srivastava, M., and Veron, R. (2005). Seeing the State: Governance and Governmentality in India (Cambridge: Cambridge University Press). Das, R. (2012). ‘The politics of economic reform in communist West Bengal’. Unpublished PhD dissertation, London School of Economics. DeLong, J. (2003). ‘India Since Independence: An Analytic Growth Narrative’ in D. Rodrik (ed.) In Search of Prosperity:  Analytic Narratives on Economic Growth, pp. 184–​ 204 (Princeton, NJ: Princeton University Press). Dollar, D. and Kraay, A. (2002). ‘Growth is good for the poor’. Journal of Economic Growth 7: 195–​202. Donaldson, J. (2008). ‘Growth is good for whom, when, how? Economic growth and poverty reduction in exceptional cases’. World Development 36: 2127–​2143. Dreze, J. and Sen, A. (2013). An Uncertain Glory:  India and its Contradictions (Princeton, NJ: Princeton University Press). Easterly, W. (2006). The White Man’s Burden: Why the West’s Efforts to Aid the Rest Have Done So Much Ill and So Little Good (Oxford: Oxford University Press). Easterly, W. and Levine, R. (2003). ‘Tropics, germs and crops: how endowments influence economic development’. Journal of Monetary Economics 50: 3–​39. Eastwood, R. and Lipton, M. (2000). ‘Rural–​urban dimensions of inequality change’. Working Paper 200 (Geneva: UN University/​WIDER). Fearon, J. and Laitin, D. (2003). ‘Ethnicity, insurgency and civil war’. American Political Science Review 97: 75–​90. Fielding, D. and Torres, S. (2008). ‘Cows and conquistadors: a contribution on the colonial origins of comparative development’. Journal of Development Studies 44: 1047–​1066. Global Health Data Exchange (1999). ‘India SRS Statistical Report 1999’ http://​ghdx.healthdata.org/​record/​india-​srs-​statistical-​report-​1999 (last accessed 2 July 2017). Grindle, M. (2000). Audacious Reforms:  Institutional Invention and Democracy in Latin America (Baltimore, MD: The Johns Hopkins University Press). Harriss-​ White, B. (2003). India Working:  Essays on Economy and Society (Cambridge: Cambridge University Press).

Economic Growth and Poverty Reduction in Contemporary India    111 Hausmann, R., Pritchett, L., and Rodrik, D. (2005). ‘Growth accelerations’. Journal of Economic Growth 10: 303–​329. Himanshu (2010). ‘Towards new poverty lines for India’. Economic and Political Weekly 45: 43–​59. International Institute for Population Sciences (2000) ‘National Family Health Survey (NFHS-​ 2) 1998–​9’ http://​www.dhsprogram.com/​pubs/​pdf/​FRIND2/​FRIND2.pdf (last accessed 27 March 2017). Jenkins, R. (1998). ‘The Developmental Implications of Federal Political Institutions in India’ in M. Robinson and G. White (eds) The Democratic Developmental State, pp. 187–​214 (Oxford: Oxford University Press). Jenkins, R. (2011). ‘The Politics of India’s Special Economic Zones’ in S. Ruparelia, S. Reddy, J. Harriss, and S. Corbridge (eds) Understanding India’s New Political Economy:  A  Great Transformation? pp. 49–​65 (London: Routledge). Kale, S. (2014). Electrifying India:  Regional Political Economies of Development (Stanford, CA: Stanford University Press). Kohli, A. (2006). ‘Politics of economic growth in India, 1980–​2005: part I, the 1980s’. Economic and Political Weekly 41: 1251–​1265. Lange, M. and Mahoney, J. (2006). ‘Colonialism and development: a comparative analysis of Spanish and British colonies’. American Journal of Sociology 111: 1412–​1462. McCartney, M. (2009). Political Economy, Liberalisation and Growth in India, 1991–​2008 (London: Routledge). Mamdani, M. (1996). Citizen and Subject:  Contemporary Africa and the Legacy of Late Colonialism (Princeton, NJ: Princeton University Press). Nayak, P. (2010) ‘Privatization: Indian Experience Since 1991’ in K. Basu and A. Maertens (eds) The Concise Oxford Companion to Economics in India, pp. 369–​371 (New Delhi:  Oxford University Press). Office of the Registrar General & Census Commissioner, India (2001) ‘Census data online-​ 2001’ http://​censusindia.gov.in/​2011-​common/​censusdataonline.html (last accessed 27 March 2017). Panagariya, A. (2008). India: The Emerging Giant (Oxford: Oxford University Press). Przeworski, A., Alvarez, M., Cheibub, J., and Limongi, F. (1996). ‘What makes democracies endure?’ Journal of Democracy 7: 39–​55. Rodrik, D. (2003). ‘Introduction: What Do We Learn from Country Narratives?’ in D. Rodrik (ed.) In Search of Prosperity: Analytical Narratives on Economic Growth, pp. 1–​19 (Princeton, NJ: Princeton University Press). Rodrik, D., Subramanian, A., and Trebbi, F. (2004). ‘Institutions rule: the primacy of institutions over geography and integration in economic development’. Journal of Economic Growth 9: 131–​165. Sachs, J. (2005). The End of Poverty: Economic Possibilities for Our Time (New York: Penguin). Sachs, J., Mellinger, A., and Gallup, J. (2001). ‘The geography of poverty and wealth’. Scientific American, 284: 70–​76. Shah, A. (2013). ‘The intimacy of insurgency: beyond coercion, greed or grievance in Maoist India’. Economy and Society 42: 480–​506. Sharma, P. (2014). Democracy and Transparency in the Indian State (New Delhi: Routledge). Sharma, S. (2009). China and India in the Age of Globalization (Cambridge: Cambridge University Press).

112   Corbridge Sinha, A. (2004). ‘The changing political economy of federalism in India: a historical institutionalist approach’. India Review 3: 25–​63. Spence, M. (2011). The Next Convergence: The Future of Economic Growth in a Multispeed World (New York: Farrar, Strauus and Giroux). Subramanian, A. (2007). ‘The evolution of institutions in India and its relationship with economic growth’. Oxford Review of Economic Policy 23: 196–​220. Topolova, P. (2008). ‘India: is the rising tide lifting all boats?’ IMF Working Paper 08/​54 (Washington, DC: IMF). UNDP (2001). ‘Human development report 2001’ http://​hdr.undp.org/​en/​content/​human-​ development-​report-​2001 (last accessed 27 March 2017). Varshney, A. (1995). Democracy, Development and the Countryside: Urban–​Rural Struggles in India (Cambridge: Cambridge University Press). Winters, A. and Yusuf, S. (2007). Dancing with Giants: China, India and the World Economy (Washington, DC: World Bank). Witsoe, J. (2013). Democracy Against Development: Lower-​caste Politics and Political Modernity in Postcolonial India (Chicago, IL: University of Chicago Press). World Bank (2006). India: Inclusive Growth and Service Delivery (Washington, DC: World Bank).

Chapter 6

Cri sis and Au st e ri t y in Action:  G re e c e Maria Tsampra Introduction The 2008–​09 global financial-​turned-​to-​sovereign debt crisis in Europe has shaken the structural and institutional configuration of the Eurozone and the very process of European integration, as it put in question the access to, and the consolidation of, prosperity for many European Union (EU) member states. The crisis exposed the structural weaknesses of the most vulnerable Eurozone economies, while the implemented-​for-​recovery austerity policy has led to prolonged recession with dramatic implications. Despite similarities, the impact of the crisis has been more negative in the southern EU region, where the steepest decline and highest job and income losses are recorded. In effect, the gap between the ‘core’ and the ‘periphery’ of Europe has deepened, as existing inequalities among countries and regions are aggravated and new inequalities emerge. Greece, along with other countries of the Eurozone southern periphery (Portugal, Spain, Italy), is a case clearly demonstrating the geographically divergent outcomes of the crisis. Bound to International Monetary Fund (IMF)/​European Commission (EC)/​European Central Bank (ECB) troika’s fiscal austerity and regulatory reform memoranda (since its first ‘bailout’ in May 2010), the country has suffered much more drastic and long economic contraction than the EU average economy. This chapter seeks to illuminate the place-​specific causes of the Greek economy’s increased vulnerability to the crisis and provide insights on issues critical for recovery. To this purpose, the focus is on the factors underlying the EU core-​periphery contradiction and diversifying the outcomes of restructuring processes and austerity policy. Such factors are traced in Greece’s path-​dependent growth patterns, depicted in inherited and evolving industrial and entrepreneurial structures. Over time, the Greek economy has evolved from semi-​Fordist late industrialization, to post-​Fordist deindustrialization; following EU accession in 1981, it has undergone restructuring towards non-​exporting and consumption-​dependent sectors; and shifted to finance-​ led growth since the European Monetary Union (EMU) in 2001. Still, as revealed by the present economic shock, restructuring and EU integration processes have not achieved to

114   Tsampra transform ‘peripherality’ structures to a more sustainable and competitive growth trajectory. On the contrary, under the pressure of financial capital, inherent structural weaknesses have been transformed in a new pattern that reproduces core-​periphery divergence. The latter was particularly accentuated by fiscal austerity measures implemented in Greece (and all debt-​ridden countries) to increase competitiveness by internal devaluation. This policy has led to a six-​year economic contraction, collapse of the Greek economy’s prominent sectors and related small-​and medium-​sized enterprise (SME)-​based entrepreneurship, dramatic employment decline, and skyrocketing public debt-​to-​gross domestic product (GDP) ratio. The following section delineates the theoretical background for the exploration of the chapter’s key-​issue, that is, the factors defining economic vulnerability and adjustment to economic shock. Emphasis is put on the significance of path-​dependent processes shaping the economic landscape and defining an economy’s place in the international division of labour. As argued, inherited and evolving structures are incorporated into emerging growth patterns through corresponding regulatory adjustments. The section ‘The Divergent Impact of the Crisis in Europe’ provides empirical background on the spatially divergent impact of the crisis in the EU. Data for employment, unemployment, long-​term and youth unemployment, poverty, and at-​risk-​of-​poverty rates establish the severe degradation of the southern Eurozone economies and the destructive impact of austerity policy, particularly in Greece. In the section ‘Path-​Dependent Growth and Vulnerability in Greece’, the Greek economy’s vulnerability is outlined with a historical perspective on the country’s semi-​Fordist legacy and peripheral position within the European economy. Structural weaknesses are identified in industrial specialization (in non-​exporting sectors dependent on household demand and credit) and business patterns (of prominent independent small and microenterpises), which have been shaped by inherited attributes of strong self-​employment, extensive atypical labour, weak welfare, dense family and social networks, and by evolving EU integration processes. The outcome of austerity policy is identified in abrupt restructuring that has led to the destruction of the established growth patterns, in a scale that undermines any prospect for socio-​economic recovery. The penultimate section, ‘Discussion: Crisis and Austerity in Action’, seeks to link more explicitly the impact of the crisis in Greece to austerity policy and induced regulatory adjustments. The issues raised concern the nature of the crisis as an effect of inherent core-​ periphery contradictions in Europe, and the impact of austerity measures in reproducing such contradictions. In the final section, it is concluded that EU policy addressing the crisis has not resolved the structural weaknesses of the Greek economy (and Eurozone periphery). Austerity, implemented as a one-​size-​fits-​all policy, has instead disintegrated the path-​ dependent structures that have for long sustained resilience and socio-​economic cohesion.

Theoretical Background and Conceptual Framework Economic geography theory of neoclassical background has tried to explain uneven regional development by concepts of transaction costs and agglomeration externalities. Yet, despite the importance of such issues, the acknowledgement of the role that historical circumstances

Crisis and Austerity in Action: Greece    115 play in economic development and its spatial outcome is crucial for an adequate explanation of regional divergence (Martin and Sunley, 1996). Marxist approaches regard the geography of accumulation as a consequence of historical and structural conditions governing the organization of capital (Harvey, 1975; Massey and Meagan, 1978; Walker, 1978; Harvey, 1985; Massey, 1984; Dunford, 1989, 1990). In structural analyses, changes in the labour demand lead to changes in investment patterns: industrial relocations, plant closings, and new plant establishments adjust to labour supply. Massey’s Spatial Divisions of Labour (1984) goes further by suggesting that a place’s industrial specialization and restructuring reflects the specific combination of past ‘layers’ of investment found there. For Massey, capitalist social relations are spatialized, just as the spatial is societalized once it is embedded within capitalism. Places react and influence future rounds of investment that determine an economy’s spatial division of labour (Massey, 1984). For instance, the deindustrialization and economic restructuring of advanced Western economies since the crisis of late 1970s signified not only the decline of previously successful regions, but also the capability of others to withstand and adjust to the negative shock. In many countries the contraction of traditional industries resulted into factory closings and massive unemployment (UK, France), while in others (Germany, Scandinavian countries) a new successful development model was built on increased productivity (Lipietz, 1986). The institutional structure of capital relations is highlighted by the French Regulationists (Aglietta, 1979; Boyer, 1979; Lipietz, 1986; Leborgne and Lipietz, 1988; Moulaert and Swyngedouw, 1989). They argue that structural shocks affect the capitalist accumulation regime as much as the mode of regulation. In their view, industrial restructuring in advanced economies since the 1970s (i.e. the end of old mass-​production forms) was triggered by drop in gain rates, caused by the rigidities of the Fordist–​Keynesian configuration. The outcome was a new international division of economic and political power, established on new growth patterns of flexible industrialization. The emergent, post-​Fordist, regime of accumulation was institutionalized by a mode of regulation that disintegrated the ‘rigid’ social pacts of the past and induced ‘flexibility’ in order to enable the recovery of gain rates (at the expense of labour’s participation in income distribution). Namely, economic activities are socially embedded and socially regulated as required for the stability of the accumulation regime, which ‘depends on specific social modes of economic regulation that complement the role of market forces in guiding capitalist development’ (Jessop, 2008, p.  24). In this view, wars and major crises are crucial episodes in the change of accumulation regimes, which initiate regulatory, or else, social (Boyer, 2013)  experiments with credit money, income tax, health and retirement insurance, new mechanisms defining salaries, public–​private cooperation, and so on. The role of the state in this process is essential, but the contradictions and crises generetated by globalized capital accumulation today are governed by supranational institutions (IMF, ECB, or EC). Still, the need to differentiate between national situations has been acknowledged in the regulation approach, despite criticism for overlooking globalization processes (Jessop, 2001; Becker and Weissenbacher, 2015). Lipietz (1987) focused on the dynamics of national capitalism, to accentuate historical and local specificities of capitalist development that differentiate nation from nation within the global context of accumulation. His concept of ‘peripheral Fordism’ particularly addressed spatially uneven development and the relative position of countries in the global (core-​ periphery) political economy (Lipietz, 1982). Arguably, the internal structures of peripheral economies did not enable core-​type Fordist development. This resulted in the periphery’s

116   Tsampra incomplete industrialization, specialized in capital goods and dependent on imported machinery. Spatial specialization was polarized between the export-​oriented industrialized core and the periphery, providing low-​cost labour and raw materials, and importing high-​value industrial products. After the Fordist crisis in late 1970s, new divergent economic spaces emerged, shaped by pre-​existing structures and restructuring processes: centres of technological innovation and finance of highly qualified labour and high wages; traditional industrial areas of qualified labour and lower wages; and spaces of unqualified labour and low wages. Since the 1980s, the significance of historical circumstances and processes in the formation of the economic landscape is acknowledged in the work of many more scholars (Smith, 1984; Harvey, 1985, 2006; Grabher, 1993; Storper, 1997; Cooke and Morgan, 1998; Boschma, 2004; Feldman, 2005; Gertler, 2005; Hassink, 2005). This concept forms the core of ‘path-​ dependence’ ideas, which have been elaborated and largely employed by economic geographers for the interpretation of place-​based production structures, inherited and evolving specializations, and persisting regional growth disparities. ‘Place dependence’ is essential in path dependence, as the latter is actually a locally contingent and locally emergent process, or effect (Cox, 1996). It can be understood as encompassing the legacy of the past and elements of the new in shaping the economic landscape: local assets, infrastructures, firms, industries, agglomeration externalities, technologies, institutions, social capital, and their interdependencies. More recently, the ‘evolutionary turn’ in economic geography has assigned particular theoretical and empirical meaning to the place-​specific dimension of path-​dependence processes, with regard to the evolution of the economic landscape in space and time (Martin and Sunley, 2006, 2007). The idea of ‘path-​dependent economic evolution’ (Scott, 2006, p. 85) seeks to interpret the evolution of a region’s growth pattern either to a ‘lock-​in’ phase of maturation and rigidification, or to a ‘path-​dissolution’ phase of adaptation and restructuring (Sydow et al., 2005; Martin and Sunley, 2011). Economic adjustment is a path-​dependent process shaped by inherited and evolving structures: industrial specialization, entrepreneurial and employment patterns, and institutional and regulatory arrangements define regional vulnerability or resilience (Martin, 2010). Hence, our first argument is that the response of a national/​regional economy to crisis is determined by path-​dependent growth patterns. An economy of strong underlying growth dynamic is likely to be less vulnerable to crisis and recession, or even to readjust successfully its structure and resume dynamic. But an economy of weak underlying growth dynamic will, conversely, be affected by economic downturn: it may recover to its pre-​crisis growth level, but on a lower long-​term trend in output and/​or employment; or, it may instead decline futher if the destructive impact of the crisis exceeds compensating growth (Hassink, 2010; Martin, 2010, 2012). Yet, the comprehensive interpretation of crisis-​induced economic restructuring requires the consideration of changes in the mode of regulation, as well. The globalized finance-​led growth regime that emerged in the 1980s has proceeded with respective institutional con­ figurations to accomplish stock market and credit growth. Following the 2008–​09 crisis, those configurations are being drastically transformed, with diverse implications for national economies. Thus, our second argument is that policy is decisive for the place-​specific outcomes of the crisis and for economic adjustment. Effective policy and planning would mitigate the regionally differentiated effects of international upheaval (Davies, 2011).

Crisis and Austerity in Action: Greece    117 National policies tightly confined in the imperatives of supranational governing institutions (e.g. through international assistance programs, and so on) tend to disregard the socio-​economic specificities of the individual country that would enable protection and recovery from the shock.

The Divergent Impact of the Crisis in Europe The global financial crisis hit Europe in a period of progressing regional convergence largely attributed to the dynamics of the new EU member states. After 2009, that trend reverted to divergence and the socio-​economic outcomes of the crisis became geographically polarized. The severe decline exacerbated existing inequalities among EU countries and regions, while new inequalities emerged. With regard to employment, divergence ranges from intact levels in the northern ‘core’ countries, to recovery in the Baltic republics, and decline in the southeastern ‘periphery’ (Figure 6.1). In a similar pattern, unemployment is high and increasing in most countries, but stable or significantly lower (even declining) in just a few (Figure 6.2). Regional divergence is wider: in forty-​nine EU regions (most in Germany and Austria) unemployment was 5.4 per cent or less in 2013 (i.e. half the EU28 average). At the other end, in twenty-​seven regions (most in Spain and Greece) unemployment was higher than 21.6 per cent (double the EU28 average). In more than a quarter of EU regions, most of the unemployed have been out of work for at least a year. Since 2008, long-​term unemployment rates in Greece, Spain, and Cyprus have skyrocketed, while in 2014, Greece and Spain recorded the highest long-​term unemployment rates in EU28 (followed by other peripheral countries) (Figure 6.3). The member 75 70

Countries where employment rates continued to fall

Countries where employment recovery was insufficient to raise pre-crisis levels

65

Countries that have exceeded pre-crisis employment rates

60 55 50 45 40 35 Greece Spain Ireland Poland Belgium Portugal Eurozone France Slovenia EU-27 Cyprus Denmark Netherlands Italy Bulgaria Slovakia Romania Lithuania Latvia Czech Republic Estonia Finland United Kingdom Sweden Hungary Malta Luxembourg Germany Austria

30

2008Q3

2010Q3

2012Q3

Figure 6.1   Employment (15–​64 years) Rates (%) EU27, 2008, 2010, and 2012. Sources: Eurostat (n.d. (a)); IILS, 2013.

118   Tsampra 30

Countries where unemployment rates have continued to increase since 2010

25

Countries where unemployment rates have decreased since 2010

20 15 10

0

Greece Spain Portugal Ireland Cyprus Bulgaria Eurozone Italy Hungary EU-27 France Poland Slovenia Denmark Netherlands Luxembourg Austria Slovakia Latvia Lithuania Estonia Sweden Belgium Finland United Kingdom Czech Republic Romania Malta Germany

5

2008 Feb

2010 Feb

2013 Feb

Figure 6.2   Unemployment (15–​64 years) Rates (%) EU27, 2008, 2010, and 2013.

20 18 16 14 12 10 8 6 4 2 0 AT SE LU DK FI DE UK CZ MT RO NL EE HU PL BE FR LV LT EU28 SI IE BG CY IT PT SK HR ES EL

LTU rate (% of active population)

Sources: Eurostat (n.d. (a)); IILS, 2013.

2014

2008

Max 2008–14

Figure 6.3   Long-​term Unemployment (LTU) Rates, 2008 and 2014. Sources: European Commission (2015).

states of the southern EU have also suffered higher rates in youth unemployment (Figure 6.4): between 2008 and 2013, youth unemployment increased by 38.4 per cent, reaching 59 per cent in Greece, and by 56.1 per cent, reaching 32.2 per cent in Spain. In 2015, Greece still suffered unsustainable rates of youth unemployment (51.8%), followed by Spain (48.6%), Croatia (43.1%), and Italy (40.5%). In sharp contrast, the lowest rates of youth unemployment are recorded in Germany (7.0%), Malta (8.7%), and Estonia (9.5%). Namely, the steepest fall in employment and raise in unemployment, long-​ term unemployment, and youth unemployment is ascribed to the countries where sovereign-​debt

Crisis and Austerity in Action: Greece    119 70

Very high increase on 2008

% of active population 15–24 years

60 Low increase

Medium increase

High increase

50 40 30 20

2013

2008

BG PT CY IT HR ES EL

SI IE PL SK

EE CZ UK LT LV EU28 RO BE EA18 FR HU

AT NL DK MT LU FI SE

0

DE

10

2012

Figure 6.4   Youth Unemployment (15–​24 years) Rates (Percentage Labour Force) EU27, 2008, 2012, and 2013. Source: Eurostat (n.d. (a)).

sustainability and credit-​ worthiness came under question since the outbreak of the Eurozone crisis. The spatial pattern of growth divergence has been transformed accordingly, illustrating the most severe implications of the crisis in the debt-​ridden economies. Under the pressure of financial capital and the threat of public bankruptcy, Greece was the first Eurozone country to receive from the IMF/​EC/​ECB-​troika a bailout package and adjunct memorandum in 2010 (shortly followed by Ireland, Portugal, Spain, and Cyprus). The troika’s programmes implemented restrictive fiscal adjustment measures addressing the crisis as a consequence of budgetary imbalances and public debts. In effect, austerity policy focused on drastic cuts of public expenditure and on regulatory reforms flexibilizing wage and labour relations. In Greece, budgetary cuts signified dramatic wage and pension reductions (by nearly 30%) rather than cutbacks in welfare spending, which inherently has been poor in southern European peripheral economies (and lower than the European average). The drastic income drop and retreat of household consumption, combined with credit restrictions, has led to severe demand contraction and recession. In this way, austerity has pushed the disadvantaged Eurozone economies into a vicious circle of massive bankruptcies and layoffs (UN, 2012). Moreover, family networks—​typical in Greek society—​have become unable to provide income support and protection (to compensate welfare shortages). Yet, the degradation of the already weak social provision structures has proceeded on the institutional level, mainly through more flexible regulations concerning severance pay and notification period, benefit allowances, and collective labour agreements.

120   Tsampra Rising poverty has been an inevitable outcome of fiscal austerity throughout Europe, but especially in the most harmed EU economies (EU-​SILC, 2014). Even before the crisis, the highest rates of at-​risk-​of-​poverty were recorded in the Mediteranean and Baltic EU periphery (20% in Greece, Italy, Spain, and Lithuania; 23% in Romania; and 26% in Latvia) as expected for countries of weak welfare state and poor social provisions. After 2009, the crisis affected all European countries, but the EU peripheral economies suffered higher rates of poverty and at-​risk-​of-​poverty or social exclusion, as a result of deteriorating social and living conditions. In 2012, Greece recorded the EU28 highest at-​risk-​of-​poverty raise (35.8%) anchored in 2008, responding to 23.1 per cent of population. At the other end, only in six EU28 member states (Slovenia, Finland, Denmark, the Netherlands, and Czech Republic) was the at-​risk-​of-​poverty population less than 13.0 per cent (Figure 6.5). Greece also ranked first in EU28, with 73.1 per cent of its population reporting having difficulties making ends meet (followed by Bulgaria, Hungary, Croatia, Latvia, and Romania). In contrast, less than 10 per cent of the population in Germany, Finland, Sweden, and Norway reported such difficulties (EU-​SILC, 2014). During 2010–​12, GDP per capita in Greece declined by an average of 7 per cent every year, accumulating to a total loss of about 20 per cent (World Bank ECA, 2015). In the same period, household disposable income per capita fell 14 per cent annually, almost double the GDP contraction. As a matter of fact, the Greek economy has been the most harmed by the crisis: contraction (4% annual decline in GDP per capita) has lasted longer (six years of negative growth) and has far exceeded the contraction in other peripheral Eurozone economies (Spain, Portugal, Italy, or Ireland) over the same period. This has accumulated to a reduction of output of nearly 26 per cent, equivalent to that of over a decade back (in 2000).

Path-​Dependent Growth and Vulnerability in Greece Within the Fordist context of increased spatial specialization, Greece has been considered a ‘semi-​peripheral’ economy of Europe, or else a ‘peripheral-​Fordism’ economy (Selwyn, 1979; King, 1982; Lipietz, 1986), characterized by late industrialization, labour-​intensive subcontracting, resilient agriculture, dependence on imported capital goods and technology, outmigration and extensive atypical labour, strong family ties, and social networks sustaining entrepreneurship and employment (and largely mitigating welfare inefficiencies). Such characteristics, widespread in Mediterranean Europe, have not enabled core-​type growth in countries that today consistute the Eurozone’s economic periphery. Based on pre-​existing regional disparities, the spatial division of labour that emerged from the Fordist crisis of the late 1970s reproduced the core-​periphery polarization in a new pattern: the key economic functions remained concentrated in the core areas of Europe, while the peripheral economies partially deindustrialized and specialized in construction, real estate, and tourism (Stathakis, 2010; Becker and Weissenbacher, 2015). In Greece, the erosion of manufacturing and the turn to non-​exporting construction, real estate, and trade sectors was backed by increasing financialization soon after the country’s European Economic Community accession in 1981 (followed by Spain and

50

45

40

35

30

20

15

10

Bulgaria Romania Greece Latvia Hungary Lithuania Croatia Ireland Italy Cyprus Portugal Spain Poland UK Malta Estonia Euro area Belgium Slovenia Germany Slovakia Luxembourg Denmark Austria France Sweden Finland Netherlands Czech Rep. EU-28 AVERAGE

2009

2012

2013

2014

Figure 6.5   At Risk of Poverty or Social Exclusion (Percentage of Total Population).

Source: Eurostat (n.d.(b)).

Crisis and Austerity in Action: Greece    121

25

122   Tsampra 6 4 2 0 –2 –4 –6

% Employment

11

10

20

09

20

08

20

07

20

06

20

05

20

04

20

03

20

02

20

01

20

00

20

99

20

98

19

97

19

19

19

96

–8

% GDP

Figure 6.6   Gross Domestic Product (GDP) and Employment Rates in Greece, 1996–​2011. Source: EIEAD (2013).

Portugal in 1986). Strongly depending on household income and credit, these sectors have been producing fiscal and external deficits since the 1980s. In the 1990s, further restructuring processes were initiated, paving the country’s way to the EMU. Induced adjustments and reforms aimed at enhancing the international integration of the Greek economy. Large multinational corporations invaded all major sectors of the internal market—​trade, construction, real estate, and tourism—​competing with small-​scale (and, commonly, family owned) domestic enterprises. Yet, it was after 2009 that Greek SMEs in these sectors massively collapsed, owing to the financial crisis and austerity-​induced recession. The outcomes of the current crisis have added up to a ten-​year contraction of the country’s economic output and employment (Figure 6.6). Along with other European peripheral countries (Portugal, Hungary, Spain, Croatia, Lithuania, Latvia, and Romania), Greece has been facing decline of economic activity since the recession of the early 2000s, which affected developed countries and, particularly, the EU. Hence, we argue that the vulnerability of the Greek economy is ascribed to its path-​dependent peripherality, defined by inherited structures and evolving processes. The 2009 Eurozone crisis further exposed the structural weaknesses of vulnerable economies, while fiscal austerity, implemented in a context of deepening EU integration, has led them to downfall. Seeking to explain the disastrous outcomes of austerity policy in Greece, this chapter focuses particularly on the country’s industrial structures and patterns of entrepreneurship. Data on employment’s pre-​ crisis concentration and post-​ crisis shift across economic sectors depict the Greek economy’s industrial specialization and restructuring. In 2000–​08, the country’s ten top employment economic sectors accounted for 70 per cent of total national employment (Table 6.1). All sectors in the 2000 top ten participate in the 2008 top ten, too, except for food and drink manufacturing (displaced by motor vehicle trade). Retail trade outpaced agriculture, which had a significant employment drop (6%) from 2000 to 2008, although still held strong. All other sectors increased their share in national total employment—​particularly construction, public administration, education, and other (tertiary) business. The high and increasing employment shares of the top-​ranking sectors

Crisis and Austerity in Action: Greece    123 Table 6.1 Top Employment Sectors in Greece, Percentage 2000 and 2008 2000

2008

Rank

Sector

Share %

Sector

1

Agriculture

16.9

Retail trade

11.9

0.8

2

Retail trade

11.1

Agriculture

10.9

–​6.0

3

Public administration

7.4

Construction

8.4

1.2

4

Construction

7.2

Public administration

8.3

0.9

5

Tourism

6.7

Tourism

7.1

0.4

5

Education

6.1

Education

7.0

0.9

7

Health and social care

4.6

Other business

5.6

1.3

S

Other business

4.3

Health and social care

5.1

0.5

9

Wholesale trade

3.4

Wholesale trade

3.8

0.4

10

Food and drinks industry 2.9

Trade of motor vehicles

2.8

Total

70.6

Total

Share %

2000–​2008% change

70.9

0.3

Sources: INE GSEE (2011); POKE (2011); author’s calculations.

manifest the country’s pre-​crisis economic specialization in non-​exporting industries and non-​traded service provision, highly dependent on internal demand (even in the ‘extrovert’ tourism sector: accounting for 70 per cent of total demand in 2008) (McKinsey & Company, 2011). At the regional level, the five top employment economic sectors indicate a more diverse pattern of sectoral composition and specialization in 2008 (Table 6.2). The national structures coincide with the structures depicted in the capital region of Attiki, where employment accounted for 1.434 million people (one-​third of national employment in 2008) and was highly concentrated in retail trade, followed by manufacturing, public administration, and construction. Retail trade and construction were among the five most significant sectors for employment in nearly all thirteen Greek regions; public administration in eight and manufacturing in just five. The prevalence of agriculture as the top employer in eight out of thirteen regions, given the sector’s low (and declining) contribution (3.8%) to the national GDP, indicates regional lock-​in to structures of persisting peripherality, hindering restructuring prospects. Overall, Greece’s industrial structure and specialization since the 1980s has been largely based on private consumption (higher by 20% than the EU average) and domestic demand funded by expanding bank borrowing, forming a growth trajectory of low domestic investments and increasing deficits. After 2009, credit restrictions and fiscal austerity struck heavily the prevailing industries, resulting to steep employment decline: 900,000 jobs were lost in 2009–​13, mainly ascribed to construction (192,200), manufacturing (178,000), and trade (172,000). Agriculture seems more resilient, but despite much lower job losses the sector’s aforementioned structural weaknesses and declining performance do not suggest any compensating alternative (Table 6.3).

Greek regions

First sector

Second sector

Third sector

Fourth sector

Fifth sector

All sectors

Employment share

East Macedonia and Thrace

Agriculture

Public administration

Manufacturing

Retail trade

Construction

147,809

62.6%

Central Macedonia

Retail trade

Agriculture

Manufacturing

Construction

Other services

399,286

52.3%

Western Macedonia

Agriculture

Construction

Retail trade

Public administration

Education

53,311

54.4%

Epirus

Agriculture

Construction

Retail trade

Public administration

Education

73,881

51.7%

Thessaly

Agriculture

Retail trade

Public administration

Education

Manufacturing

174,904

58.4%

Ionian Islands

Tourism

Agriculture

Retail trade

Construction

Other services

64,790

68.4%

Western Greece

Agriculture

Retail trade

Construction

Education

Public administration

165,185

58.7%

Central Greece

Agriculture

Manufacturing

Retail trade

Construction

Public administration

130,403

58.3%

Attiki

Retail trade 207,530

Manufacturing 164,598

Public administration 154,850

Other business 133,537

Construction 128,288

788,803

45.6%

Peloponnese

Agriculture

Retail trade

Public administration

Construction

Education

161,201

64.0%

North Aegean

Public administration

Agriculture

Retail trade

Education

Construction

42,450

58.7%

South Aegean

Tourism

Retail trade

Construction

Public administration

Other services

76,682

61.0%

Crete

Agriculture

Tourism

Retail trade

Construction

Other trade

149,399

57.2%

Source: POKE (2011).

124   Tsampra

Table 6.2 Top Employment Sectors in the Regions of Greece, 2008

Crisis and Austerity in Action: Greece    125 Table 6.3 Top Employment Sectors in Greece, 2009 and 2013 2009

2013

2009–​13

2009–​13

Rank

Sector

(000)

Sector

(000)

Change

Change

1

Wholesale and retail trade

828.2

Wholesale and retail trade

656.2

–​172.0

–​20.8

2

Agriculture

529.5

Agriculture

494.0

–​3.5.6

–​6.7

3

Manufacturing

516.0

Manufacturing

338.0

–​178.0

–​34.5

4

Public administration

378.5

Public administration

333.4

–​45.2

–​11.9

5

Construction

356.5

Education

284.1

–​47.8

–​14.4

6

Education

331.9

Tourism

265.3

–​49.5

–​15.7

7

Tourism

314.8

Construction

172.4

–​192.2

–​52.4

3632.2

–​899.7

–​19.9

Total

4531.9

Total

Sources: POKE (2011); author’s calculations.

Table 6.4 Top-​six Employment-​loss Sectors (NACE rev2.0) Greece and EU27, 2008Q1–​2011Q1 Greece NACE

EU27 000s

NACE

000s

43

Specialized construction –​68.9 activities

43

Specialized construction –​1864.4 activities

41

Construction

–​66.5

41

Construction

18

Manufacture: printing and recorded media

–​23.0

46

Wholesale except motor –​500.7 vehicles

45

Wholesale/​retail of motor vehicles

–​21.2

25

Manufacture: fabricated –​469,1 metal products

14

Manufacture: clothing

–​14.2

47

Retail except motor vehicles

–​417.4

71

Architectural and engineering activities

–​13.3

14

Manufacture: clothing

–​407.5

–​947.8

NACE, Nomenclature statistique des activités économiques dans la Communauté européenne. Source: Eurofound (2011).

The pre-​and post-​crisis industrial/​sectoral patterns of employment in Greece offer a better insight into economic vulnerability when compared with respective EU patterns. Data for 2008 and 2011 depict the top-​six employment-​losing and employment-​gaining sectors in Greece and the EU27 (Eurofound, 2011). As observed, job loss illustrates a homogenous

126   Tsampra Table 6.5 Top-​six Employment-​gain Sectors (NACE rev2.0), Greece and EU27, 2008Q1–​2011Q1 Greece

EU 27

NACE

00os

NACE

000s

97

Domestic services

13.9

58

Publishing

8.5

85

Education

591.6

86

Health

581.0

82

Office and business support activities

7.2

87

Residential care

497.2

42 74

Civil engineering

6.7

88

Social work

377.6

Other professional and scientific activities

5.7

42

Civil engineering

326.6

1

Agriculture

5.6

70

Head officer management consultancy

269.3

NACE, Nomenclature statistique des activités économiques dans la Communauté européenne. Source: Eurofound (2011).

sectoral pattern in Greece and the EU27, mainly affecting construction, trade, and manufacturing (Table 6.4). However, differences are identified in employment gains (Table 6.5): the sectors of higher employment increase in the EU27 are education, health, and residential care services; while in Greece, domestic services (of low-​skilled atypical labour) record the highest employment recovery. Evidently, employment gains do not compensate for employment losses, either in Greece or the EU27; but the trend of job recovery in Greece is much lower than in EU27 (23% vs 57%, respectively). In other words, the Greek economy demonstrates a weak underlying growth dynamic, which undermines the prospects to resume pre-​crisis growth. Post-​crisis industrial and employment contraction proceeds along with drastic entrepreneurial restructuring: SMEs have had the highest losses all over Europe and especially in the EU periphery (EC, 2013). Still, more than 60 per cent of total EU employment and 50 per cent of gross value added (GVA) was attributed to SMEs in 2012 (ECORYS, 2012). In the beginning of the crisis, the steep reduction in the number of SMEs was due to the sharp retreat of consumer demand, and was particularly acute in the southern EU, where extreme austerity policies led to dramatic income drop. The decline was aggravated as the crisis evolved, owing to the restricted accessibility of SMEs to finance. Eventually, the gap between the EU core and periphery economies increased, as well as the gap between big enterprises and SMEs. As the SMEs constitute the bulk of the business sector in Greece, economic vulnerability to the crisis is traced in business patterns of sectoral specialization, size, ownership, and employment. Path-​dependent processes have differentiated Greek SMEs from their European counterparts in many aspects (SBA, 2013): in 2009, 84.8 per cent of private employment is concentrated in the SME sector and 54.5 per cent (double the EU27 average) in microenterprises (0–​9 employees). At that time, large firms accounted just for 15.2 per cent of private employment (half the EU27 average). The total share of very small, small, and medium enterprises

Crisis and Austerity in Action: Greece    127 in GVA (69%) was also considerably higher than that of large firms (31%) in Greece. Comparatively to the EU27, the share of SMEs in GVA was 57.4 per cent and that of large firms was 42.6 per cent. As indicated, the role of SMEs and particulary of micro-​firms has been very important in the Greek economy, against the lower contribution of large firms. The significance of small entrepreneurship has been nourished by the economy’s semi-​ Fordist legacy of petty ownership, flexible industrialization, atypical labour in the form of self-​employment without employees, social networking, and family assistance. Inherited and evolving structures of the Greek business sector have for long sustained broad socio-​economic cohesion and resilience. Even in 2012, amidst the crisis, the prevalence of SMEs in the Greek economy was still indicated by high shares in national employment (85.1%) and GVA (69.9%) (EC, 2013). Nevertheless, the dynamic trajectory of the Greek SMEs was reversed by the financial crisis. The drastic reduction of internal purchasing power and access to cash liquidity has led to prolonged recession, destructive for small entrepreneurship. Over 2008–​12, approximately 250,000 SMEs have permanently vanished from the country’s business map. The loss is estimated at 25 per cent of Greek enterprises, equivalent to the loss of national GDP and employment over that period (EC, 2013). Estimates for up to 2014 illustrate the much steeper decline of Greek SMEs number, employment, and GVA since 2008, compared with EU SMEs (Figure 6.7). Today, after six years of recession, employment in SMEs is estimated to have fallen by more than 450,000 between 2008 and 2014, to 1.8 million; aggregate value added fell by 33 per cent (SBA, 2015). In effect, SMEs have suffered profoundly and disproportionately more than large enterprises from the downturn. On the one hand, the increased vulnerability of the majority of Greek enterprises to the crisis is ascribed to path-​dependent attributes, such as the predominance of micro-​entrepreneurship, the high rates of sole proprietorship, the strong tradition of self-​employment, and family ownership. To establish a few: only 220,000 out of 730,000 Greek SMEs had employees in 2012 (SBA, 2013). The surviving 580,000 Greek SMEs in 2014 generate €93 billion turnover; four of five of them are sole proprietorships and generate 40 per cent of SMEs’ turnover (National Bank of Greece, 2014). In other words, the same attributes that have previously sustained business resilience in Greece are liable for SMEs’ structural weaknesses: shallow specialization and high dependence on household demand, low internationalization and innovation, poor investments, self-​sustained employment, obsolete corporate forms and management, weak productivity, and competitiveness. On the other hand, the drastic blow to Greek enterprises is also a consequence of strong specialization in sectors severely hit by the crisis and economic downturn. At the EU28 level, the most important sector for SMEs—​in terms of employment, number of enterprises, and value added—​is trade, followed by manufacturing and construction (SBA, 2015). In Greece, SMEs concentrate 80 per cent of employment in manufacturing, 90 per cent in trade and 95 per cent in the construction sector (far exceeding the EU average, especially in manufacturing and trade). In construction, employment has spectacularly increased by 30.2 per cent in the 2000–​08 period of growth after the country’s EMU, based on the development of infrastructures and mega-​projects, real estate investments, and increased credit flow. The housing market had a significant contribution in sectoral growth, indicated in residential property prices raise by 80.5 per cent over 2000–​07 (against 52.7% in the Eurozone) (https://​www.ecb. europa.eu/​stats/​services/​sdw/​html/​index.en.html/​).

SME employment 110

110 100

100

90

90

80

80

70

70

60

128   Tsampra

SME number

2008

2009

2010

2011e

2012e

2013e

60

2014e

2008

2009

2010

2013e

2014e

2011e

2012e

2013e

2014e

SME gross value added 110 100 90 80 70

2008

2009

2010

2011e

Greek SMEs

2012e

EU SMEs

Figure 6.7   Greek and European Union (EU) Small-​and Medium-​sized Enterprise (SME) Performance, 2008–​14 (estimates (e) for 2011 onwards (2008 = 100)) Sources: Cambridge Econometrics (n.d.)/​ECORYS (2011); IME GSEVEE (2012); SBA (2014).

Crisis and Austerity in Action: Greece    129 Nonetheless, rapid increase was abruptly reversed after 2009, as the housing market collapse has been a particular implication of the financial crisis in Greece (Spain and Ireland). Contraction resulted to housing foreclosures and vacant building stock, comprising unsold or unrented real estate, residential, and commercial property. Non-​performing housing loans exceeded 27 per cent of total non-​ performing loans in 2013 (i.e. €16.8 billion) and keep rising (National Bank of Greece, 2014). Over 2008–​12, housing prices fell by 70 per cent as credit restrictions and high taxation forced property owners to sell low (National Bank of Greece, 2012). Decline in construction was big, but SMEs suffered the most as they were exclusively focused on private demand. Fiscal austerity also hit public-​ funded construction, turning large enterprises towards small-​scale projects and eventually putting SMEs out of business. The number and combined turnover of SMEs halved during that period, value added dropped by 40 per cent and employment by 43 per cent (SBA, 2015). The effect of this policy was the higher ranking of Athens among other ‘cities offering the best real estate investment prospects’ (PwC-​ULI, 2015). In manufacturing, Greek SMEs have traditionally specialized in labour-​intensive production and low-​cost subcontracting (Clark et al., 2004). Decline has started since the deindustrialization of the 1980s and peaked amidst the crisis of early 2000s, as production relocated to lower-​cost Balkan regions (Labrianidis, 2000). As a result, the sector’s share in national GDP fell from its highest 20 per cent in the mid-​1970s to 10 per cent in the 2000s. Since the 2008 crisis, manufacturing SMEs lost 20 per cent of their population, 29 per cent of their employment, and 23 per cent of value added. Big manufacturing enterprises did not perform better: their employment decreased by 27 per cent and added value by 35 per cent (SBA, 2014). Business outmigration has been rekindled: 1500–​2000 Greek SMEs relocated to neighbouring countries between 2010 and 2011 (ΙΝΕΜΥ ESEE, 2010; IME GSEVEE, 2012). Yet, considering pre-​and post-​crisis job loss in manufacturing, the share ascribed to business bankruptcy and closure has increased over the share ascribed to business off-​shoring or relocation (Eurofound, 2013). In trade, losses have also been heavy as the Greek economy suffered consumption decline bigger than the EU average (12.4% vs 6.7%). The total value added decreased by 14.6 per cent in 2011 (vs 3.8% in 2010), against an increase of 1.3 per cent in the Eurozone (ΙΝΕΜΥ ESEE, 2013). In 2013, trade SMEs recorded a 12.9 per cent decrease in sales and 13.4 per cent in gross profits for the fourth year in a row. Losses were higher for smaller firms, as turnover decreased by more than 45 per cent and profit by more than 84 per cent. From 2008 to 2014, approximately 130,000 commercial firms closed down and 200,000 jobs were lost (GSEVEE et al., 2014). Despite losses, trade is still the top employment sector in Greece, accounting for 18.1 per cent of total and 20.9 per cent of non-​agricultural employment in 2013: 31 per cent are employers, 34.7 per cent are self-​employed, 44 per cent are assisting family members, and just 15 per cent are waged employees (Table 6.6). The legacy of micro-​entrepreneurship, reflected in the high shares of employers, self-​ employed, and assisting family members in the major sectors of SME specialization, is a structural characteristic commonly found in all peripheral economies of Europe. According to Eurostat (n.d. (a)), the southern and eastern EU member states have the greatest share of self-​employed in EU24, with the highest rate occurring in Greece (30%). Greece also records one of the lowest EU24 rates of self-​employed with employees. This pattern is related to the

130   Tsampra Table 6.6 Occupational Structure of Employment (000s) in Trade, Greece, 2008–​13 2008

2013

Employers (000s)

105.6

68.5

–​38.1

Self-​employed (000s)

219.4

174.4

–​45.0

Waged employees (000s)

439.3

374.2

–​65.1

Assisting family members (000s) Total (000s)

Job losses 2008–​13

57.3

39.1

–​28.2

832.6

656.20

–​176.4

Sources: Eurostat (n.d. (a), n.d. (b)); author’s calculations.

prominence of agricultural, service-​based, and informal work in countries such as Greece, Italy, Poland, and Spain. Self-​employment is also a way out of unemployment, sustained by strong family and social relations in labour markets of poor conditions. The prolonged recession has therefore hit the business owners (employers and self-​employed) rather than the employees. On this ground, austerity-​induced restructuring of the business sector has actually demolished the predominant small entrepreneurship in Greece, with no prospect for recovery.

Discussion: Crisis and Austerity in Action Path-​dependence and austerity policy are considered in this chapter as critical determinants of the vulnerability and the adjustment capacity of national/​regional economies to crisis. Growth disparities are attributed to place-​based growth patterns, shaped by inherited and evolving production structures and institutional arrangements. As the global financial crisis unfolded in Europe, the weaker economies of the southern Eurozone periphery have been hit more severely than the core economies, and Greece has suffered the most steep and prolonged decline. The vulnerability of the Greek economy lies in its path-​dependent structural weaknesses traced in industrial specialization and entrepreneurial patterns. The fiscal austerity policy implemented to tackle economic disruption in Greece (and all debt-​ridden EU economies) has further exposed vulnerable structures and aggravated the impact of the crisis. Greece’s growth trajectory within the broader European economy has been defined by semi-​Fordist structures, which evolved to post-​Fordist patterns of deindustrialization and restructuring following the country’s European Economic Community (EEC) accession in 1981. Yet, the mode of EU integration and internationalization of the late-​industrialized economies of Greece, Portugal, and Spain has not eliminated inherent ‘peripherality’ structures. The emergent spatial division of labour was based on a less diversified sectoral specialization where key functions remained in the core economies of Europe; and the peripheral economies of the south were incorporated mainly through infrastructure-​led

Crisis and Austerity in Action: Greece    131 growth. Greece’s post-​EEC accession production structures and regulatory configurations comprised flexible industrial specialization in demand-​dependent sectors and in low-​cost subcontracting, small-​ scale entrepreneurship, traditional self-​ employment, extensive atypical labour and outmigration to advanced industrial countries (Germany, Belgium), medium-​level wages, and weak welfare state. Accelerated globalization processes and the opening of Eastern European economies in the 1990s affected the Greek economy in two ways: firstly, the massive offshoring of traditional (flexible and low-​cost) manufacturing SMEs in low-​cost neighbouring regions was triggered; secondly, further restructuring processes towards the country’s integration in the Eurozone (EMU) were initiated. Place-​based path-​dependent economic and institutional structures were reconfigured accordingly, by industrial policies and regulatory reforms promoting Greece’s deeper EU integration through compliance with European patterns. This process signified the entry of large multinationals in the Greek market and increased competition in all major sectors of the national economy. Yet, the EU-​induced changes in the mode of regulation have not achieved the dissolution of peripherality patterns: the Greek econ­ omy’s specialization in non-​exporting and demand-​dependent industries was intensified. The country’s participation in the EMU since 2001 signalled the turn to finance-​led growth, through cheap credit and consumption boost in the prevailing sectors of trade, construction, and real estate. Multinational corporate groups consolidate their place in the Greek market and gradually take over (or put out of business) ‘traditional’ independent (often family-​owned or/​and sustained) small and micro enterprises in these sectors. However, the 2008–​09 financial turmoil triggered new transformations of the established growth patterns. The crisis can be thus considered as an episode in the evolution of the global accumulation regime, stimulating regulatory experimentations to secure its stability. Such experimentations have been promoted through austerity policies implemented all over Europe by supranational institutions as the IMF and the EC, to overcome the crisis by launching a new socio-​economic configuration. Within the outlined context (depicted in Table 6.7), a major issue of discussion concerns the very nature of the crisis. The consideration of its origins and driving forces allows for a comprehensive explanation of its spatially differentiated patterns and outcomes. The shift of the global financial crisis to sovereign debt crisis in the Eurozone has been considered itself a cause of increasing inequality: public debt-​to-​GDP ratios range from 171.8 per cent in Greece, 132.9 per cent in Italy, and 128.7 per cent in Portugal to 10.0 per cent in Estonia, 17.3 per cent in Bulgaria, and 27.7 per cent in Luxemburg; Greece and Portugal also had the highest relative increase of public debt-​to-​GDP ratio during the crisis (Eurostatn.d.(a)). As a result, the crisis hit more severely the countries of problematic public finances, namely the southern Eurozone member states. Still, for many authors, the argument that the crisis was caused by different forms of imbalances and public debts ignores the deeper core-​periphery contradictions of the European economy (Becker and Weissenbacher, 2012). The economic landscape is shaped by place-​specific historical circumstances, inherited and emerging structures and processes. The evolving spatial division of labour incorporates past growth patterns within new growth trajectories. Accordingly, we sought to explain the spatially divergent impact of the crisis in Europe by associating the much more severe decline and economic contraction in the southern EU member states, with inherent structures of ‘peripherality’. The structural weaknesses of these late-​industrialized countries of ‘peripheral Fordism/​capitalism’ were dramatically exacerbated by the crisis and materialized

The 1980s following EU accession

The 1990s on the way to EMU

The 2000s in the Eurozone

Since the 2009 financial/​sovereign debt crisis

• Restructuring towards infrastructure-​led growth

• Financialization of the economy

• Consumption collapse in trade, construction, real estate, and manufacturing

• Strong agriculture

• Manufacturing decline

• Growing trade, tourism, construction, services

• Growing trade, construction and real estate, tourism, services

• Investments in trade, construction and real estate, tourism and leisure

Industrial structures • Partial de-​industrialization and specialization • Low-​cost manufacturing subcontracting

Entrepreneurship /​ • Predominance of small and business structures microentreprises • Sole proprietorship, family ownership

• SME resilience • Manufacturing business offshoring in lower-​cost countries

• Credit boost of consumption

• FDI in trade, real estate and construction, tourism and leisure • Privatization of public/​national assets: real estate property, energy, and telecommunications and transport infrastructures (ports, airports, highways)

• Consolidation of MNC groups • Massive business closures in the Greek market • Housing market collapse • Downfall of the traditional SME sector • Large MNCs take over SMEs

• Increased international competition Occupational/​ employment patterns

• Strong self-​employment

• Strong self-​employment

• Atypical employment in agriculture, tourism, and subcontracting

• Migration inflows from Eastern Europe

• Typical employment forms prevail

• Increasing atypical employment forms

• Resilient self-​employment • Institutionalization of labour flexibility

• Massive layoffs and employment decline • Escalating unemployment, long-​term and youth unemployment • Decreasing self-​employment • Expanding atypical employment

• Outmigration to industrial countries EU, European Union; EMU, European Monetary Union; SME, small-​and medium-​sized enterprise; MNC, multinational corporation; FDI, foreign direct investment. Source: Author.

132   Tsampra

Table 6.7 Greece’s Path-​dependent Growth Patterns from the 1980s to 2008–​09 Crisis

Crisis and Austerity in Action: Greece    133 into increased public deficits and excessive sovereign debts. The ensuing fiscal polarization within the Eurozone indicates that, despite restructuring processes towards EU integration over the last decades, the peripheral member states have not accomplished core-​type growth patterns. Under the pressure of financial capital, peripherality has been reproduced in a new form undermining the very substance of the EU integration. On this ground, the author of this chapter adopts the view that increased public deficits were the effect rather than the cause of the economic crisis (Becker and Jäger, 2010). The destructiveness of the 2009 crisis in Greece is related to the national mode of growth (production and consumption patterns) defined by the national economy’s role in the international division of labour (Massey, 1984; Lipietz, 1986). The structural weaknesses and vulnerability of the Greek economy, explored in the previous section, are ascribed to inherited peripherality patterns and to patterns outlined through the country’s EU and Eurozone integration. As revealed, inherent peripherality and European integration processes have so far not enabled the adjustment of Greece’s path-​dependent pattern to a more dynamic and sustainable growth trajectory. On the contrary, in the exigency of the crisis and public bankruptcy, the country has been subjected to bailout programmes of structural and regulatory reforms with disastrous implications for socio-​economic stability. No other EU country has had a GDP decline, or income and wage drop to such low levels as Greece. From a state of growth convergence to the EU average, the Greek economy shifted to deprivation reflected in business and employment collapse. Hence, the issue raised here concerns the policy addressing the crisis, its objectives, and outcomes. Since 2010, the troika’s programmes for Greece have pursued ‘improv[ing] competitiveness through internal devaluation’ (IMF, 2010). The key element of this strategy was a very restrictive fiscal policy cutting public expenditures and reducing social spending. Fiscal austerity was eventually adopted in all debt-​ridden Eurozone economies (Ireland, Portugal, Spain, and Cyprus) through bailout packages. Austerity measures in Greece (and Ireland and Portugal) particularly focused on drastic budgetary cuts, wage and pension severe reduction, and on the flexibilization of the wage relation. It is noteworthy that the change of the wage relation is a fundamental prerequisite of the EC/​IMF programmes: in this way, national arrangements of wage policies and social spending passed under supranational supervision and regulation—​especially in the indebted Eurozone countries. Gradually, a mode of non-​spatialized governance and non-​localized policies has been established through mechanisms that circumvent democratic processes and public accountability. That is the case of the ‘atypical’ Eurogroup decision-​making authority, or the troika’s technical evaluation authority. Under the austerity restrictions and the burden of public deficit, the EC and the IMF have also demanded an extended privatization programme of national assets in Greece (and Portugal and other peripheral EU economies) to attract investments. Following the aforementioned pattern, an ‘independent’ organization—​supervised by the Eurogroup and the EC—​was institutionalized to this purpose: the Hellenic Republic Asset Development Fund (HRADF) was established in 2011, under the medium-​term fiscal strategy, by a new law that aimed to restrict governmental intervention in the privatization process, and its further development within a fully professional context (L.3986/​2011). Adopting the troika’s priorities, the HRADF’s privatization portfolio comprises real estate property (mainly beachfronts and coastal areas, Olympic assets, historic buildings, and hotels), transportation (ports,

134   Tsampra and airports, railways and motorways), and other (e.g. water supply and sewerage) infrastructures and corporate organizations (Hellenic Post, etc.). The programme has so far completed privatizations mostly related to investments in residential and commercial development, tourism, and leisure. A quite revealing fact is that in 2014, the highest foreign direct investment (FDI) inflow in Greece was attributed to the privatization of Hellenikon in metropolitan Athens—​a coastal area intended for tourism, shopping, leisure, and residential land use. A year earlier, it was the Canadian Fairfax Financial Holding Ltd for equity investments in Eurobank Properties. Many more ‘strategic investments’ of such orientation are also among the top FDIs in Greece, considered as the successful outcome of ‘national assets development policy’. Today, even higher FDIs are expected from the privatization of the country’s fourteen regional airports (acquired by Fraport for €1.23 billion), also related to tourism; and from the agreement with Cosco for the central port of Pireaus (expected to be the main entrance of Asian commercial exports to Central Europe). Yet, investments in productive industries rank very low. After six years of recession, the debate on crisis and austerity in Greece is still going on, as reflected in sociopolitical upheaval. The main question concerns the effectiveness of austerity policy in terms of the Greek economy’s recovery. For many, this policy has merely served the country’s abrupt socio-​economic restructuring and enforced compliance to EU and global accumulation imperatives. To date, the outcome has been the demolition of Greece’s path-​ dependent SME-​ centred patterns of industrial specialization, entrepreneurship, employment, and socio-​economic cohesion. Massive business closures and layoffs (both in public and private sectors), income contraction, extreme unemployment, and increasing poverty have undermined the national economy’s capacity to resume growth. Policy measures adopted to allegedly increase the competitive integration of the Greek economy in the Eurozone have dismantled labour relations and social rights, leading to the verge of a humanitarian crisis. Institutional mechanisms controlling public income and expenditures, adjusting pensions and wages, and enhancing employment and wage flexibilization have gradually formed a low-​cost, low-​wage market in the Eurozone periphery.

Concluding Remarks This chapter has addressed the issue of economic vulnerability and uneven prosperity in Europe, in the context of the 2008–​09 financial crisis and the austerity policy addressing it. The turn of the global financial crisis to sovereign debt crisis in the EU has exposed the structural weaknesses of the most vulnerable Eurozone member states. The Greek economy has been in turmoil as soon as the country’s public debt sustainability and credit worthiness was questioned. Yet, increased public deficits have been rather the effect than the cause of the unfolding crisis in a context of intense financial pressure and underlying growth polarization. The extreme fiscal austerity policy that was implemented to tackle the crisis has instead aggravated its impact and accentuated disparities between the EU core and peripheral economies. Southern Eurozone member states have had the sharpest decline, and Greece, in particular, being the first EU country to accept a bailout package and adjunct memorandum in 2010, has suffered the most severe economic contraction and prolonged recession. This fact has

Crisis and Austerity in Action: Greece    135 raised questions about the causes of economic vulnerability and adjustment capacity to the economic shock. The answer is sought in Greece’s path-​dependent patterns of growth that have been defined by a semi-​peripheral legacy and processes of integration in the EU and the EMU. Preceding and evolving industrial specializations, and entrepreneurial and employment patterns have been incorporated through regulatory configurations and formed Greece’s socio-​ economic setting and capacity to adjust. The path-​dependent attributes of the Greek economy are epitomized in non-​exporting distributive and credit-​dependent industrial specialization; prominent petty ownership and micro-​entrepreneurship; traditional self-​employment and extensive atypical labour. High dependence on trade and construction, the sectors most heavily hit by recession, has resulted in market collapse, entrepreneurial downfall, and massive unemployment. As revealed, restructuring processes that have deepend the country’s integration within the European economy have not dissolved ‘peripherality’ structures: private consumption had been further promoted to sustain retail trade and leisure, and speculatory investments in real estate had been reinforced. This strategy was common in all debt-​ridden member-​states of the EU south (Portugal, Spain, Cyprus) and overlooked the inherent contradiction between ‘core’ and ‘periphery’ economies. On this ground, austerity policy has sought to tackle economic disruption through restrictive fiscal measures and abrupt regulatory reforms. Given the specificities of the Greek ecomy, austerity has accelerated the collapse of consumption, entrepreneurship and employment, and the rise of unemployment and poverty. Fiscal austerity measures inhibit development initiatives and result in falls in GDP and employment to a permanently lower trajectory (EC, 2009). The deep and prolonged recession has dangerously downgraded the country’s capacity to resume growth: a depressed economy cannot provide favourable conditions for sectoral restructuring, or for investments enhancing productivity. Long-​term unemployment tends to lead to permanent loss of skills; low investment implies the devaluation of production equipment and infrastructure; and vast youth unemployment threatens the socio-​economic cohesion. As indicated, the outcome of the austerity measures and reforms, being part and parcel of rescue packages and adjunct memoranda in Greece, is demonstrated by the emergent economic and institutional context, which is favourable for FDI and other large-​scale investments in the previously prominent and now degraded sectors of (wholesale and retail) trade, construction and real state, and tourism and leisure services. In other words, austerity-​ induced restructuring has readjusted the ‘mode of regulation’ to better serve the evolving finance-​led global ‘accumulation regime’. Still, the structural weaknesses of the Greek economy—​identified in the need for knowledge-​intensive specialization, entrepreneurial modernization, and competitive internationalization—​have not been addressed. The bailout programmes even had a negative impact on the debt of Greece; but, despite failures, their objectives are being adopted throughout the Eurozone. The issue of transferring decision-​making processes and governance to supranational organizations (i.e. IMF/​ECB/​EC troika) has been central in the implementation of austerity policy. This strategy has downgraded the political role of legitimate national and regional authorities, and has resulted in one-​size-​fits-​all and picking-​the-​winner policies. The disregard of the Greek economy’s history and path-​dependent specificities has not stimulated a more competitive structural adjustment to EU and global developments. On the contrary, austerity policy applied a drastic blow to the country’s vulnerable patterns:

136   Tsampra acute budgetary and credit cuts along with further ‘flexibilization’ reforms have demolished the established socio-​economic formation—​which, despite weaknesses, has for long sustained resilience, cohesion, and inclusive prosperity in Greece. The catastrophic outcomes of crisis and austerity may permanently disintegrate the historically formed alliance between government, domestic business elites, and the middle class in Greece, leading to sociopolitical upheaval.

References Aglietta, M. (1979). A Theory of Capitalist Regulation: The US Experience (London: New Left Books). Becker, J. and Jäger, J. (2010). ‘Development trajectories in the crisis in Europe’. Debatte: Journal of Contemporary Central and Eastern Europe 18: 5–​27. Becker, J. and Weissenbacher, R. (2012). ‘Berlin Consensus and Disintegration. Monetary Regime and Uneven Development in the EU’. 18th Conference on Alternative Economic Policy in Europe (EuroMemo Group). Poznan, Polen http://​www2.euromemorandum.eu/​ uploads/​becker_​weissenbacher_​berlin_​consensus_​and_​disintegration.pdf (last accessed 15 March 2016). Becker, J., Jäger, J., and Weissenbacher, R. (2015). ‘Uneven and Dependent Development in Europe: The Crisis and its Implications’ in Jäger, J. and Springler, E. (eds.) Asymmetric Crisis in Europe and Possible Futures. Critical Political Economy and Post-​Keynesian Perspectives. RIPE Series in Global Political Economy, pp. 81–​97 (Abingdon: Routledge). Boschma, R. (2004). ‘Competitiveness of regions from an evolutionary perspective’. Regional Studies 38: 1001–​1014. Boyer, R. (1979). ‘La crise actuelle: une mise en perspective historique’. Critiques de l’économie politique 7–​8: 5–​112. Boyer, R. (2013). ‘Capital in the twenty-​first century: a régulationist view’. Revue de la régulation http://​regulation.revues.org/​10618 (last accessed 15 March 2016). Cambridge Econometics (n.d.). ‘European regional data’. https://​www.camecon.com/​european-​regional-​data/​ (last accessed 23 May 2017). Clark, G.L., Palaskas, T., Tracey, P., and Tsampra, M. (2004). ‘Globalization and competitive strategy in Europe’s vulnerable regions: firm, industry and country effects in labour-​ intensive industries’. Regional Studies 38: 1085–​1100. Cooke, P. and Morgan, K. (1998). The Associational Economy (Oxford: University Press). Cox, K. (1996). ‘Period and Place, Capitalist Development, and the Flexible Specialization Debate’ in Knudsen, D.C. (ed.) The Transition to Flexibility, pp. 155–​177 (London: Kluwer). Davies, S. (2011). ‘Regional resilience in the 2008–​2010 downturn: comparative evidence from European countries’. Cambridge Journal of Regions, Economy and Society 4: 369–​382. Dunford, M. (1989). ‘State-​industry relations, inter-​firm relations and regional development’. Geography 74: 20–​28. Dunford, M. (1990). ‘Theories of regulation’. Environment and Planning D 8: 297–​321. EC (2009). ‘Economic crisis in Europe:  causes, consequences and responses’. European Economy 7 (Brussels: European Commission). EC (2013). ‘A recovery on the horizon?’ Annual Report on European SMEs 2012/​ 2013 (Brussels: European Commission).

Crisis and Austerity in Action: Greece    137 ECORYS (2011). ‘Are EU SMEs recovering from the crisis?’ Annual Report on EU Small and Medium sized Enterprises 2010/​2011 (Rotterdam: ECORYS). ECORYS (2012). EU SMEs in 2012: At the Crossroads. Annual Report on Small and Medium-​ sized Enterprises in the EU, 2011/​12 (Rotterdam: ECORYS). EIEAD (2013). Labour and Employment in Greece:  Annual Report 2012 (Athens:  National Institute of Labour and Human Resources) (in Greek). European Commission (2015). ‘Employment and social developments in Europe 2015’ http://​ ec.europa.eu/​social/​main.jsp?catId=738&langId=en&pubId=7859&furtherPubs=yes (last accessed 23 May 2017). EU-​SILC (2014). Living Conditions in Europe (Brussels: EU). Eurofound (2011). Public Instruments to Support Restructuring in Europe:  European Restructuring Monitor (Luxembourg: Publications Office of the European Union). Eurofound (2013). Monitoring and Managing Restructuring in the 21st Century:  European Restructuring Monitor (Luxembourg: Publications Office of the European Union). Eurostat (2012). Eurostat (n.d. (a)). ‘Labour market and labour force survey (LFS) statistics’ http://​ec.europa. eu/​eurostat/​statistics-​explained/​index.php/​Labour_​market_​and_​Labour_​force_​survey_​ (LFS)_​statistics (last accessed 23 May 2017). Eurostat (n.d. (b)). ‘European Union statistics on income and living conditions (EU-​SILC)’. http://​ec.europa.eu/​eurostat/​web/​microdata/​european-​union-​statistics-​on-​income-​and-​ living-​conditions (last accessed 23 May 2017). Feldman, M. (2005). ‘The Entrepreneurial Event Revisited: Firm Formation in a Regional Context’ in S. Breschi and F. Malerba (eds) Clusters, Networks and Innovation, pp. 136–​168 (Oxford: Oxford University Press). Gertler, M. (2005). ‘Tacit Knowledge, Path Dependency and Local Trajectories of Growth’ in G. Fuchs and P. Shapira (eds) Rethinking Regional Innovation and Change, pp. 23–​42 (New York: Springer Verlag). Grabher, G. (1993). ‘The Weakness of Strong Ties: The Lock-​in of Regional Development in the Ruhr Area’ in G. Grabher (ed.) The Embedded Firm, pp. 255–​277 (London: Routledge). GSEVEE, ESEE, KEEE, TFGR, SETE (2014). ‘The development of SMEs in Greece. Policy Document’ http://​www.gsevee.gr/​press/​mme_​eng.pdf (last accessed 23 May 2017). Harvey, D. (1975). ‘The geography of capitalist accumulation: a reconstruction of the Marxian theory’. Antipode 7(2): 1–​61. Harvey, D. (1985). ‘The Geopolitics of Capitalism’ in D. Gregory and J. Urry (eds) Social Relations and Spatial Structures, pp. 128–​163 (London: Macmillan). Harvey, D. (2006). Spaces of Global Capitalism: A Theory of Uneven Geographical Development (London: Verso). Hassink, R. (2005). ‘How to unlock regional economies from path dependency? From learning region to learning cluster’. European Planning Studies 13(4): 521–​535. Hassink, R. (2010). ‘Regional resilience: a promising concept to explain differences in regional economic adaptability?’ Cambridge Journal of Regions, Economy and Society 3(1): 45–​58. IILS (2013). World of Work Report 2013:  Repairing the Economic and Social Fabric (Geneva: International Labour Organization). IME GSEVEE (2012). Trends of Economic Climate (Athens:  Hellenic Confederation of Professionals, Craftsmen and Merchants). IMF (2010). ‘Greece: request for stand-​by arrangement’. International Monetary Fund Country Report 10/​111, May.

138   Tsampra ΙΝΕΜΥ ESEE (2010). Annual Greek Trade Report (Athens:  Hellenic Confederation of Commerce and Entrepreneurship). ΙΝΕΜΥ ESEE (2013). Annual Greek Trade Report (Athens:  Institute for Commerce and Services of the Hellenic Confederation of Commerce and Entrepreneurship). Jessop, B. (2001). ‘Capitalism, the Regulation Approach, and Critical Realism’ in A. Brown, S. Fleetwood, and J.M. Roberts (eds) Critical Realism and Marxism, pp. 88–​115 (London: Routledge). Jessop, B. (2008). State Power: A Strategic–​Relational Approach (Cambridge: Polity). King, R. (1982). ‘Southern Europe: dependency or development?’ Geography 67: 221–​234. Labrianidis, L. (2000). ‘Are Greek Companies that Invest in the Balkans in the 1990s Transnational Companies?’ in A. Mitsos and E. Mossialos (eds) The Contribution of a Changing Greece to the European Union, pp. 457–​482 (London: Ashgate Press). Leborgne, D. and Lipietz, A. (1988). ‘New technologies, new modes of regulation: some spatial implications’. Environment and Planning D 6: 263–​280. Lipietz, A. (1982). ‘Towards global Fordism?’ New Left Review I/​132: 33–​47. Lipietz, A. (1986). ‘New Tendencies in the International Division of Labor:  Regimes of Accumulation and Modes of Regulation’ in A. Scott and M. Storper, M. (eds) Production, Work, Territory, pp. 16–​40 (London: Allen and Unwin). Lipietz, A. (1987). Mirages and Miracles: The Crises of Global Fordism (London: Verso). McKinsey & Company (2011). Greece 10 Years Ahead: Defining the New National Development Model. Summary (Athens: McKinsey & Company). Martin, R. (2010). ‘Rethinking regional path dependence:  beyond lock-​in to evolution’. Economic Geography 86: 1–​27. Martin, R. (2012). ‘Regional economic resilience, hysteresis and recessionary shocks’. Journal of Economic Geography 12: 1–​32. Martin, R. and Sunley, P. (1996). ‘Paul Krugman’s geographical economics and its implications for regional development theory: a critical assessment’. Economic Geography 72: 259–​292. Martin, R. and Sunley, P. (2006). ‘Path dependence and regional economic evolution’. Journal of Economic Geography 6: 395–​437. Martin, R. and Sunley, P. (2007). ‘Complexity thinking and evolutionary economic geography’. Journal of Economic Geography 7: 573–​601. Martin, R. and Sunley, P. (2011). ‘The New Economic Geography:  Credible Models of the Economic Landscape?’ in R. Lee, A. Leyshon, L. McDowell, and P. Sunley (eds) The SAGE Handbook of Economic Geography, pp. 53–​72 (London: SAGE). Massey, D. (1984). Spatial Divisions of Labour: Social Structures and the Geography of Production (New York: Methuen). Massey, D. and Meegan, R. (1978). ‘Industrial restructuring versus the regions’. Urban Studies 15: 273–​288. Moulaert, F. and Swyngedouw, E. (1989). ‘A regulation approach to the geography of flexible production systems’. Environment and Planning D 7: 327–​345. National Bank of Greece (2012). ‘The economic crisis in Greece, causes and implications’. Report (Athens: National Bank of Greece, Strategic Planning and Research Division). National Bank of Greece (2014). ‘Survey of Greek SMEs’. Report (Athens: National Bank of Greece, Strategic Planning and Research Division). POKE (Observatory of Economic and Social Developments of INE-​ GSEE) (2011). Restructurings and Sectoral Specialisation of the Greek Economy (Athens:  INE GSEE) (in Greek).

Crisis and Austerity in Action: Greece    139 PwC-​ULI (2015). ‘Emerging Trends in Real Estate: Europe’ http://​www.pwc.com/​gx/​en/​industries/​financial-​services/​asset-​management/​emerging-​trends-​real-​estate.html (last accessed 16 March 2016). SBA (Small Business Act for Europe) (2013). 2013 SBA Fact Sheet: Greece (Brussels: European Commission). SBA (Small Business Act for Europe) (2014). 2014 SBA Fact Sheet: Greece (Brussels: European Commission). SBA (Small Business Act for Europe) (2015). 2015 SBA Fact Sheet: Greece (Brussels: European Commission). Scott, A. (2006). Geography and Economy (Oxford: Oxford University Press). Selwyn P. (1979). ‘Some Thoughts on Cores and Peripheries’ in D. Seers, B. Schaffer, and M.L. Kiliunen (eds) Underdeveloped Europe:  Studies in Core-​Periphery Relations, pp. 35–​44 (Atlantic Highlands, NJ: Humanities Press). Smith, N. (1984). Uneven Development (Oxford: Blackwell). Stathakis, G. (2010). ‘The fiscal crisis of the Greek economy’. Kurswechsel 3: 109–​114. Storper, M. (1997). The Regional World (London: Guilford Press). Sydow, J., Schreyögg, G., and Koch, J. (2005). ‘Organizational Paths: Path Dependency and Beyond’. Paper at 21st EGOS Colloquium 30 June–​2 July 2005, Berlin, Germany. UN (2012). World Economic Situation and Prospects (New York: United Nations). Walker, R. (1978). ‘Two sources of uneven development under advanced capitalism: spatial differentiation and capital mobility’. Review of Radical Political Economy 10: 28–​39. World Bank ECA (2015). Low Commodity Prices and Weak Currencies. Economic Update October 2015 (Washington, DC: The World Bank).

Pa rt  I I

C ON C E P T UA L F OU N DAT ION S

Chapter 7

E c onomic Grow t h a nd Ec onom ic Deve l op me nt : Geo graph i c a l Dimensions, De fi ni t i on, and Dispa ri t i e s Maryann P. Feldman and Michael Storper Bringing Geography and Economics to the Same Table Economists have asked why certain places grow, prosper, and attain a higher standard of living as early as Adam Smith’s The Wealth of Nations in 1776. Smith was motivated to understand the reasons why England had become wealthier than continental Europe. While Smith is widely considered the father of modern economics, his most important theorems originated in geography. When he said, ‘the division of labor is limited by the extent of the market’, he was referring to the geographical extension of market areas in Scotland as transport costs declined, which, in turn, allowed larger-​scale and more geographically concentrated production, organized in the form of the factory system. The transition from artisanal production to a modern industrial economy, with a 4800 per cent increase in productivity, was intrinsically geographic. The transition that Smith analysed was profound:  artisans disappeared; production became more centralized in large factories and towns, creating a geography of winning and losing places; industrial capitalists saw their incomes increase, while a new industrial working class faced lower incomes than artisans and more difficult working conditions. Still, there was a long-​term take-​off of per capita income that ended centuries of economic stagnation in the West (Maddison, 2007). Critically, Smith (and others) showed that the division of labour inside the new factories was not only key to the astonishing productivity gains of the factory system, but that it also picked winners and losers in terms of individual and social relationships, and geographical places. Smith was not only concerned with the positive

144   Feldman and Storper aggregate economic effects of the new system, but also the more complex picture of human and geographical development (Phillipson, 2010). The processes of change that motivated Adam Smith are still at work and are no less complex or profound. As in the industrial revolution, the much-​heralded growth of the knowledge economy is creating significant wealth, but the distribution of benefits is highly skewed. Indeed, there are elements of a winner-​take-​all tournament that favours the lucky and highly skilled, which serves to only further increase income disparities. Many individuals who have invested in high levels of human capital face unprecedented economic insecurity and diminished career perspectives. These dilemmas are not new: from the time that Smith wrote in the mid-​eighteenth century, through Marx’s reflections of the mid-​nineteenth century, income disparities were so great that the viability of the whole industrial market (or, for Marx, ‘capitalist’) system was called into question. In the twentieth century, these conditions spawned political instability witnessed by unrest, and the rise of nationalism, fascism, and communism. Yet, considering the long sweep of history, capitalism has generated increases in standards of living never before imaginable for the majority of the world’s population. Even in the worst of times, there were very wealthy local economies; just as in the best of times, there were pockets of stagnation and poverty. The objective of this chapter is to provide a review of the intellectual history of economic geography as it relates to economic growth and economic development. We show that economic development always has a complex interplay of winners and losers. The progress of the modern capitalist economy always begins in specific places; it does not emerge uniformly across space, but instead diffuses across the economic landscape. Less-​successful people and places represent underutilized capacities. Yet this pattern is not immutable: the relevant question is, what advice can scholars give policymakers to enhance economic prosperity. After investigating the geographical dynamics of economic growth, this chapter defines some new approaches to mitigate the downsides of these dynamics. To do so, we will challenge some of the sacred cows of economic theory and policy to make a new meal—​or even a feast—​of future possibilities. Conventional wisdom tinkers at the margins of the growth process but does little to address the ways that the economy picks winning people and places, and underutilizes the capacities of other people and places. By contrast, it will be shown that with a deeper understanding of the geographical wellsprings of innovation and entrepreneurship in capitalism, there are opportunities for higher growth and, most importantly, better development for both people and places.

The Interrelationship of Growth, Development, and Geography The relationship between the quantity of growth and the quality of economic development is complex (Feldman et  al., 2016). In policy circles, however, growth and development are frequently conflated and incorrectly interchanged. Economic growth is a primary focus of macroeconomists, who rely on quantifiable metrics, such as gross national product or aggregate income. Economic development was relegated to the

Economic Growth and Economic Development    145 domain of practitioners, often focusing on infrastructure, public health, or education in poorer places. For much of the twentieth century, economic development relied on specific outcome measures like number of jobs that, while relevant to policy, could not convincingly reflect any broader picture or reveal any longer-​term pathways of qualitative improvement. Investments in education have not always led countries to long-​term growth, and in some countries economic growth led to significant increases in education. This leads back to the core debates about directions of causality and the need for systemic understanding of these relationships. The strong correlation, in the range of 0.95, between per capita income and the Human Development Index, suggests that development and growth are interrelated (McGillivray and White, 1993). Taking one extreme, some argue that the same ingredients that generate aggregate growth also deliver improvements in human welfare. Professional practice and policy tend to emphasize kick-​starting growth, based on the implicit assumption that aggregate growth can be counted on to deliver qualitative improvements in human welfare (Easterly, 2012). Others argue that the sequence of improving income—​in time and across space—​must first start with improving human welfare (Dasgupta and Ray, 1986; Barro, 1991) that will, in turn, deliver improvements in per capita income, and subsequently increase human welfare. The hubris that once existed in the economic development field—​ which assumed that the path of economic development was linear, increasing, and always positive—​is gone (Dasgupta, 1993). With larger samples of growth and development experiences to study, the lesson is that growth does not automatically occur and continuously improve human welfare. Moreover, even when processes of economic growth and development appear relatively robust, there is an uneven geographical distribution of the benefits. All places do not rise, or fall, at the same time; indeed, there are frequently contrasting processes happening at the same time across contiguous neighbourhoods, cities, regions, and countries. This realization led to an explosion of interest in the microeconomic foundations of development, which consider the economies of places as products of history and local institutions and as differently structured environments where people live, work, and invest. This opens up a completely original line of inquiry into the relationship of growth and development. It is not only any set of contributing ‘factors’ that enable growth or development, nor how these factors flow or ‘sort’ into countries and regions, but rather how these factors come together to interact in intricate ways that differ across space and time because of the variation in human rules, institutions, habits, norms, and conventions.

Geography as a Fundamental Ingredient in Economics The relationship of geography and economic development presents itself in different ways in different places. In very poor places, development cannot start without basic institutions such as property rights, a solid legal system, and an infrastructure that make commerce possible (World Bank, 2009). In the majority of the world market economies, these basic

146   Feldman and Storper conditions are already in place, yet significant disparities in income and human development persist. The rest of this chapter is addressed to the middle-​and upper-​income regions of the world, as a very different discussion is required to address policy in the poorest places (Collier, 2007). There was a time not too long ago when economists were preoccupied with models that rendered spatial disparities as uninteresting temporary disequilibrium, while geographers focused on complex phenomena described in detailed case studies. There were also notable differences in the normative perspectives of these disciplines. Many economists were not fundamentally worried about geographical disparities in development, while geographers tended to be more radical, with a focus on social concerns and left-​behind places. Data were a limitation, as were empirical methods and visualizations. Yet, as frequently happens in scientific disciplines, fields converge and recombine to form new fields of inquiry. This has happened over the last thirty years with economics and geography. Paul Krugman (1991a, 1991b), unsatisfied with the observation that per capita income had not converged between places—​a prediction that was at odds with neoclassical growth theory—​launched a new research trajectory, declaring that ‘I have spent my whole professional life as an international economist thinking and writing about economic geography, without being aware of it’ (Krugman, 1991b, p. 1). Geographical differences in development, Krugman observed, had been of secondary importance because existing economic models could not address them as a central component of the market economy. Instead, economists tended to use models that assumed away distance or relegated economic disparities to temporary disequilibrium from frictions due to factor mobility (Krugman, 1991b). The founders of the new geographical economics in the early 1990s—​Krugman, Fujita, Thisse, and Venables—​showed that by incorporating economies of scale, product differentiation, and trade costs into models of firm location, it would be perfectly natural for a market economy to concentrate firms together. Following this, it would be perfectly natural for people—​in their dual roles as both workers in firms and consumers—​to concentrate as well (Fujita et al., 1999; Fujita and Thisse, 2002). Agglomeration economies, clustering, and urbanization are not temporary imperfections of the modern capitalist economy, but are rather part of its essence. This is not a new insight, but a more rigorous formulation of long-​standing wisdom. Examining Britain at the height of its industrial power, Alfred Marshall (1919) referred to localization as a phenomenon that can be observed throughout human history—​the right place at the right time. At any given moment, the most developed regions or countries specialize in the most advanced industries, which, in turn, take the form of their spatial concentration. The recognition that agglomeration is hard-​wired into capitalism gave rise to a problem for the pre-​existing conventional wisdom about spatial equilibrium. Rather than factor mobility leading to an even distribution of production and incomes across the economic landscape, powerful agglomerative forces would actually increase discrepancies. Thus, agglomeration goes against the grain of contemporary general spatial equilibrium models (see Glaeser, 2008). It also opens up a major normative debate in economic geography: while aggregate efficiency may come from strong agglomeration, it may possibly also come at the price of inequity. In this way, the geography of development entered the very heart of the economics of development.

Economic Growth and Economic Development    147

The Process of Development: The Nouvelle Cuisine of Economics and Geography The closer relationship between geography and economics does not stop with the observation that there is a deep tension between development and territorial equity or convergence. Most importantly, the unexplored mechanisms for creating wealth in more places and for increasing its diffusion or spreading the benefits has moved to centre stage. The core of all this is the economics of knowledge (Stephan, 1996) and the geography of innovation (Feldman, 1994). In the classical definitions of growth, from David Ricardo (1891) to Robert Solow (1956), the economy is conceptualized as a machine that produces economic output as a function of various inputs (including capital, labour, and technology). Solow showed that the different factors considered in growth models—​such as augmented capital and labour, and the inclusion of more education, better infrastructure, and better health—​only explained a relatively small part of the observed economic growth since the Industrial Revolution. He concluded that technological innovation was generating more output per unit of input over time, and that this was leading to greater total factor productivity. Yet even if innovation were a possible cause of greater efficiency in certain industries, it would still be very costly to the overall economy, owing to the diminishing marginal returns of augmenting the traditional inputs of labour and capital needed to realize innovation. Robert Lucas (1988) and Paul Romer (1986) solved this paradox by: challenging the classic assumption of constant or decreasing returns, and observing that knowledge is different from other inputs to the economy. True knowledge, they argued, has increasing returns to scale because of the externalities inherent in its creation and application. Rather than diminishing over time, the value of knowledge actually increases with use due to network effects, cumulative reapplication, path dependencies, non-​exclusivity, and spillovers (the recombination through leakage). This all leads to more knowledge over time and better uses of that knowledge. This insight explains why, from the year 1820 onward, capitalism has been able to spring the Malthusian trap of the stagnation in worldwide per capita income that existed from the year 1000 until the Industrial Revolution (Maddison, 2007). Moreover, since 1820, global per capita income has steadily increased against a world population boom. However, the modern era’s astonishing growth has distributed unevenly across people and places. There are periods of retrenchment, as well as economic booms. The agglomeration models of the New Economic Geography imply a fundamental trade-​off between efficiency and inter-​place convergence However, the new economics of growth, which centre on innovation, suggest that there are alternative possibilities. The forces that create innovation also create far-​flung production chains that spread knowledge, diffusing it away from the places where it was initially created (Grossman and Helpman, 2005; Iammarino and McCann, 2013). If some places are better at innovating than others—​and are, hence, wealthier—​why not think about a new type of development policy, based on spreading innovation capacities or creating them in more places? This approach might offer hope for income convergence, which is not offered by factor mobility between places (the core recipe of traditional models in regional and urban economics), or simple liberalization of trade (the core recipe of international development economics).

148   Feldman and Storper We will show that the investments in capacity that generate innovation have increasing returns for the regions, firms, and workers who exercise them. Virtuous self-​reinforcing cycles of economic development that are also widely spread in geographical terms can more widely share the desired social and economic outcomes of prosperity and more sustainable economic growth. An innovative, place-​based development policy approach counters the potentially negative spiral of geographically restricted development in three ways: firstly, it starts with investment in basic capacities that are essential to a dignified and creative life (as argued by Amartya Sen in Feldman et al., 2016); secondly, it expands the sources of creativity and satisfaction that are good, in and of themselves, on human grounds; thirdly, it works towards the overall goal of having non-​routine (innovative) functions in the economic mix of more and more economies.

Back to Fundamentals: The States and Markets Debate The relationship between government—​or the State—​and development requires greater theoretical development. Mainstream economic theory is wary of government intervention in markets, only justifying public policy to correct market failures (Laffont and Tirole, 1993). Market failure takes many forms:  externalities, market power that inhibits competition, information asymmetries that prevent efficient transactions, and incomplete provision of certain kinds of goods and services. In the specific field of industrial policy, the most widely accepted rationale for public action are the positive social externalities in R & D and knowledge creation. Firms cannot fully appropriate all the benefits of their investment in knowledge because some of the benefits accrue to other firms or sectors. Thus, the social return on investment in R & D and knowledge creation is larger than the private return. As a result, the private R & D effort will be lower than that which is socially optimal. Consquently, the public sector has a role to fund R & D, or to enhance the incentives of private firms to invest in knowledge creation. While market failure leads some economists to admit a theoretical role for a mix of regulatory and investment policies, others claim that these measures lead to government failure, where the medicine is worse than the ailment. In their view, government is intrinsically beset by rigid bureaucracy, entrenched interest groups and inadequate information, such that interventions become ineffective or actively harmful. The empirical evidence is much more nuanced, with cases of public stimulus of firms resulting in subsequent private success (Mazzucato, 2013). Reality certainly lies with detailed empirical analysis of markets to determine what is required and when to withdraw public supports (Avnimelech and Teubal, 2006). High-​quality public administration is necessary so that government policies and programmes are well executed. The real policy world, however, often does not respect the fine points of what theory and evidence say about dealing with market failures. Starting in the 1980s, the Reagan–​Thatcher agenda was blindly hostile to regulation and public goods. This economic realm is sometimes referred to as ‘neo-​liberal’, a pejorative label for an extreme laissez-​faire political philosophy (Fawcett, 2014). For decades, it has failed to protect the public from predatory economic

Economic Growth and Economic Development    149 behaviour in the form of monopolies, crony capitalism, and rent-​seeking behaviour. The private provision of certain goods is lower quality and more costly than public provision. There is an inherent tension between private firms’ incentive to generate profits by reducing costs and the public need for high-​quality, universally available, and reliable services. Of course, the highest profits are made in essential services for which demand is inelastic and there is no functioning market. Yet, as of now, there is little agreement on the need for government intervention, and the specific policies to implement, and investments for the government to make. In the USA, there is still a strong contest between proponents of austerity and minimalist government (this is supposedly a way to stimulate entrepreneurial energy at the local level), and traditional macroeconomic Keynesianism (as a way to stimulate development via demand). However, neither of these perspectives responds to the issues that are specific to the ongoing process of economic development, nor to its geography. Hence, we now turn to some new microeconomic foundations of innovation and production, and their geography.

An Alternative Definition of Economic Development Inspired by Sen (1990), Feldman et al. (2016) argue to define economic development as the development of capacities that expand economic actors’ capabilities. This new definition of development involves a twofold difference with standard models in economics. On the one hand, this definition departs from the strict Benthamite utilitarianism, which is interested in simply maximizing the sum of so-​called ‘utilities’ in the form of income and consumption possibilities. This definition goes beyond this hard side of the economy, explicitly incorporating a humanistic vision of the economy as a source of human fulfilment, where people create, explore possibilities, earn self-​respect, and develop a good life for themselves through well-​distributed opportunities (Phelps, 2013). Once this perspective is adopted, then the mechanics of a desirable growth process itself are also different from standard models, going beyond factor augmentation to better production through innovation (the theme that is threaded throughout every section of this chapter). Thus, development can be regarded as fortifying autonomy and substantive freedom, which promotes individuals’ participation in economic life (Sen, 1999). Economic development occurs when individuals have the opportunity to actively engage and contribute to society, and are likely to realize their potential. This promotes the advancement of the whole society. Why is this the case? Part of today’s malaise is due to the increasingly unequal distribution of income causing large parts of society to see stagnating material welfare in the midst of overall plenty (Katz, 1999; Piketty and Saez, 2001). This is, however, only part of the problem. There have been other periods, as for example the height of American mass production in the 1950s, when incomes were advancing rapidly for much of the population, but with a sense of frustration due to the deadening and hierarchical character of work. Even the sense that the ‘next generation’ was expected to be wealthier did not entirely compensate for the constraining industrialized lifestyle, leading to the social unrest of the 1960s, and to sociological critiques with titles such as ‘The Joyless Economy’ (Scitovsky, 1976). Today’s temptation is to think that all we need to do is restore high-​enough wages and low-​enough

150   Feldman and Storper unemployment to have a good-​enough economy. However, it is important not to miss the currently difficult conjuncture of high inequality, low employment creation, and stagnating median wages, when thoroughly re-​thinking development, and how to best generate it. A broader perspective on development suggests that we need a better geographical and social distribution of the capacity to realize opportunity. In this sense, the expansion of capacities provides the basis for the realization of individual, firm, and community potential, which, in turn, contributes to the ability of the economy to prosper—​materially, through innovation, and non-​materially, through widespread improvements in human experience, striving, and creativity. The latter may be called ‘entrepreneurialism’, rather than the frequently reductionist notion today of ‘starting up a firm’. As Edmund Phelps (2013, p. 14) noted in Mass Flourishing, development occurs not just through spectacular inventions, but when ‘people of ordinary ability can have innovative ideas’. In nineteenth-​century America, ‘even people with few and modest talents … were given the experience of using their minds: to seize an opportunity, to solve a problem, and think of a new way or a new thing’ (Phelps, 2013, p. 15). Rather than simple counts of jobs or rates of output growth, economic development is concerned with the quality of any such growth. There are many ways to measure the quality of growth. Often, the starting point is the rate of change in per capita personal income and convergence towards the wealthiest places. But, if per capita income is very unequally distributed, the majority of people do not benefit. The quality of employment, which is, in turn, a manifestation of the skills of those employed and hence the wages those skills command, is another consideration. Even this, however, does not fully capture development as the overall dynamic is about an economy in relationship to its principals—​the people who work and live in an area. True development includes increasing: the calibre of business practices, the distribution and density of social capital, and many other things that fortify the ability of the economy to improve economic welfare continually over time. These are themes that we need to explore in more detail. This notion of development does not accord easily with classical economics, but there are bridges that can easily be built. According to Schumpeter (1934), economic development involves relocating capital from already established methods to new and innovative methods, which further enhance productivity. For instance, not only did mass production drive the textiles industry in the industrial revolution, but it also influenced other complementary sectors and, in turn, diffused widely, thus increasing quality of life. While economic growth is measured by returns to inputs or factor augmentation, in reality sustained economic growth changes the dominant forms of organization, work, market coordination, needed skills, attitudes and beliefs, and norms for how things get done. Throughout all this, there is immense learning-​by-​doing on the part of individuals and organizations (Arrow, 1962), and a cumulative process of technological change through incremental tweaking and continuous improvement (Meisenzahl and Mokyr, 2011). It is through this complex process of change that activities that have become simple and repetitive are replaced with higher value-​added, non-​routine activities (Levy and Murnane, 2005; Aghion, 2006). In this updated Schumpeterian view, economic development entails a fundamental systemic transformation of an economy, including the industrial structure, the educational and occupational characteristics of the population, and the entire social and institutional framework. This point has been revived in the idea that an economy of widespread creativity and innovation requires institutions that facilitate its reorganization (Rodrik et al., 2004).

Economic Growth and Economic Development    151 Institutions, following North and Thomas (1973), promote productive activities, capital accumulation, skill acquisition, invention, and technology transfer. Effective institutions help individuals and businesses make investment decisions by reducing certain forms of uncertainty through stable and predictable rules that encourage risk-​taking. Thus, to further build the definition, economic development requires institutions that promote norms of openness, tolerance for risk, appreciation of diversity, and confidence in the realization of mutual gain for the public and private sectors (Feldman et al., 2016). These institutions do not come easily; they are socially constructed and painstakingly generated over time. However, these institutions provide the foundation for building basic capabilities for sustainable economic development. These institutions have often been ignored because they have been evasive to study, but the time for detailed research on the role of institutions in entrepreneurism and economic development is now ripe (Feldman and Lowe, 2015). Among institutions, the public sector is arguably the only current actor in the economy with the required long-​term perspective and sufficient command of resources to make large-​scale investments in infrastructure and education, and to coordinate effectively economic systems. Moreover, government—​as the agent for its citizen—​has the mandate to ensure that the resulting benefits are fairly and widely distributed.

Place-​based Innovation Capacities: A New Vision of the Geography of Development At a time when market fundamentalism has come to guide US policy debates, the public sector has actually become more and more immersed in the economy through policies related to technology and, more particularly, to innovation (Block and Keller, 2009). When we move from generic capacities to the specific precursors of innovation, there is also evidence of a growing role for public institutions and investments (Block and Keller, 2009; Mazzucato, 2013). This is partially because the nature of scientific research has changed, increasingly taking the form of decentralized industrial networks or open innovation (Nelson and Winter, 1982; Lundvall and Johnson, 1994). R & D and innovation are thus no longer confined to the laboratories of large corporations or government, but are now collaborative activities, embedded in networks between both public and private institutions, and large and small firms. This degree of decentralization fosters a greater dependence on government programs to coordinate the operations of these networks and limit moral hazards (Schrank and Whitford, 2009). In more technical terms, knowledge spillovers among firms are a conduit for innovation, but such spillovers are a capacity that must be built and sustained over time and are not an automatic dimension of rational economic behaviour. Regional economists have long asked whether such spillovers are better encouraged by a regional economy focused on a few similar industries (‘specialized’), or one with many different industries (‘diversified’). This is sometimes captured, in our view quite imprecisely, as the difference between Marshallian externalities, defined as spillovers between firms in the same sector, and Jacobs externalities, defined as spillovers between firms in seemingly unrelated sectors. There is, however, no convincing evidence that either specialization or diversity is key to better long-​term

152   Feldman and Storper economic performance (Kemeny and Storper, 2015). The deeper issue is figuring out how to create a local context where there is a dynamic exchange of knowledge, widespread experimentation, and minimal penalties for failure, and where there are institutions that facilitate recombination into new and better products and processes. Regardless of the level of specialization or diversification in the local economic base, what counts most is the local context for these processes. So what can policy do to strengthen these desirable aspects of local context? This is where the policy debates engender another significant controversy. Many economists are sceptical of place-​based economic development strategies (Einiö and Overman, 2012; Cheshire et al., 2013). If economic development policy is place-​based, in the sense of redistributing resources to specific places, then it might reduce the optimal level of agglomeration by dissipating activity, which results in a reduction in total productivity and output growth of the national economy. Standard urban economics widely defines place-​based to include such things as land-​use housing regulations and environmental regulations, individual stimulus, or any place-​based payments to people or place-​based worker training. This framework leads the World Bank (2009), for example, to advocate a spatially blind or people-​based approach as the most effective way of generating efficiency, guaranteeing equal opportunities, and improving the lives of individuals where they live and work. A key to this approach is the assumption that geographical factor mobility will lead to the best aggregate outcome and to income convergence across places: human mobility increases individual income and productivity, while depleting unproductive regions of their surplus populations, and hence leads to a smoother geographical distribution of wealth, also known as general spatial equilibrium (Glaeser, 2008). This is a powerful argument, but it is nonetheless incomplete in two ways: it overestimates the potential for factor mobility to achieve the ends of aggregate economic growth and geographical convergence, and it underestimates the importance and potential of widely spread capacities for innovative, creative mass economic flourishing. It seems unlikely that substantially higher levels of migration of skilled labour, reductions in the basic agglomerative tendencies of the economy, and substantially more even economic development can be achieved simply by de-​regulating housing markets (Kemeny and Storper, 2012). This is, to us, like the ‘tail wagging the dog’ of economic development. In this light, the scepticism expressed about place-​based approaches can be turned on its head. The major contribution of the new growth theories is to recognize that knowledge benefits from increasing returns to scale rather than the constant or decreasing returns associated with physical commodities (Romer, 1986). Activities that create knowledge—​and encourage the sharing of knowledge—​support increasing returns that lead to improved national welfare. Agglomeration, with its various forms of returns to scale, is key to this process. However, there is no evidence that such agglomeration must take a particular national distribution—​such as a highly hierarchical national urban system, with a small number of Silicon Valley-​type supernova agglomerations and the rest of the nation left behind—​which would result in steep territorial inequalities. Indeed, the benefits of agglomeration may be achieved through a more even distribution of middle-​sized agglomerations, that is, on the exact spatial layout and distribution of agglomeration benefits (Crescenzi et al., 2007, 2012). The notion that any attempt to widely distribute innovation capacities is going to somehow kill the benefits of agglomeration is not sustained by theory, nor by any robust empirical evidence at this point.

Economic Growth and Economic Development    153 Indeed, economic development policy should be sensitive to the need for agglomeration to occur in as many places as possible (Duranton and Puga, 2001). The reason is owing to the inherent uncertainty of creativity—​to the what and where of future innovation. Economic development officials and government planners dream of being able to define long-​term strategies, but they typically fail at this task. It is impossible to predict scientific discoveries, important new technologies, and the ongoing tweaks that transform our lives. Few predicted the potential of the Internet and how it would change the way we communicate and access information. Even private firms, such as IBM (once the industry leader), underestimated the potential of the computer, creating an opportunity for new firms to enter the market and form new industries. Moreover, successful entrepreneurs make their own luck, adjusting and adapting to survive. Instead of wisely considered, far-​sighted solutions, entrepreneurial activity is by necessity messy, adaptive, and unpredictable. The biggest problem is that it is impossible to predict which technologies are going to yield any pay-​off and when. By the time a new industry—​for example, biotechnology or nanotechnology—​is on its way to becoming a household name, it is probably too late for other places to participate as major centres (Storper, 2013). The best economic development strategy is therefore to enable as many actors as possible to participate productively in the economy to the fullest of their ability. This prioritizes improving quality of life and well-​being by enhancing capabilities and ensuring that agents have the capacities and freedom to achieve their potential (Feldman et al., 2015). Hence, economic development strategies need to be adaptive and need to maximize the diversity of the people, firms, and places involved (Feldman and Lowe, unpublished). Diversity is the most powerful tool of success in the open probability game of innovation and economic creativity (Kemeny, 2017). Effective policy is intricate to design because regional economies are complex systems, which are notoriously difficult to model and influence. There is no reason to believe that optimizing the performance of any one component of a complex system will optimize or even necessarily improve the performance of the system overall. Current thinking is that economic development is not brought about by discrete projects or programmes, but rather emerges from the development of interactive and dynamically adaptive ecosystems (Hwang and Horowitt, 2012). Ecosystems have many different parts and many redundancies. Ecosystems also evolve in unpredictable ways, with multiple positive unexpected outcomes. The knowledge spillovers discussed earlier are the key internal flows and connective tissue of economic ecosystems, while institutions are its organic structure. The problem in most existing policies is that they use economic impact studies that do not fully capture the returns to a wide range of public economic development investments. Moreover, the amount of funding provided for economic development initiatives, while important to recipients, is miniscule in relation to the size of a regional economy. Claims that attribute positive outcomes to any specific programmes or projects are probably more about good luck, publicity, and hype, and are oftentimes not supported by sound economic analysis. There simply is no magic recipe. Moreover, external shocks to wider economic conditions (e.g., major technological changes and macroeconomic policies or cycles) may wipe out any hard-earned local gains. In this light, policymakers cannot afford to wait for perfect predictability and a world free of error. As Kline and Moretti (2013, p. 34) conclude, ‘Second best may, in practice, be very attractive relative to the status quo’. And, second-​best may be first-​best in the long-​run, if it promotes widespread capacities that are the basis for flourishing in ways that cannot be predicted in the short run.

154   Feldman and Storper

On to the Feast Throughout this chapter, we have attempted to slay some sacred cows, that is, received conventional ideas about economics, growth, development, and geography. Economic development occupies our collective imagination, but the term is often not well defined, or defined in a limited manner that does not accommodate the situation of the full range of places faced with restructuring and economic uncertainly. All too often, the emphasis is on innovation as an end in itself rather than as a means to the end of widely shared prosperity and human fulfilment. Alternatively, there are mechanical policy frameworks that focus rigidly on generating income convergence between places through redistribution, and others with rigid emphases on generating more employment and output, while ignoring their highly unequal social and geographical distributions. The starting points are different in the different parts of the world, and even between regions within nations. In the US, for example, basic infrastructure and public goods are lacking in many states and regions, leading to large parts of the population with limited capacities, even when the culture of risk and openness is present. In the high-​income areas of Western Europe, infrastructure and basic goods are well distributed, but cultures of openness and risk-​taking are—​in many regions—​not present. In the eastern regions of the European Union (EU), educational levels tend to be high, but the basic infrastructure of connectedness is still being put into place, and old cultures of cronyism and corruption must be dismantled. In many southern parts of the European Union, low educational levels and stagnant demography are combined with rules that are inimical to risk taking and open sharing of information. The needs that government must address are thus different in these areas, but in all cases, the quality of government is an overriding concern, especially as government must evolve along with the changing regional context. Indeed, as the regional context moves forward, government is often left behind doing the same old thing. The question then becomes how to develop institutions and systems appropriate for different places and how to motivate ongoing innovation and adaptation to changing external conditions in the public sector. Cutting across a wide variety of different contexts, a set of universally important tasks can be identified, although they must be addressed in context-​specific ways. The first is entrepreneurship, a staple of discourse about economic development. There is, however, a difference between entrepreneurship that leads to development (through sustained build up of innovative productive capacity in a region), and mere firm creation. Industry-​ building entrepreneurship leads to the creation of: regional agglomeration, networks of producers, knowledge exchange, the growth of new types of dealmakers and intermediaries, and ongoing waves of creativity (Feldman, 2001, 2014). A second element is the existence of networks of all kinds: between producers, producers and workers, government and industry, among leaders, between leaders and community groups. These networks are what creates what Granovetter (1973) called the ‘strength of weak ties’, reducing transaction costs and increasing confidence without creating cronyism and clubs. They are the key untraded interdependencies of a dynamic regional economy (Storper, 1995), and when they fail through predatory and rent-​seeking behaviours, or failures in communication, there are negative consequences (Whitford and Schrank, 2011; Storper

Economic Growth and Economic Development    155 et al., 2015). A third, closely related focus for policy is to help the region’s actors create the informal conventions that enable coordination under uncertainty. Rules are valuable in creating broad and stable framework conditions for orderly development, and they are the province of an active government. The successful use of rules, however, under changing circumstances takes place at the level of informal norms and conventions, yet sometimes these are the wrong ones for a dynamic process of growth (Storper and Salais, 1997). Linked to this is a fourth actionable domain of policy: beliefs and goals. Nobel Prize-​winning economist Douglass North argues that ‘the dominant beliefs—​those of political and economic entrepreneurs in a position to make policies—​over time result in the accretion of an elaborate structure of institutions that determine economic and political performance’ (North, 2006, p. 2). Beliefs and goals can only be changed through a broadly based regional ‘conversation’ that is inclusive and confidence-​building, effectively changing perceptions of who we are and what is possible, and that we are in the process together (Lowe and Feldman, 2008; Storper et al., 2015). And, finally, for every newly supplied capacity created in a regional economy, there must be demand. Steve Casper (2009) showed, for example, that Los Angeles was similar to the San Francisco Bay Area in the production of university-​based scientific outputs related to information technology, but that the market for such outputs was much greater in the Bay Area, where there is a community of IT commercial start-​ups creating effective regional demand for university-​based inventions. To summarize a wide body of theory and evidence, economic development can be enhanced via a longer-term and more expansive perspective that continuously works towards measureable increases in regional capacity. The best policies to harness the natural tendency of innovative activity to cluster may be policies and investments that allow economic agents—​in as many places as possible, and across as many types of people as possible—​the capacity to be creative and fully engaged in the economy and society. This expansive view of economic development necessitates important participation of the public sector as the agent of collective investment in capacity and suggests that businesses that benefit from knowledge spillovers and local capacity are key partners in building such public institutions (Feldman, 2014). The geography of this perspective is also more inclusive than winner-​take-​all agglomeration geography, although it builds on the essential microeconomics of geographical concentration as a fundamental source of innovation and development. Thus, our emphasis on creating the capacities for humanly fulfilling and widely distributed innovation is motivated by both humanism and good economics. At regional, national, and world scales, this way of thinking offers a different programme for economic development policy, and a different set of aspirations, from the conventional ones. To implement such policies, much hard work lies ahead. We will have to abandon the existing sacred cows, in the forms of the standard metrics of growth, innovation, convergence, and well-​being. We will have to operationalize new metrics for development as the broad process defined here (Bartik, 2012). And, finally, we will have to abandon and redefine many of the politically expedient practices that shape the field of economic development policy and the local politics of development. The hopeful news is that the economics and geography of development now provide ingredients in order to better understand these processes, and hence to create this new feast.

156   Feldman and Storper

References Aghion, P. (2006). ‘A primer on innovation and growth’. Bruegel Policy Brief 6: 1–​8. Arrow, K.J. (1962). ‘The economic implications of learning by doing’. The Review of Economic Studies 29: 155–​173. Avnimelech, G. and Teubal, M. (2006). ‘Creating venture capital industries that co-​evolve with high tech: insights from an extended industry life cycle perspective of the Israeli experience’. Research Policy 35: 1477–​1498. Barro, R. (1991). ‘Economic growth in a cross-​section of countries’. Quarterly Journal of Economics 106: 407–​444. Bartik, T.J. (2012). ‘The future of state and local economic development policy: what research is needed’. Growth and Change 43: 545–​562. Block, F. and Keller, M.R. (2009). ‘Where do innovations come from? Transformations in the US economy, 1970–​2006’. Socio-​Economic Review 7: 459–​483. Casper, S. (2009). ‘The marketplace for ideas: can Los Angeles build a successful biotechnology cluster’. A Report to the John Randolph Haynes Foundation, Keck Graduate Institute of Applied Life Sciences. Cheshire, P., Overman, H.O., and Nathan, M. (2013). Urban Economics and Urban Policy (Cheltenham: Edward Elgar). Collier, P. (2007). The Bottom Billion (Oxford: Oxford University Press). Crescenzi, R., Rodríguez-​Pose, A., and Storper, M. (2007). ‘The territorial dynamics of innovation:  a Europe–​ United States comparative analysis’. Journal of Economic Geography 7: 673–​709. Crescenzi, R., Rodríguez-​Pose, A., and Storper, M. (2012). ‘The territorial dynamics of innovation in China and India’. Journal of Economic Geography 12: 1055–​1085. Dasgupta, P. (1993). An Inquiry into Well-​Being and Destitution (Oxford: Clarendon Press). Dasgupta, P. and Ray, D. (1986). ‘Inequality as a determinant of malnutrition and unemployment: theory’. The Economic Journal 96: 1011–​1034. Duranton, G. and Puga, D. (2001). ‘Nursery cities: urban diversity, process innovation, and the life cycle of products’. American Economic Review 91: 1454–​1477. Easterly, W. (2012). White Man’s Burden: Why the West’s Efforts to Aid the Rest Have Done so Much Ill and so Little Good (New York: Oxford University Press). Einiö, E. and Overman, H. (2012). ‘The effects of spatially targeted enterprise initiatives: evidence from UK LEGI’ https://​ideas.repec.org/​p/​wiw/​wiwrsa/​ersa12p164.html (last accessed 12 May 2017). Fawcett, E. (2014). Liberalism: The Life of an Idea (Princeton, NJ: Princeton University Press). Feldman, M. (1994). The Geography of Innovation (Boston, MA: Kluwer Academic Publishers). Feldman, M.P. (2001). ‘The entrepreneurial event revisited: an examination of new firm formation in the regional context’. Industrial and Corporate Change 10: 861–​891. Feldman, M.P. (2014). ‘The character of innovative places: entrepreneurial strategy, economic development and prosperity’. Small Business Economics 43: 9–​20. Feldman, M. and Lowe, N. (2008). ‘Consensus from controversy: Cambridge’s biosafety ordinance and the anchoring of the biotech industry’. European Planning Studies 16: 395–​410. Feldman, M. and Lowe, N. (2015). ‘Triangulating regional economies: realizing the promise of digital data’. Research Policy 44: 1785–​1793. Feldman, M., Hadjimichael, T., Kemeny, T., and Lanahan, L. (2016). ‘The logic of economic development: a definition and model for investment’. Environment and Planning C Government and Policy 34: 5–​21.

Economic Growth and Economic Development    157 Fujita, M., Krugman, P., and Venables, A.J. (1999). The Spatial Economy: Cities, Regions and International Trade (Cambridge, MA: MIT Press). Fujita, M. and Thisse, J.-​F. (2002). Economics of Agglomeration (Cambridge:  Cambridge University Press). Glaeser, E.L. (2008). Cities, Agglomeration and Spatial Equilibrium (Oxford: Oxford University Press). Granovetter, M. (1973). ‘The strength of weak ties’. American Journal of Sociology 78: 1360–​1380. Grossman, G. and Helpman, E. (2005). ‘Outsourcing in a global economy’. Review of Economic Studies 72: 135–​159. Hwang, V.W. and Horowitt, G. (2012). The Rainforest: The Secret to Building the Next Silicon Valley (Los Altos, CA: Regenwald). Iammarino, S. and McCann, P. (2013). Multinationals and Economic Geography:  Location, Technology and Innovation. (Cheltenham: Edward Elgar Publishing). Katz, L.F. (1999). ‘Changes in the wage structure and earnings inequality’. Handbook of Labor Economics 3: 1463–​1555. Kemeny, T. (2017). ‘Immigrant diversity and economic performance in cities’. International Regional Science Review 40: 164–​208. Kemeny, T. and Storper, M. (2012). ‘The sources of urban development: wages, housing and amenity gaps across American cities’. Journal of Regional Science 52: 85–​108. Kemeny, T. and Storper, M. (2015). ‘Is specialization good for regional economic development?’ Regional Studies 49: 1003–​1018. Kline, P. and Moretti, E. (2013). ‘People, places and public policy: some simple welfare economics of local economic development programs’. National Bureau of Economic Research, No. w19659. Krugman, P. (1991a). ‘Increasing returns and economic geography’. Journal of Political Economy 99: 483–​499. Krugman, P. (1991b). Geography and Trade (Cambridge, MA: MIT Press). Laffont, J.J. and Tirole, J. (1993). A Theory of Incentives in Procurement and Regulation (Cambridge, MA: MIT Press). Levy, F. and Murnane, R. (2005). The New Division of Labor: How Computers are Creating the Next Job Market (Princeton, NJ: Princeton University Press). Lowe, N. and Feldman, M. (2008). ‘Constructing entrepreneurial advantage: consensus building, technological uncertainty and emerging industries’. Cambridge Journal of Regions, Economy and Society 1: 265–​284. Lucas, R.E. (1988). ‘On the mechanics of economic development’. Journal of Monetary Economics 22: 3–​42. Lundvall, B.Ä. and Johnson, B. (1994). ‘The learning economy’. Journal of Industry Studies 1: 23–​42. McGillivray, M. and White, H. (1993). ‘Measuring development? The UNDP’s human development index’. Journal of International Development 5: 183–​192. Maddison, A. (2007). The World Economy Volume 1:  A  Millennial Perspective Volume 2: Historical Statistics (Paris: OECD). Marshall, A. (1919). Industry and Trade (London: Macmillan). Mazzucato, M. (2013). The Entrepreneurial State: Debunking Public vs. Private Sector Myths (London: Anthem Press). Meisenzahl, R.R. and Mokyr, J. (2011). ‘The Rate and Direction of Invention in the British Industrial Revolution:  Incentives and Institutions’ in J. Lerner and S. Stern (eds) The Rate and Direction of Inventive Activity Revisited, pp. 443–​479 (Chicago, IL: University of Chicago Press).

158   Feldman and Storper Nelson, R.R. and Winter, S.G. (1982). An Evolutionary Theory of Economic Change (Cambridge, MA, and London: Harvard University Press). North, D.C. (2006). Understanding the Process of Economic Change (Princeton, NJ, and Oxford: Princeton University Press). North, D.C. and Thomas, R.P. (1973). The Rise of the Western World: A New Economic History (Cambridge: Cambridge University Press). Phelps, E.S. (2013). Mass Flourishing: How Grassroots Innovation Created Jobs, Challenge, and Change (Princeton, NJ, and Oxford: Princeton University Press). Phillipson, N. (2010). Adam Smith: An Enlightened Life (London: Penguin UK). Piketty, T. and Saez, E. (2001). ‘Income inequality in the United States, 1913–​1998  (series updated to 2000 available)’. National Bureau of Economic Research, No. w8467. Powell, W.W. and Sandholtz, K.W. (2012). ‘Amphibious entrepreneurs and the emergence of new organizational forms’. Strategic Entrepreneurship Journal 6: 94–​115. Ricardo, D. (1891). Principles of Political Economy and Taxation (London: G. Bell and Sons). Rodrik, D., Subramanian, A., and Trebbi, F. (2004). ‘Institutions rule: the primacy of institutions over geography and integration in economic development’. Journal of Economic Growth 9: 131–​165. Romer, P.M. (1986). ‘Increasing returns and long-​run growth’. The Journal of Political Economy 94: 1002–​1037. Schrank, A. and Whitford, J. (2009). ‘Industrial policy in the United States: a neo-​Polanyian interpretation’. Politics & Society 37: 521–​553. Schumpeter, J.A. (1934). The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest, and the Business Cycle (Vol. 55) (Piscataway, NJ: Transaction Publishers). Scitovsky, T. (1976). The Joyless Economy: An Inquiry into Human Satisfaction and Consumer Dissatisfaction (Oxford: Oxford University Press). Sen, A. (1990). ‘Development as Capability Expansion’ in K. Griffin and J. Knight (eds) Human Development and the International Development Strategy for the 1990s (London: MacMillan). Sen, A. (1999). Commodities and Capabilities (Oxford: Oxford University Press). Solow, R.M. (1956). ‘A contribution to the theory of economic growth’. The Quarterly Journal of Economics 70: 65–​94. Stephan, P. (1996). ‘The economics of science’. Journal of Economic Literature 34: 1199–​1235. Storper, M. (1995). ‘The resurgence of regional economies, ten years later the region as a nexus of untraded interdependencies’. European Urban and Regional Studies 2: 191–​221. Storper, M. (2013). Keys to the City: How Economics, Institutions, Social Interaction, and Politics Shape Development (Princeton, NJ: Princeton University Press). Storper, M. and Salais, R. (1997). Worlds of Production: The Action Frameworks of the Economy (Cambridge, MA: Harvard University Press). Storper, M. and Venables, A.J. (2004). ‘Buzz:  face-​to-​face contact and the urban economy’. Journal of Economic Geography 4: 351–​370. Storper, M., Kemeny, T., Makarem, N., and Osman, T. (2015). The Rise and Fall of Urban Economies: Lessons from San Francisco and Los Angeles (Redwood City, CA: Stanford University Press). Whitford, J. and Schrank, A. (2011). ‘The Paradox of the Weak State Revisited: Industrial Policy, Network Governance, and Political Decentralization’ in F. Block and M.R. Keller (eds) State of Innovation: The US Government’s Role in Technology Development, pp. 261–​281 (Boulder, CO: Paradigm). World Bank (2009). World Development Report 2009: Reshaping Economic Geography (Washington, DC: World Bank).

Chapter 8

Heterod oxy as Orthod oxy: Prolegom en on for a Geo graphical P ol i t i c a l Ec onomy Eric Sheppard Introduction It has been over two decades since mainstream economists rediscovered economic geography (Krugman, 1991). In that period, and notwithstanding attempts to produce spaces for engaged pluralism (The Journal of Economic Geography; Clark et al., 2000), it has become a peculiar kind of boundary object (see Star, 1989) in the English literature. There is an ongoing struggle over how economic geography, and thus capitalism, is to be interpreted: by whom, and to what ends. While an imperfect dualism, these struggles can be framed as a disagreement between how scholars working from mainstream economics and those working from mainstream geography think about the capitalist space economy. Mainstream economics is an avowedly canonical discipline, with a well-​defined orthodoxy (micro-​foundations, equilibrium, mathematics as theory-​language). A wide range of topics of longstanding interest also within economics (historical, institutional, ecological, Marxist, feminist, etc.) align poorly with this paradigm, generating a self-​described heterodox economics tradition where these can be discussed (www.worldeconomicsassociation.org​; Lawson, 2005). However, such heterodoxy has failed to gain traction in contemporary economics’ centres of calculation. Remarkably, even the seemingly existential crisis for mainstream theory triggered by the 2008 meltdown, occurring as it did in the geographical and institutional cores of capitalism, has had little discernable impact on orthodoxy in economics (Fine, 2002; Mirowski, 2013). Heterodoxy in mainstream economics counts as orthodoxy among mainstream (critical human) geographers; however, they tend to dismiss economists’ orthodoxy as outside

160   Sheppard their canon. Geographers’ scholarship on economic geography is notably heterogeneous and promiscuous, inclined to sally forth expectantly into other domains of geography, and beyond. As Sheppard et al. (2012, p. 18) write: ‘Economic geography has become a peculiarly open-​ended sub-​discipline [sic], one that has tended to privilege the analysis of rapidly changing phenomena, studied in real time. It is an anti-​canonical project; it is open-​ended and will remain so, repeatedly breaking out of the boundaries created for itself ’. Nevertheless, there is a broadly shared dismissal of equilibrium, micro-​foundations, and mathematical theory, and discomfort with the promises of capitalism itself. This shared Weltanschauung makes it plausible to gather this heterogeneity under the label of geographical political economy (Sheppard, 2006a, 2011a). This is economic geographers’ orthodoxy. Rather than compare (again) economic geography with geographical economics, in this chapter I seek to lay out the lineaments of what it means to think geographically about economic phenomena. My overarching claim is that thinking geographically undermines the plausibility of what passes as orthodoxy in mainstream (geographical) economics. For geographers, this argument is largely preaching to the choir. For mainstream economists and fellow travellers, as my engagements over the years with them repeatedly drive home, it remains unconvincing.1 One reason for this is a seemingly profound difference in theory culture. Economists insist that social science requires the theory-​language of mathematics, combined with an empiricist epistemology based in statistical inference; geographers, after dabbling in location theory during the 1960s and 1970s, insist on qualitative approaches, from dialectics to post-​structural ontologies and qualitative empirical analysis—​rejecting, to boot, the possibility of value-​free inductive empirical research (Sheppard, 2014). These are, of course, false binaries: it is not necessary that ‘science’ (Wissenschaft) be value-​free or quantitative (Longino, 2002), and political economy need not be qualitative (Morishima, 1973). At issue here are self-​representations and otherings of disciplinary cultures and identities, whose essentialized characteristics are belied by the heterogeneity of participants’ practices. Writing from my position as a quantitative (and qualitative) geographical political economist—​a geographer by inclination and profession—​in what follows I lay out the consequences of thinking geographically about the capitalist space-​economy in the form of nineteen propositions (a full explication can be found in Sheppard, 2016). To those aligned with mainstream economics’ orthodoxy, I can reassure you that these propositions can be logically grounded in the theory-​language of mathematics (albeit derived from very different assumptions about the world). To those aligned with mainstream geography’s orthodoxy, these propositions largely confirm geographers’ shared Weltanschauung. What does it mean to think geographically about the economy, capitalism, or perhaps anything (Sheppard, 2015)? Firstly, thinking geographically means thinking through spatiality. Geographical space (in its various manifestations: territory, distance, scale, etc.) is not simply ‘out there’, an exogenous backdrop that shapes economic possibilities. Rather, it is constructed—​constituted through societal (and biophysical) processes. These constructed spaces then have causal effect on things economic—​the socio-​spatial dialectic (proposition 1). But geography is not simply the discipline of space (contra Kant). Thus, thinking geographically also is to think through the lens of an ‘interdisciplinary discipline’: To attend to how economic, political, cultural, and biophysical processes (to name but a few) are co-​constitutive of one another. For a geographer, the economy cannot be

Heterodoxy as Orthodoxy    161 studied in isolation, nor is it defensible to claim that other processes are dominated by, or reducible to, economic processes (economism). The following propositions derive from this starting point. It is conventional to argue that any economic system involves three kinds of processes: production; exchange and consumption; and the distribution of economic costs and benefits. Following this convention, I subdivide the propositions into geographies of production, exchange, and distribution. Yet a fourth cluster is necessary, covering the two ways in which economic geographical thinking exceeds the study of capitalism: how economic processes co-​evolve with other processes (taking us beyond the economic), and how capitalism’s inherent contradictions require consideration of alternative economic systems (beyond capitalism).

Geographies of Production Proposition 1: Geography is Produced (Not Exogenous): The Socio-​spatial Dialectic It has been said that geography ‘is as exogenous a determinant as an economist can hope to get’ (Rodrik et al., 2004, p. 133). Not so! Geographies are produced (Lefebvre, 1974 [1991]; Smith, 1984): places are transformed, the nature and significance of geographical scales altered, and distance and connectivities reconstructed. In a capitalist space economy, its geographies also are brought to market. For example, transportation and communications are commodities, manufactured and sold by industries specializing in these activities. Indeed, considerable private-​and public-​sector effort is invested in improving their productivity because of the economic benefits associated with the accelerated mobility of capital, labour, and commodities (the attempted annihilation of space by time). Further, recognizing their distinctive role within a multi-​sectoral economy disrupts foundational propositions of both mainstream and Marxian economic theory (Sheppard and Barnes, 1990). Discourses of the Anthropocene reference how features of the more-​than-​human world (climate, soils, access to water) have also long been shaped by economic processes. ‘Nature’ is steadily being commodified and thereby transformed, through activities ranging from fracking, mining, and agriculture, to genetic modification, environmental services, carbon markets, and human organ trading. Once we acknowledge that geography is produced, some would conclude that geography thus has no causal effect, obviating any need to theorize the impact of geography on economic processes. Again: not so! We have moved well beyond such thinking, acknowledging that a phenomenon does not have to be exogenous to exert agency. Edward Soja (1980) dubbed this the socio-​spatial dialectic, by which he means that economic (and other societal) processes shape the geography in which they find themselves, even as they are themselves influenced by those produced geographies: [S]‌patial structures shape spatial interdependencies, but in turn are shaped by those same interdependencies. Human agency shapes structure, but broader structural changes may undermine the efficacy of agency. Individuals share interests across class and space lines

162   Sheppard (not to mention gender, race, etc.) that can result in collective action and social conflict. Markets cannot automatically arbitrate these, and market-​based outcomes need not be socially beneficial. [The capitalist] space economy [is] a complex, non-​linear system; one in which space is no longer Newtonian and time is an emergent property (Plummer and Sheppard, 2006, p. 625).

Proposition 2: The Capitalist Space Economy is not Reducible to Micro-​foundations This proposition disputes two overly simplistic models of causation that have dominated the social sciences: that individual human agency is the only relevant causal force (methodological individualism—​mainstream economics’ microfoundations), or that humans are deeply constrained by—​even dupes of—​broader social structures (structuralist versions of Marxism, also found in anthropology and linguistics). Yet Karl Marx (2008 [1897], p. 15) famously quipped: ‘Men [sic] make history, but they do not make it as they please’. This aphorism captures a central idea in social theory: structuration (Giddens, 1984). Structuration conceptualizes how human agency shapes broader-​scale phenomena—​structures (including spatial organization)—​whereas those produced structures, in turn, shape the conditions of possibility of human agency. Structuration is the corollary of proposition 1: a defining property of complex, non-​linear dynamical systems is that they generate emergent structures that alter the context in which smaller-​scale processes and actants operate. Micro-​ foundations, and structuralism, are ruled out. The logic of structuration is exactly that of the socio-​spatial dialectic, and thus can also be termed a dialectical relationship of causality (irreducible to the specifications used in most econometric models). Dialectical analysis focuses on the relations between entities: ‘Dialectical reasoning emphasizes the understanding of processes, flows, fluxes, and relations over the analysis of elements, things, structures, and organized systems’ (Harvey, 1996, p. 49). Under dialectical reasoning, entities are neither stable nor well defined; they are heterogeneous phenomena that emerge through relations. This is not the inclination of mainstream economics, which conceptualizes individuals (but also firms, regions, and nations) as quasi-​autonomous entities with stable, well-​defined properties. Economic geography has undergone a relational turn, often framed as an alternative to (dialectical) Marxian political economy (Yeung, 2005). Caught between these schools of thought, dialectical reasoning is unpopular. This is a mistake. Dialectical reasoning can be mathematical and consistent with both complex systems theory and post-​structural assemblage theory (Rosser Jr, 2000; Sheppard, 2008).

Proposition 3: The Economy is Multi-​sectoral Writing before Adam Smith (1776) penned The Wealth of Nations, François Quesnay (1753–​ 58) noted that the economy is multi-​sectoral. It cannot be reduced to firms acquiring inputs to make commodities, which then are sold directly to households. Rather, a considerable component of the economy is capital goods—​commodities sold by one firm to another as an

Heterodoxy as Orthodoxy    163 input to commodity production. Organizing the multitude of firms into economic sectors, Russian émigré Wassily Leontief (1928) captured this complexity through input–​output analysis (tables depicting the flow of commodities between economic sectors, as well as from sectors to households). Inter-​regional input–​output modelling was a staple of regional science in the 1950s and 1960s (Isard, 1951), but is largely absent from the ‘new’ geographical economics. This is a mistake; the complexities of a multi-​sectoral economy challenge the conventions of mainstream (geographical) economic theory. Ricardo and Marx, constructed simple multi-​sectoral models of capitalism. The so-​called capital controversies of the 1960s laid bare the implications of taking multi-​sectorality seriously. It was demonstrated that the foundational claims of conventional macroeconomics, about scarcity, marginal productivity, and the capacity of factor markets to determine societally appropriate wages and profit rates, are deeply problematic as they depend on an untenable assumption about multi-​sectorality (Harcourt, 1972).2 Mainstream economics has sought to avoid this by assuming the existence of ‘regular’ input–​output structures (i.e. the subset of such structures that is consistent with neoclassical production functions), or by turning to micro-​foundations (but see proposition 2). Setting aside such pretzel-​twisting assumptions, attention to multi-​sectorality calls into question the validity of some foundational parables of mainstream economics: the free-​trade doctrine, self-​adjusting growth models, the ‘highest and best use’ principle for land markets, and the setting of wages and profit rates via market mechanisms (Steedman, 1979; Sheppard and Barnes, 1990).

Proposition 4: A Capitalist Space Economy is Grounded in Commodity Production, not Exchange When it comes to understanding the functioning of the capitalist space economy, geographers typically begin with places of production, not markets. The process of commodity production necessarily takes time (unlike the instantaneous translation of the production function), but also occurs across space.3 Political economists always have recognized the necessary time lag between the moment when capital is advanced to finance production, purchasing inputs (even if labour is paid ex post) in the expectation of realizing a profit, and the moment, after the commodity has been manufactured, distributed to markets and sold, when profits should be realized: the turnover time. The resulting rate of profit (generally positive in a going capitalist concern) is calculated per unit of time (typically per annum), and depends not only on the difference between revenues and costs, but also on the turnover time. Production also extends across space, however: commodities (and inputs) have to be moved from where they are produced to where they should be sold. Transcending space takes time and effort, often entailing enhanced risk because of uncertainties about how to market commodities successfully in distant markets. Under globalizing capitalism, this challenge of overcoming spatial barriers becomes increasingly important: the logistics necessary to make this happen (produced as transportation and communications commodities) remain a neglected area of study by economic geographers (but see Cowen, 2014).

164   Sheppard

Proposition 5: Commodity Production is Political It is well known that politics shapes outcomes within places of commodity production. The apparent equity and rationality of a free labour market dissolves as purchaser and seller of labour power enter the place of production: ‘The one with an air of importance, smirking, intent on business; the other, timid and holding back’ (Marx 1967 [1867], p. 176). On the one hand, capitalists can enforce labour discipline, enhancing profitability by extending working hours, speeding up production, and reducing hourly wages. On the other, workers can organize (more easily in a large production facility), protest, work to rule, and utilize other ‘weapons of the weak’ (Scott, 1985), to lower profit margins. This implies that technological change cannot be reduced to the ratio of labour to capital, to machinery, or the size of the production facility. Labour relations, governed by these unequal politics of the workplace, and the micropolitics of work are also influential in shaping productivity. Indeed, wages and profits are always inversely related in a capitalist space economy, ranging along a wage-​profit frontier (Sraffa, 1960; Sheppard and Barnes, 1990). The monetary surplus produced annually under capitalism (aggregate net revenue) is a pie to be divided between workers and capitalists (as well as landlords and resource owners). There is no rational, market-​based outcome, whereby factor prices match marginal productivity. The position on this frontier, defining real wages relative to the mean profit rate, depends on such non-​economic factors as class power and cultural politics (Mann, 2007).

Proposition 6: There is More to Value than Price The default measure of the value of a commodity is market price: this is what all economic actors calculate and respond to. Conceptually, prices fluctuate around long-​term prices of production—​a centre of gravity that depends on technology, competition, wages, and profits. Firms seek to set prices for their commodities, seeking to maximize their rate of profit over costs of production (Lee, 1998). Production factors—​labour, capital, and land (US Federal Reserve sometimes also adds entrepreneurship)—​also are valued through market prices (wages, profit and interest rates, rent). All these prices vary across space and time. In a space economy, price setting is complicated by how the price of accessibility (transportation/​communications) as a commodity, itself varying across space and time, shapes all other prices, sometimes in unexpected ways. Yet profound questions remain about the adequacy of price as a measure of the value of commodities; indeed, questions about whether value can be adequately quantified at all (Barnes, 1996). Use value, the idiosyncratic question of what an object means to a particular person in a particular place and time, is qualitative and subjective.4 As Marx noted, exchange happens only when participants assess the use values being exchanged differently:  at the agreed-​on price, each values the other person’s commodity higher than their own. Marx also develops a sophisticated framework of labour value, correcting inconsistencies in earlier versions popularized by Smith and Ricardo. For an aggregate multi-​sectoral economy in equilibrium, the average monetary profit rate is only positive if workers are exploited in labour value terms—​Morishima’s ‘fundamental Marxian theorem’

Heterodoxy as Orthodoxy    165 (Morishima, 1973). Empirical calculations of labour value closely correlate with prices of production, and keep open the question how attention to labour value can bring distinctive insights into capitalism (see Henderson, 2013). But labour values also vary geographically, inter alia as a result of incorporating the labour value of accessibility into calculations of the labour value of all commodities, raising questions about the validity of claims made on the basis of Marx’s aspatial value theory (Sheppard, 2004). This also opens space to consider other measures of value—​features common to all commodities like energy and carbon, bringing their own optics to the study of capitalism (Georgescu-​Roegen, 1971; Bergmann, 2013).

Proposition 7: Disequilibrium is the Norm It is possible, in principle, to calculate a general dynamic equilibrium for any capitalist space economy. For given technologies, there exists a unique geography of production, relative quantities of output distributed across space, and associated production prices (and labour values) that ensure that the economy would grow smoothly (as long as labour grows at the same rate). But it does not follow that the actions of individual capitalists and workers, mediated by market mechanisms, will converge on this equilibrium. Even in an aspatial capitalist economy, this only happens if we are willing to assume that all individuals are self-​interested rational actors, who already know in advance where the equilibrium should be (Duménil and Lévy, 1991).5 Yet the dynamical complexity is such that human actors would need to have the reasoning capacity of Turing machines, which we self-​evidently do not. Take the elementary example of a two-​sector (wage and capital goods), two-​region economy, where capitalists make profit-​rate maximizing decisions about how much capital to invest in commodity production in each time period. Unaware of conditions for equilibrium, they make their best guesses based on what they know at the time. Even when transport costs are zero, wages fixed, the politics of production is ignored, and no technological change, equilibrium is far from certain and dynamical complexity a distinct possibility (Bergmann et al., 2009). The spatiality of capitalism compounds this complexity, increasing the likelihood that individual capitalists’ self-​interested actions produce unintended aggregate outcomes undermining their goals (e.g. actions believed to enhance individual profitability undermine aggregate average profitability) (Sheppard and Barnes, 1990). For example, Okishio (1961) postulated that technical changes, calculated as cost-​effective by firms, would always enhance the average rate of profit. Not necessarily, in a capitalist space economy: technical change need not enhance profitability, wide ranges of technologies coexist, and different regions pursue quite distinct technological trajectories (Rigby, 1990; Webber et al., 1992; Rigby and Essletzbichler, 2006). Indeed, regional economies of firms are evolutionary in nature (Boschma and Martin, 2010; Martin and Sunley, 2015). For a multi-​sectoral, multi-​regional economy (including transportation), Bergmann (2012) computes the dynamics of technical change in each sector and region with fixed real wages. Firms choose stochastically from sets of possible technologies developed by other firms, seeking to raise their profit rate above the social average. Dynamical equilibrium is possible, but so is regional collapse, inter-​regional divergence, and divergence followed by reconvergence.

166   Sheppard Suppose, nevertheless, that the economy were to find itself in dynamical equilibrium (by sheer chance, or state policy): would it stay there? No! Any such equilibrium includes a wages–​profit rate trade-​off along the wage–​profit frontier, which self-​interested capitalists and workers would rationally seek to destabilize. Gains for either social class are possible at the expense of the other. Even in a simple spatial market, with firms competing with one another to sell to consumers, any spatial price equilibrium is unstable because some firms can gain by departing from it (Plummer et al., 2012). Thus a capitalist space economy has no Nash equilibria—​stable states where everyone is contented and that no one would rationally disrupt. It is a complex dynamical system that characteristically is out of equilibrium, with conflicts of interest and crises the norm, not the exception.

Geographies of Exchange Proposition 8: Perfectly Competitive Spatial Markets are an Oxymoron In a capitalist space economy, the mainstream ideal of a perfectly competitive market is a never-​never land. This has also been long recognized in geographical economics. Starrett (1978) developed what has been dubbed the spatial impossibility theorem: ‘If space is homogeneous, transport is costly and preferences are locally nonsatiated, then there is no competitive equilibrium involving transportation’ (Ottaviano and Thisse, 2004, p. 2571). Effects of local spatial monopoly are known to undermine the societal benefits presumed to accompany perfect competition among retailers (e.g. Denike and Parr, 1970; Sheppard and Curry, 1982). Krugman’s (1991) monopolistic spatial competition between firms only generates quasi-​ perfectly competitive conditions (equilibrium, zero profits, utilities maximized) when firms have identical technologies and the world is flat (no relative locational advantage). Thus, in actually existing, spatially differentiated economies, the following proposition 9 holds.

Proposition 9: Profit Rates are Positive The capitalist space economy is not some halcyon market where spatial price equilibria prevail, in which firms in competition, seeking to maximize total profits, never actually make a profit (at least net of costs) (contra Lösch, 1954 [1940]; Krugman, 1991). In a spatially differentiated market, net profits are positive (and spatially variable, depending on locational advantage). In fact, existing firms do not maximize total profits: they seek to maximize the rate of profit to be made on the capital advanced prior to production—​a rate that depends on both the profit margin and the turnover time (Lee, 1998; Moudud et al., 2013). But this is not just an empirical claim. In a spatially differentiated market, rational firms would act to maximize their rate of profit, not total profits (Sheppard et al., 1992). The fundamental Marxian theorem follows (proposition 6).

Heterodoxy as Orthodoxy    167

Proposition 10: Markets are Emergent Seeking to present capitalism as making good on Adam Smith’s (1776) invisible hand principle—​that markets enable the self-​interested actions of individuals to advance the general societal interest—​economists defined the norm of a perfect market, as well as other market structures departing from this (e.g. monopolistic competition, duopoly, hierarchy, networks). These categories leached into economic geography (e.g. Greenhut et al., 1987; Nagurney, 1987; Scott, 1988; Mulligan and Fik, 1989), notwithstanding concerns about whether they survive the disruptive addition of spatiality (proposition 8). The rub always has been that really existing markets never fit these categories; debates about the possibilities of free trade and perfect competition become like those about how many angels can dance on the head of a pin. In the rare case that a perfect market can be identified ‘in the wild,’ it is not because capitalism naturally produces them, but because of the imaginaries and actions of people in place (Garcia-​Perpet, 2007). Thus, actually existing markets are socio-​spatial constructs, producing economic spatiality (proposition 1). By now, a substantial literature, particularly in economic sociology, examines ‘marketization’—​how markets are constituted (Çaliskan and Callon, 2010): familiar territory also for economic geographers. Attending to these processes does not simply mean examining the markets that emerge, and the spatialities they reflect and produce (Berndt and Boeckler, 2012). It also means examining how the nature of really existing markets reflects the economic—​and geographical—​imaginaries of those empowered to make them (Mackenzie et al., 2008; Mackenzie, 2009). Markets are not ‘out there’; they are made, as economists and economic actors construct the world they aspire to (Mitchell, 2005). In short, capitalism is not legitimated by the invisible hand; its proponents use this myth to perpetuate its legitimacy.

Proposition 11: Unrestricted Inter-​territorial Commodity Trade Reproduces Spatial Inequality The free trade doctrine, stating that territories specializing in and exporting the appropriate commodities will mutually benefit from unrestricted commodity trade, is the macro-​geographical equivalent of perfect competition, transferring Smithian invisible hand discourses to the global scale. Its plausibility is undermined by the multi-​sectorality and spatialities of capitalism. Even for a conventional two-​region, two-​commodity trade model, the problems identified by Sraffa imply that standard trade theorems do not hold (Steedman, 1979; Steedman and Metcalfe, 1979; Wong, 1995). Attending to the spatiality of commodity production further undermines these theorems, raising questions about whether it is even possible to identify appropriate specializations. The unintended consequences of individual capitalists’ specialization decisions also make it unlikely that such actions would generate trading patterns satisfying the doctrine (Sheppard and Barnes, 1990; Sheppard, 2012).

Proposition 12: Uneven Geographical Financialization Matters Living in an era of capitalism where the finance sector has vastly expanded its influence over both the economy and daily life, economic geographers are paying increasing attention to

168   Sheppard geographies of financialization. Finance is not readily incorporated into mathematical models of a capitalist space economy, but poses no impediments to geographical research.6 Responding to misguided claims that the flattened landscape of global finance implies an ‘end of geography’ (O’Brien, 1992), geographers are documenting the geographical differentiations and inequities associated with financialization. These range from localized studies of financial exclusion, to financial labour, urban and regional finance, the persistence of financial centres notwithstanding the dissipative forces of digital trading, and to sovereign wealth funds, Islamic finance, sovereign wealth funds, and global financial markets (e.g. Pollard, 1996; Leyshon and Thrift, 1997; McDowell, 1997; Dymski and Li, 2003; Clark and Wójcik, 2007; Pollard and Samers, 2007; Christophers, 2013; Clark et al., 2013; Dixon, 2014).

Geographies of Distribution Proposition 13: A Capitalist Space Economy Entails Uneven Geographical Development A core conclusion of the trajectory of geographical research into the nature of a capitalist space economy is that processes of commodity production, exchange, and consumption do not level the playing field, thereby creating equal opportunities for all, regardless of geographical location. Far from mitigating the socio-​spatial inequalities preceding capitalism, it reproduces and enhances such inequality. It is not the case that the socio-​spatially disadvantaged are doomed to impoverishment. Nevertheless, the spatial dynamics of capitalism tend to create spaces of relative wealth, whose prosperity depends, inter alia, on unequal (exploitative) relations connecting them with spaces of impoverishment—​relations that enhance rather than mitigate inequities between ‘north’ and ‘south’. Phase shifts occasionally happen, undermining prosperity in core spaces and enriching selective peripheral spaces (think of suburbanization, the rise of the sunbelt in the USA, or China’s advancement). Yet the result is not regional convergence, but rewritten geographies of inequality: uneven geographical development (e.g. Harvey, 1982; Smith, 1984; Storper and Walker, 1989; Harvey, 2014). Such processes concatenate across all geographical scales, consistent with evidence that neoliberal globalization has been an era of expanded socio-​spatial inequality (Milanovic, 2011; Chatty et al., 2014; Piketty, 2014 [2013]).

Proposition 14: Labour Geographies Matter Contestations along the wage–​profit frontier (proposition 5) are key to questions of distribution. Notwithstanding the decline of organized labour—​a result of both neoliberalization and shifting geographies of production (abandoning places with a history of independent unions, casualizing labour, and reorganizing production into smaller, flexible, and mobile units)—​working populations retain the capacity to reshape economic geographies. Those studying such labour geographies seek to understand the spatial strategies of labour organizations, playing out across various interrelated scales that range from places of production to regional-​and national-​scale organizing movements and labour markets, to global struggles

Heterodoxy as Orthodoxy    169 (Martin et al., 1996; Mitchell, 1996; Herod, 1998, 2001). It also involves examining the strategies and impact of more informal social movements that, inter alia, seek to shift the wage profit frontier, and the cultural politics surrounding the wage (Mann, 2007; Featherstone, 2012). These more ​specific struggles themselves reflect and are shaped by the complex, co-​ evolving, geographies of social class—​geographies that immensely complicate Marx’s ‘workers vs capitalists’ narrative (Sheppard and Glassman, 2010).

Proposition 15: The State Matters An inescapable consequence of the aforementioned propositions is that the capitalist space economy is neither harmonious nor equilibrium-​oriented. The politics and agencies of the state, and legal rulings, are necessary for mitigating politically unsustainable inequalities, regulating commodity production and exchange, mediating the politics of production, and addressing the multitude of ‘externalities’ associated with capitalist markets. States (and legal systems) have their own spatialities, shaping the entanglements of politics with a capitalist space economy (Blomley et al., 2001; Brenner et al., 2003; Agnew, 2012). Geographers have devoted much effort to understanding the spatialities of state governance, seeking to illumi­ nate how these vary across space, as well as over time, and across geographical scales (Tickell and Peck, 1992; Brenner, 2004; Peck and Theodore, 2007). They also analyse how policy mobilizes through space-​time, is transformed along the way, and articulates with local conditions where it comes to ground (Roy, 2010; Peck and Theodore, 2015). The broader goal is to illuminate how shifting political agendas are co-​implicated with capitalist spatial dynamics.

Exceeding the Capitalist Space Economy Proposition 16: Capitalism is Cultural With the emergence of cultural industries, economic geographers have taken a lead in studying how cultural practices are commodified and brought to market, and the cognitive labour involved (Scott and Power, 2004; Markusen et al., 2008). But there is more to culture and economy than this (Barnes, 1995). Culture is about the relationship between identity, imaginary, and practice. Economic agents are not quasi-​autonomous individuals driven by exogenous preference functions. They are species-​beings whose norms, desires, and ethics emerge through geographically and socially differentiated processes of socialization. Their distinct socio-​spatial positionalities emerge relationally and intersectionally (Valentine, 2007; McDowell, 2015), and through power-​laden interactions with other (human and non-​ human) agents. Seemingly persistent, positionalities also can shift unexpectedly (Sheppard, 2006b). They can be reshaped by economic forces (e.g. advertising), but emergent identities (gendered, raced, spaced, sexualized, etc.) can also shape commodity production and exchange—​the dialectic again at work. Beyond this are broader processes through which hegemonic discourses shape economic activities. Understanding the emergence, power, and impact of discourses, including their spatio-​temporal differentiation, is the domain of cultural theory (Foucault, 1971; Derrida, 1976). Yet these have powerful economic implications

170   Sheppard (Thrift and Olds, 1996; Thrift, 2005). Consider, for example, how shifting discourses catalysed the dramatic shift in economic norms from Keynesianism to neoliberalism in geographically variegated ways (Mirowski and Plehwe, 2009; Peck, 2010).

Proposition 17: The Economy is More than Human Economic geographers have long attended to the biophysical environment, usually reducing it conceptually to a material input for the economy: agricultural and resource geographers have examined how resources are extracted from what was conventionally seen as an inert ‘nature’, exogenous to the economy (Rees, 1990; Grigg, 2003; Auty, 2007). But we now know better: humans reside in a more-​than-​human world that is as embedded in us as we are in it (Haraway, 1997; Whatmore, 2001; Braun, 2009). Economic (and other societal) processes and the more-​than-​human world are mutually constitutive. The nature and driving forces of geological, geomorphological, biological, and meteorological processes are quite distinct from those of economic and societal processes. Their articulations are not reducible to economic logics, but neither are they determined by immutable natural laws. The capitalist space economy seeks to domesticate the more-​than-​human world to its logics, through such processes as commodification, full-​cost pricing, carbon markets, and technological innovations to control and shape nature. At the same time, more-​than-​human processes consistently exceed and escape such attempts at their containment, reshaping economic processes (O’Connor, 1991; de Landa, 1997). The immense complexity of entanglements between the economy and the more-​than-​human world raise profound questions about the capacity of mainstream (geographical) economic theory to account for the evolution of a capitalist space economy, but are an active area of economic geographical scholarship (e.g. Harvey, 1996; Castree, 2005; Bakker and Bridge, 2008; Bumpus and Liverman, 2008; Robertson and Wainwright, 2013).

Proposition 18: Economic Analysis is a Moral Task Notwithstanding genuflections to the supposed goal of social science as an impartial and eventually truthful understanding of the world, questions of morals and ethics cannot be set aside from studying the economy (Sayer, 2007). There is an inescapably value-​laden frame for all research, whether or not we seek to elevate it by using the adjective ‘scientific,’ particularly given the failure of all attempts to identify a philosopher’s stone that is capable of turning base research into golden truth. Logical positivism, often resorted to by those seeking to claim scientific impartiality, failed in its attempt to identify such a foolproof method (Passmore, 1967; Sheppard, 2014). The best we can do, then, is much more modest—​a mutual critical engagement between different explanations and interpretations of the world, each shaped by the experiences and norms of differently positioned investigators. Views about justice, imaginaries of what constitutes a good world, and the nature and extent of our moral community infuse the very labels framing contemporary geographical research into the economy (radical, critical, feminist, etc.). But, equally, they are central to mainstream economic orthodoxy, whose

Heterodoxy as Orthodoxy    171 theories legitimize claims about the societal benefits of globalizing capitalism, at least in principle. Thus, there can be no single, privileged theory of the economy—​a monism that is no longer subject to question. There will always be multiple, differently situated theoretical perspectives (in and beyond geography) that should be prepared to learn from one another (Longino, 2002; Barnes and Sheppard, 2010). Beyond this are the vital and always-​ fraught questions of professional ethics, particularly with respect to academics’ relations with the world that we seek to illuminate and change. The anticanonical nature of geography has the merit of forcing us to attend to such issues, although much remains to be done to problematize this.

Proposition 19: Alternatives to Globalizing Capitalism are Necessary We live in a world where it is taken for granted that the capitalist space economy is global. The logics of commodification and marketization seem ubiquitous, and operative at every geographical scale—​pervasive in the public, as well as the private, sector, and seeping through the pores of everyday life. Along with this are powerful discourses, associated particularly with this neoliberal era of globalization, that globalized capitalism is the best (or least worst) form of economy—​capable, in principle, of enabling anyone to prosper and realize their desires. Failure to succeed is characteristically laid at the feet of individuals (unable or unwilling to make the most of the opportunities available to them), or places (condemned by bad governance, culture, or geography). Geographical political economy reaches a different conclusion, however. Not only is capitalism far less ubiquitous than it seems, but the possibilities associated with such alternatives should be taken seriously. Reiterating a long-​standing interest by geographers in diverse economic systems stretching back to the early-​twentieth-​century era of what they termed commercial geography, geographical political economists have undertaken important and influential work, documenting how the capitalist space economy is far less prevalent than it seems. Notwithstanding appearances, our world is also constituted through all kinds of non-​or more-​than-​capitalist economic processes and logics. Recognizing this diversity of economies creates critical space to imagine and practice alternatives to capitalism (Gibson-​Graham, 1996, 2006; Gibson-​ Graham et al., 2013). But this is about much more than an existence claim that diverse economies are ever-present. If socio-​spatial inequality is endemic to any capitalist space economy (proposition 13), then it is vital to experiment with alternatives, attending also to how they articulate with capitalism, if we are to move closer to a world where socio-​spatial justice, and justice to our more-​than-​human cohabitants, is realizable (Low and Gleeson, 1998; Harvey, 2000; Sheppard, 2011b).

Conclusion I have sought to lay out the broad lineaments of a shared vision of geographical political economy, clarifying what differentiates now-​orthodox geographical thinking about economic geography from now-​orthodox mainstream (geographical) economics. The immense

172   Sheppard diversity of the scholarship gathered under geographical political economy, in topics, philosophical claims, and methodologies, is simultaneously its great strength and a potential weakness (Sheppard and Barnes, 2000; Tickell et al., 2007; Barnes et al., 2012). It is replete with passionate theoretical debates and seemingly intransigent divides, supplemented by a wealth of provocative and illuminating empirical studies reflecting these different positions. Many of its participants might well resist the label ‘geographical political economy’ at all, seeing their work as quite distinct from Marxian political economy (often personified as disagreements with David Harvey and his ilk). My claim is that the eclectic community of critical economic geographers, notwithstanding their differences, would nevertheless be willing to endorse these nineteen propositions (albeit prioritizing these claims very differently to that implicit in my sequencing). In this sense, geographical thinking about the economy has a shared Weltanschauung; a commonality to return to when things seem to be falling apart at the seams (Barnes and Sheppard, 2010). Even if I have been somewhat successful in identifying a set of core claims, it goes without saying that much remains to be done. My priorities would include: a deeper engagement with transportation and communications—​the commodities and logistics that hold globalizing capitalism together, and their mobile workforces (Cowen, 2014); the digitizing economy (the Internet of things, the misnamed ‘sharing economy,’ the availability to firms of individualized data on consumption and shopping behaviour, three-​dimensional printing, etc.); the relationship between the economy and the more-​ than-​human world—​a defining interest of our discipline still plagued by the spectre of environmental determinism; and more engagement with the economic geography of the postcolonial world—​including the crucial question whether a geographical political economy, honed through studies of the North Atlantic realm, is applicable to the very differently positioned societies of the postcolony (Pollard et al., 2011). Others will have much to add to this list, a symptom of the ongoing vitality and promise of geographical research on economies.

Notes 1. Related experiences, notwithstanding the occasional counter-​example (e.g. Garretsen and Martin, 2010), have pushed some economic geographers to advocate abandoning altogether any attempts at such engagement (Amin and Thrift, 2000; Peck, 2012). 2. This assumption is that all sectors have identical input requirements—​the same assumption that is necessary for Marx’s transformation problem to work. 3. Neoclassical economics reduces production to a quasi-​exchange, whereby capital goods, labour, and technology, demanded by the firm, are translated into goods supplied (output) by means of a production function. 4. Mainstream economics equates market price with marginal utility, but this necessitates adopting the profoundly unrealistic presumption (proposition 16) that consumers are rational, autonomous individuals with exogenously determined preferences, are fully informed, and seek to maximize personal utility. 5. Krugman’s spatial equilibria face the same problem (Fowler, 2011). 6. Finance and money are conspicuous by their absence from canonical texts of geographical economics (see Combes et al., 2008; Brakman et al., 2009; Duranton et al., 2015).

Heterodoxy as Orthodoxy    173

References Agnew, J.A. (2012). ‘Putting Politics Into Economic Geography’ in T.J. Barnes, J. Peck, and E. Sheppard (eds) The Wiley-​Blackwell Companion to Economic Geography, pp. 567–​580 (New York: Wiley). Amin, A. and Thrift, N. (2000). ‘What kind of economic theory for what kind of economic geography?’ Antipode 32: 4–​9. Auty, R.M. (2007). ‘Natural resources, capital accumulation and the resource curse.’ Ecological Economics 61: 627–​634. Bakker, K. and Bridge, G. (2008). ‘Material worlds? Resource geographies and the matter of nature’. Progress in Human Geography 30: 5–​27. Barnes, T. (1996). Logics of Dislocation: Models, Metaphors, and Meanings of Economic Space (New York: Guilford Press). Barnes, T. and Sheppard, E. (2010). ‘Nothing includes everything. Towards engaged pluralism in Anglophone economic geography.’ Progress in Human Geography 34: 193–​214. Barnes, T.J. (1995). ‘Political economy I: “the culture, stupid” ’. Progress in Human Geography 19: 423–​431. Barnes, T.J., Peck, J., and Sheppard, E. (eds) (2012). The Wiley-​Blackwell Companion to Economic Geography (New York: Wiley-​Blackwell). Bergmann, L. (2012). ‘A coevolutionary approach to the capitalist space economy’. Environment and Planning A 44: 518–​537. Bergmann, L. (2013). ’Bound by chains of carbon: ecological–​economic geographies of globalization’. Annals of the Association of American Geographers 103: 1348–​1370. Bergmann, L., Sheppard, E., and Plummer, P. (2009). ‘Capitalism beyond harmonious equilibrium: mathematics as if human agency mattered.’ Environment and Planning A 41: 265–​283. Berndt, C. and Boeckler, M. (2012). ‘Geographies of Marketization’ in T.J. Barnes, J. Peck, and E. Sheppard (eds) The Wiley-​Blackwell Companion to Economic Geography, pp. 199–​212 (Oxford: Wiley-​Blackwell). Blomley, N., Delaney, D., and Ford, R.T. (eds) (2001). The Legal Geographies Reader: Law, Power and Space (Oxford: Blackwell). Boschma, R.A. and Martin, R. (eds) (2010). The Handbook of Evolutionary Economic Geography (London: Edward Elgar). Brakman, S., Garretsen, H., and von Marrewijk, C. (2009). A New Introduction to Geographical Economics (Cambridge: Cambridge University Press). Braun, B. (2009). ‘Nature’ in N. Castree, D. Demeritt, D. Liverman, et al. (eds) A Companion to Environmental Geography, pp. 29–​36 (Oxford: Blackwell). Brenner, N. (2004). New State Spaces:  Urban Governance and the Rescaling of Statehood (Oxford: Oxford University Press). Brenner, N., Jessop, B., Jones, M., et al. (eds) (2003). State/​Space: A Reader (New York: Wiley). Bumpus, A.G. and Liverman, D. (2008). ‘Accumulation by decarbonization and the governance of carbon offsets.’ Economic Geography 84: 127–​155. Çaliskan K. and Callon M. (2010). ‘Economization, part 2: a research programme for the study of markets.’ Economy and Society 39: 1–​32. Castree, N. (2005). Nature (London: Routledge). Chatty, R., Hendren, N., Kline, P., et al. (2014). Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States (Cambridge, MA: National Bureau of Economic Research).

174   Sheppard Christophers, B. (2013). Banking Across Boundaries:  Placing Finance in Capitalism (New York: John Wiley). Clark, G.L. and Wójcik, D. (2007). The Geography of Finance (Oxford:  Oxford University Press). Clark G.L., Gertler, M.S., and Feldman, M.P. (eds) (2000). The Oxford Handbook of Economic Geography (Oxford: Oxford University Press). Clark, G.L., Dixon, A.D. and Monk, A.H.B. (2013). Sovereign Wealth Funds:  Legitimacy, Governance, and Global Power (Princeton, NJ: Princeton University Press). Combes, P.-​P., Mayer, T., and Thisse, J.-​F. (2008). Economic Geography: The Integration of Regions and Nations (Princeton, NJ: Princeton University Press). Cowen, D. (2014). The Deadly Life of Logistics: Mapping Violence in Global Trade (Minneapolis, MN: University of Minnesota Press). de Landa, M. (1997). A Thousand Years of Nonlinear History (New York: Zone Books). Denike, K. and Parr, J. (1970). ‘Production in space, competition and restricted entry.’ Journal of Regional Science 10: 49–​63. Derrida, J. (1976). Of Grammatology (Baltimore, MD: Johns Hopkins University Press). Dixon, A.D. (2014). The New Geography of Capitalism: Firms, Finance and Society (Oxford: Oxford University Press). Duménil, G. and Lévy, D. (1991). ‘Micro adjustment toward long-​term equilibrium.’ Journal of Economic Theory 53: 369–​395. Duranton, G., Henderson, V., and Strange, W. (eds.) (2015). Handbook of Regional and Urban Economics (Amsterdam: Elsevier). Dymski, G.A. and Li, W. (2003). ‘The macrostructure of financial exclusion: mainstream, ethnic, and fringe banks and money/​space.’ Espaces, Populations, Societes 1: 183–​201. Featherstone, D. (2012). Solidarity:  Hidden Histories and Geographies of Internationalism (London: Zed Press). Fine, B. (2002). ‘Economics imperialism and the new development economics as Kuhnian paradigm shift?’ World Development 30: 2057–​2070. Foucault, M. (1971). The Archeology of Knowledge (London: Tavistock). Fowler, C.S. (2011). ‘Finding equilibrium: how important is general equilibrium to the results of geographical economics?’ Journal of Economic Geography 11: 457–​480. Garcia-​Perpet, M.-​F. (2007). ‘The Social Construction of a Perfect Market: The Strawberry Auction at Fontaines-​en-​Sologne’ in D. Mackenzie, F. Muniesa, and L. Siu (eds) Do Economists Make Markets? pp. 20–​53 (Princeton, NJ: Princeton University Press). Garretsen, H. and Martin, R. (2010). ‘Rethinking (new) economic geography models: taking geography and history more seriously’. Spatial Economic Analysis. 5: 127–​160. Georgescu-​ Roegen, N. (1971). The Entropy Law and the Economic Process (Cambridge, MA: Harvard University Press). Gibson-​Graham, J.K. (1996). The End of Capitalism (As We Knew It) (Oxford: Blackwell). Gibson-​ Graham, J.K. (2006). A Postcapitalist Politics (Minneapolis, MN:  University of Minnesota Press). Gibson-​Graham, J.K., Cameron, J., and Healy, S. (2013). Take Back the Economy, Any Time, Any Place (Minneapolis, MN: University of Minnesota Press). Giddens, A. (1984). The Constitution of Society: Outline of the Theory of Structuration (Berkeley, CA: University of California Press). Greenhut M.L., Norman G., and Hung C.-​ S. (1987). The Economics of Imperfect Competition: A Spatial Approach (Cambridge: Cambridge University Press).

Heterodoxy as Orthodoxy    175 Grigg, D. (2003). An Introduction to Agricultural Geography (London: Routledge). Haraway, D. (1997). Modest_​Witness@Second_​Millenium.FemaleMan©_​Meets_​OncoMouse™ (London: Routledge). Harcourt, G.C. (1972). Some Cambridge Controversies in the Theory of Capital (Cambridge: Cambridge University Press). Harvey, D. (1982). The Limits to Capital (Oxford: Basil Blackwell). Harvey, D. (1996). Justice, Nature and the Geography of Difference (Oxford: Basil Blackwell). Harvey, D. (2000). Spaces of Hope (Berkeley, CA: University of California Press). Harvey, D. (2014). Seventeen Contradictions and the End of Capitalism (Oxford: Oxford University Press). Henderson, G. (2013). Value in Marx: The Persistence of Value in a More-​than-​Capitalist World (Minneapolis, MN: University of Minnesota Press). Herod, A. (ed.) (1998). Organizing the Landscape (Minneapolis, MN: University of Minnesota Press). Herod, A. (2001). Labor Geographies:  Workers and the Landscapes of Capitalism (New York: Guilford). Isard, W. (1951). ‘Interregional and regional input-​output analysis: a model of a space-​economy’. The Review of Economics and Statistics 33: 318–​328. Krugman, P. (1991). Geography and Trade (Cambridge, MA: MIT Press). Lawson, T. (2005). ‘The nature of heterodox economics’. Cambridge Journal of Economics 30: 483–​505. Lee, F.S. (1998). Post Keynesian Price Theory (Cambridge: Cambridge University Press). Lefebvre, H. (1974 [1991]). The Production of Space (Oxford: Blackwell). Leontief, W. (1928). ‘Die Wirtschaft als Kreislauf ’. PhD, University of Berlin. Leyshon, A. and Thrift, N. (1997). Money/​Space:  Geographies of Monetary Transformation (London: Routledge). Longino, H. (2002). The Fate of Knowledge (Princeton NJ: Princeton University Press). Lösch, A. (1954 [1940]). The Economics of Location (New Haven, CT: Yale University Press). Low, N. and Gleeson, B. (1998). Justice, Society and Nature (London: Routledge). McDowell, L. (1997). Capital Culture: Money, Sex and Power at Work (Oxford: Blackwell). McDowell, L. (2015). ‘The lives of others: body work, the production of difference, and labor geographies’. Economic Geography 91: 1–​23. Mackenzie, D. (2009). Material Markets:  How Economic Agents are Constructed (Oxford: Oxford University Press). Mackenzie, D., Muniesa, F., and Siu, L. (eds) (2008). Do Economists Make Markets? On the Performativity of Economics (Princeton, NJ: Princeton University Press). Mann, G. (2007). Our Daily Bread: Wages, workers, and the Political Economy of the American West (Chapel Hill, NC: University of North Carolina Press). Markusen, A., Wassall, G.H., DeNatale, D., et al. (2008). ‘Defining the creative economy: industry and occupational approaches.’ Economic Development Quarterly 22: 24–​45. Martin, R. and Sunley, P. (2015). ‘Towards a developmental turn in evolutionary economic geography?’ Regional Studies 49: 712–​732. Martin, R., Sunley, P., and Wills, J. (1996). Retreat and the Regions: The Shrinking Landscape of Organised Labour (London: Jessica Kingsley). Marx, K. (1967 [1867]). Capital: A Critique of Political Economy, Vol. 1 (New York: International Publishers). Marx, K. (2008 [1897]). The 18th Brumaire of Louis Bonaparte (Rockville, MD: Wildside Press).

176   Sheppard Milanovic, B. (2011). Worlds Apart: Measuring International and Global Inequality (Princeton, NJ: Princeton University Press). Mirowski, P. (2013). Never Let a Serious Crisis go to Waste: Now Neoliberalism Survived the Financial Meltdown (London: Verso). Mirowski, P. and Plehwe, D. (eds.) (2009). The Road from Mount Pelerin: The Making of the Neoliberal Thought Collective (Cambridge, MA: Harvard University Press). Mitchell, D. (1996). The Lie of the Land:  Migrant Workers and the California Landscape (Minneapolis, MN: University of Minnesota Press). Mitchell, T. (2005). ‘The work of economics:  how a discipline makes its world’. European Journal of Sociology 46: 297–​320. Morishima, M. (1973). Marx’s Economics:  A  Dual Theory of Value and Growth (Cambridge: Cambridge University Press). Moudud, J.K., Bina, C., and Mason, P.L. (eds) (2013). Alternative Theories of Competition (London: Routledge). Mulligan, G.F. and Fik, T.K. (1989). ‘Price variation in spatial oligopolies’. Geographical Analysis 21: 32–​46. Nagurney, A. (1987). ‘Competitive equilibrium problems, variational inequalities and regional science’. Journal of Regional Science 27: 503–​518. O’Brien, R. (1992). Global Financial Integration: The End of Geography (London: The Royal Institute of International Affairs). O’Connor, J. (1991). ‘On the two contradictions of capitalism’. Capitalism Nature Socialism 2: 107–​109. Okishio, N. (1961). ‘Technical changes and the rate of profit’. Kobe University Economic Review 7: 85–​99. Ottaviano, G. and Thisse, J.-​F. (2004). ‘Agglomeration and Economic Geography’ in J.V. Henderson and J.-​F. Thisse (eds) Handbook of Urban and Regional Economics, Vol. 4, pp. 2564–​2608 (Amsterdam: Elsevier). Passmore, J. (1967). ‘Logical Positivism’ in P. Edwards (ed.) The Encyclopedia of Philosophy, pp. 52–​57 (New York: Macmillan). Peck, J. (2010). Constructions of Neoliberal Reason (Oxford: Oxford University Press). Peck, J. (2012). ‘Economic geography: island life’. Dialogues in Human Geography 2: 113–​133. Peck, J. and Theodore, N. (2007). ‘Variegated capitalism’. Progress in Human Geography 31: 731–​772. Peck, J. and Theodore, N. (2015). Fast Policy:  Experimental Statecraft at the Thresholds of Neoliberalism (Minneapolis, MN: University of Minnesota Press). Piketty, T. (2014 [2013]). Capital in the Twenty-​first Century (Cambridge, MA: Belknap Books). Plummer, P. and Sheppard, E. (2006). ‘Geography matters: agency, structures and dynamics’. Journal of Economic Geography 6: 619–​637. Plummer, P., Sheppard, E., and Haining, R.P. (2012). ‘Rationality, stability and endogenous price formation in spatially interdependent markets’. Environment and Planning A 44: 538–​559. Pollard, J.S. (1996). ‘Banking at the margins: a geography of financial exclusion’. Environment and Planning A 28: 1209–​1232. Pollard, J. and Samers, M. (2007). ‘Islamic banking and finance: postcolonial political economy and the decentring of economic geography’. Transactions of the Institute of British Geographers 32: 313–​330. Pollard, J., McEwan, C., and Hughes, A. (eds.) (2011). Postcolonial Economies (London: Zed Books).

Heterodoxy as Orthodoxy    177 Quesnay, F. (1753–​58). Tableau Économique (Versailles: privately printed). Rees, J.A. (1990). Natural Resources: Allocation, Economics and Policy (London: Routledge). Rigby, D. (1990). ‘Technical change and the rate of profit: an obituary for Okishio’s theorem’. Environment and Planning A 22: 1039–​1050. Rigby, D. and Essletzbichler, J. (2006). ‘Technological variety, technological change and a geography of production techniques’. Journal of Economic Geography 6: 45–​70. Robertson, M.M. and Wainwright, J. (2013). ‘The value of nature to the state’. Annals of the Association of American Geographers 10​3: 890–​905. Rodrik, D., Subramanian, A., and Trebbi, F. (2004). ‘Institutions rule: the primacy of institutions over geography in economic development’. Journal of Economic Growth 9: 131–​165. Rosser, Jr, J.B. (2000). ‘Aspects of dialectics and non-​linear dynamics’. Cambridge Journal of Economics 24: 311–​324. Roy, A. (2010). Poverty Capital (New York: Wiley-​Blackwell). Sayer, A. (2007). ‘Moral economy as critique’. New Political Economy 12: 261–​270. Scott, A.J. (1988). New Industrial Spaces:  Flexible Production Organization and Regional Development in North America and Western Europe (London: Pion). Scott, A.J. and Power, D. (eds) (2004). Cultural Industries and the Production of Culture (London: Routledge). Scott, J.C. (1985). Weapons of the Weak: Everyday Forms of Peasant Resistance (New Haven, CT: Yale University Press). Sheppard, E. (2004). ‘The spatiality of limits to capital.’ Antipode 36: 470–​479. Sheppard, E. (2006a). ‘The Economic Geography Project’ in S. Bagchi-​Sen and H. Lawton Smith (eds) Economic Geography: Past, Present and Future, pp. 11–​ 23 (New York: Routledge). Sheppard, E. (2006b). ‘Positionality and Globalization in Economic Geography’ in G. Vertova (ed.) The Changing Economic Geography of Globalization, pp. 45–​72 (London: Routledge). Sheppard, E. (2008). ‘Geographic dialectics?’ Environment and Planning A 40: 2603–​2612. Sheppard, E. (2011a). ‘Geographical political economy’. Journal of Economic Geography 11: 319–​331. Sheppard, E. (2011b). ‘Geography, nature and the question of development’. Dialogues in Human Geography 1: 3–​22. Sheppard, E. (2012). ‘Trade, globalization and uneven development:  entanglements of geographical political economy’. Progress in Human Geography 36: 44–​7 1. Sheppard, E. (2014). ‘We have never been positivist’. Urban Geography 35: 636–​644. Sheppard, E. (2015). ‘Thinking geographically: globalizing capitalism, and beyond’. Annals of the Association of American Geographers 105: 1113–​1134. Sheppard, E. (2016). Limits to Globalization:  The Disruptive Geographies of Capitalist Development (Oxford: Oxford University Press). Sheppard, E. and Barnes, T.J. (1990). The Capitalist Space Economy: Geographical Analysis After Ricardo, Marx and Sraffa (London: Unwin Hyman). Sheppard, E. and Barnes, T.J. (2000). A Companion to Economic Geography (Oxford: Blackwell). Sheppard, E. and Curry, L. (1982). ‘Spatial price equilibria’. Geographical Analysis 14: 279–​304. Sheppard, E. and Glassman, J. (2010). ‘Social Class’ in J. Agnew and D. Livingstone (eds) The SAGE Handbook of Geographical Knowledge, pp. 401–​416 (London: SAGE). Sheppard, E., Barnes, T.J., and Peck, J. (2012). ‘The Long Decade:  Economic Geography, Unbound’ in T.J. Barnes, J. Peck, and E. Sheppard (eds) The Wiley-​Blackwell Companion to Economic Geography, pp. 1–​24 (Oxford: Wiley-​Blackwell).

178   Sheppard Sheppard, E., Haining, R.P., and Plummer, P. (1992). ‘Spatial pricing in interdependent markets’. Journal of Regional Science 32: 55–​75. Smith, A. (1776). An Inquiry Into the Nature and Causes of the Wealth of Nations (London: A. Strahan and T Cadell). Smith, N. (1984). Uneven Development:  Nature, Capital and the Production of Space (Oxford: Basil Blackwell). Soja, E. (1980). ‘The socio-​spatial dialectic’. Annals of the Association of American Geographers 70: 207–​225. Sraffa, P. (1960). The Production of Commodities by Means of Commodities (Cambridge: Cambridge University Press). Star, S.L. (1989). ‘The Structure of Ill-​structured Solutions: Heterogeneous Problem-​solving, Boundary Objects and Distributed Artificial Intelligence’ in M. Huhns and L. Gasser (eds) Distributed Artificial Intelligence 2, pp. 37–​54 (Menlo Park, CA: Morgan Kauffmann). Starrett, D. (1978). ‘Market allocations of location choice in a model with free mobility’. Journal of Economic Theory 17: 21–​37. Steedman, I. (1979). Trade amongst Growing Economies (Cambridge: Cambridge University Press). Steedman, I. and Metcalfe, J.S. (1979). ‘Reswitching, Primary Inputs and the Hecksher–​Ohlin–​ Samuelson Theory of Trade’ in I. Steedman (ed.) Fundamental Issues in Trade Theory, pp. 38–​46 (New York: St. Martin’s Press). Storper, M. and Walker, R. (1989). The Capitalist Imperative:  Territory, Technology and Industrial Growth (Oxford: Basil Blackwell). Thrift, N. (2005). Knowing Capitalism (London: SAGE). Thrift, N. and Olds, K. (1996). ‘Refiguring the economic in economic geography’. Progress in Human Geography 20: 311–​338. Tickell, A. and Peck, J. (1992). ‘Accumulation, regulation and the geographies of post-​ Fordism: missing links in regulationist research’. Progress in Human Geography 16: 190–​218. Tickell, A., Sheppard, E., Peck, J., et  al. (eds.) (2007). Politics and Practice in Economic Geography (London: SAGE). Valentine, G. (2007). ‘Theorizing and researching intersectionality: a challenge for feminist geography.’ The Professional Geographer 59: 10–​21. Webber, M., Sheppard, E., and Rigby, D. (1992). ‘Forms of technical change’. Environment and Planning A 24: 1679–​1709. Whatmore, S. (2001). Hybrid Geographies: Natures, Cultures, Spaces (Thousand Oaks, CA: SAGE). Wong, K.-​Y. (1995). International Trade in Goods and Factor Mobility (Cambridge, MA: MIT Press). Yeung, H.W.-​C. (2005). ‘Rethinking relational economic geography’. Transactions of the Institute of British Geographers NS 30: 37–​51.

Chapter 9

Re l ational Re se a rc h Design in Ec onomi c Geo gra ph y Harald Bathelt and Johannes Glückler Introduction: Why a Relational Research Design? In recent years, the field of economic geography has exhibited tendencies not only towards segmentation and disintegration, but also towards expansion and reintegration. Within the context of geography, increasing separations between economic and cultural orientations and quantitative and qualitative approaches can be observed. At the same time, a powerful, yet careful, convergence of research interests has occurred across disciplines such as regional economics, economic sociology, and business and management studies, all of which have become more interested in questions of unequal distribution and development of economic activities across space. While being committed to their own research traditions, a gradual dialogue has emerged which crosses disciplinary boundaries, for example in the Journal of Economic Geography, which was launched in 2000 with the specific goal of driving such debates. The ‘relational approach’ is one conceptual tool that aims to overcome separations between different disciplinary perspectives and support cross-​disciplinary engagement (Bathelt and Glückler, 2011). It facilitates productive conversation across academic fields by embodying a broader social science perspective that applies economic, social, institutional, cultural, and political dimensions to the study of economic action (Yeung, 2005). Relational economic geography analyses economic processes from a spatial perspective, which generates immediate positionality in studying the specificity and diversity of places and the flows and interdependencies between them, as well as the spatiality of process outcomes. It is this analysis of economic and social processes from a spatial perspective that has attracted interest from other disciplines. The goal of this chapter is to illustrate how the relational approach enables us to conceptualize questions and problems in economic geography and how this can facilitate links to

180   Bathelt and Glückler studies in neighbouring fields that focus on similar research objects. Those questions that bring these different research agendas together are fundamental to economic and social life because they focus on both the processes that drive economic action and interaction and on their outcomes across space. This is shown in this chapter by investigating one specific research question, namely: What are the different ways in which firms and industries benefit from geographical agglomeration? The goal is to illustrate key principles of the relational approach and to characterize elements of what we refer to as a relational research design. The relational approach is not a theory of economic geography per se, but a comprehensive research perspective grounded in principles of relationality. The remainder of this chapter will trace the implications of these principles for the design of relational research practices.

Fundamentals of the Relational Economy As opposed to conventional approaches in economic geography that use spatial structures or spatial variables as a starting point for analyses, the relational approach focuses on the actors most relevant to the problem or question under investigation (Bathelt and Glückler, 2003; Clark and Tracey, 2004). For instance, instead of using data about spatial cost structures (alongside other variables) to conduct aggregate statistical analyses and draw inferences about the decline of regional economies, the relational approach investigates the specific technological and institutional contexts and processes that affect the performance of the corresponding economic actors. By studying corporate linkages and the relationships between firms in the value chain, the relational approach identifies the conditions under which firms make localized and trans-​local decisions and how this depends on the presence of complementary and competing businesses in other regions. There are three different principles that guide a relational research design in fundamental ways (Bathelt and Glückler, 2011): context, path dependence, and contingency. Firstly, it is crucial to recognize the context within which specific producer–​user relations unfold. It is this context that drives technological change and generates global competitiveness. Economic interaction is always fundamentally contextual in nature. Secondly, and relatedly, firms base their decisions on pre-​existing structures that have resulted from prior decisions and from their evaluation of these decisions. The resulting reflexivity of economic decision-​making translates into path-​dependent economic processes, in the course of which pre-​existing economic structures impact future developments. Thirdly, this does not imply, however, that pre-​existing structures or economic contexts determine the outcomes of economic decision-​making (Sayer, 1992, 2000). In fact, they do not. Firms may operate under the same conditions as their nearby competitors, yet they may draw different conclusions from their rivals about how to differentiate themselves or address a distinct customer clientele. Specialization and differentiation are fundamental principles of competitive behaviour, which is why economic decisions and their outcomes are contingent in nature and not predetermined. A relational research design is thus built around the fundamental principles of contextuality, path dependence, and contingency of economic action and interaction, and how these principles translate into spatial and cross-​spatial structures and linkages.

Relational Research Design in Economic Geography    181 This perspective resonates with concepts of the milieu school (Crevoisier and Maillat, 1991), the global production networks approach (Dicken et  al., 2001; Yeung and Coe, 2015), and practice theory (Jones and Murphy, 2011; Faulconbridge, 2017), as well as other approaches in economic geography that explicitly focus on the interdependencies between actors and the situated nature of social and economic relations (e.g. Dicken and Malmberg, 2001; Hayter, 2004; Hudson, 2004). Recently, and in a similar way, Lagendijk (2017) conceptualizes regional development through processes of circulation, trajectories, and logics, fundamentally drawing on a multiple-​agent perspective. In all these approaches, economic action is understood to occur in networks and structures of social relations that develop in dynamic ways, influenced—​yet not determined—​by prior action.

Contextual Theorization and the Cluster Approach One important field in which relational approaches have been widely applied is cluster research. Based on Porter’s (1990) seminal work, the cluster approach has signalled an important step forward in management studies by recognizing that a firm’s competitiveness does not merely rely on its own capabilities, but also depends—​sometimes crucially so—​ on linkages with suppliers and users and the competitive environment in the industry. The cluster approach utilizes a relational framework to understand the competitiveness of firms (Delgado et al., 2016) and extends the analysis of economic action from a corporate to a territorial perspective, which can be national or regional in character depending on the specific analytical focus. The cluster approach has generated manifold opportunities for cross-​disciplinary exchanges and created linkages between economics, management, and economic geography (Porter, 2000). However, as will be shown in the following sections, there is no single cluster concept that establishes a general theory of agglomeration. The explanatory framework to understand clustering processes needs to be adjusted to the specific sectoral and technological contexts under investigation. To illustrate this, this chapter uses a regional cluster approach and applies it to two different sectoral contexts: manufacturing and professional services.

Manufacturing Clusters and Know-​how Dynamics The conventional cluster approach has primarily been applied to manufacturing industries. To understand why manufacturing clusters exist, how they grow, and through which mechanisms they reproduce themselves (Malmberg and Maskell, 2002), value-​chain-​based explanations related to Porter’s (1990) original conceptualization have been developed. This work has shown that three conceptual levels of cluster relationships can be distinguished, each of which is crucial for cluster development and the corresponding learning and interaction processes. These levels can be distinguished according to (i) the type of value-​chain relationships with other firms (i.e. vertical vs horizontal); (ii) the spatial context of relationships (i.e.

182   Bathelt and Glückler local/​regional vs cross-​local/​cross-​regional); and (iii) the institutional context of reproduction (Bathelt and Glückler, 2011). A crucial feature of the first conceptual level of cluster relationships is the diversity of the learning mechanisms in the cluster’s value-​chain linkages. On the one hand, firms are vertically linked through supplier–​producer–​user relationships that are complementary in character (Maskell and Malmberg, 1999; Gordon and McCann, 2000). They produce complementary products and benefit from interactive learning processes in generating new products. This may take the form of co-​development of product designs with core suppliers or systematic feedbacks from co-​located users (von Hippel, 2001; Gertler, 2004). The learning mechanisms in place here rely on repeated interaction and direct contact between technical specialists. Collaborating with cluster firms has clear advantages in this case, as it reduces transaction costs (Scott, 1988). On the other hand, firms in clusters also benefit from being close to competing firms that produce the same or similar products and are horizontally related in the value chain. These firms typically have no intention to collaborate. Rather, they keep developments and new product strategies secret from their competitors. Being co-​located in a cluster, however, creates all sorts of cross-​organizational knowledge flows about new product developments or technological changes, based on rumours, gossip, and personal relationships. This supports the emergence of a specific knowledge ecology, or local ‘buzz’ (Storper and Venables, 2004). Cluster firms benefit from these knowledge flows as they operate under similar cost structures and can quickly identify the reasons why competitors are more or less successful. This generates rivalry and enables horizontal learning based on observation and monitoring (Glückler, 2013; Li, 2017), which facilitates product differentiation. Essentially, it is the combination of both horizontal and vertical relations that drives cluster growth. These arguments suggest that studies of manufacturing clusters should not only be based on a transaction perspective alone, but they also need to attend to related knowledge flows. Indeed, recent conceptualizations have emphasized the role of knowledge creation and transfer (Pinch et al., 2003; Bathelt et al., 2004). This links to the second conceptual level of cluster relationships, which concerns their spatial context. More precisely: while cluster-​ based, internal linkages and knowledge flows may be very effective, owing to localized vertical and horizontal learning processes, important technological breakthroughs or fundamental changes in consumption patterns may take place in other regional/​national markets or competitive environments. These are the changes that require firms to be open to developments outside the cluster. To avoid lock-​in into regional technology paths and suboptimal localized solutions and strategies, which could cause competitive problems in the future (Grabher, 1993; Clark and Tracey, 2004), cluster firms need to engage actively in building relationships with partners in other regions and countries. In other words, they need to establish trans-​local or ‘global knowledge pipelines’ (Bathelt et al., 2004; Owen-​Smith and Powell, 2004). The need for local cohesion and relationship building, as well as trans-​local openness to access different knowledge ecologies, also has consequences for the third conceptual level of cluster relationships. This directs our attention to the institutional context (Glückler and Bathelt, 2017). Regional clusters benefit from shared institutional settings that have developed over time based on common experiences and continuous interaction within the same technological context, supported by specific cluster policies, research networks, and training programmes. Such coherent settings support the development of highly specialized labour markets, allow for efficient knowledge exchanges with co-​located partners, and generate

Relational Research Design in Economic Geography    183 manifold opportunities for interaction and learning. However, if such institutional contexts are too narrow, too exclusive, and too self-​referential, cluster actors may become over-​ embedded in the localized knowledge ecology (Uzzi, 1997) and easily miss out technological and market opportunities that are related to institutional settings outside the cluster (Bathelt et  al., 2004; Owen-​Smith and Powell, 2004). This danger may be exacerbated by strong regional power asymmetries and rigid hierarchical power relations, which limit opportunities to break out of dominant routines and institutional arrangements and constrain technological change (Dicken et al., 2001; Allen, 2003). Related conceptions have been used to explain the competitive advantage of manufacturing clusters that rely on a value-​chain-​based organization. Such analyses emphasize different dimensions of knowledge generation and suggest that clusters can grow and reproduce themselves by capitalizing on synergies across these dimensions and avoiding trade-​offs between them (Bathelt and Glückler, 2011). Inter-​firm relationships in the context of manufacturing clusters develop in order to access technological know-​how by mobilizing both local and trans-​local knowledge flows.

Professional Services Clusters and Know-​who Professional and knowledge-​intensive services are highly concentrated in certain cities, some of which have become leading international centres of finance, media, advertising, legal, accounting, and consulting services. The growth of these services favours existing urban clusters over other locations (Glückler and Hammer, 2011). To account for the development of dynamic professional services clusters, a different explanatory approach is needed than that discussed for manufacturing because some assumptions underlying conventional cluster research do not apply to professional services. For example, compared with manufacturing, professional services firms maintain limited vertical value-​chain relations with suppliers. Transaction costs and vertical learning processes are thus less important. Additionally, despite the fact that geographical proximity is important to customers, professional services firms often find the majority of their clients outside of their immediate catchment area. To understand why knowledge services firms cluster in urban centres, it is necessary to take these differences into account through context-​specific theorization. The focus of analysis has to shift from backward production linkages to forward market relationships (Keeble and Nachum, 2002), from cost considerations to business opportunities (Johannisson, 1990), and from technological know-​how to relational know-​who (Lundvall and Johnson, 1994). Firstly, instead of linear and sequential transactions along vertical value-​chain linkages, knowledge services follow the logic of so-​called ‘value shops’. The value shop framework (Stabell and Fjeldstad, 1998) that does not rely on vertical relationships captures the problem-​specific alignment of resources and scheduling of activities in response to client problems. In the context of professional services, problem-​solving activities often need to be carried out precisely in the locations where they are later applied. This impedes the exploitation of economies of scale through a spatial division of labour in upstream activities. And although the value shop model puts much emphasis on external collaboration, the required expertise and choice of partners varies with each specific problem at hand. This need for flexibility runs counter to the rationale for establishing durable relations among co-​located professional services firms. What is important instead is the ‘communication nodality’ (Keeble

184   Bathelt and Glückler and Nachum, 2002) of large and interconnected cities, which facilitates mobility and access to remote partners and clients. Secondly, while conventional understandings of manufacturing clusters identify local externalities from a cost perspective, conceptualizations of professional services clusters need to focus on business opportunities. This is done in the concept of ‘economies of overview’ (Johannisson, 1990) that applies a demand-​side perspective to highlight important urbanization advantages. These derive from the business opportunities that emanate from a comprehensive awareness of markets, actors, and recent developments in the field. Such opportunies increase with the number, diversity, and density of nearby market participants, intermediaries and multipliers, generating substantial agglomeration advantages for large cities. However, if transactions are not important at the local level, the question remains as to why many knowledge-​intensive services firms cluster in metropolitan areas. It is here where a shift to a relational perspective is needed, extending cluster theory by moving from local to trans-​local externalities that result from a city’s connectivity with the international city system. The utility of an urban business location can be viewed as a firm’s access to strategic partners and clients in other locations. This view suggests that cities generate urbanization advantages because they enable firms to realize trans-​local externalities in the global network economy (Moulaert and Djellal, 1995). Thirdly, we need to fit our conceptualization to the type of innovation that dominates within the specific context at hand. Knowledge services do not produce the same kind of innovation as manufacturing firms. Since the crucial challenge for firms is how to acquire new clients for highly specialized services, know-​who (Lundvall and Johnson, 1994) becomes a particularly important form of knowledge. It becomes crucial to mobilize information, referrals, and reputation through networks of interconnected actors across place and space. The entrepreneurial value of know-​who relates to the fact that it yields business opportunities by utilizing economies of overview. And urban clustering generates the size, density, and internal diversity of a pool of specialized firms required to maximize reputational spillovers. In fact, compared with other services firms, those located in urban services clusters have higher shares of clients and revenues from other regions and win new clients more frequently through reputational spillovers (Glückler, 2007). This evidence underlines the importance of know-​who and the economic significance of geographies of reputation in such contexts. Moreover, it suggests that the better connected a city is with other cities in the urban network, the higher the likelihood of local firms finding business opportunities in other places. The two cases of manufacturing and professional services clusters suggest that cluster conceptions need to be designed in a relational way. Cluster developments are not uniform processes and should not be expected to fit a ‘global’ theory. Depending on the context, conceptual adjustments have to be made to integrate the specific conditions of economic action and interaction and to highlight relationships between those variables that matter.

Reflexivity of Interaction in Spatial Perspective: The ‘Geographical Lens’ The considerations on cluster conceptions illustrate the importance of incorporating a spatial perspective in a relational approach, as a ‘geographical lens’ (Bathelt and Glückler,

Relational Research Design in Economic Geography    185 2003) that aims to understand economic action and interaction and their consequences. Individual actors such as managers and workers are localized in certain places, defined by the location of their firms and homes, and this is reflected in the different logics by which they operate in their roles as entrepreneurs, innovators, suppliers, and consumers. As they perform multiple professional and private roles at the same time, the place where they operate becomes a focal point where connections with other actors intersect, overlap, and are bundled. And as these connections are also social in character, the patterns of relationships that develop do not follow purely economic rationales, but involve elements such as trust, feelings, and emotions (Bathelt and Glückler, 2011; Jones and Murphy, 2011). Firstly, economic actors perform different roles and in each role they develop networks of relationships with other actors in order to benefit from specialization within the social division of labour. For instance, as innovators they specialize in certain aspects in the research process while relying for the supply of parts on the technological expertise of other actors. At least initially, these kinds of relationships may be localized—​especially if the different roles are performed routinely in everyday practices. As a consequence, the same actors are involved in a number of different and spatially overlapping activity networks. This creates interdependencies, as resources from one type of relationship can be utilized in different functional networks in the same localized context. Secondly, as several networks of relationships of the same type co-​develop around different actors, interdependencies emerge between them. For instance, competitive relationships develop with little direct interaction between these networks, even though these relationships draw on the same resources and generate opportunities for knowledge spillovers. This supports the development of specialized labour markets and local supplier networks. However, co-​location may also generate opportunities to create localized ‘networks of networks’ leading to broader collaborative practices and economies of scale. In any case, whether by design or otherwise, knowledge flows also develop across multiple co-​located networks of the same type. Thirdly, economic actors can extend their activity space by connecting with actor networks in other places through mobile practices such as business travel (Yeung, 2005). Through the use of modern transportation and communication technologies such places and their networks can be accessed over large distances (Bathelt and Henn, 2014). But even if they are relatively close, they may be characterized by different institutional, cultural, and political settings, and different sets of material resources may shape their functional networks. As a result, different technology and knowledge bases exist in these places and trigger different kinds of labour market dynamics, specialization tendencies, and institutional practices. The generation of linkages across different contexts requires effective understanding and the generation of new professional and/​or personal trust (Ettlinger, 2003). While this may be difficult and risky at first owing to limited experience and a lack of oversight and local power (Allen, 2003), it may become more of a routine process over time as mechanisms are developed to reduce uncertainties in trans-​local linkages. All of these processes generate fundamental interdependencies between similar and different functional networks across space. Finally, an even higher level of complexity in spatial networks develops as migration and relocation create permanent linkages across distant localities and territories, sometimes on a global basis (Dicken et al., 2001; Faulconbridge, 2017). Firms merge with potential partners in other regions worldwide and develop multiple local identities. This generates fundamentally interlinked and closely integrated global knowledge networks that extend the

186   Bathelt and Glückler advantages of co-​present relationships based on ‘relational proximity’ (Amin and Cohendet, 2004) across territorial boundaries. This enables firms to combine aspects of familiarity with otherness, to connect ‘here’ and ‘there’, and to integrate the economy at the micro and macro scale. It is this level of integration that creates permanently interlinked economic spaces of relationships that are simultaneously characterized by interlinked local, trans-​local, and global dynamics. In short, the relational approach requires a spatial perspective as economic action and interaction do not occur in a spaceless world. They take place somewhere, are grounded in and across specific territories, and are embedded in multiple and sometimes-​distant knowledge and resource bases, creating linkages and interdependencies across these places (Bathelt and Glückler, 2011).

Creating Linkages with Other Fields It is this recognition of the importance of a geographical lens in relational approaches that has drawn related social sciences to adopt a spatial perspective and develop an interest in economic geography research. With increasing globalization, this has become quite pronounced in recent years. Aside from fields such as cluster and global value-​chain research and innovation, global finance, and entrepreneurship studies, where close ties with economic geography have already existed for some time, numerous other research programmes have established closer links with economic geography through relational conceptualization. This has created new trans-​disciplinary research agendas.

International Business Studies Some exchange between the fields of international business and economic geography has always taken place. Economic geographers have, for instance, used findings from international business studies to analyse the spatial consequences of foreign direct investments in developing contexts (Phelps, 2007). And economic geography conceptions of global production networks (Dicken and Malmberg, 2001; Dicken et al., 2001) have developed from prior work on global commodity chains (Gereffi and Korzeniewicz, 1994). Despite these shared research interests, however, both fields have only slowly engaged in more substantive interdisciplinary exchanges. In the context of cluster research in economic geography, for instance, the emphasis on localized learning processes has been broadened to include fundamental interdependencies between local and global knowledge flows (Maskell and Malmberg, 1999; Bathelt et  al., 2004)—​a conceptualization that has drawn the attention of international business scholars. In globalization research, economic geography studies have focused on the role of trans-​regional networks and on the regional consequences of internationalization processes (Phelps, 2007), while international business analyses, following Ghoshal and Bartlett (1990) and Dunning (1993), have concentrated on the development of intra-​organizational networks at the cross-​national level (Cantwell and Mudambi, 2011). However, interactions between economic geography and international business studies have become more intensive in recent years, especially with the rise of the global

Relational Research Design in Economic Geography    187 knowledge economy and converging relational research designs (Beaverstock, 2004; Bathelt and Cohendet, 2014; Giuliani, 2017). This has generated opportunities for cross-​disciplinary debate (Lorenzen and Mudambi, 2013; Cohendet et al., 2014).

Industrial Marketing Similar convergence processes can be observed between the fields of industrial marketing and economic geography, for instance in research on trade fairs (Bathelt et al., 2014; Rinallo et  al., 2017). Industrial marketing studies traditionally portray trade fairs as events that enable producers to advertise and sell their products to buyers. In contrast, conventional economic geography research focuses on the spatial patterns of trade show activity and their regional economic impact. Both research streams have shown that linkages between trade fairs and their regional/​national economic environment become weaker over time. With the observation that the importance of transactions at these events has declined (Borghini et al., 2006), while their role as facilitators of knowledge generation processes has become crucial, knowledge-​based conceptualizations of trade fairs have developed, which have triggered cross-​fertilization between both fields. In economic geography, trade fairs have been conceptualized as ‘temporary clusters’ that enable producers to meet with actors from faraway and learn about trends in global industry and technology fields (Maskell et al., 2006). Through interaction and observations, participants can draw important conclusions for product development and initiate the development of networks with distant partners. In industrial marketing, a complementary discourse about ‘temporary markets’ has emerged, which explains how interaction patterns between sellers and buyers vary according to different trade fair types and how the knowledge strategies of trade fair organizers affect these events (Rinallo and Golfetto, 2011). The combination of both perspectives has led to a comprehensive knowledge-​based conceptualization of trade fairs (Bathelt et al., 2014) and triggered cross-​disciplinary research (Gibson and Bathelt, 2014; Li, 2014; Maskell, 2014). In particular, this conceptualization does justice to the variety and complexity of the relationships that develop between trade show specialization and territorial economic specialization. Rinallo et al. (2017) identify a number of ways in which economic geography and industrial marketing perspectives can engage in broader cross-​disciplinary research agendas.

Economic Sociology The 1990s and 2000s have also witnessed a convergence towards relational thinking (Fourcade, 2007)  in economic sociology research on network analysis, actor networks, and organizational fields. Emirbayer’s (1997) early manifesto for a relational sociology has become a central point of departure that has been echoed across the field in recent years (Pachucki and Breiger, 2010; Mische, 2011; Powell and Dépelteau, 2013). Proponents of relational sociology reject both holistic (macro) and individualistic (micro) thinking and rather emphasize the need to understand social life by studying social relations. A belief shared in these approaches is the ‘anti-​categorical imperative’ (Emirbayer and Goodwin, 1994), which rejects a substantialist understanding of social phenomena as monadic entities and instead conceives them as rooted in social relations. Relational thinking thus shifts the analytical

188   Bathelt and Glückler focus from attributes and categories to context, process, and emergence (Mutch et al., 2006). While conceptual debates, for example regarding the understanding of networks (Grabher, 2006), have sometimes developed independently in the two fields, the relational perspective has become a source of exchange and commonality between them. Both are interested in the emergence and effects of social structures, such as culture, power, institutions, trust, reputation, and social capital, on economic development in spatial perspective.

Organization and Network Studies Another point of cross-​fertilization is the intersection of the fields of organization studies and economic geography, which share a common interest in how new knowledge is created and mobilized (Glückler and Doreian, 2016). Scholars in organization studies have adopted network theories (Brass et al., 2004) to study processes of collective learning and innovation in intra-​and inter-​organizational relations (Powell et al., 1996; Uzzi, 1997; Zaheer et al., 2010). Network theories of creativity, learning, and innovation that conceptualize ‘structural holes’ (Burt, 1992), ‘structural wholes’ (McGrath and Krackhardt, 2003), or ‘structural folds’ (Vedres and Stark, 2010) emphasize that social outcomes are contingent on both the relational positions of actors in and the overall connectivity of a social network. While this line of research originally ignored geographical aspects or simply viewed the region as a container confining the focal agents’ actions and interactions, recent approaches have identified geography as a source of contingency for organizational change and social outcomes (Whittington et al., 2009; Glückler et al., 2017). The geographical dimension has been found as crucially moderating the association between network structure and innovation (Owen-​ Smith and Powell, 2004) or directly affecting the structure and dynamics of network formation (Powell et  al., 2005). Relational economic geography has responded by applying network theory to challenges, such as learning in project ecologies (Grabher, 2004), controversial innovation in organizational and geographical peripheries (Glückler, 2014), or localized knowledge spillovers (Breschi and Lissoni, 2009; Ter Wal and Boschma, 2009). A recent area of convergence between both fields is the attempt to understand the dynamic ‘rewiring’ of networks in spatial perspective (Panitz and Glückler, 2017) by theorizing multilevel networks across the spatial scales of local and global markets and across social scales, such as interpersonal and inter-​organizational relations, in permanent (Lazega, 2017) or temporary proximity (Brailly et al., 2016).

Political Economy Studies in the political economy literature on varieties of capitalism have shown how capitalist economies develop specific institutional arrangements that trigger distinct national economic development paths. Adopting a ‘relational perspective of the firm’, which focuses on the actions and interactions of economic agents within a national setting, the varieties of capitalism approach draws particular attention to ‘deliberative institutions’, which provide the basis for ongoing interaction and exchange (Hall and Soskice, 2001). It uses a microscale perspective to generate explanations at the macro-​level about disparities in capitalist development (Peck and Theodore, 2007) and thus offers manifold opportunities to link

Relational Research Design in Economic Geography    189 with relational perspectives in economic geography. While related research sometimes still remains committed to a macro-​perspective and has been characterized as deterministic (Crouch, 2005; Faulconbridge, 2008), the varieties of capitalism approach provides an important explanation of why economic agents may find it easier to interact with partners from the same institutional context. This work is therefore closely linked with investigations by innovation scholars and economic geographers on national systems of innovation (Lundvall, 2017). This interconnection is illustrated in the work of Gibson (2018), who uses a bottom-​up perspective similar to Hall and Thelen (2009) to combine political economy and economic geography research questions. Gibson shows that, rather than promoting convergence of industrial structures between national varieties of capitalism, global exchange processes during trade fairs instead support ongoing specialization and sustained differentiation in technology use and adaptation between them. To sum up, researchers from various fields across the social sciences increasingly share relational perspectives in developing theory. They agree (i)  that social relations between people and organizations are key to understanding the contemporary economy, (ii) that economic processes rest on the spatial and temporal interplay between regional and global networks, and (iii) that innovation and learning depend on simultaneous inter-​firm, intra-​ organizational, and community-​ based interactions and relations (Bathelt and Glückler, 2011).

Conclusion: Communalities of a Relational Research Design The selection of a research design is a fundamental element of empirically grounded, reliable, and ‘useful’ theory. The purpose of this selection is to lay out the conceptual foundations for an empirical research programme that allows research questions to be answered as clearly as possible. Research design deals with logical rather than logistical problems (De Vaus, 2001), and demands consistency between the research questions, theoretical categories and arguments, operational measures, and empirical observations for the sake of both internal and external validity. In this chapter we have shown that a relational research design embraces the richness of context and the indetermination of change and that it participates in the increasing conversation among related fields about the space economy. The analysis presented suggests that there is no single-​best relational approach. Rather, different modes of enquiry suit different disciplinary foci and research contexts. These relational approaches enable cross-​fertilization between disciplinary debates based on a number of communalities:

(i) A  relational research design assumes that it is necessary to study social relations between actors to understand social and economic outcomes. Accordingly, society and economy do not ‘have’ relations per se, but are constituted through them. (ii) Consequently, a relational design is contextual in nature and therefore sensitive to a rich set of conditions and meanings embedded in time-​space-​situated phenomena.

190   Bathelt and Glückler (iii) A  relational research design presumes outcomes as necessarily contingent and thus prioritizes case-​sensitive empirical analysis over the search for universal laws. Because variation is endogenous to context the outcomes of situated practice and specific relations need to be conceived as open-​ended and undetermined. (iv) A relational research design is interested in analysing process rather than causality and traces the way in which interactions and relational conditions lead to certain outcomes over time. As process is conceived as undetermined, social and economic development is theorized as path-​dependent in historical perspective and contingent with respect to the future. A relational design therefore prefers evolutionary concepts of change over deterministic or life-​cycle frameworks. (v) A relational research design relies on a spatial perspective, as a geographical lens (Bathelt and Glückler, 2003), through which research questions are framed and the economic process is traced across places and spaces and between scales. Geography is neither reified as an ontological identity or a container of characteristics, nor is physical distance used as an explanatory variable of social interaction and economic outcomes. Instead, a geographical lens leads to research questions about the density, distance, diversity, and disparity of social and economic phenomena. (vi) Finally, a relational research design is non-​idiosyncratic in that it aims to generate abstract and transferable findings. Although relational theories embrace context and are non-​universal, they must involve conceptual abstraction if the concepts, mechanisms, and processes they employ are to be applicable across multiple contexts. It is therefore important to distinguish between the necessary and contingent conditions of a particular context (Sayer, 2000) in order to increase the external validity of emergent theorization. Relational theories can thus be characterized as ‘theories of the middle range’ (Merton, 1949) that are close enough to the empirical case to ensure richness and internal validity (authenticity), and, at the same time, abstract enough to develop transferrable theories (structuration) that account for variation, context, and contingency, while resisting universal, scale-​free, or deterministic expectations.

Acknowledgements We wish to thank Gordon Clark for his encouragement, and Daniel Hutton Ferris for editorial support.

References Allen, J. (2003). Lost Geographies of Power (Malden, MA, and Oxford: Blackwell). Amin, A. and Cohendet, P. (2004). Architectures of Knowledge:  Firms, Capabilities, and Communities (Oxford and New York: Oxford University Press). Bathelt, H. and Cohendet, P. (2014). ‘The creation of knowledge: local building, global accessing and economic development—​ toward an agenda’. Journal of Economic Geography 14: 869–​882.

Relational Research Design in Economic Geography    191 Bathelt, H. and Glückler, J. (2003). ‘Toward a relational economic geography’. Journal of Economic Geography 3: 117–​144. Bathelt, H. and Glückler, J. (2011). The Relational Economy:  Geographies of Knowing and Learning (Oxford: Oxford University Press). Bathelt, H., Golfetto, F., and Rinallo, D. (2014). Trade Shows in the Globalizing Knowledge Economy (Oxford: Oxford University Press). Bathelt, H. and Henn, S. (2014). ‘The geographies of knowledge transfers over distance: toward a typology’. Environment and Planning A 46: 1403–​1424. Bathelt, H., Malmberg, A., and Maskell, P. (2004). ‘Clusters and knowledge:  local buzz, global pipelines and the process of knowledge creation’. Progress in Human Geography 28: 31–​56. Beaverstock, J.V. (2004). ‘Managing across borders: knowledge management and expatriation in professional service legal firms’. Journal of Economic Geography 4: 157–​179. Borghini, S., Golfetto, F., and Rinallo, D. (2006). ‘Ongoing search among industrial buyers’. Journal of Business Research 59: 1151–​1159. Brailly, J., Favre, G., Chatellet, J., and Lazega, E. (2016). ‘Embeddedness as a multilevel problem: a case study in economic sociology’. Social Networks 44: 319–​333. Brass, D.J., Galaskiewicz, J., Greve, H.R., and Tsai, W. (2004). ‘Taking stock of networks and organizations: a multilevel perspective’. Academy of Management Journal 47: 795–​817. Breschi, S. and Lissoni, F. (2009). ‘Mobility of skilled workers and co-​invention networks: an anatomy of localized knowledge flows’. Journal of Economic Geography 9: 439–​468. Burt, R.S. (1992). Structural Holes:  The Social Structure of Competition (Cambridge, MA: Harvard University Press). Cantwell, J. and Mudambi, R. (2011). ‘Physical attraction and the geography of knowledge sourcing in multinational enterprises’. Global Strategy Journal 1: 206–​232. Clark, G.L. and Tracey, P. (2004). Global Competitiveness and Innovation: An Agent-​Centred Perspective (Basingstoke: Palgrave Macmillan). Cohendet, P., Grandadam, D., Simon, L., and Capdevila, I. (2014). ‘Epistemic communities, localization and the dynamics of knowledge creation’. Journal of Economic Geography 14: 929–​954. Crevoisier, O. and Maillat, D. (1991). ‘Milieu, Industrial Organization and Territorial Production System:  Towards a New Theory of Spatial Development’ in R. Camagni (ed.) Innovation Networks: Spatial Perspectives, pp. 13–​34 (London and New York: Belhaven Press). Crouch, C. (2005). Capitalist Diversity and Change: Recombinant Governance and Institutional Entrepreneurs (Oxford: Oxford University Press). De Vaus, D. (2001). Research Design in Social Research (London: SAGE). Delgado, M., Porter, M.E., and Stern, S. (2016). ‘Defining clusters of related industries’. Journal of Economic Geography 16: 1–​38. Dicken, P. and Malmberg, A. (2001). ‘Firms in territories: a relational perspective’. Economic Geography 77: 345–​363. Dicken, P., Kelly, P.F., Olds, K., and Yeung, H.W.-​c. (2001). ‘Chains and networks, territories and scales: towards a relational framework for analysing the global economy’. Global Networks 1: 89–​112. Dunning, J.H. (1993). Multinational Enterprises and the Global Economy (Reading: Addison-​Wesley). Emirbayer, M. (1997). ‘Manifesto for a relational sociology’. American Journal of Sociology 103: 281–​317. Emirbayer, M. and Goodwin, J. (1994). ‘Network analysis, culture, and the problem of agency’. American Journal of Sociology 99: 1411–​1454.

192   Bathelt and Glückler Ettlinger, N. (2003). ‘Cultural economic geography and a relational and microspace approach to trusts, rationalities, networks, and change in collaborative workplaces’. Journal of Economic Geography 3: 145–​172. Faulconbridge, J. (2008). ‘Managing the transnational law firm: a relational analysis of professional systems, embedded actors, and time-​ space-​ sensitive governance’. Economic Geography 84: 185–​210. Faulconbridge, J. (2017). ‘Relational Geographies of Knowledge and Innovation’ in H. Bathelt, P. Cohendet, S. Henn, and L. Simon (eds) The Elgar Companion to Innovation and Knowledge Creation, pp. 671–684 (Cheltenham and Northampton, MA: Edward Elgar). Fourcade, M. (2007). ‘Theories of markets and theories of society’. American Behavioral Scientist 50: 1015–​1034. Gereffi, G. and Korzeniewicz, M. (eds) (1994). Commodity Chains and Global Capitalism (Wesport, CT: Praeger). Gertler, M.S. (2004). Manufacturing Culture: The Institutional Geography of Industrial Practice (Oxford and New York: Oxford University Press). Ghoshal, S. and Bartlett, C.A. (1990). ‘The multinational corporation as an interorganizational network’. Academy of Management Review 15: 603–​625. Gibson, R. (2018). ‘Dynamic capitalisms? Understanding national specialization patterns through inter-​firm interaction at international trade fairs’ (PhD thesis: University of Toronto). Gibson, R. and Bathelt, H. (2014). ‘Field configuration or field reproduction? The dynamics of global trade fair cycles’. Zeitschrift für Wirtschaftsgeographie 58: 216–​231. Giuliani, E. (2017). ‘Industrial Clusters in Global Networks’ in H. Bathelt, P. Cohendet, S. Henn, and L. Simon (eds) The Elgar Companion to Innovation and Knowledge Creation, pp. 360–371 (Cheltenham and Northampton, MA: Edward Elgar). Glückler, J. (2007). ‘Geography of reputation: the city as the locus of business opportunity’. Regional Studies 41: 949–​962. Glückler, J. (2013). ‘Knowledge, networks and space: connectivity and the problem of non-​ interactive learning’. Regional Studies 47: 880–​894. Glückler, J. (2014). ‘How controversial innovation succeeds in the periphery? A network perspective of BASF Argentina’. Journal of Economic Geography 14: 903–​927. Glückler, J. and Bathelt, H. (2017). ‘Institutional Context and Innovation’ in H. Bathelt, P. Cohendet, S. Henn, and L. Simon (eds) The Elgar Companion to Innovation and Knowledge Creation, pp. 121–137 (Cheltenham and Northampton, MA: Edward Elgar). Glückler, J. and Doreian, P. (2016). ‘Editorial:  social network analysis and economic geography—​ positional, evolutionary and multi-​ level approaches’. Journal of Economic Geography 16: 1123–​1134. Glückler, J. and Hammer, I. (2011). ‘A pragmatic service typology: capturing the distinctive dynamics of services in time and space’. Service Industries Journal 31: 941–​957. Glückler, J., Lazega, E., and Hammer, I. (eds) (2017). Knowledge and Networks. Knowledge and Space Series, Vol. 11 (Berlin: Springer). Gordon, I.R. and McCann, P. (2000). ‘Industrial clusters: complexes, agglomeration and/​or social networks?’. Urban Studies 37: 513–​532. Grabher, G. (1993). ‘The Weakness of Strong Ties: The Lock-​in of Regional Development in the Ruhr Area’ in G. Grabher (ed.) The Embedded Firm: On the Socioeconomics of Industrial Networks, pp. 255–​277 (London: Routledge). Grabher, G. (2004). ‘Temporary architectures of learning: knowledge governance in project ecologies’. Organization Studies 25: 1491–​1514.

Relational Research Design in Economic Geography    193 Grabher, G. (2006). ‘Trading routes, bypasses, and risky intersections:  mapping the travels of “networks” between economic sociology and economic geography’. Progress in Human Geography 30: 163–​189. Hall, P.A. and Soskice, D. (2001). ‘An Introduction to Varieties of Capitalism’ in P.A. Hall and D. Soskice (eds) Varieties of Capitalism: The Institutional Foundations of Comparative Advantage, pp. 1–​68 (Oxford and New York: Oxford University Press). Hall, P.A. and Thelen, K. (2009). ‘Institutional change in varieties of capitalism’. Socio-​Economic Review 7: 7–​34. Hayter, R. (2004). ‘Economic geography as dissenting institutionalism:  the embeddedness, evolution and differentiation of regions’. Geografiska Annaler 86 B: 95–​115. Hudson, R. (2004). ‘Conceptualizing economies and their geographies: spaces, flows and circuits’. Progress in Human Geography 28: 447–​471. Johannisson, B. (1990). ‘Economies of overview: guiding the external growth of small firms’. International Small Business Journal 9: 32–​44. Jones, A. and Murphy, J.T. (2011). ‘Theorizing practice in economic geography: foundations, challenges and possibilities’. Progress in Human Geography 35: 366–​392. Keeble, D. and Nachum, L. (2002). ‘Why do business service firms cluster? Small consultancies, clustering and decentralization in London and Southern England’. Transactions of the Institute of British Geographers 27: 67–​90. Lagendijk, A. (2017). ‘Innovation, Regional Development and Relationality’ in H. Bathelt, P. Cohendet, S. Henn, and L. Simon (eds) The Elgar Companion to Innovation and Knowledge Creation, pp. 490–505 (Cheltenham and Northampton, MA: Edward Elgar). Lazega, E. (2017). ‘Organized Mobility and Relational Turnover as Context for Social Mechanisms: A Dynamic Invariant at the Heart of Stability from Movement’ in J. Glückler, E. Lazega, and I. Hammer (eds) Knowledge and Networks, pp. 119–​142 (Berlin: Springer). Li, P.-​F. (2014). ‘Global temporary networks of clusters: structures and dynamics of trade fairs in Asian economies’. Journal of Economic Geography 14: 995–​1021. Li, P.-​F. (2017). ‘Horizontal Learning’ in H. Bathelt, P. Cohendet, S. Henn, and L. Simon (eds) The Elgar Companion to Innovation and Knowledge Creation, pp. 392–404 (Cheltenham and Northampton, MA: Edward Elgar). Lorenzen, M. and Mudambi, R. (2013). ‘Clusters, connectivity and catch-​up: Bollywood and Bangalore in the global economy’. Journal of Economic Geography 13: 501–​534. Lundvall, B.-​ Å. (2017). ‘National Innovation Systems and Globalization’ in H. Bathelt, P. Cohendet, S. Henn, and L. Simon (eds) The Elgar Companion to Innovation and Knowledge Creation, pp. 472–489 (Cheltenham and Northampton, MA: Edward Elgar). Lundvall, B.-​Å. and Johnson, B. (1994). ‘The learning economy’. Journal of Industry Studies 1: 23–​42. McGrath, C. and Krackhardt, D. (2003). ‘Network conditions for organizational change’. Journal of Applied Behavioral Science 39: 324–​336. Malmberg, A. and Maskell, P. (2002). ‘The elusive concept of localization economies: towards a knowledge-​based theory of spatial clustering’. Environment and Planning A 34: 429–​449. Maskell, P. (2014). ‘Accessing remote knowledge—​the roles of trade fairs, pipelines, crowdsourcing and listening posts’. Journal of Economic Geography 14: 883–​902. Maskell, P., Bathelt, H., and Malmberg, A. (2006). ‘Building global knowledge pipelines: the role of temporary clusters’. European Planning Studies 14: 997–​1013. Maskell, P. and Malmberg, A. (1999). ‘Localised learning and industrial competitiveness’. Cambridge Journal of Economics 23: 167–​185. Merton, R.K. (1949). Social Theory and Social Structure (New York: Simon and Schuster, Free Press).

194   Bathelt and Glückler Mische, A. (2011). ‘Relational Sociology, Culture, and Agency’ in J. Scott and P.J. Carrington (eds) The SAGE Handbook of Social Network Analysis, pp. 80–​99 (London: SAGE). Moulaert, F. and Djellal, F. (1995). ‘Information technology consultancy firms: economics of agglomeration from a wide-​area perspective’. Urban Studies 32: 105–​122. Mutch, A., Delbridge, R., and Ventresca, M. (2006). ‘Situating organizational action: the relational sociology of organizations’. Organization 13: 607–​625. Owen-​Smith, J. and Powell, W.W. (2004). ‘Knowledge networks as channels and conduits: the effects of spillovers in the Boston biotechnology community’. Organization Science 15: 5–​21. Pachucki, M.A. and Breiger, R.L. (2010). ‘Cultural holes: beyond relationality in social networks and culture’. Annual Review of Sociology 36: 205–​224. Panitz, R. and Glückler, J. (2017). ‘Rewiring global networks in local events: congresses in the stock photo trade’. Global Networks 17: 147–​168. Peck, J. and Theodore, N. (2007). ‘Variegated capitalism’. Progress in Human Geography 31: 731–​772. Phelps, N.A. (2007). ‘Foreign direct investment and the regional economy’. Growth and Change 38: 738–​739. Pinch, S., Henry, N., Jenkins, M., and Tallmann, S. (2003). ‘From “industrial districts” to “knowledge clusters”: a model of knowledge dissemination and competitive advantage in industrial agglomerations’. Journal of Economic Geography 3: 373–​388. Porter, M.E. (1990). The Competitive Advantage of Nations (New York: Free Press). Porter, M.E. (2000). ‘Locations, Clusters, and Company Strategy’ in G.L. Clark, M.P. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 253–​274 (Oxford: Oxford University Press). Powell, C. and Dépelteau, F. (eds) (2013). Relational Sociology: Ontological and Theoretical Issues (New York: Palgrave Macmillan). Powell, W.W., Koput, K.W., and Smith-​Doerr, L. (1996). ‘Interorganizational collaboration and the locus of innovation: networks of learning in biotechnology’. Administrative Science Quarterly 41: 116–​145. Powell, W.W., White, D., Koput, K.W., and Owen-​Smith, J. (2005). ‘Network dynamics and field evolution: the growth of interorganizational collaboration in the life sciences’. American Journal of Sociology 110: 1132–​1205. Rinallo, D., Bathelt, H., and Golfetto, F. (2017). ‘Economic geography and industrial marketing views on trade shows: collective marketing and knowledge circulation’. Industrial Marketing Management 61: 93–103. Rinallo, D. and Golfetto, F. (2011). ‘Exploring the knowledge strategies of temporary cluster organizers:  a longitudinal study of the EU fabric industry trade shows (1986–​2006)’. Economic Geography 87: 453–​476. Sayer, A. (1992). Method in Social Science (London: Routledge). Sayer, A. (2000). Realism and Social Science (London: Sage). Scott, A.J. (1988). New Industrial Spaces: Flexible Production Organization and Regional Development in North America and Western Europe (London: Pion). Stabell, C. and Fjeldstad, Ø. (1998). ‘Configuring value for competitive advantage: on chains, shops, and networks’. Strategic Management Journal 19: 413–​437. Storper, M. and Venables, A.J. (2004). ‘Buzz:  face-​to-​face contact and the urban economy’. Journal of Economic Geography 4: 351–​370. Ter Wal, A.L.J. and Boschma, R.A. (2009). ‘Applying social network analysis in economic geography: framing some key analytic issues’. The Annals of Regional Science 43: 739–​756.

Relational Research Design in Economic Geography    195 Uzzi, B. (1997). ‘Social structure and competition in interfirm networks: the paradox of embeddedness’. Administrative Science Quarterly 42: 35–​67. Vedres, B. and Stark, D. (2010). ‘Structural folds: generative disruption in overlapping groups’. American Journal of Sociology 115: 1150–​1190. von Hippel, E. (2001). ‘Innovation by user communities: learning from open-​source software’. MIT Sloan Management Review 42: 82–​86. Whittington, K.B., Owen-​Smith, J., and Powell, W.W. (2009). ‘Networks, propinquity, and innovation in knowledge-​intensive industries’. Administrative Science Quarterly 54: 90–​122. Yeung, H.W.-​ c. (2005). ‘Rethinking relational economic geography’. Transactions of the Institute of British Geographers 30: 37–​51. Yeung, H.W.-​c. and Coe, N.M. (2015). ‘Toward a dynamic theory of global production networks’. Economic Geography 91: 29–​58. Zaheer, A., Gözübüyük, R., and Milanov, H. (2010). ‘It’s the connections: the network perspective in interorganizational research’. Academy of Management Perspectives 24: 62–​77.

Chapter 10

B ehaviour i n C ont e xt Gordon L. Clark Introduction Human behaviour is fundamental to our understanding of economy and society. What people do, where they do it, why they do it, and the costs and benefits of behaviour, individually and collectively, are key topics of research. It could hardly be otherwise. Being self-​ conscious, responsive to the environment, and having regard for the consequences of our actions in relation to others’ actions and interests, human behaviour is both subject and object. There are various ways of explaining human behaviour, some of which objectify behaviour as in the rational agent of conventional economic theory (Becker, 1962). Other explanations emphasize the diversity of human experience, giving priority to the ‘context’ or ‘environment’ in which action takes place (Pettit and McDowell, 1986). As such, context stands for the cultural, economic, and social formations that frame but do not necessarily determine behaviour. There are reasons to be cautious of the reach of agent-​centred models of behaviour. It is easy enough to begin with the individual, attribute preferences, what they hope to achieve, and then explain how they go about realizing goals and objectives. This is the standard mode of reasoning in much of microeconomics (Almenberg and Dreber, 2013). There are, however, limits to what people can achieve on their own account. These limits are to be found in their cognitive capabilities, learning and reflective capacities, and material resources. Where they recognize the constraints on their prospects, people may seek to transcend these limitations in various ways. For example, collecting better information and knowledge may give people a deeper understanding of their options and prospects, thereby enabling them to rethink their future plans. Still, the context in which people live can impose hard constraints on human prospects. Indeed, how they think about themselves and others in context can systematically frame behaviour and expectations. Not surprisingly, researchers often focus upon the formal and informal mechanisms that frame people’s options. Social science researchers often acknowledge that their own frames of reference can also be important in characterizing observed behaviour—​our position, our goals, and objectives, and the communities of scholarship in which we work are all significant in this regard. Over the years, sharp lines have sometimes been drawn between

Behaviour in Context    197 economists and geographers in relation to the significance to be attributed to ‘context’ when seeking to explain observed behaviour. Tools such as indifference curves, utility functions, and systematic models of behaviour have enabled economists to cut through the clutter of everyday life to impose order on observed behaviour. If universal in its ambition, critics contend that this strategy fails to do justice to the specific circumstances (history and geography) of behaviour (Clark, 1998). It can also give rise to misleading recipes for public policy (see Easterly’s critique of the Washington consensus (Easterly, 2014)). Economic geography tends to treat context as the centrepiece of explanation. Here, there is much less about the ‘rational agent’ and much more about how circumstances and socio-​ demographic factors such as gender, race, and income interact to shape patterns of behaviour. For example, Clark and Whiteman’s (1983) model of US urban labour market behaviour demonstrates that a central-​city location combined with limited resources systematically shapes job-​search behaviour so as to reinforce spatial patterns of income inequality and unemployment. In a way, this type of model foreshadowed more recent attempts to explain the consequences of co-​location for information acquisition and assessment (see Storper and Venables, 2004). At the limit, this approach to behaviour in context subsumes behaviour by context such that individuals’ goals and aspirations mirror context, leaving ciphers rather than active agents. In effect, the objects of research are meant to stand for the richness or otherwise of specific kinds of situations. In this chapter, I review the relevant literature on behaviour in context with due regard to recent developments in both economics and geography. In doing so it is assumed that people are generally rational. This presumption is grounded in human evolution and related research (see Hurley and Nudds, 2006). But, as indicated in the next section, the definition and attributes of rationality individually and collectively are subject to debate. In any event, just because we are rational does not mean to say that we are always effective decision-​ makers. Here, I consider the implications of recent research on individual decision-​making in time and space. Throughout, a distinction is made between rationality, an attribute of human beings, and people’s inherited and learnt decision-​making competence given the nature and scope of the problems that people must deal with in their everyday lives. Whereas much of the literature focuses upon decision-​making in time, I  suggest that decision-​making in space is more challenging and, perhaps, more important than decision-​ making in time.1 There is little doubt that the behavioural turn in economics was given legitimacy and momentum by the path-​breaking research of economists and psychologists such as Herbert Simon, Daniel Kahneman, and Amos Tversky. Many others have been involved in the research programme, which has extended over more than fifty years. To the extent that this research programme bears upon risk and uncertainty, the work of Keynes (1921) and Knight (1921) can be seen as the foundations for the modern treatment of the meaning and significance of time. In geography, behaviouralism has been an ever-​present thread found, for example, in the early work of Hagerstrand (1967), Webber (1972), and Wolpert (1980). Indeed, Golledge et al.’s (1972) research programme on behaviour is representative of an important stream of research on spatial cognition shared by geographers and psychologists alike. In economics and geography, behaviouralism now makes important claims for recognition (Strauss, 2008, 2009), reinforced by the utility of behaviouralism for understanding behaviour in global financial markets (Clark, 2011).

198   Clark

Rationality A core assumption underpinning cognitive science and psychology is that human beings are rational; that is, humans have an instinct for self-​preservation. Drawing on evolutionary theory, cognitive scientists believe that this instinct is characteristic of human beings whatever the context or environment in which they are situated. Taking the argument one step further, it is conceivable that this shared trait has come to dominate the human species because it is consistent with species’ persistence and reproduction; that is, it has come to dominate via the process of natural selection.2 By definition, this is a statement about the world as observed rather than a claim made about how people ought to behave or should behave (a normative standard) (see Pliske and Klein, 2003). Many social scientists categorize behaviour as more or less rational as if there is a continuum or gradation in levels of rationality. Here, there appear to be two basic approaches. One approach is concerned about the means by which a priori specified goals and objectives are realized. Here, the issue is whether the means chosen by which to achieve those goals and objectives are consistent with individuals’ interests—​more precisely, whether they are the best available means such that the optimal outcome, or set of outcomes, is realized. So, at this level, means and ends can be expressed through an objective function (ends), a set of instruments or mechanisms by which those ends are achieved (means), and a set of constraints or limits on realizing those ends. If we assume, as many economists do, that people maximize their utility, then rational agents are those who choose the best possible combination of means to realize desired ends. Using a normative test of rationality, observers of human behaviour can evaluate whether the means chosen to achieve desired ends are more or less effective, given the available options. So, for example, given a choice of means in relation to a priori specified ends, observers should be able to determine whether a chosen instrument, or set of instruments, is superior to another instrument or sets of instruments in realizing those ends. Of course, it could be that people have insufficient information to assess adequately the relative performance of various instruments. As a consequence, they may be led to certain actions that, upon receipt of better information about the relative performance of other instruments, prompt them to abandon old instruments in favour of better ways of realizing their goals and objectives. A significant body of research in economics, and to some extent in economic geography, has been concerned with the causes and consequences of variable information for human behaviour and social welfare (see Stiglitz, 2000). Another test of the rationality of behaviour is whether chosen goals and objectives are consistent with individuals’ self-​interest: whether they enhance welfare as opposed to being self-​defeating, and are either neutral or positive with respect to the opportunities of others to similarly pursue their interests. Inconsistency of behaviour is sometimes interpreted as arbitrary decision-​making and, ultimately, self-​defeating behaviour (Thaler and Sunstein, 2008). But, of course, being consistent in the face of changing circumstances can also be self-​ defeating. Harming or, worse, exploiting others could be advantageous in the short term but self-​defeating in the long term. So, rationality also involves self-​reflection, learning, and adaptation. In a well-​functioning society, arbitrary and exploitive behaviours are likely to violate social norms and conventions associated with cooperation and collaboration—​both of which may be necessary if individuals are to realize their separate but interrelated goals. In this context, cooperation may be a precondition for individual action, not the product of behaviour.

Behaviour in Context    199 Just as we can evaluate individual behaviour against tests of internal rationality, we can also evaluate individual behaviour in relation to others against tests of external rationality. Whether these tests of rationality are plausible depends upon the assumptions we make about the context in which behaviour takes place. It may be difficult to realize desired goals and objectives if information relevant to human decision-​making is unevenly distributed, has scarcity value, and is costly to verify. In some situations, information relevant to decision-​making may have such value that it becomes a pawn in games of self-​seeking and competitive behaviour. There may be situations in which individuals can only realize their objectives at the expense of others. In these circumstances, governments can play a significant role in regulating the context of behaviour such that the rules that define legitimate behaviour are consistent with individuals realizing their separate but related interests. By this account, context could matter a great deal—​witness recent research on the institutional preconditions for economic growth and development (Geif, 2006). Herbert Simon (1956) suggested that observed behaviour is produced at the intersection between cognition and context. By his assessment, neither is sufficient as an explanation of behaviour. His scissors metaphor stands for an enormous body of research in psychology and related disciplines on the processing of signals from the environment. Simon made other substantial contributions to behaviouralism. He and his colleagues at Carnegie Mellon showed that there are physiological limits to humans’ capacity to process signals from the environment. Referencing the nascent science of computer processing, he suggested that there are limits to humans’ capacity to store information, retrieve past data, and process new information against relevant benchmarks.3 Playing off of economists’ commitment to a strong version of human rationality, Simon suggested it be replaced with a more realistic conception of ‘bounded rationality’. Given the limits of human cognition and the importance of context, he argued that utility maximization was implausible; people make decisions as best they can, oftentimes ‘satisficing’ rather than maximizing utility. Unfortunately, Simon’s argument has been misunderstood. For some analysts, bounded rationality is treated as a special case and the conventional standard of rationality holds sway. For other analysts, satisficing has been treated as something less than ideal—​an unwitting compromise that can be corrected given sufficient time and effort. Simon was at pains not to discount the rational intentions of human agents. By his account, the outcomes of decision-​ making are subject to all manner of complications, some of which reflect limited cognitive capacity in the context of risk and uncertainty. For other analysts, the idea that humans adapt to acknowledged cognitive limits and the options and resources of their immediate environment would seem to discount the possibility of a universal model of human nature. Just around the corner, in fact, is a form of cognitive-​cum-​environmental determinism which would return economic geography to a bygone era. It would also deny the importance of human aspirations, the context, and the possibility of transcending context.

Coping with Time Notwithstanding the significance of Herbert Simon, the behavioural revolution in economics was driven by Daniel Kahneman and Amos Tversky (1979). Much has been written about their path-​breaking research on individual decision-​making and the systematic

200   Clark anomalies and biases identified thereof (see Baron, 2008). Like many cognitive psychologists, Kahneman and Tversky did not dispute whether humans are ipso facto rational, but focused upon the competency of individual decision-​making in the context of risk and uncertainty (time). They, as well as others following in their footsteps, have shown that many people heavily discount the future, are loss averse, and are subject to status quo bias. They also showed that how problems are framed tends to affect how people solve them; that is, instead of looking through problems to their underlying properties, people are seduced by their perception of problems in relation to other problems (see Clark et al., 2006, 2007). Kahneman and Tversky set off a far-​reaching search in cognitive science and behavioural psychology for other anomalies and biases. Recent surveys of research in the area have identified over sixty such traits (see Krueger and Funder, 2004). It is not my intention here to identify and explain the nature and scope of these anomalies and biases. Rather, I focus upon two specific attributes of human experience: all people, whatever their cultures or socio-​ demographic characteristics, act in time and space. Inevitably, individual decision-​making is located at a particular time, with regard to past experience and expectations about the future, and in a particular place, with regard to local circumstances and other possible action spaces. Being located at a particular time and in a specific place provides individuals with an experiential legacy, as well as a set of possible options and resources (Shafir et al., 1997). In this section, I deal with time and in the next section, I deal with space. This separation is necessarily arbitrary for the purpose of exposition. For many people, everyday life here and now dominates decision-​making. The actions they take in relation to the choices available are well defined in that there are few salient options. In any event, most choices are marginal to people’s established patterns of living. In these circumstances it is easy enough to predict with a high degree of confidence the costs and benefits (consequences) of taking one action relative to other options. At one level, it is arguable that the degree of certainty varies by social position, stage in life, and institutional setting (Sharpe, 2007). At another level, it is arguable that as most people are loss averse, the choices they make over a sequence of related courses of action tend to place a premium on certainty over uncertainty. That is, current decisions integrate selected valued attributes of the past with the desired attributes of the immediate future. Even so, there are some people who have a preference for more rather than less risk; a minority are willing to gamble in ways that put at risk their future well-​being (Clark et al., 2009). When facing similar or analogous choices, people tend to formulate ad hoc rules that come to stand in place of an analysis of each and every situation. These rules can be formally constituted as in if X and Y then decision Z but are more likely heuristics, which are rough-​ and-​ready approximations of a known world (Gigerenzer et al., 1999). If we were to utilize an optimizing framework to explain individuals’ decision-​making processes, the adoption of formal as opposed to informal decision rules could be explained by the likely costs and benefits of each calibrated against past experience with respect to expectations of the future. However, cognitive psychology disputes the realism of such a formulation. Assuming they trust intuition over reflection, people are likely to adopt rough-​and-​ready decision rules without regard to the full range of past experience and an assessment of current and expected circumstances (Kahneman, 2003). People tend to recall past experience in a selective manner; having made a decision to act in a certain way people look back for evidence legitimating such a decision. They allocate scarce cognitive resources to more difficult issues.

Behaviour in Context    201 There are, of course, more challenging situations. For example, financial markets are subject to risk and uncertainty. In normal circumstances, risk can be approximated by a probability distribution (although the choice of the probability distribution may be a significant issue in its own right). By contrast, uncertainty refers to a class of events that are either unknown or, if known, for which their likelihood of occurrence is unknown (Keynes, 1921; Knight, 1921). Whereas financial markets may experience episodes of relative stability where risk metrics provide a reasonable basis for investment decision-​making, market stability may be disturbed by unanticipated events or shifts in behaviour that undercut the utility of past practices. As recent evidence has shown, these types of events or behaviour can come from outside of financial markets or be an endogenous property of financial markets. One response to unanticipated market events is to assess whether and to what extent investors should adapt risk-​related decision-​making protocols to accommodate these circumstances. An immediate response runs the risk of overreacting to an event, which is shown later to be mere ‘noise’ in the context of the known range of market events (Stambaugh, 2014). Equally, an immediate response could be an appropriate reaction to an event, which is shown later to be indicative of a shift in the underlying distribution of market events. There is a risk of under​reacting, as well as overreacting, to market signals. The standard way of dealing with this issue is to assume market agents are Bayesian theorists; that is, they continuously evaluate market signals against the inherited distribution of market events and update their expectations accordingly. There is no doubt that some, perhaps a large number of, investment professionals in financial markets use Bayes’s theorem in this manner. However, many people do not revise their expectations (financial or otherwise) in a Bayesian manner, being preoccupied with new information (events) as opposed to its relative significance (Jones and Love, 2011). While people can be taught Bayesian reasoning, it appears that they are not intuitively Bayesian (Gigerenzer et al., 1999). Our research suggests that how people cope with risk and uncertainty (time) depends upon the circumstances in which these issues are pertinent and the domain-​specific skills and expertise of those that must operate in these circumstances. So, for example, being knowledgeable about probability, having experience in applying these statistical tools to a range of relevant events, and having calibrated and re-​calibrated models of market movements can provide these types of decision-​makers with the skills and expertise needed to cope with the unexpected. This does not mean that they are always successful in coping with risk and uncertainty. But it does mean that, over the long term, these types of decision-​ makers are more effective in coping with risk and uncertainty than neophytes (Clark et al., 2012). Going on recent experience, financial markets can be considered ‘extreme’ environments, quite foreign to the everyday circumstances of most people. However, as national governments have discounted the expected value of state pensions and have sought to encourage people to save for retirement either directly or indirectly through financial instruments, many millions of people have been drawn into the problem of managing time. This involves managing short-​term decisions in relation to long-​term interests, adapting expectations and behaviour to shifts in financial markets and performance, and judging recent events in relation to the past and future where knowing whether a recent event is ‘singular’ can only be known in the future.

202   Clark

Home Bias As indicated in the previous section, much of the research spawned by the behavioural revolution has treated human behaviour in the context of risk and uncertainty as an expression of how we cope with time. Research on how we discount the future has brought to light the fact that while many people heavily discount the future, others value highly both the immediate and distant future but discount the medium term (Ainslie, 2001; Laibson, 2003). Just as importantly, it can be shown that some people value the immediate future but are unable to provide a coherent valuation scheme (discount function) beyond it (Clark et al., 2006). People are temporally myopic. They are also spatially myopic in that they tend to value opportunities close at hand and discount opportunities further away. Distance decay functions are expressions of this apparent fact of everyday life. In much of the behavioural finance literature, this phenomenon is summarized as home bias: an apparent local bias in financial decision-​making, despite better opportunities for profit or benefit located further away. For example, Hong et al. (2004, 2005) show that the structure and composition of mutual fund investment products designed and managed in Boston and offered to US retail investors have evident similarities consistent with a local or neighbourhood effect (taking into account asset class and investment style). Huberman (2001) shows that financial institutions’ investment strategies tend to be biased in favour of their jurisdictions of origin, notwithstanding the risk-​related benefits of geographical diversification. Early research in economic geography on the diffusion of innovation assumed that the rate of adoption of an innovation from a point of origin is best characterized as a distance decay process. In a similar vein, tendencies of co-​location have been associated with preferential access to richer sources of information associated with spatial agglomeration (Storper and Venables, 2004). The standard way of explaining home bias is to invoke the costs associated with searching for, and validating, information relevant to individual decision-​making. If we assume that economic agents maximize profit (and minimize costs), if we assume that the per unit costs of information acquisition and validation increase with increasing distance from the site of the decision-​maker, and if we assume that signalling costs on the buy side and sell side of the market also increase from the site of the decision-​maker then home bias can be seen as a response to the spatial configuration of the market. Arrow (1984, p. 171) claimed that learning is always more effective if done locally: ‘an explorer in hitherto unknown territory will find it easier to explore new areas near to those he has already covered. Geographical propinquity is but a special case.’ Behavioural approaches provide a rather different take on the problem. It is easy enough to show that where a person is located frames the search process such that the available local options tend to squeeze out other similar options located further away and deny the salience of different options not directly observable in the local community. Risk aversion is also an important element in the story. It is easy enough to show that local options may be perceived to carry less risk than similar options offered by vendors located further away. Even if it can be shown that distant options offer greater benefits than local options, Kahneman and Tversky’s (1979) formulation helps explain why, in these circumstances, decision-​makers may prefer local commitment over the possible upside of distant but untried options. In this

Behaviour in Context    203 way, the preference for the local over the global can be explained by reference to status quo bias: the preference for repeating past behaviour in a known situation rather than initiating the assessment and choice of alternative courses of action elsewhere (Samuelson and Zeckhauser, 1988). At the heart of behavioural psychology is an assumption that behaviour is experiential. That is, how people frame issues, assess options, and make decisions is based upon past experience, the use of information from the immediate environment, and expectations of the near future (in time and space). See, for example, Shafir et al. (1997) on the framing of inflationary expectations and Clark (2012) on why people prefer investing in local property over financial products that offer a diverse portfolio of global stocks. Being experiential, we can imagine situations that are rich in beneficial possibilities and situations that are poor in beneficial possibilities (Smith and Easterlow, 2005). To the extent that people are trapped in the latter situation by virtue of limited resources and the lack of choice and decision conditions that encourage risk-​taking (boot-​strapping) in favour of a desired future, then people can be knowingly myopic; that is, they are self-​ conscious of their limited prospects. Social and behavioural scientists tend to present time as a linear sequence running from the past, to the present, and to the future. The past is our (selective) memory of behaviour and events, the present is essentially the events we must deal with or accommodate, and the future is actually a set of expectations or beliefs based on the past and (especially) the present extended into the future. We cannot return to the past, directly. But we may face repeated instances of past events. Equally, we may see in new events sufficient similarities to past events that allow for the use of past experience to inform contemporary decision-​making. At the limit, Keynes (1921) suggested that reasoning by analogy provides a way of systemising decision-​making even if people tend not to use Bayes’s theorem. However, people may face a series of events that are not amenable to such decision procedures. In these situations, loss aversion may paralyse decision-​making. Geography is more complicated. Even if experience of other places may not be as important as in situ experience, experience of other places can prompt revision of preferences and expectations. Indeed, given the significance often attributed to isolated events, vivid experience of other places can assume considerable importance notwithstanding the habits associated with local living. As Keynes (1921) intimated, there is an issue of salience: the degree to which experience elsewhere can be applied to local circumstances. But geography is also an issue of scale (Tversky, 1992). That is, the local in relation to the regional, the national, and the international. Indeed, what counts as a local event may be an expression of processes that originate not only elsewhere, but also further up in the spatial hierarchy. Inflation is a case in point. It can be expressed in terms of local consumer prices, but it typically originates in global commodity markets, national monetary policy, and regional housing markets. Quite clearly, people experience local consumer prices and the turbulence or otherwise of regional housing markets. The implication of behavioural psychology is that if people were to frame expectations about global commodity markets and national monetary policy, they would do so through their experience of local consumer prices and housing markets. But, as is well appreciated, the causal pathway between local and global economic processes tends to run from the global to the local, less often from the local to the global. There are exceptions. For example, the London property market could affect UK price inflation and, hence, UK monetary policy.

204   Clark But how do we know? Implied is a causal chain linking property with prices and policies. At the highest level of abstraction, theoretical principles based upon established points of reference would drive the interpretation. At a lower level of abstraction, any such linkage would be estimated empirically via an econometric model. And yet, there are reasonable disagreements among experts about the theory of price formation and the direction of causation. Likewise, estimating empirically any such relationship is only a game played by experts or policymakers. Either way, few people are able to make a reasoned assessment of the issue. So, most people fall back on experience, more often than not the most recent experience rarely past experience. In this sense, experience is always partial and the lessons drawn from experience need not always lead to effective decision-​making. In fact, experience can paralyse rather than enhance decision-​making, especially in circumstances where the costs of decision-​making are entirely personal (Benartzi, 2015). To illustrate the costs of experiential learning, consider the following. Naive property investors tend to overemphasize local property prices and often fail to appreciate how local prices are associated with property prices in adjacent suburbs except when a certain property appears a ‘bargain’ or as ‘expensive’ (Clark, 2012). Importantly, naive investors focus on the idiosyncratic attributes of their chosen property, failing to recognize that the causal linkage between property prices and monetary policy means that, sometime in the future, the cost of borrowing will affect their capacity to repay the loan used to purchase the property and the market price of the property itself. Naive property investors also fail to appreciate the costs and consequences of treating property as a ‘local’ phenomenon when housing markets become a traded financial product, just like stocks and bonds, commodities, and infrastructure (Clark et al., 2012). In essence, behaviour in space and time requires mental models which put options in context. Here, however, is a conundrum: mental models may well impose a selective view of what is relevant to decision-​making.

Context of Behaviour In recent years, behavioural psychology has been preoccupied with cognition. In part, this reflects the ‘universal’ aspirations of the research programme: to say something meaningful about individual decision-​making in the context of space and time. Following in the footsteps of Herbert Simon, this research programme has also sought a deeper understanding of the ‘nature’ of the rational agent of conventional microeconomic theory. Critics of the behavioural revolution have suggested that, notwithstanding the success of this research programme in terms of the insights gleaned about anomalies and biases, it runs the risk of essentialism: this is an implied or even explicit argument to the effect that cognition trumps behavioural and social psychology when it comes to explaining observed behaviour (Pykett, 2013). No doubt there are instances that justify this type of criticism. There is another important point of dispute regarding the implications to be drawn from the behavioural revolution. Considering the testing procedures that underpin the search for behavioural anomalies and biases, it is notable that many of the problems and puzzles that are used to assess behavioural predispositions do so as one-​off instances. Standard problems and puzzles are designed to elicit certain types of responses, while guarding against any triggers that would prompt recognition of the problem posed and its related solution

Behaviour in Context    205 (Baron, 2008). Most people, most of the time, encounter familiar problems and carry with them information and knowledge of the effectiveness of past solutions. Whereas Kahneman (2011) emphasizes intuition and discounts learning, deliberation is arguably a fundamental characteristic of human reasoning (Doherty, 2003). So, are we to be optimistic or pessimistic about people’s deliberative and reasoning abilities? The answer to this question has important implications for public policy (Thaler and Sunstein, 2008). More generally, it is a truism that individual decision-​making is always embedded in some context or situation. It could hardly be otherwise, unless we are talking about those rare individuals who lack empathy, a sense of reciprocity, and a commitment to mutual survival and thus have no cognitive appreciation of others. But, of course, what counts as a person’s context or environment can be quite complex and multidimensional (Clark, 2014). For example, a person’s immediate social environment could include their family, workplace, community, state, and nation. Likewise, their relevant time horizons could include their own (now and the realization of some goal in the future), their immediate family’s future plans, and commitments made to workplace colleagues in relation to shared goals and objectives. Importantly, a person’s reference point or sense of others they wish to emulate could be found in spatially extensive networks of family and friends, as well as icons found through social media, which wraps around the globe (Clark, 2013). The idea that individual decision-​making is embedded in time and space is unproblematic for Simon (1956) and Kahneman and Tversky (1979), and for many other behavioural psychologists who take seriously the interaction between cognition and context. But in economic geography, being embedded is more than recognition of time and space; it is at once a critique of the ‘atomised view of economic agents’, which dominates conventional economics (Bathelt and Glückler, 2011, p. 28) and an ontological statement summarizing a research programme that finds favour in a number of other social science disciplines, including sociology (Granovetter, 1985) and political economy (Hodgson, 1988). As Grabher (1993, p. 4) noted, ‘ “embeddedness” refers to the fact that economic action and outcomes, like all social action and outcomes, are affected by actors’ dyadic relations and by the structure of the overall framework of relations’. In a similar fashion, Bathelt and Glückler (2011, p. 42) note that ‘interactions between individual and collective actors in economic contexts’ are a crucial element of their ‘relational economic geography’. Translating this argument into an analytical framework relevant to study of decision-​ making is more challenging than designing a problem or set of problems to test the competence of individual decision-​makers. Hogarth (2001) tackles the issue by situating learning in different types of situations, distinguishing between situations that do not reward learning from sequential decision-​making and situations that facilitate learning. Here, Hogarth brings together three threads of research found in decision theory. Recognizing the importance of intuition and the interaction between cognition and context, Hogarth provides an analytical framework whereby behaviour can be situated in time and space at a general level, and in specific kinds of situations with certain types of institutions. Hogarth also identifies a key role for deliberation in context, going beyond the shortcomings of non-​reflexive intuition. Moreover, his model is particularly useful in situations where people combine knowledge and understanding about their part of the world with the rules or algorithms used by others to make decisions in that domain. Extending this argument further, analytical frameworks can be designed that combine the nature and scope of individual behaviour with the attributes of different kinds of situations.

206   Clark Imagine that individuals are required to make two kinds of decisions, one that is rather simple and incremental in effect, and one that requires a level of imagination and commitment that goes beyond routine decision-​making. The first kind of decision-​making relies upon tried-​and-​true heuristics informed by fresh information which either reinforces intuition or prompts minor adjustment. The second kind of decision-​making requires individuals to take stock of that which is inherited in relation to the challenge before them in time and space. Let us also assume that there are two kinds of environments, where one region is poor in knowledge and information, while the other region is rich in knowledge and information. These two types of decisions and the two regions are set out in schematic form in Figure 10.1 (a version of Figure 5.1 in Clark et al., 2012). Quite obviously, the institutional fabric of the two regions (poor and rich) that can aid individual decision-​making can be quite significant in terms of the sophistication of response by the average person to decisions that require either incremental or intuitive reasoning as opposed to complex or judgement-​related reasoning. In effect, the gap between the average person in region one compared with region two with respect to the consequences of incremental reasoning may be relatively modest, although in all likelihood would favour the average person in region two. By contrast, the gap between the average person in region one compared with region two with respect to the consequences of complex reasoning may be relatively large, clearly favouring the average person in region two (Clark et al., 2012). If we add to the argument the possibility that social position is a significant factor in the quality of financial decision-​making (Sharpe, 2007), and that regions one and two are each composed of two groups of people—​one group that is relatively poorly educated and financially illiterate, whereas the other group has the skills and education consistent with financial literacy—​then it is easy enough to imagine why the financially illiterate group located in region one is likely to fall behind the welfare of the financially illiterate group located in region two. That is, a rich institutional structure consistent with the nature and scope of problems faced by decision-​makers can compensate for shortfalls in individual decision-​ making competence and consistency (Clark, 2013). This may be particularly important in

Poor

Rich

Incremental

A

B

Complex

Decision (nature)

Environment (resources)

C

D

Figure  10.1   Interaction Between the Environment (Resources) and the Nature of the Decision Problem. Source: author.

Behaviour in Context    207 covering the downside risks associated with decision-​making, even if on the upside of decision-​making, those most equipped by virtue of their social position are able to take advantage of opportunities in ways that others are not able to either conceptualize or carry through by virtue of their limited skills and expertise. Here, then, is the most significant implication of this argument. Whereas the thought experiment above refers to the inherited decision-​related resources of two regions, recognizing the vulnerability of the financially illiterate to complex financial decisions, the communities of both regions might reasonably consider how best to ensure the welfare of this group in the face of the costs and consequences of poor decision-​making (Thaler and Sunstein, 2008; Campbell et al., 2011). Clearly, this is a complex issue involving social issues of mutual respect and common commitment, as well as issues of institutional design and performance. These are the subjects of a much larger research programme that crosses over and joins social science disciplines with philosophy. As such, consideration of institutional design goes beyond the purpose of this chapter. Nonetheless, the implication is clear: understanding the interaction between cognition and context requires much more than an account of individual behaviour; it requires a theory of how and why people are embedded in space and time, as well as a theory of institutional formation and performance.

Conclusions For many years, the standard economic model of individual behaviour assumed a rational actor who maximized utility subject to budget constraints. While people were assumed to differ in terms of their preferences and their resources, it was assumed that they share the same process whereby decisions are made. If people were to deviate from the optimizing framework used to characterize the decision-​making process, it was widely assumed that markets would penalize those who deviated from convention. As such, incentives were assumed to drive people to rationality. Given a shared decision framework, theorists were then able to sum up individuals to the economy as a whole, providing the microeconomic foundations for macroeconomic phenomena (Weintraub, 1979). It was a short step, indeed, to rational expectations (Lucas, 1972). Whereas economic geographers and economic sociologists railed against the essentialism underpinning this model, invoking instances that seemed to go against the idealized notions of rationality, too often these challenges were side-​lined into a debate about rationality versus irrationality. The behavioural revolution initiated by Simon (1956) and given force by Kahneman and Tversky (1979) is less about rationality per se and more about whether people are competent decision-​makers in a given spatial and temporal context. Most importantly, the behavioural revolution has sought to show that identified behavioural anomalies and biases are systemic rather than simply isolated instances of aberrant behaviour. The behavioural revolution has provided the empirical foundations for a revised theory of behaviour which, when summed to the economy as a whole, undercuts the plausibility of conventional theorizing. These behavioural insights have proven important in understanding recurrent financial crises (Clark, 2011; Haldane and May, 2011).

208   Clark In this chapter, space and time have been identified as fundamentally important descriptors of the situation in which people must make decisions. Many people who live and work in Western capitalist democracies must juggle short-​term considerations with long-​term commitments. The short term, for most people, is rather unproblematic in that the decisions that must be made are, more often than not, incremental rather than transformative. But the long term is more problematic, especially when they must cope with risk punctuated by systemic shocks which unexpectedly shift the basis upon which to make judgements about the future. Likewise, the here and now is that which is inherited from the past, whereas individual prospects are bound up with the place of their community, region, and nation in the global economy. Recognizing the significance of risk and uncertainty in individual financial decision-​ making has prompted governments to reconsider policies that require those least able to cope in this situation to bear consequences of their decision-​making (Clark, 2012). One branch of institutional theory is content to focus on the rules of the game, in effect the institutional framework for individual behaviour, cooperation and collaboration, and reconciliation of conflict (North, 1991). Another branch of institutional theory is not so sanguine about the power of rules, and seeks to reinforce the significance of established organizations like national welfare agencies that protect those least able to be effective in space and time while enabling those better able to cope with incentives to carry forward their own ambitions (see Rodrik, 2013 on the relevance of the nation state). Whereas much of economic theory has focused upon the representative agent, only obliquely recognizing the diversity of behaviour around the mean (Becker, 1962), others have sought a better understanding of the diversity of behaviour per se while holding to an empirically informed theory of human nature (see Henrich et al., 2004). Once we accept the intimate relationship between cognition and context, empirical research on the systematic impact of context on observed behaviour must inevitably flourish. Here, there are many signposts for the future. For example, Fehr and Schmidt (1999) have sought to integrate evidence on the significance of moral commitments with economic reasoning. Their project is indicative of the importance of Hacker’s (2007, p. 4) argument that ‘while human being is a biological category, person is a moral, legal and social one. To be a person is, among other things, to be a subject of moral rights and duties. It is to be not only an agent, like other animals, but also a moral agent, standing in reciprocal relations to others.’ Recognizing the importance of institutions in framing the context of behaviour, and recognizing that institutional design is a deliberative act in cooperation with others, economic geography may be precisely the right venue through which to make good on institutionalism in general and in particular (Bathelt and Glückler, 2014).

Acknowledgements This chapter was first presented at Seoul National University. It bears the imprint of a long-​ term research programme on savings behaviour in conjunction with the late John C. Marshal, Dorothee Franzen, Heribert Karch, Emiko Caerlewy-​Smith, Janelle Knox-​Hayes, and Kendra Strauss. Initial support for the project was provided by the National Association of Pension Funds, Mercer, the Economic and Social Research Council, and MetallRente GmbH. More

Behaviour in Context    209 recently, this work has been supported by the Monash–​CSIRO Superannuation Research Project with data from Mercer (Australia). In this regard, I am pleased to acknowledge collaboration with Christine Brown, Maurizio Fiaschetti, Paul Gerrans, and David Knox. Dane Rook provided penetrating comments on a preliminary draft of the chapter; it is better for his intervention even if I could not accommodate all of his criticisms. Cody McCoy and Sarah McGill provided advice on successive drafts of the chapter. None of the above should be held responsible for any views or opinions expressed herein.

Notes 1. Cognitive scientists emphasize the importance of spatial awareness, noting that the visual processing of information from the environment is fundamental to behaviour. As such, it is arguable that the cognitive processing of space comes before time and is in that sense more meaningful for understanding observed behaviour (Bor, 2012). In this respect, time and space are substantive rather than relational; that is, they exist independent of our place in the world (Simons, 2016). Nonetheless, the status and significance of time and space are framed by the institutional context in which human beings must make decisions. This issue is considered in the penultimate section of the chapter. 2. Selection operates at the level of the population, mediated by the imperatives faced by people in specific situations. In some accounts, selection is suggested to function at the level of the individual in that rational people are assumed to act in a manner consistent with ‘surviving a very harsh competitive world’ (Becker, 1962, p. 1). This suggests the necessity of a certain kind of rationality, rather than an empirically justified statement about the nature and scope of actual behaviour. 3. It was standard practice to suppose that the brain stores and retrieves information in the same linear fashion as computers code, store, access, and use data. Cognitive science suggests that humans perform these functions simultaneously and memory is selective in that we tend to retrieve past experience so as to reinforce immediate commitments (D. Rook, ‘Aligning representation: doxastic distances and fiduciary delegation’, draft).

References Ainslie, G. (2001). Breakdown of Will (Cambridge: Cambridge University Press). Almenberg, J. and Dreber, A. (2013). ‘Economics and Evolution: Complementary Perspectives on Cooperation’ in M.A. Nowak and S. Coakley (eds) Evolution, Games, and God: The Principle of Cooperation, pp. 132–​149 (Cambridge, MA: Harvard University Press). Arrow, K.J. (1984). Collected Papers of K. J. Arrow Vol. 3: Individual Choice Under Certainty and Uncertainty (Oxford: Blackwell). Baron, J. (2008). Thinking and Deciding, 4th ed. (Cambridge: Cambridge University Press). Bathelt, H. and Glückler, J. (2011). The Relational Economy:  Geographies of Knowing and Learning (Oxford: Oxford University Press). Bathelt, H. and Glückler, J. (2014). ‘Institutional change in economic geography’. Progress in Human Geography 38: 340–​363. Becker, G.S. (1962). ‘Investment in human capital: a theoretical analysis’. Journal of Political Economy 70: 9–​49.

210   Clark Benartzi, S. (2015). The Smarter Screen (New York: Penguin). Bor, D. (2012). The Ravenous Brain:  How the New Science of Consciousness Explains our Insatiable Search for Meaning (New York: Basic Books). Campbell, J.C., Jackson, H., Madrian, B., and Tufano, P. (2011). ‘Consumer financial protection’. Journal of Economic Perspectives 25: 91–​114. Clark, G.L. (1998). ‘Stylised facts and close dialogue: methodology in economic geography’. Annals, Association of American Geographers 88: 73–​87. Clark, G.L. (2011). ‘Myopia and the global financial crisis: context-​specific reasoning, market structure, and institutional governance’. Dialogues in Human Geography 1: 4–​25. Clark, G.L. (2012). ‘Property or pensions?’ Environment and Planning A 44: 1185–​1199. Clark, G.L. (2013). ‘Mapping financial literacy’. Geografiska Annaler B: Human Geography 95: 131–​145. Clark, G.L. (2014). ‘Roepke lecture in economic geography—​financial literacy in context’. Economic Geography 90: 1–​23. Clark, G.L., Caerlewy-​Smith, E., and Marshall, J.C. (2006). ‘Pension fund trustee competence:  decision making in problems relevant to investment practice’. Journal of Pension Economics and Finance 5: 91–​110. Clark, G.L., Caerlewy-​Smith, E., and Marshall, J.C. (2007). ‘The consistency of UK pension fund trustee decision-​making’. Journal of Pension Economics and Finance 6: 67–​86. Clark, G.L., Caerlewy-​Smith, E., and Marshall, J.C. (2009). ‘Solutions to the asset allocation problem by informed respondents: the significance of the size-​of-​bet and the 1/​n heuristic’. Risk Management and Insurance Review 12: 251–​271. Clark, G.L., Strauss, K., and Knox-​Hayes, J. (2012). Saving for Retirement: Intention, Context, and Behaviour (Oxford: Oxford University Press). Clark, G.L. and Whiteman, J. (1983). ‘Why poor people do not move: job search behaviour and disequilibrium amongst local labor markets’. Environment and Planning A 15: 85–​104. Doherty, M.E. (2003). ‘Optimists, Pessimists, and Realists’ in S.L. Schneider and J. Shanteau (eds) Emerging Perspectives on Judgement and Decision Research, pp. 643–​679 (Cambridge: Cambridge University Press). Easterly, W. (2014). The Tyranny of Experts: Economists, Dictators, and the Forgotten Rights of the Poor (New York: Basic Books). Fehr, E. and Schmidt, K.M. (1999). ‘A theory of fairness, competition, and cooperation’. Quarterly Journal of Economics 114: 817–​868. Geif, A. (2006). Institutions and the Path to the Modern Economy (Cambridge:  Cambridge University Press). Gigerenzer, G., Todd, P.M., and the ABC Research Group (1999). Simple Heuristics That Make Us Smart (New York: Oxford University Press). Golledge, R., Brown, L.A., and Williamson, F. (1972). ‘Behavioural approaches in geography: an overview’. Australian Geographer 12: 59–​79. Grabher, G. (1993). ‘Rediscovering the Social in the Economics of Interfirm Relations’ in G. Grabher (ed.) The Embedded Firm: On the Socioeconomics of Industrial Networks, pp. 1–​31 (London: Routledge). Granovetter, M. (1985). ‘Economic action and social structure: the problem of embeddedness’. American Journal of Sociology 91: 481–​510. Hagerstrand, T. (1967). Innovation Diffusion as a Spatial Process. Trans. A. Pred. (Chicago, IL: University of Chicago Press). Haldane, A. and May, R. (2011). ‘Systemic risk in banking systems’. Nature 469: 351–​355. Hacker, P.M.S. (2007). Human Nature: The Categorical Framework (New York: Wiley-​Blackwell).

Behaviour in Context    211 Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., and Gintis, H. (eds) (2004). Foundations of Human Sociality: Economic Experiments and Ethnographic Evidence from Fifteen Small-​ Scale Societies (Oxford: Oxford University Press). Hodgson, G. (1988). Economics and Institutions (Cambridge: Polity Press). Hogarth, R.M. (2001). Educating Intuition (Chicago, IL: University of Chicago Press). Hong, H., Kubik, J., and Stein, J. (2004). Social interaction and stock-​market participation. Journal of Finance 59: 137–​163. Hong, H., Kubik, J., and Stein, J. (2005). ‘Thy neighbor’s portfolio: word-​of-​mouth effects in the holdings and trades of money managers’. Journal of Finance 60: 2801–​2824. Huberman, G. (2001). ‘Familiarity breeds investment’. Review of Financial Studies 14: 659–​680. Hurley, S. and Nudds, M. (2006). ‘The Question of Animal Rationality: Theory and Evidence’ in S. Hurley and M. Nudds (eds) Rational Animals? pp. 1–​83 (Oxford: Oxford University Press). Jones, M. and Love, M.S. (2011). ‘Bayesian fundamentalism or enlightenment? On the explanatory status and theoretical contributions of Bayesian models of cognition’. Behavioral and Brain Sciences 34: 169–​187; 215–​231. Kahneman, D. (2003). ‘Maps of bounded rationality: psychology for behavioural economics’. American Economic Review 93: 1449–​1475. Kahneman, D. (2011). Thinking Fast and Slow (London: Allen Lane). Kahneman, D. and Tversky, A. (1979). ‘Prospect theory: an analysis of decision under risk’. Econometrica 47: 263–​291. Keynes, J.M. (1921). A Treatise on Probability (London: Macmillan). Knight, F.H. (1921). Risk, Uncertainty, and Profit (Boston, MA: Houghton Mifflin). Krueger, J.I. and Funder, D.C. (2004). ‘Towards a balanced social psychology: causes, consequences, and cures for the problem-​seeking approach to social behavior and cognition’. Behavioral and Brain Sciences 27: 313–​328. Laibson, D. (2003). ‘Golden eggs and hyperbolic discounting’. Quarterly Journal of Economics 62: 443–​477. Lucas, R.E. (1972). ‘Expectations and the neutrality of money’. Journal of Economic Theory 4: 103–​124. North, D.C. (1991). ‘Institutions’. Journal of Economic Perspectives 5: 97–​112. Pettit, P. and McDowell, J. (eds) (1986). Subject, Thought, and Context (Oxford:  Oxford University Press). Pliske, R. and Klein, G. (2003). ‘The Naturalistic Decision-​making Perspective’ in S.L. Scheider and J. Shanteau (eds) Emerging Perspectives on Judgment and Decision Research, pp. 559–​587 (Cambridge: Cambridge University Press). Pykett, J. (2013). ‘Neurocapitalism and the new neuros: using neuroeconomics, behavioural economics and picoeconomics for public policy’. Journal of Economic Geography 13: 845–​869. Rodrik, D. (2013). ‘Roepke lecture in economic geography—​who needs the nation‐state?’ Economic Geography 89: 1–​19. Rook, D. (2014). ‘Aligning representation: doxastic distances and fiduciary delegation’. Draft. Samuelson, W.A. and Zeckhauser, R. (1988). ‘Status quo bias in decision making’. Journal of Risk and Uncertainty 1: 7–​59. Shafir, E., Diamond, P., and Tversky, A. (1997). ‘Money illusion’. Quarterly Journal of Economics 112: 341–​374. Sharpe, W.F. (2007). Investors and Markets:  Portfolio Choices, Asset Prices and Investment Markets (Princeton, NJ: Princeton University Press). Simon, H.A. (1956). ‘Rational choice and the structure of the environment’. Psychology Review 63: 129–​138.

212   Clark Simons, P. (2016). ‘External Relations, Casual Coincidence, and Contingency’ in A. Marmodoro and D. Yates (eds) The Metaphysics of Relations, pp. 113–​137 (Oxford: Oxford University Press). Smith, S. and Easterlow, D. (2005). ‘The strange geography of health inequalities’. Transactions of the Institute of British Geographers NS30: 173–​190. Stambaugh, R.F. (2014). ‘Presidential address:  investment noise and trends’. The Journal of Finance 69: 1415–​1453. Stiglitz, J.E. (2000). ‘The contributions of the economics of information to twentieth century economics’. Quarterly Journal of Economics 115: 1441–​1478. Storper, M.J. and Venables, A.J. (2004). ‘Buzz: face-​to-​face contact and the urban economy’. Journal of Economic Geography 4: 351–​370. Strauss, K. (2008). ‘Re-​ engaging with rationality in economic geography:  behavioural approaches and the importance of context in decision-​ making’. Journal of Economic Geography 8: 137–​156. Strauss, K. (2009). ‘Cognition, context, and multi-​method approaches to economic decision making’. Environment and Planning A 41: 302–​317. Thaler, R. and Sunstein, C. (2008). Nudge:  Improving Decisions about Health, Wealth and Happiness (New Haven, CT: Yale University Press). Tversky, B. (1992). ‘Distortions in cognitive maps’. Geoforum 23: 131–​138. Webber, M.J. (1972). Impact of Uncertainty on Location (Canberra:  Australian National University Press). Weintraub, E.R. (1979). Microfoundations:  The Compatibility of Microeconomics and Macroeconomics (Cambridge: Cambridge University Press). Wolpert, J. (1980). ‘The dignity of risk’. Transactions, Institute of British Geographers NS5: 391–​401.

Chapter 11

Evolu tionary E c onomi c Geo gra ph y Ron Boschma and Koen Frenken Introduction From its start, evolutionary economic geography (EEG) has aimed to contribute to the understanding of several topics in economic geography, as to why industries concentrate in space, how networks evolve in space, and why some regions grow more than others. These core topics are by no means new in economic geography, but theorized and analysed from an evolutionary perspective, EEG provides new and additional insights, and in some cases alternative explanations (Boschma and Lambooy, 1999; Boschma and Frenken, 2006; Martin and Sunley, 2006; Boschma and Martin, 2010; Kogler, 2015; special issues on EEG in Journal of Economic Geography 2007, Economic Geography 2009, Regional Studies 2015, and Journal of Economic and Social Geography 2015). The main objective of this chapter is to provide a short outline of key contributions of EEG with respect to industrial clusters (‘Clustering as an Evolutionary Process’), geography of networks (‘Networking as an Evolutionary Process’), urban and regional development (‘Regional Development as an Evolutionary Process’), and the role of institutions (‘Institutions and Evolutionary Economic Geography’).

Evolutionary Economic Geography Key to the emergence of EEG has been the establishment of evolutionary economics (Nelson and Winter, 1982; Dosi et al., 1988), which presented itself in the 1980s as an alternative to orthodox or mainstream neoclassical economics, rejecting the representative agent, optimal decision-​making, and equilibrium analyses. Because the evolutionary critique on mainstream economics appealed to many economic geographers, while evolutionary economics itself was basically a-​spatial (Boschma and Lambooy, 1999), economic geographers started

214   Boschma and Frenken to develop new concepts based on evolutionary thinking, including ‘windows of locational opportunity’ (Storper and Walker, 1989), ‘technology districts’ (Storper, 1992), and ‘regional innovation systems’ (Cooke, 1992). This led to a lively debate about how to apply evolutionary principles to economic geography in a coherent manner. Several bodies of literature have served as key sources of inspiration to evolutionary scholars in economic geography in particular (Boschma and Martin, 2010): generalized Darwinism (Essletzbichler and Rigby, 2010), path dependency theory (Martin and Sunley, 2006), complexity theory (Martin and Sunley, 2007), and geographical political economy (MacKinnon et al., 2009). What these evolutionary approaches to economic geography have in common is a focus on historical processes that explain the uneven development and transformation of the economic landscape. This spatial pattern is perceived as the outcome of largely contingent, yet path-​dependent and place-​dependent historical processes. Since the late 1990s, EEG slowly developed into an alternative to equilibrium-​based perspectives dominant in the ‘new economic geography’ and in large parts of the regional science literature (Boschma and Frenken, 2006). At this time, most work in EEG built on biological notions as variety, selection, retention, mutation, and adaptation, as well as economic notions such as path dependence, lock-​in, and proximity (Boschma and Frenken, 2003, 2006; Martin and Sunley, 2006). It often has a strong micro-​foundation, taking organizational routines (Nelson and Winter, 1982) as a point of departure. Routines are capabilities on the basis of which organizations, like firms, try to adapt and survive in a dynamic environment. Economic evolution is understood as the creation and selective transmission of new routines in space. This transmission process is imperfect because of bounded rationality, as actors have no full access nor a perfect ability to respond to information, and because of spatial conditions (like institutions) that favour (or not) the creation and diffusion of routines. This leads to uneven geographical patterns, as embodied in agglomerations, centre-​periphery patterns, clusters, and networks. The main focus of EEG is on how these spatial structures of the economy emerge from and are transformed by the micro-​behaviour of individual and collective agents, and why and how these processes of change are themselves path-​and place-​dependent (Boschma and Frenken, 2006). Within this alternative framework to equilibrium-​based perspectives, EEG has made a number of contributions to the field of economic geography. Firstly, EEG has challenged the Marshallian view on industry clustering in economic geography, in which firms in clusters would automatically benefit from Marshallian externalities, and clusters are conceived to exist and persist because of that. Secondly, EEG has proposed a dynamic proximity view on the geography of knowledge networks that has led to additional insights in the cluster and network literatures. Thirdly, EEG has made new contributions to the agglomeration externalities literature with the introduction of the concept of related variety, and it has initiated a new literature on regional diversification, known as regional branching. Fourthly, EEG has taken a critical view on the way institutions have been analysed in economic geography, by drawing attention, for example, to the contingent role of (local) institutions, as exemplified by the existence and persistence of heterogeneity of actors in the same institutional environment (Boschma and Frenken, 2009). In what comes next, we elaborate on each of these key contributions of EEG one by one.

Evolutionary Economic Geography    215

Clustering as an evolutionary process A classic question in economic geography is why some industries are concentrated in space. Marshall (1920) raised this question a century ago when he investigated the clustering of the metals industry in Sheffield and South Yorkshire in the UK (Potter and Watts, 2011). His explanation of such spatial clustering of an industry has dominated the field of economic geography for a long time: as soon as an industry locates somewhere, economic benefits can be derived from the co-​location of firms in that industry, provided by a local pool of specialized knowledge, labour, and suppliers, also known as ‘localization economies’ or ‘Marshallian externalities’. What attracted special attention was Marshall’s remark that the ‘mysteries of trade’ in an industry were ‘in the air’ in clusters. His ideas implied that knowledge spillovers are geographically bounded, that intra-​industry knowledge is accessible almost exclusively to firms in clusters, and that cluster firms are expected to perform better than firms outside clusters. This dominant Marshallian thinking has been challenged, and EEG has made important contributions here. Sorenson and Audia (2000) and Stuart and Sorenson (2003) claimed that clusters emerge and exist because of a self-​perpetuating process of local entry: the more local firms in a new industry, the more local entry. This has been called ‘cognitive legitimacy’ in organizational ecology, in which a high rate of local entry generates information that diffuses to potential entrepreneurs inducing them to create their own firms (Aldrich and Fiol, 1994; Maggioni, 2002; Wenting and Frenken, 2011). More local entry also generates more local competition rendering it harder for cluster firms to survive. In this ecological view, clusters decrease entry costs, while increasing the chances of exit at the same time. Klepper (2007) challenged Marshallian thinking even more by providing an alternative theory on spatial clustering from an industry life-​cycle perspective. The point of departure is an evolutionary micro-​perspective in which firms are depicted as being heterogeneous in their routines and capabilities because of bounded rationality, and therefore firms show differential growth rates. Firms have different routines because their pre-​entry experience differs. In particular, spin-​off companies inherit superior capabilities from successful parents from the same or related industries and therefore tend to outperform other types of entrants. As firms often locate in the founder’s home region and rarely relocate (Stam, 2007; Dahl and Sorenson, 2012), a cluster can simply emerge through a local self-​reinforcing spin-​off process, and there is no need for Marshallian externalities to make that happen. Klepper (2007) found evidence for his spin-​off thesis for the US automobile industry that concentrated in the Detroit region. He showed that survival rates depended primarily on the quality of pre-​entry experience, not on Marshallian externalities (here, being located in Detroit). This finding has been replicated for other industries as diverse as semiconductors, tyres, fashion design, and book publishing. However, some studies, especially on creative industries, found that also Marshallian externalities, besides pre-​entry experience, positively affected firm performance (for surveys, see Boschma, 2015a; Frenken et al., 2015). Such an evolutionary approach explains not only why industries concentrate in space, but also why an industry historically emerged in one particular place, rather than another. As a cluster can, theoretically, emerge from a single successful firm generating many spin-​offs, a cluster can emerge anywhere where this firm happens to locate (e.g. in the entrepreneur’s

216   Boschma and Frenken home region). The question holds to what extent such a path-​dependent process can be said to be also place-​dependent (Martin and Sunley, 2006). Although the location of a cluster is rather unpredictable owing to the path-​dependent, self-​reinforcing logic of the spin-​off process, there is substantial evidence that the first generation of successful firms in an emerging industry are often firms diversifying, or spinning off, from local related industries (Boschma and Wenting, 2007; Klepper, 2007; Buenstorf and Klepper, 2009). This is line with evidence that regions hosting industries that are related to a new industry have a higher probability of giving birth to it (Neffke et al., 2011). Combined with Klepper’s findings, recent studies suggest that clusters are characterized by positive related-​industry externalities and (possibly) negative Marshallian externalities. In the conventional definitions of Marshallian externalities, as well as Porter’s (1990) original cluster concept, intra-​industry externalities and related-​industry externalities are not analytically separated. This distinction, however, is very relevant because the two types of externalities tend to have opposite effects on firm survival. Related-​industry externalities among local firms are expected to be positive, arising from knowledge spillovers and the mobility of skilled people, while intra-​industry externalities among local firms are expected to be mainly negative owing to competitive pressure and involuntary knowledge spillovers (Boschma, 2015a; Frenken et  al., 2015). In particular, intra-​industry externalities mostly harm well-​performing cluster firms that have the most to lose and least to gain from other cluster firms, while young and small firms may still benefit from intra-​industry externalities to compensate for their weak internal capabilities (Rigby and Brown, 2015). Similarly, firms are also heterogeneous to the extent to which they profit from the local presence of multinationals, with the most internationalized firms benefitting most (Crescenzi et al., 2015). Thus, given that firms are heterogeneous in their capabilities, the extent to which firms suffer or profit from co-​location is expected to vary accordingly. Another branch in the EEG literature is the ‘cluster life-​cycle’ approach that studies the evolution of clusters, in particular the endogenous dynamics that may turn successful clusters into declining ones (Pouder and St. John, 1996; Brenner, 2004; Iammarino and McCann, 2006; Belussi and Sedita, 2009). It is crucial to underline that the life-​cycle notion should be understood here in a non-​deterministic, evolutionary manner (Martin and Sunley, 2011), as a cluster can renew itself, for instance. Menzel and Fornahl (2010) proposed a cluster life-​ cycle model in which firms enter and exit clusters, capabilities of firms interact and evolve, and inter-​organizational linkages are formed and dissolved within and beyond clusters. When a cluster emerges, the heterogeneity of firms’ capabilities initially increases but subsequently decreases, as firms engage in competition, inter-​firm learning, and networking (Rigby and Essletzbichler, 1997; Vicente and Suire, 2007). If this convergence continues, the recombinant potential of the cluster decreases and its principal activities will decline. Menzel and Fornahl (2010) argue that a declining cluster can revive itself by upgrading its knowledge base through inflow of new knowledge from outside the cluster (‘adaptation’), by integrating various local knowledge bases (‘renewal’), or by diversifying into new activities while building on the local knowledge base (‘transformation’). Clusters can only adapt if firms and other agents proactively engage in such a change process, but this is far from easy, given their proximity to local networks and institutions (Glasmeier, 1991). This ‘lock-​in’ may be reinforced when public policy is responsive primarily to demands from vested interests (Grabher, 1993; Hassink, 2005), and local actors hold on to a collective identity (Staber and Sautter, 2011).

Evolutionary Economic Geography    217 Life-​cycle dynamics may also stem from herding behaviour in location decisions, indicative of hypes. Suire and Vicente (2009) developed a model of cluster emergence and stability that takes into account such herding effects. Firms may locate in clusters not for alleged Marshallian externalities associated with co-​location, but for reasons of what Appold (2005) called ‘geographical charisma’. Some clusters have a strong reputation owing to very visible and successful firms that attract other firms to the cluster. Here, being located in a cluster with successful firms acts as a signal to their stakeholders that they are present ‘where the action takes place’, hereby legitimating the location choice. The model by Suire and Vicente (2009) shows that if legitimation effects prevail, a cluster can grow very fast, but remains fragile, as the pattern of co-​location is not based on positive externalities. As a result, once the reputed firm would lose its reputation, or would relocate to another location, the cluster is likely to break down. To conclude, the main contribution of EEG to the topic of spatial clustering of industries is that the dominant explanation of industry clustering resulting from Marshallian externalities has been challenged: (i) clusters can emerge despite the absence of localization economies; (ii) clusters can emerge and exist because of a self-​reinforcing process of local entry, in particular the entry of successful spin-​offs; (iii) emerging clusters tend to be characterized by positive related-​industry externalities; (iv) not all firms perform equally in clusters—​some have better routines, partly owing to the pre-​entry background of the entrepreneur, and firms differ in their ability to exploit positive externalities and cope with negative externalities in clusters; (v) emergent clusters produce new institutions or adapt existing institutions by the collective action of agents; (vi) declining clusters can revive and overcome lock-​in, but not necessarily so.

Networking as an Evolutionary Process In economic geography, it is well known that spatial clustering provides opportunities to make connections between people and organizations. Firms not only compete, but also interact and collaborate with a range of organizations like other firms, banks, research institutes, or universities. As geographical distance often forms a barrier, organizations in the same region are more likely to connect, but not necessarily all of them. EEG is well equipped to incorporate these relational issues because it reasons from the heterogeneity of agents (Boschma and Frenken, 2010). As capabilities differ between organizations, they do not easily connect, nor do they easily learn from each other. This is exactly why networks in general, and knowledge and innovation networks in particular, are not randomly structured but skewed: that is, some organizations are more connected than others (Powell et al., 1996; Giuliani, 2007). Such an evolutionary take on the geography of knowledge networks has led to additional insights in the cluster literature. Contrary to what the literature often suggests, being part of a cluster does not necessarily mean that all cluster firms are connected with each other. Indeed, there is strong evidence that some cluster firms are highly connected in (local) knowledge networks, while other cluster firms are poorly connected, or not connected at all. Giuliani and Bell (2005) showed in a seminal study on a Chilean wine cluster that firms with a high absorptive capacity occupy a more central position in the local knowledge network.

218   Boschma and Frenken Such firms are attractive partners to connect to and capable of absorbing knowledge from other firms in and outside the cluster. This means that only a few firms in clusters (the most connected) have access to crucial local knowledge. This goes against the Marshallian view that knowledge is ‘in the air’ in clusters, in which all cluster firms are perceived to have equal access to local knowledge because they share the same location and the same norms and values. As Giuliani (2007) put it, knowledge networks are not pervasive but selective, and networks in clusters are no exception to that rule. Besides individual features of firms, like absorptive capacity, proximities are also key drivers of network tie formation, and this is where the proximity literature and EEG clearly meet (Boschma and Frenken, 2010). As actors differ, they show a strong bias towards which firms they interact and collaborate with, preferring those with whom they share similar knowledge (cognitive proximity), norms and values (institutional proximity), the same location (geographical proximity), social ties (social proximity), or organizational boundaries (organizational proximity) (Boschma, 2005; Breschi and Lissoni, 2009). As other forms of proximity can substitute for geographical proximity, the proximity concept can explain why networks within clusters are not pervasive, and why some cluster firms, sometimes acting as gatekeepers (Morrison, 2008), have most of their relations with firms outside the cluster. The relationship between proximity and firm performance is more ambiguous. While proximity promotes actors to collaborate, it does not necessarily increase collaborative performance, and may even turn out to be harmful. This has been referred to as the ‘proximity paradox’ (Broekel and Boschma, 2012). For instance, cognitive proximity not only facilitates communication and knowledge transfer between firms, but it also reduces the scope for learning and enhances the risk of involuntary knowledge leakage. Moreover, one expects proximity in relationships to increase over time, as interacting agents tend to become more similar as a result of social interaction and interactive learning (Balland et al., 2015). This has led to a search for ‘optimal’ proximity to cope with the negative aspects of proximity (Boschma, 2005). For geographical proximity, scholars have underlined the importance of a combination of local and non-​local knowledge linkages for the long-​term development of clusters (Asheim and Isaksen, 2002; Bathelt et al., 2004), or the importance of temporary proximity between agents who meet at conferences or trade fairs where they exchange knowledge (Torre, 2008). A questionable assumption in most proximity studies, however, holds the assumption of symmetry. A future challenge is to take up and integrate power and asymmetric relations in the proximity framework, as an actor can be proximate to another actor but not necessarily vice versa. The evolution of networks has been a subject of recent research in EEG (Ter Wal and Boschma, 2011). Balland et al. (2013) studied network dynamics in the global video game industry by looking at collaborations of co-​developers of new video games. Their study demonstrated that geographical proximity became a more important driver of network tie formation as the industry evolved. This increasing tendency of inter-​firm collaboration at shorter geographical distances could be explained by the increasing technological complexity of video games (Sorenson et al., 2006) and the project-​based nature of video game production in which ‘local buzz’ and ‘who-​knows-​who’ are key inputs (Grabher, 2006). Ter Wal (2014) found the opposite result in biotech: geographical proximity became less important as driver of co-​inventor networks, possibly owing to the increasing codification of biotech knowledge. However, there still is little understanding of how spatial networks change: little is known of how proximities in networks evolve over time (Balland et al., 2015), how network

Evolutionary Economic Geography    219 structures in clusters change, to what extent network dynamics exhibit path dependence (Gluckler, 2007), and how network dynamics affect the evolution of a cluster (Cantner and Graf, 2006; Hendry and Brown, 2006; Balland, 2012). In sum, the contributions of EEG to the topic of spatial knowledge networks are, so far, the following: (i) knowledge is not ‘in the air’, but channeled through networks that are uneven and selective in clusters; (ii) networks are selective, because firms and other agents have different capabilities and routines; (iii) various proximities, including geographical proximity, are important drivers of network formation, but proximities do not necessarily increase the performance of firms; (iv) while geographical and institutional proximity may drive network tie formation in clusters, not all cluster firms will connect and perform equally, despite being part of the same local institutional environment; (v) network relations in clusters have a tendency to become more inward-​looking over time; (vi) non-​local linkages, or temporary proximities, are crucial for the competitiveness of cluster firms, but they require other forms of proximities to enable effective transmission of knowledge.

Regional Development as an Evolutionary Process EEG has devoted attention to how regions can secure their long-​term development by developing new industries or new growth paths. A source of inspiration has been Schumpeter’s (1912) description of innovations as new combinations. This has been further developed in the notion of recombinant innovations, which emerge from recombining parts of pre-​ existing technologies or services in new way’s (Fleming, 2001). When recombinant innovations are the rule rather than the exception, this implies that the existing variety in a region conditions the scope for innovation. This builds on the seminal work by Jacobs (1969, p. 59), who argued that ‘the greater the sheer numbers and varieties of divisions of labor already achieved in an economy, the greater the economy’s inherent capacity for adding still more kinds of goods and services. Also the possibilities increase for combining the existing divisions of labor in new ways’. This idea was taken up by Glaeser et al. (1992), who tested whether diversified or specialized regions tend to grow more. Diversified regions should be more innovative owing to Jacobs’ externalities, while specialized regions could benefit from Marshallian externalities. Glaeser’s study was followed by many others, but, despite massive empirical efforts, there is conflicting evidence for both hypotheses: there are almost as many studies proving that regions benefit from variety as there are studies showing that regions benefit from specialization (De Groot et al., 2009, 2015). A possible reason for the weak evidence on Jacobs’ externalities is that many technologies and services cannot be meaningfully combined. Rather, one expects that recombinant innovations more often stem from related industries that share similar knowledge and skills. Frenken et al. (2007, p. 687) argued that for variety to be supportive in innovation processes, variety must be related (i.e. cognitively close), as related variety ‘improves the opportunities to interact, copy, modify, and recombine ideas, practices and technologies across industries giving rise to Jacobs externalities’. This motivated studies to test whether related variety increases regional employment growth. The evidence collected so far indicates by and large a

220   Boschma and Frenken positive effect of related variety on employment growth (Essletzbichler, 2007; Frenken et al., 2007; Quatraro, 2010), especially in knowledge-​intensive industries (Bishop and Gripiaos, 2010; Hartog et al., 2012). The question of new industry formation is associated with the concept of related variety. Frenken and Boschma (2007) depicted local industry formation as a branching process in which the local presence of industries that are related to a new industry increases the probability for a new industry to occur, given that related industries provide the main source for knowledge, capabilities, and potential entrepreneurs (Klepper, 2007). The more related the variety of industries is vis-​à-​vis the new industry, the more likely a region can be successful in that new industry. Hence, the existing set of industries conditions the likelihood of new industries emerging, and in that sense there exists ‘regional path dependence’ (Iammarino, 2005; Martin and Sunley, 2006; Fornahl and Guenther, 2010). Empirically, the branching phenomenon has been analysed at the level of countries by Hidalgo et al. (2007), who demonstrated that countries tend to develop new export products that are related in ‘product space’ with existing export products. The product space specifies the relatedness between products based on the frequency of co-​occurrence of products in countries’ export portfolios. The same reasoning has been applied to understand the development of regions becoming active in new markets. Neffke et al. (2011) found that an industry had a higher probability of entering a region when technologically related to pre-​existing industries in that region. Studies have confirmed relatedness driving regional diversification in new industries (Boschma et al., 2013; Essletzbichler, 2015), new technologies (Kogler et al., 2013; Rigby, 2015), and new eco-​technologies (Tanner, 2014; van den Berge and Weterings, 2014). What these studies tend to show is that related diversification in regions is the rule and unrelated diversification the exception. That unrelated diversification is a more rare event does not come as a surprise, as it is more uncertain and risky to recombine previously unrelated domains. It is a crucial question whether regions can keep relying on related diversification to sustain long-​term development, or whether they need to diversify in unrelated activities now and then. Studies have started investigating the conditions that make regions more likely to diversify into unrelated activities. Castaldi et al. (2015) found that unrelated variety is associated with high rates of breakthrough innovations in US states. In the rare cases that recombination innovations between unrelated technologies or services succeed, they become related (Desrochers and Leppälä, 2011). Such a radical new combination not only opens up new markets and innovation opportunities, but it might also provide the basis for long-​lasting competitive advantage, as other regions will face difficulties in coping with such radical change. A similar issue is analysed in the expanding literature on new growth paths (Garud et al., 2010) in which new path creation is defined as the emergence of entirely new sectors or products, while path renewal occurs when local activities switch to different but related activities (Isaksen and Trippl, 2014). To break with path dependence and create new growth paths, regions will have to rely more on knowledge and resources residing in other regions. Hence, the presence of multinationals, the immigration of entrepreneurs, and a targeted government policy are all elements that come into play in explaining new path creation (Binz et al., 2013; Dawley, 2014; Neffke et al., 2017). The contributions of EEG on regional development so far are: (i) related variety is a key concept in EEG that has shed new light on the MAR (Marshall–​Arrow–​Romer) versus Jacobs’ externalities debate—​there is emerging evidence of positive externalities stemming

Evolutionary Economic Geography    221 from the co-​presence of firms in related industries; (ii) EEG has shed light on how regions diversify over time—​regional development is depicted as a branching phenomenon in which new recombinations stem from related activities that share similar knowledge and skills, and in which local capabilities in existing industries or technologies conditions the set of industries and technologies that are more likely to emerge; (iii) unrelated diversification, recombining previously unrelated fields, is expected to be a more rare event, and tends to rely more on the inflow of resources and capabilities from other regions.

Institutions and Evolutionary Economic Geography A recurrent critique to evolutionary scholars in economic geography has been the perceived neglect of the role of institutions in firm behaviour and economic development processes (MacKinnon et al., 2009). This critique is understandable given that many empirical studies in EEG did not pay explicit attention to the institutional contexts in which economic processes take place, or ‘bracketed’ such processes in dummy variables or ‘fixed effects’. However, this relative empirical neglect says little about the theoretical possibilities to integrate institutional analysis into the EEG framework. EEG has engaged at length with the question of how institutional and evolutionary approaches can be combined (see e.g. Boschma and Lambooy, 1999; Boschma, 2004; Boschma and Frenken, 2009; Martin, 2010). Institutions provide incentives, but they may also form obstacles that make the development of some industries and organizational practices in some places more feasible (Malmberg and Maskell, 2010). Institutions are depicted as co-​evolving with new technologies and markets that are deemed crucial for the development of new industries (Nelson, 1994). The chances for new industry formation in a region depend on the timing and direction in which institutions are adapted in a way that supports the industry’s further development (Murmann, 2003). This requires more understanding of the conditions that favour or hamper institutional change in regions. In an attempt to come to a theory of institutional change and new industry formation, Battilana et al. (2009) argued that conditions supportive of institutional change are a common sense of urgency (e.g. due to a crisis), institutional contradictions and discontent (e.g. as new industries challenge existing categorizations), and a low degree of past institutionalization. These conditions may have a strong regional dimension, suggesting that regions are not all equally likely to engage in effective institutional change. EEG has also made progress in taking up explicitly the role of institutions in recent empirical work. There is an increasing attention paid to how local agents (private and public) engage in collective action to mobilize knowledge, resources, and public opinion as to create new or adapt existing institutions, how vested interests may be circumvented, and the key role that both regional and national governments can play in regional economic development (Feldman et al., 2005; Strambach, 2010; Sotarauta and Pulkkinen, 2011; Binz et al., 2013). In quantitative studies, the role of institutions has been highlighted as well. In their study on the local entry dynamics of fashion designers across the world, Wenting and Frenken (2011) could show that the institutional environment in Paris blocked the starting

222   Boschma and Frenken up of new firms in commercial design owing to strict regulations in Paris’s design profession, while designers in other cities did not experience such obstacles. A recent study by Boschma and Capone (2015), building on the literature on varieties of capitalism (Hall and Soskice, 2001), found that institutions associated with ‘liberal market economies’ give countries considerably more freedom to diversify in more unrelated activities than institutions associated with ‘coordinated market economies’, thus shedding light on different development logics channelled by national institutional environments. An issue that has remained little elaborated is how the concept of innovation system can be integrated in EEG. The regional innovation system (RIS) literature has explained the clustering of innovative activities by focusing on the nature of relationships between organizations such as firms, governments, universities, and non-​governmental organizations that are involved in the innovation process at the sub-​national level (Cooke, 1992; Asheim and Isaksen, 1997). Having strong evolutionary roots (Freeman, 1987), this approach has drawn attention to the importance of localized capabilities for the production and transmission of tacit knowledge (Asheim and Gertler, 2005). Given the path-​dependent nature of building up localized capabilities, as embodied in local knowledge bases and institutions, it is hard for regions to imitate ‘constructed regional advantages’ from successful regions (Asheim et al., 2011). Recently, scholars have expressed a need to go beyond a static approach that maps actors and institutions in a RIS, and to concentrate more on how RIS change in response to globalization, technological change, and societal challenges. This necessitates an understanding of how changes in RIS are initiated and implemented by agents, how changes in institutions are activated, and how relations at multiple spatial scales are constructed, managed, and utilized. Recently, there is increasing attention to the global dynamics that underlie the early formation of new innovation systems (Binz et al., 2013, 2014). The understanding of where new technologies emerge requires not only insight into the local mechanisms of capability transfer from related technologies to emerging ones, but also into the organization of knowledge production and regulatory processes at the national and international level (Morrison and Cusmano, 2015). EEG has a particular take on the role of institutions: (i) the influence of (local) institutions is contingent given the existence and persistence of heterogeneity of firms in the same institutional context; (ii) institutions have an effect on the intensity and nature of interactions between agents in RIS, and therefore they affect the process of regional diversification; (iii) new industry formation is depicted as co-​evolving with the establishment of new institutions or the adaptation of existing ones; (iv) local agents engage in collective action to create new or adapt existing institutions and challenge vested interests that may oppose such change; (v) regions may differ in their ability to induce required institutional change; (vi) there is still little understanding of what conditions at various spatial scales support or hamper institutional change.

Final Remarks This chapter has given a brief outline of recent theoretical and empirical contributions of EEG with respect to clustering, networking, urban and regional development, and the role

Evolutionary Economic Geography    223 of institutions. Give the limited space available, this outline has been partial at best. Studies in EEG have provided new but often still preliminary answers to old enduring questions in economic geography, and also bring up new questions and problems not yet explored. Although advancing, the empirical literature on EEG is work in progress. This also applies to the development of its main concepts like institutions (MacKinnon et al., 2009), path dependence (Martin, 2010), life cycle (Martin and Sunley, 2011), development (Martin and Sunley, 2015), and the use of appropriate methodologies that do justice to evolutionary principles (Hassink et al., 2014). Recently, EEG has been moving into topics like regional resilience (Simmie and Martin, 2010; Boschma, 2015b), the geography of transition (Truffer and Coenen, 2012; Patchell and Hayter, 2013), and public policy and governance, especially in the context of the smart specialization debate (Foray, 2015; McCann and Ortega-​Argiles, 2015). An EEG approach on resilience aims to leave behind an equilibrium perspective that is primarily interested in a recovery to pre-​existing or new equilibrium states, and instead proposes an evolutionary perspective in which regional resilience is redefined and analysed in terms of the impact of shocks on the capacity of regions to develop new growth paths (Boschma, 2015b). The newly emerging literature of the geography of transition (Truffer and Coenen 2012) is also promising, as it provides a comprehensive view on niche formation and the role of collective agents that has not yet been fully developed in EEG. In turn, EEG provides concepts like regional branching that describe how regions move into new green technologies, and why some regions are more successful in doing so (Tanner, 2014; van den Berge and Weterings, 2014). The smart specialization literature explores how the entrepreneurial discovery process can be organized and governed to make regions move in new specializations. To the main adherents of smart specialization strategies, related variety and regional branching, are regarded as key building blocks of smart specialization policies (Foray, 2015; McCann and Ortega-​Argiles, 2015). Without doubt, this future work on smart specialization will boost the further development of regional innovation policy in an EEG framework that is far from fully developed (Coenen et al., 2015). These continuing debates and emerging topics bear testimony to the fact that EEG is alive and evolving, and they will surely make EEG develop further into a more comprehensive approach in economic geography.

References Aldrich, H.E. and Fiol, C.M. (1994). ‘Fools rush in? The institutional context of industry creation’. Academy of Management Review 19: 645–​670. Appold, S.J. (2005). ‘Location patterns of US industrial research: mimetic isomorphism and the emergence of geographic charisma’. Regional Studies 39: 17–​39. Asheim, B.T. and Gertler, M. (2005). ‘The Geography of Innovation:  Regional Innovation Systems’ in J. Fagerberg, D. Mowery, and R. Nelson (eds) The Oxford Handbook of Innovation, pp. 291–​317 (Oxford: Oxford University Press). Asheim, B. and Isaksen, A. (1997). ‘Location, agglomeration and innovation: towards regional innovation systems in Norway?’ European Planning Studies 5: 299–​330. Asheim, B.T. and Isaksen, A. (2002). ‘Regional innovation systems. The integration of local “sticky” and global “unbiquitous” knowledge’. Journal of Technology Transfer 27: 77–​86.

224   Boschma and Frenken Asheim, B.T., Boschma, R., and Cooke, P. (2011). ‘Constructing regional advantage:  platform policies based on related variety and differentiated knowledge bases’. Regional Studies 45: 893–​904. Balland, P.A. (2012). ‘Proximity and the evolution of collaboration networks: evidence from research and development projects within the Global Navigation Satellite System (GNSS) Industry’. Regional Studies 46: 741–​756. Balland, P.A., Boschma, R., and Frenken, K. (2015). ‘Proximity and innovation: from statics to dynamics’. Regional Studies 49: 907–​920. Balland, P.A., de Vaa, M., and Boschma, R. (2013). ‘The dynamics of interfirm networks along the industry life cycle: the case of the global video game industry, 1987–​2007’. Journal of Economic Geography 13: 741–​765. Bathelt, H., Malmberg, A., and Maskell, P. (2004). ‘Clusters and knowledge. Local buzz, global pipelines and the process of knowledge creation’. Progress in Human Geography 28: 31–​56. Battilana, J., Leca, B., and Boxenbaum, E. (2009). ‘How actors change institutions: towards a theory of institutional entrepreneurship’. The Academy of Management Annals 3: 65–​107. Belussi, F. and Sedita, S.R. (2009). ‘Life cycle versus multiple path dependency in industrial districts’. European Planning Studies 17: 505–​528. Binz, C., Truffer, B., and Coenen, L. (2014). ‘Why space matters in technological innovation systems—​ mapping global knowledge dynamics of membrane bioreactor technology’. Research Policy 43: 138–​155. Binz, C., Truffer, B., Li, L., Shi, Y. and Lu, Y. (2013). ‘Conceptualizing leapfrogging with spatially coupled innovation systems—​the case of onsite wastewater treatment in China’. Technological Forecasting and Social Change 79: 155–​171. Bishop, P. and Gripaios, P. (2010). ‘Spatial externalities, relatedness and sector employment growth in Great Britain’. Regional Studies 44: 443–​454. Boschma, R.A. (2004). ‘The competitiveness of regions from an evolutionary perspective’. Regional Studies 38: 1001–​1014. Boschma R.A. (2005). ‘Proximity and innovation. A  critical assessment’. Regional Studies 39: 61–​74. Boschma, R. (2015a). ‘Do spinoff dynamics or agglomeration externalities drive industry clustering? A reappraisal of Steven Klepper’s work’. Industrial and Corporate Change 24: 859–​873. Boschma, R. (2015b). ‘Towards an evolutionary perspective on regional resilience’. Regional Studies 49: 733–​751. Boschma, R. and Capone, G. (2015). ‘Institutions and diversification: related versus unrelated diversification in a varieties of capitalism framework’. Research Policy 44: 1902–​1914. Boschma, R.A. and Frenken, K. (2003). ‘Evolutionary economics and industry location’. Review for Regional Research 23: 183–​200. Boschma, R.A. and Frenken K. (2006). ‘Why is economic geography not an evolutionary science? Towards an evolutionary economic geography’. Journal of Economic Geography 6: 273–​302. Boschma, R.A. and Frenken, K. (2009). ‘Some notes on institutions in evolutionary economic geography’. Economic Geography 85: 151–​158. Boschma, R.A. and Frenken K. (2010). ‘The Spatial Evolution of Innovation Networks. A Proximity Perspective’ in R. Boschma and R. Martin (eds) Handbook on Evolutionary Economic Geography, pp. 120–​135 (Cheltenham: Edward Elgar). Boschma, R.A. and Lambooy, J.G. (1999). ‘Evolutionary economics and economic geography’. Journal of Evolutionary Economics 9: 411–​429.

Evolutionary Economic Geography    225 Boschma, R. and Martin, R. (eds) (2010). The Handbook of Evolutionary Economic Geography (Cheltenham: Edward Elgar). Boschma, R.A. and Wenting, R. (2007). ‘The spatial evolution of the British automobile industry: does location matter?’ Industrial and Corporate Change 16: 213–​238. Boschma, R., Minondo, A., and Navarro, M. (2013). ‘The emergence of new industries at the regional level in Spain. A  proximity approach based on product-​relatedness’. Economic Geography 89: 29–​51. Brenner, T. (2004). Local Industrial Clusters. Existence, Emergence and Evolution (London: Routledge). Breschi S. and Lissoni F. (2009). ‘Mobility of skilled workers and co-​invention networks: an anatomy of localized knowledge flows’. Journal of Economic Geography 9: 439–​468. Broekel, T. and Boschma, R. (2012). ‘Knowledge networks in the Dutch aviation industry: the proximity paradox’. Journal of Economic Geography 12: 409–​433. Buenstorf, G. and Klepper, S. (2009). ‘Heritage and agglomeration: the Akron tyre cluster revisited’. The Economic Journal 119: 705–​733. Cantner, U. and Graf, H. (2006). ‘The network of innovators in Jena. An application of social network analysis’. Research Policy 35: 463–​480. Castaldi, C., Frenken, K., and Los, B. (2015). ‘Related variety, unrelated variety and technological breakthroughs. An analysis of US state-​level patenting’. Regional Studies 49: 767–​781. Coenen, L., Moodysson, J., and Martin, H. (2015). ‘Path renewal in old industrial regions: possibilities and limitations of regional innovation policy’. Regional Studies 49: 850–​865. Cooke, P. (1992). ‘Regional innovation systems: competitive regulation in the New Europe’. Geoforum 23: 365–​382. Crescenzi, R., Gagliardi, L., and Iammarino, S. (2015). ‘Foreign multinationals and domestic innovation: intra-​industry effects and firm heterogeneity’. Research Policy 44: 596–​609. Dahl, M.S. and Sorenson, O. (2012). ‘Home sweet home: entrepreneurs’ location choices and the performance of their ventures’. Management Science 58: 1059–​1071. Dawley, S. (2014). ‘Creating new paths? Offshore wind, policy activism, and peripheral region development’. Economic Geography 90: 91–​112. De Groot, H.L.F., Poot, J., and Smit, M.J. (2009). ‘Agglomeration Externalities, Innovation and Regional Growth: Theoretical Perspectives and Meta-​analysis’ in R. Capello and P. Nijkamp (eds) Handbook of Regional Growth and Development Theories, pp. 256–​281 (Northampton, MA: Edward Elgar). De Groot, H.L.F., Poot, J., and Smit, M.J. (2015). ‘Which agglomeration externalities matter most and why’. Journal of Economic Surveys 30: 756–​782. Desrochers, P. and Leppälä, S. (2011). ‘Opening up the “Jacobs Spillovers” black box:  local diversity, creativity and the processes underlying new combinations’. Journal of Economic Geography 11: 843–​863. Dosi, G., Freeman, C., Nelson, R., Silverberg, G., and Soete, L. (1988). Technical Change and Economic Theory (London: Pinter Publishers). Economic Geography (2009). Special issue: ‘Yet another turn? The evolutionary project in economic geography’. 85: 119–​182. Essletzbichler, J. (2007). ‘Diversity, Stability and Regional Growth in the United States 1975–​ 2002’ in K. Frenken (ed.) Applied Evolutionary Economics and Economic Geography, pp. 203–​229 (Cheltenham: Edward Edgar). Essleztbichler, J. (2015). ‘Relatedness, industrial branching and technological cohesion in US metropolitan areas’. Regional Studies 49: 752–​766.

226   Boschma and Frenken Essletzbichler, J. and Rigby, D.L. (2010). ‘Generalized Darwinism and Evolutionary Economic Geography’ in R. Boschma and R. Martin (eds) Handbook on Evolutionary Economic Geography, pp. 43–​61 (Cheltenham: Edward Elgar). Feldman, M.P., Francis, J., and Bercovitz, J. (2005). ‘Creating a cluster while building a firm. Entrepreneurs and the formation of industrial clusters’. Regional Studies 39: 129–​141. Fleming, L. (2001). ‘Recombinant uncertainty in technological space’. Management Science 47: 117–​132. Foray, D. (2015). Smart Specialisation. Opportunities and Challenges for Regional Innovation Policy (London and New York: Routledge). Fornahl, D. and Guenther, C. (2010). ‘Persistence and change of regional industrial activities. The impact of diversification in the German machine tool industry’. European Planning Studies 18: 1911–​1936. Freeman, C. (1987). Technology Policy and Economic Policy:  Lessons from Japan (London: Pinter). Frenken, K. and Boschma, R.A. (2007). ‘A theoretical framework for evolutionary economic geography:  industrial dynamics and urban growth as a branching process’. Journal of Economic Geography 7: 635–​649. Frenken, K., Cefis, E., and Stam, E. (2015). ‘Industrial dynamics and clusters: a survey’. Regional Studies 49: 10–​27. Frenken, K., Van Oort, F.G., and Verburg, T. (2007). ‘Related variety, unrelated variety and regional economic growth’. Regional Studies 41: 685–​697. Garud, R., Kumaraswamy, A., and Karnoe, P. (2010). ‘Path dependence or path creation’. Journal of Management Studies 47: 760–​774. Giuliani, E. (2007). ‘The selective nature of knowledge networks in clusters: evidence from the wine industry’. Journal of Economic Geography 7: 139–​168. Giuliani, E. and Bell, M. (2005). ‘The micro-​determinants of meso-​level learning and innovation: evidence from a Chilean wine cluster’. Research Policy 34: 47–​68. Glaeser, E., Kallal, H.D., Scheinkman, J.A., and Shleifer, A. (1992). ‘Growth in cities’. Journal of Political Economy 100: 1126–​1152. Glasmeier A. (1991). ‘Technological discontinuities and flexible production networks: the case of Switzerland and the world watch industry’. Research Policy 20: 469–​485. Gluckler, J. (2007). ‘Economic geography and the evolution of networks’. Journal of Economic Geography 7: 619–​634. Grabher, G. (1993). ‘The Weakness of Strong Ties—​The Lock-​in of Regional Development in the Ruhr Area’ in G. Grabher (ed.) The Embedded Firm, pp. 255–​277 (London: Routledge). Grabher, G. (2006). ‘Trading routes, bypasses, and risky intersections:  mapping the travels of “networks” between economic sociology and economic geography’. Progress in Human Geography 30: 163–​189. Hall, P.A. and Soskice, D. (2001) (eds) Varieties of Capitalism: The Institutional Foundations of Comparative Advantage (Oxford: Oxford University Press). Hartog, M., Boschma, R., and Sotarauta, M. (2012). ‘The impact of related variety on regional employment growth in Finland 1993–​2006: high-​tech versus medium/​lowtech’. Industry and Innovation 19: 459–​476. Hassink, R. (2005). ‘How to unlock regional economies from path dependency? From learning region to learning cluster’. European Planning Studies 13: 521–​535. Hassink, R., Klaerding, C., and Marques, P. (2014). ‘Advancing evolutionary economic geography by engaged pluralism’. Regional Studies 48: 1295–​1307.

Evolutionary Economic Geography    227 Hendry, C. and Brown, J. (2006). ‘Dynamics of clustering and performance in the UK opto-​ electronics industry’. Regional Studies 40: 707–​725. Hidalgo, C.A., Klinger, B., Barabasi, A.L., and Hausmann, R. (2007). ‘The product space and its consequences for economic growth’. Science 317: 482–​487. Iammarino, S. (2005). ‘An evolutionary integrated view of regional systems of innovation. Concepts, measures and historical perspectives’. European Planning Studies 13: 497–​519. Iammarino, S. and McCann, P. (2006). ‘The structure and evolution of industrial clusters. Transactions, technology and knowledge spillovers’. Research Policy 35: 1018–​1036. Isaksen, A. and Trippl, M. (2014). ‘Regional industrial path development in different regional innovation systems: a conceptual analysis’. Papers in Innovation Studies, no. 2014/​17 (Lund: Lund University, CIRCLE). Jacobs, J. (1969). The Economy of Cities (New York: Vintage Books). Journal of Economic Geography (2007). Special issue: ‘Evolutionary economic geography’. 7: 537–​672. Journal of Economic and Social Geography (2015). Special issue:  ‘Globalisation, knowledge and institutional change. Towards an evolutionary perspective to economic development’. 106: 133–​219. Klepper, S. (2007). ‘Disagreements, spinoffs, and the evolution of Detroit as the capital of the U.S. automobile industry’. Management Science 53: 616–​631. Kogler, D.F. (2015). ‘Editorial: evolutionary economic geography—​theoretical and empirical progress’. Regional Studies 49: 705–​7 11. Kogler, D.F., Rigby, D.L., and Tucker, I. (2013). ‘Mapping knowledge space and technological relatedness in US cities’. European Planning Studies 21: 1374–​1391. MacKinnon, D., Cumbers, A. Pyke, A., Birch, K., and McMaster, R. (2009). ‘Evolution in economic geography. Institutions, political economy and adaptation’. Economic Geography 85: 129–​150. Maggioni, M.A. (2002). Clustering Dynamics and the Location of High-​Tech Firms (Heidelberg and New York: Physica-​Verlag). Malmberg, A. and Maskell, P. (2010). ‘An Evolutionary Approach to Localized Learning and Spatial Clustering’ in R Boschma and R. Martin (eds) The Handbook of Evolutionary Economic Geography, pp. 391–​405 (Cheltenham: Edward Elgar). Marshall, A. (1920). Principles of Economics (London: Macmillan). Martin, R. (2010). ‘Roepke lecture in economic geography—​rethinking regional path dependence: beyond lock-​in to evolution’. Economic Geography 86: 1–​27. Martin, R. and Sunley, P. (2006). ‘Path dependence and regional economic evolution’. Journal of Economic Geography 6: 395–​437. Martin, R. and Sunley, P. (2007). ‘Complexity thinking and evolutionary economic geography’. Journal of Economic Geography 7: 573–​601. Martin, R. and Sunley, P. (2011). ‘Conceptualizing cluster evolution:  Beyond the life cycle model?’ Regional Studies 45: 1299–​1318. Martin, R. and Sunley, P. (2015). ‘Toward a developmental turn in Evolutionary Economic Geography?’ Regional Studies 49: 712–​732. McCann, P. and Ortega-​Argiles, R. (2015). ‘Smart specialization, regional growth and applications to European Union Cohesion Policy’. Regional Studies 49: 1291–​1302. Menzel, M.-​P. and Fornahl, D. (2010). ‘Cluster life cycles. Dimensions and rationales of cluster evolution’. Industrial and Corporate Change 19: 205–​238. Morrison, A. (2008). ‘Gatekeepers of knowledge within industrial districts: who they are, how they interact’. Regional Studies 42: 817–​835.

228   Boschma and Frenken Morrison, A. and Cusmano, L. (2015). ‘Introduction to the special issue: globalisation, knowledge and institutional change. Towards an evolutionary perspective to economic development’. Journal of Economic and Social Geography 106: 133–​139. Murmann, J.P. (2003). Knowledge and Competitive Advantage. The Co-​evolution of Firms, Technology, and National Institutions (Cambridge: Cambridge University Press). Neffke, F., Hartog, M., Boschma, R., and Henning, M. (2017). ‘Agents of structural change. The role of firms and entrepreneurs in regional diversification’, Economic Geography, forthcoming. Neffke, F., Henning, M., and Boschma, R. (2011). ‘How do regions diversify over time? Industry relatedness and the development of new growth paths in regions’. Economic Geography 87: 237–​265. Nelson, R.R. (1994). ‘The co-​evolution of technology, industrial structure, and supporting institutions’. Industrial and Corporate Change 3: 47–​63. Nelson, R.R. and Winter S.G. (1982). An Evolutionary Theory of Economic Change (Cambridge, MA, and London: The Belknap Press). Patchell, J. and Hayter, R. (2013). ‘Environmental and evolutionary economic geography: time for EEG2?’ Geografiska Annaler: Series B, Human Geography 95: 1–​20. Porter, M.E. (1990). The Competitive Advantage of Nations (New York: Free Press). Potter, A. and Watts, H.D. (2011). ‘Evolutionary agglomeration theory:  increasing returns, diminishing returns, and the industry life cycle’. Journal of Economic Geography 11: 417–​455. Pouder, R. and St. John, C. (1996). ‘Hot spots and blind spots: geographic clusters of firms and innovation’. The Academy of Management Review 21: 1192–​1225. Powell W.W., Koput, K., and Smith-​Doerr, L. (1996). ‘Interorganizational collaboration and the locus of innovation: networks of learning in biotechnology’. Administrative Science Quarterly 41: 116–​145. Quatraro, F. (2010). ‘Knowledge coherence, variety and productivity growth: manufacturing evidence from Italian regions’. Research Policy 39: 1289–​1302. Regional Studies (2015). Special issue:  ‘Evolutionary economic geography:  theoretical and empirical progress’. 49: 705–​898. Rigby, D. (2015). ‘Technological relatedness and knowledge space: entry and exit of US cities from patent data’. Regional Studies 49: 1922–​1937. Rigby, D.L. and Brown, W.M. (2015). ‘Who benefits from agglomeration’. Regional Studies 49: 28–​43. Rigby, D.L. and Essletzbichler, J. (1997). ‘Evolution, process variety and regional trajectories of technological change in US manufacturing’. Economic Geography 73: 269–​284. Schumpeter, J.A. (1912). The Theory of Economic Development. An Inquiry Into Profits, Capital, Credit, Interest and the Business Cycle (Cambridge, MA: Harvard University Press). Simmie, J. and Martin, R. (2010). ‘The economic resilience or regions: towards an evolutionary approach’. Cambridge Journal of Regions, Economy and Society 3: 27–​43. Sorenson, O. and Audia, P.G. (2000). ‘The social structure of entrepreneurial activity: geographic concentration of footwear production in the United States, 1940–​1989’. American Journal of Sociology 106: 424–​462. Sorenson, O., Rivkin, J., and Fleming, L. (2006). ‘Complexity, networks and knowledge flow’. Research Policy 35: 994–​1017. Sotarauta, M. and Pulkkinen, R. (2011). ‘Institutional entrepreneurship for knowledge regions: in search of a fresh set of questions for regional innovation studies’. Environment and Planning C 29: 96–​112.

Evolutionary Economic Geography    229 Staber, U. and Sautter, B. (2011). ‘Who are we, and do we need to change? Cluster identity and life cycle’. Regional Studies 45: 1349–​1361. Stam, E. (2007). ‘Why butterflies don’t leave. Locational behavior of entrepreneurial firms’. Economic Geography 83: 27–​50. Storper, M. (1992). ‘The limits to globalization: technology districts and international trade’. Economic Geography 68: 60–​93. Storper, M. and Walker, R. (1989). The Capitalist Imperative: Territory, Technology, and Industrial Growth (Oxford and Cambridge, MA: Basil Blackwell). Strambach, S. (2010). ‘Path Dependence and Path Plasticity: The Co-​evolution of Institutions and Innovation—​the German Customized Business Software Industry’ in R.A. Boschma and R. Martin (eds) The Handbook of Evolutionary Economic Geography, pp. 406–​431 (Cheltenham: Edward Elgar). Stuart, T. and Sorenson, O. (2003). ‘The geography of opportunity: spatial heterogeneity in founding rates and the performance of biotechnology firms’. Research Policy 32: 229–​253. Suire, R. and Vicente, J. (2009). ‘Why do some places succeed when others decline? A social interaction model of cluster viability’. Journal of Economic Geography 9: 381–​404. Tanner, A.N. (2014). ‘Regional branching reconsidered: emergence of the fuel cell industry in European regions’. Economic Geography 90: 403–​427. Ter Wal, A.L.J. (2014). ‘The dynamics of the inventor network in German biotechnology: geographic proximity versus triadic closure’. Journal of Economic Geography 14: 589–​620. Ter Wal, A.L.J. and Boschma, R. (2011). ‘Co-​evolution of firms, industries and networks in space’. Regional Studies 45: 919–​933. Torre, A. (2008). ‘On the role played by temporary geographical proximity in knowledge transmission’. Regional Studies 42: 869–​889. Truffer, B. and Coenen, L. (2012). ‘Environmental innovation and sustainability transitions in regional studies’. Regional Studies 46: 1–​21. van den Berge, M. and Weterings, A. (2014). ‘Relatedness in eco-​technological development in European regions’. Papers in Evolutionary Economic Geography, no. 14.13 (Utrecht: Utrecht University). Vicente, J. and Suire, R. (2007). ‘Informational cascades vs. network externalities in locational choice: evidences of “ICT clusters” formation and stability’. Regional Studies 41: 173–​184. Wenting, R. and Frenken, K. (2011). ‘Firm entry and institutional lock-​in: an organizational ecology analysis of the global fashion design industry’. Industrial and Corporate Change 20: 1031–​1048.

Chapter 12

In st itu tions, G e o g ra ph y, and Ec onomi c  L i fe Meric S. Gertler Introduction The idea that institutions play a major role in shaping economic behaviour is now widely shared. The intellectual lineage underlying this idea goes back at least to Veblen (1899), although the long reign of neoclassical thought as the dominant paradigm within economics largely suppressed a wider appreciation for the role that institutions play in shaping economic behaviour. With the emergence of post-​Keynesian and other heterodox schools of economic thought, the place of institutions in economic analysis was gradually restored. Indeed, Veblen’s own work has enjoyed something of a revival in recent years (Davenzati and Pacella, 2014; Lawson, 2015). Moving closer to the mainstream of economics, the institutional turn within the past ten to fifteen years has produced a substantial body of work examining how institutions might help explain divergent growth and development paths across different national economies (Rodrik et al., 2004; Acemoglu et al., 2005; Rodrik, 2013). This work argues that institutions may well trump other more conventional drivers of economic prosperity, such as natural resource endowments, human capital, gains from trade, and new technologies. Going further still, others have argued that state institutions have created the very foundations of societal wealth, creating the basis for private-​sector profitability (Mazzucato, 2013). Within economic geography, the openness to ideas originating outside the neoclassical tradition has fostered a healthy and growing interest in the role of institutions at least since the mid-​1990s (Amin and Thrift, 1995; Gertler, 1997; Morgan, 1997; Maskell and Malmberg, 1999; Gertler, 2004; Storper and Rodríguez-​Pose, 2006; Boschma and Frenken, 2009; Gertler, 2010; Rodríguez-​Pose, 2013; Zukauskaite, 2013; Tomaney, 2014). The central idea emerging from this literature is that the competitive advantages (or disadvantages) and innovative potential associated with particular regional economies are shaped, in large part, by their distinctive institutional configurations and the unique ‘cultures’ of economic practice they foster.

Institutions, Geography, and Economic Life    231 Notwithstanding the increasingly widespread awareness of the pervasive impact of institutions on economic decision-​making and macroeconomic performance, important questions about the precise ways in which ‘institutions matter’ remain unresolved. This chapter provides a synoptic overview of the key issues by addressing a series of core questions. How do we understand institutions—​both formal and informal—​and the way they structure and shape economic activity? How does the approach to this question vary between contemporary economics and economic geography? How do institutions at different spatial scales shape economic activity, and how do they interact with one another? And how do institutions evolve and change over time? To what extent do they evolve along paths that can be predicted? This chapter concludes by suggesting some of the most pressing areas of contention, debate, and unfinished business. It critically examines the argument that national institutions may be losing their purchase in favour of both sub-​and supra-​national institutional forms.

Institutions and the Contours of Economic Life For decades, mainstream economics revolved around the neoclassical assertion that the market as a means for exchanging goods and services and for equilibrating supply and demand was a natural phenomenon. Markets were presented as a self-​evident and universal mechanism, whose ‘perfection’ was considered an ideal state no matter what the particularities of location or history might suggest: ‘an ethically colorless state of nature in human relations’, as one commentator has recently put it (Martin, 2016, p. 13). In this world view, intervention by government was justified only as a means to correct imperfections or failures, most typically in the form of externalities in which social costs (e.g. polluting behaviour) or benefits (e.g. public goods) exceed private costs or benefits. Other forms of government action were viewed with scepticism or worse, as they served only to distort or corrupt the inherent, natural purity of the market form. One of the most trenchant and comprehensive critiques of this view comes from Karl Polanyi (1944), whose signal contribution was to demonstrate that the market was, in fact, the product of deliberate, coordinated action by state and state-​like actors. It required sustained and concerted planning and support for its birth and continued existence. Polanyi famously observed: ‘Laissez-​faire was planned, planning was not’ (1944, pp. 139–​140), having demonstrated that there was, in fact, nothing natural about the market form. Rather, Polanyi argued that it was more properly regarded as an institution that was socially constructed, with deliberate effort and conscious planning, to produce particular kinds of outcomes. This debate over the origins of the market provides a fundamental insight into the nature of institutions and their effects on economic activity. Nevertheless, this begs the obvious question: What exactly do we mean by ‘institutions’, and how should we understand their influence on economic activity?

232   Gertler

What Are Institutions and What Do They Do? While a widely accepted definition of institutions remains elusive, the work of North (1990) has been particularly influential. North likens institutions to ‘rules of the game’ that shape and guide economic behaviour. Although there is a natural tendency to equate institutions with formally codified rules, that is, laws and regulations, as well as constitutionally defined rights (e.g. property rights)—​implicit in North’s concept is the notion that many of the rules that shape our economic behaviour are informal rather than formal. They reside more in the realm of socially shared norms, conventions, traditions, and routines (Storper, 1993). They underlie pervasive expectations and shared values (Gertler, 2004). Seen in this way, institutions are much closer to Veblen’s notion of ‘settled habits of thought’ (1919, p. 239). In this view, the primary function—​and power—​of institutions is to facilitate economic transactions. Property rights confer power of ownership—​an essential precursor to exchange. Labour market and industrial relations institutions shape practices around the hiring and firing of labour, and the character of the employment relation between owners and workers. Capital market institutions define how firms typically raise funds to finance their operations and growth, as well as the relationship between lenders and borrowers, the degree of integration across different ‘pillars’ of finance, and more. Competition and trading institutions shape the nature and extent of inter-​firm interaction, the degree of ownership concentration in particular industries, the rights of consumers, and the prevalence of foreign ownership in domestic economies. Understood this way, institutions comprise the scaffolding or ‘exoskeleton’ on which economic activity unfolds (Martin and Sunley, 2015). They make particular decisions and transactions easier to perform, and, at the same time, they discourage others. However, they do not determine outcomes directly: individual economic actors do so by asserting their own agency (Rantisi, 2002). Moreover, the relationship between formal and informal institutions is complex and nuanced. Informal conventions, attitudes, and norms become formalized in regulations and legislation, and in the structures and practices of regulatory organizations. In turn, these formal structures reinforce and reproduce informal rules and norms. Gertler (2004) documents this interaction with respect to the production, adoption, and use of advanced manufacturing technologies. This study revealed the profound variation in characteristic practices between German and North American (American and Canadian) users of such technologies, during a period in the 1990s and early 2000s when digitally controlled manufacturing systems were revolutionizing production methods in many industries. Such diverging practices and attitudes could be understood superficially by pointing to differences in national industrial ‘cultures’: production managers in German firms put a higher value on workforce training and machinery maintenance, were more inclined to involve their shop-​floor workers in technology acquisition and implementation decisions, and were more ‘patient’ in being willing to wait for such investments to pay back their original outlays. North American firms spent far less on training and maintenance, rarely—​if ever—​consulted workers when making technology purchases, and insisted on much shorter payback periods for their investments in new production technologies. This study revealed how these distinctive national industrial ‘cultures’ could instead be understood on a deeper level to stem from the underlying architecture of national

Institutions, Geography, and Economic Life    233 institutional structures. German labour and industrial relations institutions foster long-​ term worker tenure and minimal workforce turnover, which boosts the incentive to invest in workplace training. They have also produced a ‘co-​determination’ model of governance, in which workers (through their unions and works councils) have a formal voice in major decisions such as adoption of new technologies. German firms raise most of their capital through private markets and bank loans, rather than public markets. As a result, the pressure for short-​term payback or return on investment is much lower. Machinery is purchased with the expectation that it will be in place for longer periods of time, creating a stronger incentive for generous spending on maintenance. As a result, German users of advanced technologies were shown to be much more successful at exploiting the full potential of such production systems, achieving higher quality, fewer defects, and fewer disruptions due to breakdowns. In sum, both formal and informal institutions shape the behaviour of economic actors. Moreover, they do so in ways that are not always obvious to the actors themselves (Gertler, 2003). Indeed, the habits, routines, conventions, and norms that produce commonplace behaviours and practices are often deployed in unreflective ways, although their roots can be traced to prevailing architectures of national institutions.

Institutional Economics versus Institutional Economic Geography As noted at the outset of this chapter, the discipline of economics has taken something of an institutionalist turn itself. The leading proponents of this approach seek to identify those institutional factors that determine aggregate economic performance and economic development over time (Acemoglu et al., 2005; Rodrik et al., 2004; Rodrik, 2013). This literature also highlights particular ‘barriers’ to successful economic performance, in the form of specific institutions that are absent, only weakly developed, or influential in ways that undermine performance. Methodologically, this work stands in contrast to much of the work of institutional economic geographers as described above, in that it is almost exclusively national in focus, and largely confines its analysis to formal institutions. Moreover, it adopts either a comparative statics approach or a comparative longitudinal analysis of time-​series data by country, using econometric tools to identify the influence of particular institutions—​such as property rights, political stability, or the rule of law—​as new ‘factors of production’. Finally, it stands apart conceptually by focusing only on the impact that formal institutions have, in the aggregate, on economic behaviour and performance. In doing so, it affords no role for individual agency as asserted by economic actors (see Farole et al., 2011 for further discussion of this issue). In comparison, economic geographers have been interested in understanding more broadly how formal and informal institutions shape and constrain all forms of economic behaviour and practices, including production, investment, innovation, and organizational change, labour and capital markets, inter-​firm relations and learning, and more. They have tended to draw on more diverse conceptual foundations, including the work of sociologists, political scientists, management theorists, and heterodox economists.

234   Gertler In terms of method, much of the geographical work on institutions has been based on case studies using ethnographic approaches, in order to understand the influence of historical-​spatial specificity (see Barnes et al., 2007). And, as we shall see in the next section, geographers have tended to be more sensitive to questions of the relative importance of institutions operating at different spatial scales, and the potential interactions between them.

Institutions and Geography As the case study documented in Gertler (2004) demonstrates, while institutions serve to facilitate economic activity, they are themselves shaped by the distinctive historical and political evolution of particular geographically defined polities—​with nation states being the primary unit of analysis. Hence, no single, universal institutional forms exist: observable institutional architectures vary from one nation state to the next, and are the outcome of locally distinctive political and historical processes. Pushing this idea further, Christopherson (2002) has argued that these distinctive national architectures of institutional form play a major role in shaping the industrial strengths of particular nation states. She demonstrates how the capital market and labour market structures of a country like Germany have produced and reinforced its traditional strengths in mechanical engineering and advanced manufacturing. The collective impact of such institutions has promoted longer time horizons to guide decision-​making and reinforced stability of tenure in the employment relation, minimizing employee turnover and inter-​firm mobility of labour. Christopherson compares this to analogous institutional forms in the USA, where very different capital market institutions, including dominant public markets, a well-​developed venture capital system, and shorter time horizons, have shaped labour market practices that foster much shorter average lengths of tenure, much higher rates of employee turnover, and much more prevalent circulation of workers between firms. Christopherson argues that these conditions have been conducive to the emergence of American competitive advantages in digital media, entertainment, advertising, information and communications technology, software, and related activities. More generally, the idea that different nation states produce distinctive constellations of institutions, which, in turn, produce and reproduce particular economic strengths and define distinctive evolutionary trajectories over time, has figured prominently in the social sciences, at least since the early 1990s, with the emergence of the Regulation School (Boyer, 1990). It is also a dominant theme in the body of work examining national systems of innovation (Lundvall, 1992; Nelson, 1993), and in the national business systems literature (Whitley, 1999). However, this idea finds its strongest expression in the extensive literature on ‘varieties of capitalism’ (Hall and Soskice, 2001), in which a set of mutually reinforcing institutional forms combine to create coherent, complementary, distinctive, and self-​reproducing national constellations. This literature emphasizes two ideal typical varieties: the liberal market economies, or LMEs (US, UK, Canada, Australia), and the coordinated market economies, or CMEs (Germany, other Northern European nations, Japan).

Institutions, Geography, and Economic Life    235 Peck and Theodore (2007) provide a balanced and comprehensive critique of this approach, noting the challenge of fitting all national cases into the binary LME–​CME scheme—​especially recently emerging forms such as China, India, and Brazil. They also highlight the static nature of this approach, which emphasizes the forces that produce distinctive national systems, but which is seemingly incapable of theorizing how such systems change over time. One of the consequences is that there seems to be little scope for individual economic agents (firms, managers, workers) to assert their agency in any meaningful way. Finally, they point out that the varieties of capitalism approach appears to fetishize the national scale over all others. In particular, it does not seem to be able to accommodate significant variations in institutional forms between different regions of the same nation state. This last element of Peck and Theodore’s critique seems particularly problematic in the face of several well-​known and influential studies that accentuate the stark differences in regional institutional forms, and their consequences in terms of shaping divergent economic performance and trajectories. Saxenian’s (1994) classic comparison of Silicon Valley and Route 128 stands out in this regard, as does Grabher’s (1993) analysis of Germany’s Ruhr Valley (in contrast to Baden-​Württemberg or Bavaria), not to mention the huge volume of work documenting the distinctive nature of Italian regions such as Emilia-​Romagna relative to other parts of Italy (see e.g. Brusco, 1982; Trigilia and Burroni, 2009). Moreover, the large literature on regional innovation systems (Cooke et al., 2004; Asheim and Gertler, 2005; Asheim et al., 2011; Wolfe and Gertler, 2016) underscores the importance of distinctive sub-​ national constellations of institutions that, in turn, shape the innovation-​generating activities of economic actors. On this important question of the possibility of sub-​national variation in institutional forms, recent years have seen a blossoming of interest. This is long overdue, as an overview of this literature several years ago concluded: … any reconstituted institutional economic geography needs to have more ‘geography’ in it. That is, it needs to illuminate the processes by which institutions are produced and reproduced at a number of spatial scales, from the local to the national to the global, as well as promoting one’s understanding of how these institutions shape and constrain (but do not determine) economic action. At the same time, it should also address the vexing issue of interscalar institutional interaction. In other words, one needs to understand far better than is done currently exactly how institutional forms and the incentives they create at any one particular scale influence, are influenced by, and interact with, the institutional architectures that are erected at other geographical scales (Gertler, 2010, p. 6).

As a result, a new geography of institutional analysis has begun to emerge, one that explores the conditions under which distinctive regional forms are produced, and how these might interact with the national institutional framework within which they are embedded (Pike et al., 2015; Ebner, 2016; Grillitsch and Rekers, 2016). Addressing one of the most glaring lacunae in the ‘varieties’ literature, Zhang and Peck (2016) consider the nature of China’s economic institutions. In doing so, they adopt an explicitly spatial perspective, providing a detailed analysis of ‘the regional subspecies of Chinese capitalism’ (p. 73). In their insightful analysis, a key feature that differentiates one regional formation from another is its type and degree of transnational economic connections. In Guangdong, capital from Hong Kong has played a critical role in spurring development. In Chongqing, foreign direct investment by the likes of Cisco and HP, and the local prominence

236   Gertler of contract manufacturer Foxconn, which is deeply embedded in global supply chains, have led the way. Having successfully identified these regionally distinctive models, Zhang and Peck nevertheless conclude that ‘seeking explanations for their (re)production in the context of ongoing scalar transformations of capitalism … is a task that must remain for future work’. They offer only the most general of explanations at this point: ‘regional variability is a feature of all “national” capitalisms … every national model internalises particular configurations of uneven geographical development’. They opine further that ‘the challenge must be to theorize across scales, not to privilege, a priori, one scale of analysis over another’ (p. 73). Schröder and Voelzkow (2016) take up this question of exploring the interaction between institutions at different spatial scales. They argue that the influence of national institutions relative to regional institutions will vary by nation state, depending on whether or not a country has a federal system of government, as well as the overall strength and effectiveness of its national government. The stronger the degree of centralization, the lower the likelihood that particular regional-​institutional forms will diverge significantly from national institutions. However, federally organized nation states are more likely to support the emergence of significant inter-​regional variation in institutional architectures. Citing the work of Elbing et al. (2009) and Crouch et al. (2009), Schröder and Voelzkow document the mechanisms by which regional-​industrial clusters begin to evolve along pathways that are not consistent with the wider national institutional environment: Television and film production firms in Cologne invent new institutions when those of the German market economy lack the flexibility that their fast project-​based workflow requires. Media firms in London similarly build regional institutions for training and financing, generating more coordination than their embedding liberal market economy would lead one to assume (2006, p. 13).

In contrast to one of the central tenets of the varieties of capitalism approach, in which internal coherence and strong complementarity of institutions are assumed, Schröder and Voelzkow reach a different conclusion: (F)irms seem to be increasingly at liberty to diverge from a national regulatory framework; they are free, that is, to build up institutions locally that cater to sectoral needs … Insofar as such divergence from a national model is productive, one can speak of “productive incoherence” … (W)hen national institutions do not provide the support that regional economies need, liberal clusters spring up in coordinated countries and coordinated clusters spring up in liberal countries (2006, p. 13).

In their view, the initiative to chart such distinctive regional paths originates with groups of firms constituting a critical mass in a local industry, leading to ‘collective rule-​breaking’, and forming ‘an institutional island, in which regulation diverges from the embedding national regime’ (2006, p. 14). This conclusion is worth considering more fully, as it suggests that the driving forces for institutional change become more obvious when one adopts a regional (vs. national) lens. Indeed, Schröder and Voelzkow’s analysis raises an interesting question: Does national institutional change actually originate in collective action organized at the local and regional scale, or do such ‘institutional islands’ at the regional scale remain just that? What potential

Institutions, Geography, and Economic Life    237 do such phenomena have to transform the prevailing, dominant national-​level institutions? Moreover, this analysis leads to other key unanswered questions. How sustainable are such ‘productive incoherencies’ at the regional level over time? And can these institutional innovations spread from one region to another? If so, what are the conditions under which such institutional diffusion is more likely to occur? Pike et al. (2015) undertake a comprehensive study of local institutional change, examining the rollout of local enterprise partnerships (LEPs) by the national government in the UK since 2010. By adopting a comparative approach that examines institutional evolution across 39 LEPs, they are able to document the distinctive paths charted by each region, defined by local histories and social dynamics. Their consciously multi-​agent and multiscalar framework reveals much about the forces shaping institutional change, including the influence of local agency in charting evolutionary pathways. They elaborate a typology of institutional change processes, including ‘layering, converting and recombining as well as dismantling and improvising’ (2015, p. 201). On the question of how regional institutional change might shape national institutional architectures, Pike et al. (2015) conclude that, notwithstanding the emergence of substantial local variation, national institutions remain dominant. At the same time, they acknowledge that this finding may not be generalizable to other national settings, demonstrating the need for further international comparative work.

Institutional Change Over Time The above discussion provides a helpful segue to one of the most fundamental shortcomings of institutionalist analysis as noted by Peck and Theodore (2007) and others—​namely, its failure to theorize how institutions change over time. Not surprisingly, this glaring gap has begun to generate some thoughtful responses in the past few years. For example, Thelen and colleagues (Streeck and Thelen, 2005; Hall and Thelen, 2009)  have made important conceptual contributions in advancing our understanding of how institutions change over time. They document a range of different pathways to institutional change, in which the locus of the actions that instigate change may be, alternatively, inside or outside of government itself. In a way that helps guide the later work of Pike et al. (2015), they also describe a path-​dependent, evolutionary dynamic, in which new institutional forms are ‘layered’ on top of pre-​existing forms whose contours continue to exert influence on the directions in which those new forms may evolve over time. This last idea anticipates subsequent conceptual developments in economic geography—​ in particular, evolutionary economic geography (EEG). While the common starting place for institutionalist approaches is to demonstrate how ‘big structures’ shape and influence (although they do not determine) the behaviour of individual economic actors, Martin and Sunley (2015) see the relationship as two-​way. In their manifesto for a ‘developmental turn’, they advocate a grand synthesis of conceptual ideas from EEG, geographical political economy, and institutional economic geography (see also MacKinnon et al., 2009; Pike et al., 2015). In such a synthesis, institutional forms and the ‘big structures’ that figure prominently in political–​economic approaches would be viewed as subject to the same kind of evolutionary dynamics that help determine the fortunes of individual firms:

238   Gertler Not only do institutions of all kinds and at all scales condition, constrain and enable the operation of evolutionary mechanisms in the economy, but also these same institutions are themselves subject to similar such evolutionary mechanisms and processes: an economy and its institutional forms and arrangements co-​evolve. Institutions are both context and consequence of economic evolution (Martin and Sunley, 2015, p. 724, original emphasis).

Although Martin and Sunley (2015) suggest that we can understand institutional evolution as being shaped by core concepts of developmental evolutionary biology, such as homeostasis (mechanisms that foster stability) and adaptive plasticity (processes of change in response to new competitive, technological or regulatory features), they stop short of outlining exactly how this might work. Such questions are of more than conceptual interest, but have very real and practical implications. If institutions are incapable of significant change over time, then the very coherence and complementarity that provided the foundation for economic success in the past could become problematic when broader competitive conditions or technological paradigms emerge—​akin to the familiar notion of institutional ‘lock-​in’ from EEG. Indeed, it is precisely such questions that have propelled much of the interest in institutional change in recent years. The work of Thelen and colleagues cited earlier has been motivated, in large part, by existential concerns over the ‘future of the German model’ in the face of global competitive pressures. The dominant narrative predicts convergence between different national economic models over time towards the Anglo-​American variety of capitalism, a process driven by mobile capital and the global organization of production systems. Moreover, there is at least preliminary empirical evidence of such convergence tendencies emerging within capital market structures and practices (Clark and Wójcik, 2003; Wójcik, 2003). In response, Thelen, Streeck, and others working in this field have argued instead that the German variety of capitalism is more resilient than it may seem. While they document the gradual change in Germany’s economic and social institutions, they make the case that the end result will not be convergence so much as continued evolution along a distinctive path—​one that differs from the evolutionary paths of other nation states (see also Bathelt and Gertler, 2005). In making this argument, they have anticipated more recent work by Boschma (2015) at the regional scale. Their arguments are also consistent with the findings of a study of production methods, workplace organization, and associative interaction adopted by German firms operating in three different regions of North America (Gertler and Vinodrai, 2005), which finds little evidence of practices learned overseas being transposed back to operations in Germany.

Conclusions The analysis presented here confirms that institutionally inflected conceptual approaches have become widespread in recent years. Both economists and economic geographers have embraced the idea that institutions shape, constrain, and enable economic behaviour in characteristic and predictable ways. Furthermore, we know that the nature of this relationship will vary according to the precise institutional architecture of particular geographical

Institutions, Geography, and Economic Life    239 spaces. While much of the literature has focused on national-​level institutions, geographers, in particular, have shown a keen interest in the role and influence of regional institutions in shaping economic behaviour and performance. Economists have, for the most part, confined their analysis to formal institutions and how they shape aggregate outcomes at the level of nation states. While the geographical literature has recently turned its attention to the pressing question of how regional institutions interact with institutions of the nation state within which they are nested, the analysis thus far—​both conceptual and empirical—​remains partial, preliminary, and tentative in its conclusions. As noted a few years ago (Barnes et al., 2007; Gertler, 2010), the scarcity of rigorous comparative analysis within economic geography may explain the slow progress on this front. An equally pressing question pertains to the sources of institutional change—​both regional and national—​and the processes by which this occurs. Here, too, early musings remain just that. While the literature has tended to emphasize exogenous sources of institutional change, more recent work (Elbing et al., 2009; Hall and Thelen, 2009; Pike et al., 2015; Grillitsch and Rekers, 2016) has attempted to understand the way in which endogenously generated institutional change might work. Nevertheless, we are still far short of a comprehensive understanding of how and when institutions evolve—​or, to use the language proposed by Martin and Sunley (2015), what determines the relative influence of homeostasis versus adaptive plasticity in institutions. Nor has our understanding of the role and influence of individual agency developed much in the past few years (Gertler, 2010). What impact do the actions of individual agents—​workers, managers, firms, and other organizations—​have in determining economic outcomes, and to what extent do institutional structures accommodate such decisions and actions? How does the agency of individual actors shape institutional change, and which actors are most influential in this regard? While some (Schröder and Voelzkow, 2016)  have suggested that collections of firms may be responsible for catalysing institutional change, there does not yet appear to be much systematic analysis of the relative importance of other types of actors in effecting such change. Once again, comparative case studies have the potential to shed useful light on this issue, as the fine-​ grained, theoretically grounded work of Vinodrai (2013) and Pike et al. (2015) so clearly demonstrates. And while some would argue that the path and direction of institutional evolution at the macro level are quite predictable (read convergence towards an Anglo-​American norm), others view this process in a far more indeterminate way. While institutions in all nation states and regions evolve continuously, the rate, direction, and overall trajectory of this evolution may itself be shaped by inherited histories and politics. And as Gertler (2010) has pointed out, the capacity for variation in evolutionary paths between countries (or cities) situated within broadly similar macro-​institutional structures, and subject to similar macroeconomic and competitive conditions, should not be underestimated: Social, economic and geographical outcomes are contested and determined through the interplay between inherited national, provincial and local institutional structures, political processes, and the agency of individual actors such as municipal leaders, civic entrepreneurs, and community activists. And there remains a surprising degree of latitude at the local scale to modify existing institutions, create new institutions, and enact policies promoting progressive social change (2010, p. 11).

240   Gertler

References Acemoglu, D., Johnson, S., and Robinson, J.A. (2005). ‘Institutions as a Fundamental Cause of Long-​run Growth’ in P. Aghion, and S.N. Durlauf (eds) Handbook of Economic Growth, pp. 385–​472 (Amsterdam: North-​Holland). Amin, A. and Thrift, N.J. (1995). ‘Globalization, Institutional Thickness and the Local Economy’ in P. Healy, S. Cameron, and A. Davoudi (eds) Managing Cities: The New Urban Context, pp. 92–​108 (Chichester: Wiley). Asheim, B. and Gertler, M.S. (2005). ‘The Geography of Innovation’ in J. Fagerberg, D.C. Mowery, and R.R. Nelson (eds) The Oxford Handbook of Innovation, pp. 291–​317 (Oxford: Oxford University Press). Asheim, B., Smith, H.L., and Oughton, C. (2011). ‘Regional innovation systems: theory, empirics and policy’. Regional Studies 45: 875–​891. Barnes, T.J., Peck, J., Sheppard, E.S., and Tickell, A. (2007). ‘Methods Matter: Transformations in Economic Geography’ in A. Tickell, E.S. Sheppard, J. Peck, and T.J. Barnes (eds) Politics and Practice in Economic Geography, pp. 1–​24 (London: SAGE). Bathelt, H. and Gertler, M.S. (2005). ‘The German variety of capitalism: forces and dynamics of evolutionary change’. Economic Geography 81: 1–​9. Boschma, R. (2015). ‘Towards an evolutionary perspective on regional resilience’. Regional Studies 49: 733–​751. Boschma, R. and Frenken, K. (2009). ‘Some notes on institutions in evolutionary economic geography’. Economic Geography 85: 151–​158. Boyer, R. (1990). The Regulation School:  A  Critical Introduction (New  York:  Columbia University Press). Brusco, S. (1982). ‘The Emilian model:  productive decentralization and social integration’. Cambridge Journal of Economics 6: 167–​184. Christopherson, S. (2002). ‘Why do national labor market practices continue to diverge in the global economy? The “missing link” of investment rules’. Economic Geography 78: 1–​20. Clark, G.L. and Wójcik, D. (2003). ‘Path dependence and financial markets: the economic geography of the German model, 1997–​2003’. Environment and Planning A 37: 1769–​1791. Cooke, P., Heidenreich, M., and Braczyk, H.-​J. (eds). (2004). Regional Innovation Systems (2nd edition) (London: Routledge). Crouch, C., Schröder, M., and Voelzkow, H. (2009). ‘Conclusions:  Local and Global Sources of Capitalist Diversity’ in C. Crouch and H. Voelzkow (eds) Innovation in Local Economies:  Germany in Comparative Context, pp. 169–​188 (Oxford:  Oxford University Press). Davenzati, G.F. and Pacella, A. (2014). ‘Thorstein Veblen on credit and economic crises’. Cambridge Journal of Economics 38: 1043–​1061. Ebner, A. (2016). ‘Editorial: exploring regional varieties of capitalism’. Regional Studies 50: 3–​6. Elbing, S., Glassmann, U., and Crouch, C. (2009). ‘Creative local development in Cologne and London film and TV production’ in C. Crouch and H. Voelzkow (eds) Innovation in Local Economies: Germany in Comparative Context, pp. 139–​168 (Oxford: Oxford University Press). Farole, T., Rodríguez-​Pose, A., and Storper, M. (2011). ‘Human geography and the institutions that underlie economic growth’. Progress in Human Geography 35: 58–​80. Gertler, M.S. (1997). ‘The invention of regional culture’ in R. Lee and J. Wills (eds) Geographies of Economies, pp. 47–​58 (London: Arnold).

Institutions, Geography, and Economic Life    241 Gertler, M.S. (2003). ‘Tacit knowledge and the economic geography of context, or the undefinable tacitness of being (there)’. Journal of Economic Geography 3: 75–​99. Gertler, M.S. (2004). Manufacturing Culture: The Institutional Geography of Industrial Practice (Oxford: Oxford University Press). Gertler, M.S. (2010). ‘Rules of the game: the place of institutions in regional economic change’. Regional Studies 44: 1–​15. Gertler, M.S. and Vinodrai, T. (2005). ‘Learning from America? Knowledge flows and industrial practices of German firms in North America’. Economic Geography 81: 31–​52. Grabher, G. (1993). ‘The Weakness of Strong Ties’ in G. Grabher (ed.) The Embedded Firm, pp. 255–​277 (London: Routledge). Grillitsch, M. and Rekers, J.V. (2016). ‘How does multi-​scalar institutional change affect localized learning processes? A  case study of the med-​tech sector in Southern Sweden’. Environment and Planning A 48: 154–​171. Hall, P.A. and Soskice, D. (eds.) (2001). Varieties of Capitalism (Oxford: Oxford University Press). Hall, P.A. and Thelen, K. (2009). ‘Institutional change in varieties of capitalism’. Socio-​Economic Review 7: 7–​34. Lawson, T. (2015). ‘Process, order and stability in Veblen’. Cambridge Journal of Economics 39: 993–​1030. Lundvall, B-​Å. (1992). National Systems of Innovation: Towards a Theory of Innovation and Interactive Learning (London: Frances Pinter). MacKinnon, D., Cumbers, A., Pike, A., Birch, K., and McMaster, R. (2009). ‘Evolution in economic geography:  institutions, political economy, and adaptation’. Economic Geography 85: 129–​150. Martin, F. (2016). ‘History’s trust fund’. New York Times Book Review, 26 June. Maskell, P. and Malmberg, A. (1999). ‘Localized learning and industrial competitiveness’. Cambridge Journal of Economics 23: 167–​185. Martin, R. and Sunley, P. (2015). ‘Towards a developmental turn in evolutionary economic geography?’ Regional Studies 49: 713–​732. Mazzucato, M. (2013). The Entrepreneurial State:  Debunking Public vs Private Sector Myths (London: Anthem). Morgan, K. (1997). ‘The learning region:  institutions, innovation and regional renewal’. Regional Studies 31: 491–​503. Nelson, R.R. (1993). National Innovation Systems: A Comparative Analysis (New York: Oxford). North, D.C. (1990). Institutions, Institutional Change and Economic Performance (Cambridge: Cambridge University Press). Peck, J. and Theodore, N. (2007). ‘Variegated capitalism’. Progress in Human Geography 31: 731–​772. Pike, A., Marlow, D., McCarthy, A., O’Brien, P., and Tomaney, J. (2015). ‘Local institutions and local economic development:  the Local Enterprise Partnerships in England, 2010–​’. Cambridge Journal of Regions, Economy and Society 8: 185–​204. Polanyi, K. (1944). The Great Transformation (Boston, MA: Beacon). Rantisi, N.M. (2002). ‘The competitive foundations of localized learning and innovation: the case of women’s garment production in New York City’. Economic Geography 78: 441–​462. Rodríguez-​Pose, A. (2013). ‘Do institutions matter for regional development?’ Regional Studies 47: 1034–​1047. Rodrik, D. (2013). ‘Roepke lecture in economic geography—​who needs the nation state?’ Economic Geography 89: 1–​19.

242   Gertler Rodrik, D., Subramanian, F., and Trebbi, F. (2004). ‘Institutions rule: the primacy of institutions over geography and integration in economic development’. Journal of Economic Growth 9: 131–​165. Saxenian, A. (1994). Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA: Harvard University Press). Schröder, M. and Voelzkow, H. (2016). ‘Varieties of regulation:  how to combine sectoral, regional and national levels’. Regional Studies 50: 7–​19. Storper, M. (1993). ‘Regional “worlds” of production: learning and innovation in the technology districts of France, Italy and the USA’. Regional Studies 27: 43355. Storper, M. and Rodríguez-​Pose, A. (2006). ‘Better rules or stronger communities? On the social foundations of institutional change and its economic effects’. Economic Geography 82: 1–​25. Streeck W. and Thelen K. (2005). ‘Introduction: Institutional Change in Advanced Capitalist Economies’ in W. Streeck and K. Thelen (eds) Beyond Continuity: Institutional Change in Advanced Capitalist Economies, pp. 1–​39 (Oxford: Oxford University Press). Tomaney, J. (2014). ‘Region and place I: institutions’. Progress in Human Geography 38: 131–​140. Trigilia, C. and Burroni, L. (2009). ‘Italy: rise, decline and restructuring of a regionalized capitalism’. Economy and Society 38: 630–​653. Veblen, T.B. (1899). The Theory of the Leisure Class: An Economic Study of Institutions (New York: Macmillan). Veblen, T.B. (1919). The Place of Science in Modern Civilization and Other Essays (New York: B.W. Huebsch). Vinodrai, T. (2013). ‘Design in a downturn? Creative work, labour market dynamics and institutions in comparative perspective’. Cambridge Journal of Regions, Economy and Society 6: 159–​176. Whitley, R. (1999). Divergent Capitalisms:  The Social Structuring and Change of Business Systems (Oxford: Oxford University Press). Wójcik, D. (2003). ‘Change in the German model of corporate governance:  evidence from blockholdings 1997–​2001’. Environment and Planning A 35: 1431–​1458. Wolfe, D.A. and Gertler, M.S. (eds.) (2016). Growing Urban Economies: Innovation, Creativity and Governance in Canadian City-​Regions (Toronto: University of Toronto Press). Zhang, J. and Peck, J. (2016). ‘Variegated capitalism, Chinese style: regional models, multi-​ scalar constructions’. Regional Studies 50: 52–​78. Zukauskaite, E. (2013). ‘Institutions and the geography of innovation: a regional perspective’, PhD thesis (Lund: Faculty of Social Sciences, Lund University).

Pa rt  I I I

I N N OVAT ION

Chapter 13

Ec onomic Ec o syst e ms Philip E. Auerswald and Lokesh Dani Introduction ‘The Mecca of the economist lies in economic biology rather than in economic dynamics’, the great economist Alfred Marshall famously wrote in 1920, in the preface to the eighth edition of Principles of Economics. ‘But biological conceptions are more complex than those of mechanics’, he continued, ‘a volume on Foundations must therefore give a relatively large place to mechanical analogies; and frequent use is made of the term “equilibrium,” which suggests something of statistical analogy’ (Marshall, 1920, p. 19). Inspired by these words of Marshall’s and the work of other foundational figures in the field of economics who similarly perceived a fundamentally biological order in the evolution of the economy,1 economists have for decades sought to represent the adaptive dynamics evident in economic systems.2 A second celebrated passage in Marshall’s Principles relates to the localization of economic activity: ‘When an industry has thus chosen a locality for itself, it is likely to stay there long: so great are the advantages which people following the same skilled trade get from near neighbourhood to one another. The mysteries of the trade become no mysteries; but are as it were in the air, and children learn many of them unconsciously.’ Emphasizing the central role of invention and innovation in geographical localization, Marshall continues, ‘Good work is rightly appreciated, inventions and improvements in machinery, in processes and the general organization of the business have their merits promptly discussed: if one man starts a new idea, it is taken up by others and combined with suggestions of their own; and thus it becomes the source of further new ideas’ (Marshall, 1920, p. 225). This observation similarly has inspired a now decades-​old literature within economics on the localization of economic activity, in general, and of inventive activity, in particular. Economists have only recently begun to connect these two Marshall-​inspired literatures, studying localized systems of innovation as ‘ecosystems’ at the sub-​national or regional level.3 This new work on regional entrepreneurial ecosystems has been prompted equally by the advance of models and empirical methods to represent the adaptive evolution of ecosystems (Holling et al., 1995; Gavrilets, 1999, 2004; Holling, 2001), a discontinuous increase in the volume and quality of data available to economic geographers (Rosenthal and Strange, 2001, 2004; Wallsten, 2001; Auerswald et al., 2007), and strong interest among policymakers (Isenberg, 2010; Auerswald, 2015).

246   Auerswald and Dani In this chapter we review the concept of ecosystems as applied to economic geography. We define an economic ecosystem as a dynamically stable network of interconnected firms and institutions within bounded geographical space. We propose that concepts familiar to economic geographers, such as ‘clusters’ (Porter, 1990)  and ‘production networks’ (Piore and Sabel, 1984), are subsystems within regional economies, and further that representing regional economic networks as ‘ecosystems’ provides analytical structure and depth to otherwise mostly ad hoc theories of the sources of regional advantage, the role of entrepreneurs in regional development, and the determinants of resilience in regional economic systems. We frame regional economic change in terms of ecosystem dynamics, with reference to ecologically derived concepts of succession, speciation, diversity, resilience, and adaptation. We seek both to provide a summary of the scholarly discussion on economic ecosystems and to sketch directions for future research. In the next section we summarize the conceptual origins of the term ‘ecosystem’ and the concept of the ‘fitness landscape’ in evolutionary biology and explain their application to the study of economic ecosystems. In the section, ‘Unit of Analysis’, we introduce the concept of the ‘production algorithm’, which is analogous in economic ecosystems to the gene in biological ecosystems. In the fourth section, ‘Structure’, we summarize the structural characteristics of economic ecosystems, with a focus on the determination of ecosystem boundaries. In the fifth section, ‘Dynamics’, we describe the hypothesized dynamics of economic ecosystems, with particular emphasis on the systemic processes that lead to speciation and ecosystem-​scale life cycles. In the penultimate section, ‘Health’, we propose some potential definitions of the health of ecosystems in terms of their resilience, adaptive capacity, diversity, and entrepreneurial dynamism. We conclude with a discussion on the future potential and direction of economic ecosystems research.

Conceptual Origins The Definition of ‘Ecosystem’ In an 1857 essay titled ‘Progress: its Law and Cause’, Herbert Spencer argued that systems of all types—​natural and social—​tended to grow from simplicity to complexity through stages of differentiation, a process he characterized as ‘an advance from homogeneity of structure to heterogeneity of structure’ (Spencer, 1857, p. 234). Presenting an array of examples of such progression from simplicity to complexity, Spencer endeavoured to establish that the law of progress is the law of all progress. Whether it be in the development of the Earth, in the development of Life upon its surface, in the development of Society, of Government, of Manufactures, of Commerce, of Language, Literature, Science, Art, this same evolution of the simple into the complex, through a process of continuous differentiation, holds throughout (Spencer, 1857, p. 234).

Nearly six decades after the publication of ‘Progress: Its Law and Cause’, Sir Arthur Tansley published a paper entitled ‘The Use and Abuse of Vegetational Concepts and Terms’ in which he introduced the term ‘ecosystem’. Tansley’s insight was that dynamically stable networks of interconnected organisms and inorganic resources constituted their own distinct domain of analysis. Evolutionary biologists in the 1930s were as naturally inclined to place

Economic Ecosystems   247 ‘the organism’ at the centre of their inquiry as economists in the 1930s were to place ‘the firm’ at the centre of production theory, but Tansley rebelled against the application of the term ‘complex organism’ to describe dynamically stable networks of interconnected organisms and inorganic resources because ‘the term [“complex organism”] is already in common use for an individual higher animal or plant, and because the biome is not an organism except in the sense in which inorganic systems are organisms’. Accordingly, a new term was required. Tansley proposed the word ‘ecosystem’, which he defined as follows: It is the systems so formed which, from the point of view of the ecologist, are the basic units of nature on the face of the earth. Our natural human prejudices force us to consider the organisms (in the sense of the biologist) as the most important parts of these systems, but certainly the inorganic ‘factors’ are also parts—​there could be no systems without them, and there is constant interchange of the most various kinds within each system, not only between the organisms but between the organic and the inorganic. These ecosystems, as we may call them, are of the most various kinds and sizes. They form one category of the multitudinous physical systems of the universe, which range from the universe as a whole down to the atom (Tansley, 1935, p. 299).

As Tansley emphasized in defining the term, ecosystems come in a variety of sizes and scales determined by internal linkages and external boundaries. Tansley included the interactions between both the organic biome and the inorganic habitat in which these organisms live in determining the scope of the ecosystem. Accordingly, biological ecosystems can be as small as ponds, or as large as forests, and collections of ecosystems can be combined into higher-​ order systems. Nationally and globally, ecosystems are classified into a hierarchy of nested geographies of interacting networks (Bailey, 2009). In this chapter we elaborate on the proposition that the firm in an economic ecosystem is analogous to the organism in a biological ecosystem. We propose that economic ecosystems are characterized by interactions among densely interconnected firms, but that such ecosystems cannot reasonably be considered ‘complex firms’. Firms within an ecosystem are generally less tightly interconnected than subunits within a firm, but more tightly interconnected than atomistic entities reacting anonymously to price signals in a market.

Evolution as the Solution to a Search Problem At about the same time that Tansley was defining the ecosystem and establishing the basis for ecology as a field of study, fellow biologist Sewall Wright was setting the stage for the ‘modern synthesis’ in evolutionary biology: the systematic integration of the ‘micro’ genetics of combination and recombination first postulated by Gregor Mendel with the ‘macro’ theory of evolution first and most famously expounded by Charles Darwin, with significant contemporaneous contributions from Herbert Spencer that described changes in the character of populations over time. Sewall Wright’s work constituted a significant advance over Darwinian theory and offered a bridge from evolutionary biology to other domains of inquiry. Wright began his 1932 paper ‘The Roles of Mutation, Inbreeding, Crossbreeding, and Selection in Evolution’ by distinguishing between two mechanisms by which genetic novelty might be introduced into particular populations. The first was the one emphasized by Darwin:  single-​point mutation, which would constitute incremental change for an

248   Auerswald and Dani offspring as compared with a parent. The second was that emphasized by Mendel: sexual, or ‘bi-​parental’, reproduction, which would constitute large-​scale, combinatorial change for an offspring as compared with parents. Wright noted the fundamentally unsatisfactory nature of mutation as a sole explanatory factor in the evolutionary process: The observed properties of gene mutation—​fortuitous in origin, infrequent in occurrence and deleterious when not negligible in effect—​seem about as unfavorable as possible for an evolutionary process. Under bi-​parental reproduction, however, a limited number of mutations which are not too injurious to be carried by the species furnish an almost indefinite field of possible variations through which the species may work its way under natural selection (Wright, 1932, p. 356).

Wright supported his claim by describing the astronomical number of combinations of genes that are possible in higher organisms, in comparison with the linearly scaling number of possible single-​point mutations. As a consequence of the enormous space of combinatorial possibilities afforded through sexual reproduction, populations of higher organisms (including humans) demonstrate tremendous genetic diversity and thus different implications for a species’ reproductive success: There is no reasonable chance that any two individuals have exactly the same genetic constitution in a species of millions of millions of individuals persisting over millions of generations. There is no difficulty accounting for the probable genetic uniqueness of each human being or other organism which is the production of bi-​parental reproduction (Wright, 1932, p. 356).

To organize inquiry regarding the way in which populations evolve over time, Wright introduced the idea of a ‘landscape’ that assigns an environmentally determined level of ‘fitness’, or reproductive success, to each genetic combination. ‘The problem of evolution’, he states, ‘is that of a mechanism by which the species may continually find its way from lower to higher peaks in such a field. In order that this may occur, there must be some trial and error mechanism on a grand scale by which the species may explore the region surrounding a small portion of the field which it occupies. To evolve, the species must not be under the strict control of [mutation-​driven] natural selection’ (Wright, 1932, p. 359). As introduced by Wright (1932), fitness landscapes are a two-​dimensional visualization of the relationship between a species’ reproductive success and its genotype. The organism’s genotype is one possible combination in the ‘genotype space’, itself represented as a hypercube with vertices composed of all possible gene combinations. When plotted as two-​ dimensional contour map representing levels sets of reproductive fitness, fitness landscapes show multiple local and global optima, as well as ‘maladaptive valleys’ (see Figure 13.1). Although the combinatorial possibility of the genotype space is nearly boundless, not all of the landscape is accessible to a given individual of a species. The geometric features of the landscape play an important role in the accessibility to higher peaks of fitness for a given species. Sexual reproduction—​which we will propose is analogous to Schumpeterian entrepreneurship and innovation in an economic ecosystem framework—​thus provides a mechanism by which the population can reach regions of the fitness landscape that lie beyond its immediate ‘neighbourhood’ composed of adjacent genetic variants. The presence of sexual reproduction alone is, however, not sufficient to sustain continued evolution. Genetic diversity is also necessary. The reason for this is that, for a closed population, inbreeding among members of the population combines with natural selection to shift

Economic Ecosystems   249

Figure 13.1   Sewall Wright’s Fitness Landscapes. Source: Wright (1932, p. 3).

the distribution of genetic combinations ‘uphill’ on the landscape towards a fixed number of peaks, where a ‘peak’ on the fitness landscape is defined as local maximum in terms of favourable adaptation to the environment. After a sufficient time has elapsed such that the population converges on such peaks, a certain stasis sets in. While mutation may continue to introduce some variation after the population has reached a set of high points on the fitness landscape, ‘the species will occupy a certain field of variation about a peak … The field occupied [on the landscape] remains constant although no two individuals are ever identical’. Under such conditions ‘further evolution can only occur by the appearance of wholly new (instead of recurrent) mutations, and ones which happen to be favorable from the first [instance it appears]’.4 Absent fortuitous encounters with entire new populations of the same species, the single most effective way out of this trap is for the species to subdivide into local subspecies that occasionally crossbreed. This allows for the regular introduction of truly new combinations that fundamentally expand the field of variation occupied by the species. Wright’s primary conclusion is that evolution requires a balance among the various mechanisms for generating novelty upon which it depends: mutation, selection, inbreeding, and crossbreeding. ‘There must be gene mutation, but an excessive rate gives an array of [maladapted] freaks, not evolution; there must be selection, but too severe a process destroys the field of variability, and thus the basis for further advance; prevalence of local inbreeding within a species has extremely important evolutionary consequences, but too close inbreeding leads merely to extinction’. In the short term, narrow specialization leads to economies of scale and increased productivity; however, in the long term, narrow specialization leads to the exhaustion of possibilities for search, and thus to evolutionary dead ends. Success for a species depends on balancing these factors.

250   Auerswald and Dani As we will see, an ecosystems perspective on economic geography suggests that the same holds for the densely interconnected firms that comprise economic ecosystems: specialization yields increased productivity, but success in the long term depends equally on the continued introduction of novelty.5

Unit of Analysis Representing the ‘DNA’ of Firms Markets exert a selection pressure on firms that is reflected in the dynamics of industries. Standard theories of industrial organization suggest that firms with greater-​than-​average productivity will grow over time within a given industry, while low-​productivity firms are likely to shrink or exit (Viner, 1932; Jovanovic, 1982; Hopenhayn 1992; Ericson and Pakes, 1995; Foster et al., 2008). However, contrary to the predictions of Viner (1932), productivity differences among industries in different geographies, among firms within industries, and even among plant within firms, are large and tend to persist over time.6 Work by Bloom and Van Reenen (2010) singled out the influence of management practices on cross-​country variations in firm productivity. These results suggest that economically relevant knowledge is generally firm-​specific and costly to transmit. Imperfect appropriability of the production process allows entrepreneurs to capture persistent rents (Aghion and Howitt, 1992; Auerswald, 2010). While surprising in the context of the knowledge-​based variants of ‘new growth’ theory that emphasize the ostensible ubiquity of ‘knowledge spillovers’,7 these results fit comfortably within an ecosystems view of economic geography. If firms are organisms, then the DNA of firms is the economically relevant knowledge embedded within the firm on which the firm’s survival depends. We will term such economically relevant knowledge embedded within the firm the firm’s ‘production algorithm’. A notable conceptual antecedent to the production algorithm is the firm-​level ‘routine’ (Nelson and Winter, 1982), or those firm-​specific functions that relate inputs to outputs given the internal context and the external environment of the business operations. Nelson and Winter (1982) proposed that routine plays ‘the role that genes play in biological evolutionary theory’ (Nelson and Winter, 1982, p. 17).8 Auerswald et al. (2000) refer to the economically relevant knowledge encoded within a firm as its ‘production recipe’, which represents ‘the complete description of the underlying engineering process’. Employing a culinary rather than biological analogy, the notion of recipes emphasizes that a firm’s production plan extends well beyond the mapping of ingredients (inputs) to outputs, that is the focus of standard production theory, to incorporating the specific list order for routinization of the production plan. More important, the production algorithm consists of the ‘how’ of the production process—​the code that specifies the distinct operations required to convert inputs into outputs.9 Invoking the Coasean notion that firms exist to internalize externalities (Coase, 1937), it follows naturally that the production recipes or production algorithms—​two terms we will use interchangeably—​that are mostly likely to survive under evolutionary pressure are complex and cannot be easily imitated.

Economic Ecosystems   251 In the ecosystems framework, therefore, production recipes/​algorithms are analogous to genes in organisms. They are the basic units of economic recombination.

Learning by Doing and Adaptive Walks on Fitness Landscapes In 1936, four years after Sewall Wright published his pioneering work introducing the concept of the fitness landscape to evolutionary biology, his brother, T.P. Wright, published a paper titled ‘Factors Affecting the Cost of Airplanes’ that set the stage for a future modern synthesis in economics, linking of systematic modifications at the scale of the production algorithm to observed outcomes at the scale of economic ecosystem. This paper was a contribution to the engineering literature that documented the manner in which the cost of airframes declined as experience accumulated. Yet what T.P. Wright had discovered—​the organizational learning curve—​turned out to have fundamental significance in fields as varied as business strategy, industrial organization, macroeconomics, and economic geography. Citing Wright (1932), Arrow (1962) proposed a growth model based on the observation that per unit costs of production can fall even in the absence of capital accumulation and R & D inputs. He attributed this productivity gain in the absence of increased inputs to ‘learning by doing’. Also citing Wright (1932), Muth (1986) began to link production algorithms to the emergence of learning curves by suggesting that higher efficiencies in the search procedure can be achieved by breaking the design problem into smaller components and systematically modifying the components individually. Building on this work, and employing the intuitive conceptualization of fitness landscapes articulated by Wright to explain the emergence of learning curves as documented by Wright (1932), and Auerswald et al. (2000) applied the production recipes approach to learning by doing as a process of systematic search in space of possibilities represented by adjacent production algorithms. The specific form of a fitness landscape employed in Auerswald et al. (2000) is Kauffman and Levin’s (1987) NK model. In the NK model, N refers to the number of traits of an organism that contribute to increasing fitness of the organism, while K refers to the number of other traits of the organism that have a bearing on its fitness. The evolutionary pathway is modelled as an ‘adaptive walk’ or a step-​wise optimization process. Genetic mutations happen at random but only those mutations that increase the species’ fitness are adopted and it is through this evolutionary process that a species traverses the fitness landscape in search of a more optimal peak. As K increases, the ruggedness of the landscape increases with the number of peaks increasing, but the typical height of the peaks decreases to reflect that an increase in the epistatic linkages increases the conflicting constraints of the fitness landscape (Kauffman, 1989). More recently, Sergey Gavrilets suggested that the properties of multidimensional landscapes can differ significantly from low-​dimension landscapes, which can have implications on how a species’ population moves from one peak to another, or crosses a valley. Gavrilets and Janko Gravner (1997) answered this question by suggesting that a population can cross maladaptive valleys to reach a higher fitness peak by traversing along ‘ridges’ if and where available. However, these ridges are determined by how peaks cluster on that landscape, and they may not be accessible. In the production recipes application of the NK model introduced in Auerswald et al. (2000), N refers to the number of individual operations in the production recipe, and K refers to the average number of interactions among operations that have a bearing on overall

252   Auerswald and Dani efficiency of the production recipe. In this framework, the difficulty of the search problem solved by the firm is determined primarily by K, ‘the richness of epistatic linkages in the system’ (Kauffman and Johnsen, 1991). Strumsky and Lobo (2003), Siggelkow and Levinthal (2003), and McNerney et al. (2010) have similarly employed landscape models to study how the co-​evolutionary patterns of organizational and design subsystems can improve efficiency of the search process while reducing costs.10 Hidalgo et  al. (2007), Neffke and Henning (2013), and Muneepeerakul et al. (2013) have further advanced work on economic ecosystems by mapping the search space of possibilities embedded in the knowledge structures of regional production processes, industries, and occupations, respectively.

Combination and Recombination The correspondence between genes and production algorithms in biological and economic ecosystems, respectively, suggests that evolution as search may occur not only as a consequence of mutation and selection—​analogous to competition among firms in the presence of learning by doing as described by Wright (1932) and Arrow (1962)—​but also as a consequence of the combination and recombination—​analogous to combination and recombination as a driver of technological advance, as emphasized by Schumpeter (1911) and Arthur (2013), among others. Schumpeter famously wrote: ‘The carrying out of new combinations we call “enterprise”; the individuals whose function it is to carry them out we call “entrepreneurs” ’. Echoing Wright (1932) and deriving conclusions from a combination of theoretical first principles and insights derived from a plethora of historical cases, Arthur (2013, p. 129) argues that novel technologies—​new technological ‘species’—​arise overwhelmingly as the consequence of the purposive recombination of existing solutions: We … have our answer to the key question of how novel technologies arise. The mechanism is certainly not Darwinian: novel species in technology do not arise from the accumulation of small changes. They arise from a process, a human and often lengthy one, of linking a need with a principle (some generic use of an effect) that will satisfy it. This linkage stretches from the need itself to the base phenomenon that will be harnessed to meet it, through supporting solutions and subsolutions … In the end the problem must be solved with pieces—​ components—​that already exist (or pieces that can be created from ones that already exist).

Learning by doing, analogous to mutation combined with selection, and recombination, broadly analogous to sexual reproduction, drive the dynamics as the scale of production algorithms that determine the evolution of economic ecosystems.

Co-​evolutionary Dynamics Species do not just evolve, of course. They co-​evolve with other species in their environments. The preceding section talked about the evolution of a species as a search for a better fit along a fixed fitness landscape. Yet, the adaptive progress of one species in an environment is likely to have implications for the evolution of a different species sharing the same environment. “Anecdotally, development of a sticky tongue by the frog alters the fitness of the fly, and what changes it must now make to increase its fitness; given the frog’s sticky tongue, the

Economic Ecosystems   253 fly should now develop slippery feet. In this framework, adaptive moves by any partner may deform the fitness landscapes of other partners.” (Kauffman and Johnsen, 1991, p. 468). The study of the evolution of a species should also incorporate the co-​evolutionary patterns of other interacting species. To model this mathematically, Kauffman and Johnsen (1991) extend the NK framework to include C, the number of traits of the other interacting species that have a bearing on its fitness. Such ‘coupled NK fitness landscapes’ have varied emergent properties and the co-​evolutionary process applies particular pressures on the evolutionary pathways of different species. Key developments in this field that further carry over to the economic ecosystem perspective is that ecosystems are not completely connected, but instead each species in the ecosystem only interacts with a subset of other species in the ecosystem, forming a web of interactions. Co-​evolutionary patterns of industries and technologies within economic ecosystems are driven by specific and relevant interactions that can be identified and evaluated in the context of different search strategies.

Structure Spatial Agglomeration and the Definition of Ecosystem Boundaries In economic geography the periphery is often contrasted with the core. Agglomeration arises from the locational choices of manufacturing firms in the presence of transport costs, thereby determining how the core and periphery grow over time (Krugman, 1991). More generally, economic ecosystems are defined within specific geographies by internal linkages and external boundaries. Just as biological habitats comprise ecosystems, which, in turn, make up biomes, economic ecosystems are also nested within larger hierarchies of regional, national, and global systems. While ecologists are able to rely on the physical and topographical characteristics of space to define the boundaries of ecosystems (Bailey, 2009), these methods have limited applicability when dissecting abstract strategic networks that comprise economic ecosystems. The ability to identify internal linkages and define boundaries is thus a necessary condition to applying the concept of the ‘ecosystem’ to economics. To specify interactions that contribute to defining ecosystem boundaries, they must be complemented with methods to map knowledge networks, strategic alliances, and other outcomes of processes that develop social, cultural, and economic ties.11 Gunderson and Holling (2001) recommend a method of mapping ecosystems that is similar to the biological approach of the ‘controlling factors method’ (Bailey, 2009). They suggest that ‘the complexity of living systems of people and nature emerges not from a random association of a large number of interacting factors rather from a smaller number of controlling processes. These systems are self-​organized and a small set of critical processes create and maintain this self-​organization’ (Holling, 2001, p. 391). The Gunderson and Holling (2001) approach has developed alongside theoretically similar applications of evolutionary theories in the multi-​level perspective, particularly to the study of technological transitions (Geels, 2002; Genus and Coles, 2008). In an outcome-​based approach to specifying the internal

254   Auerswald and Dani interactions that correlate with the overall performance of economic ecosystems, Stangler and Bell-​Masterson (2015) recommend evaluating the overall performance of economic ecosystems based on a set of four regional entrepreneurship-​specific indicator variables: density; fluidity; connectivity; and diversity. Ecosystem boundaries are characterized by an exchange of energy and information between neighbouring ecosystems and are termed ‘transition zones’ (Banks-​Leite and Ewers, 2009). In natural ecosystems, these transition zones may be abrupt, such as the boundary between a marine and terrestrial ecosystem, or a field and an adjoining forest. Alternately, the transition zones may be more gradual, incorporating a series of overlapping structures, such as in estuaries and marshes. While the width of the transition zone depends largely on the geography of the region, these zones provide spaces through which one ecosystem influences another, what is termed an ‘edge effect’ (Murcia, 1995). These edge effects are primarily driven by abiotic conditions and pose a strong influence on the environmental conditions of the transition zones, creating unique habitats to which species are specifically adapted. Accordingly, the greater the contrast between the habitats sharing an edge, the stronger will be the edge effect. Boundaries of economic ecosystems resemble transition zones in biological ecosystems. In the most direct geographical sense, the political boundaries reflect sharp contrasts in the transition zones making the difference between ecosystem structures sharply visible. For instance, Figure 13.2 shows an aerial view of a section of the USA/​Mexico border, one of the most controlled and frequently crossed borders in the world. The edge of the two national ecosystems is very apparent in the image. In another example, metropolitan statistical area (MSA) definitions in the USA are updated every few years to reflect population changes along border counties (Office of Management and Budget, 2013). These changes can include the reassigning of counties from one MSA to an adjacent one based on commuting zones, or the addition or exclusion of a

Figure 13.2   Ecosystem Boundaries—​US/​Mexico Border. Source: Google Earth satellite image.

Economic Ecosystems   255 county from an MSA based on population change. Although MSA definitions are political boundaries, they reflect the socio-​economic relationships between interior and boundary regions of a common physical system.

Interdependencies Firms in economic ecosystems are not uniformly distributed, and a firm’s location within the ecosystem has strong implications on its productivity and evolution. Many new economic combinations fail to survive in the market because complementary factors or vital inputs for production and commercialization may not yet be available. Fagerberg (2005) gives the example of Leonardo da Vinci, who had presented designs of advanced technologies, including airplanes, but he lacked the adequate materials or production processes to realize them during his time. A contemporary example is the recent explosion in the fields of computational sciences. Although much of the mathematics behind pattern-​recognition algorithms was well established more than a century ago, it took the computational power of modern computers to allow researchers to apply fully the programming methods that today are branching out new technological fields in augmented reality, artificial intelligence, and cybernetics. In this sense, it is hard to conceive of the structure of innovation without also considering how the structure evolves over time. When looking at a static representation of the innovation system we would be hard pressed to identify the relevant components that have led up to the current opportunity for commercialization. Spencer (1857) long ago described how different components of an economy, when connected, become mutually dependent and begin to differentiate themselves by assuming different functions: ‘When roads and other means of transit become numerous and good, the different districts begin to assume different functions, and to become mutually dependent.’ Economic complexity in this sense is a result of increased interdependence within systems and where more complex interactions imply more complex social arrangements. These complex interdependencies in ecosystems are multidimensional and can be measured in a number of ways. Most often they are studied in terms of the number of parts to a technological artifact (Strumsky et al., 2012), but they may also be reflected in terms of organizational complexity of production processes within firms (Auerswald et al., 2000), the diversity of teams required to develop new technological innovations (Kash and Rycroft, 1999; Adams et al., 2005), as well as the in the intricacies of buyer–​supplier networks and peer-​production networks (Appleyard, 2003; Auerswald and Branscomb, 2008).

Dynamics Succession Ecosystem succession refers to the process by which the structure of the system evolves over time. In broad terms, succession in biological ecosystems is represented by the emergence of a new biological community following a large disturbance. These changes in biological

256   Auerswald and Dani composition of the ecosystem, termed ‘succession’, are analogous to the progressive development of practices within an industry or local economy. Tansley describes succession as follows: Succession is a continuous process of change in vegetation which can be separated into a series of phases. When the dominating factors of change depend directly on the activities of the plants themselves (autogenic factors) the succession is autogenic: when the dominating factors are external to the plants (allogenic factors) it is allogenic. The successions (priseres) which lead from bare substrata to the highest types of vegetation actually present in a climatic region (progressive) are primarily autogenic. Those which lead away from these higher forms of vegetation (retrogressive) are largely allogenic, though both types of factor enter into all successions (1935, p. 306).

Just as succession can be either autogenic or allogenic, the evolution of industries in economic ecosystems can be either endogenously driven or exogenously driven. Furthermore, just as Tansley defines successions that lead towards greater biological complexity as progressive change in biological systems, so we are suggesting that the evolution of the capabilities of a city or region towards greater complexity constitutes progressive change in economic systems.

Speciation Gavrilets (2004, pp. 399–​400) proposes that ‘speciation can be visualized as the process of formation and subsequent divergence of clusters of organisms in genotype space accompanied by the evolution of RI [reproductive isolation] between the emerging clusters’. In developing this higher-​order complexity of fitness landscapes, Gavrilets coined the term ‘holey adaptive landscapes’, defining them as ‘an adaptive landscape where relatively infrequent well-​fit combinations of genes form a contiguous set that expands throughout the genotype space’ (Gavrilets, 1997, 1999). These holey adaptive landscapes, being less contiguous and offering fewer pathways for adaptive walks, have properties that can result in subdivision of populations, leading to speciation.12

Ecosystem Life Cycles Adaptive Capacity or the Resilience of the System As Figure 13.3 illustrates, the biological adaptive cycle alternates between short periods of systemic restructuring triggered by a disturbance (release and reorganization), followed by longer periods of accumulation and transformation of resources (exploitation and conservation). The development of the ecosystem from the exploitation phase to the conservation phase captures the traditional notions of ecological succession. Organisms rapidly colonize a disturbed space to accumulate stores of energy and form complex interdependencies. The shorter period beginning with the release is often referred to as the phase of ‘creative destruction’ to parallel Schumpeterian entrepreneurship (Schumpeter, 1911, 1939). It occurs when the over-​connected dependencies resulting from the conservation phase collapse under some external disturbance, such as a fire or disease. The result is a release of a great

Economic Ecosystems   257 4. REORGANISATION

2. CONSERVATION Succession Consolidation

STORED NUTRIENTS

Accessible carbon and nutrients

1. EXPLOITATION

3. RELEASE Disturbance: Fire, storm, pest

Pioneer Opportunist

CONNECTEDNESS

Figure 13.3   Succession and Reorganization of Ecosystems. Source: Bengtsson et al. (2000).

amount of stored energy, potentially creating new opportunity for more complex reorganization with more diverse inputs. These features are paralleled in social and economic ecosystems as well and can be observed in the entrepreneurial activity that drives change in the structure of the ecosystem. To summarize the argument presented by Gunderson and Holling (2001) in their book Panarchy: Understanding Transformations in Human and Natural Systems, we can consider the exploitation phase to be crowded by entrepreneurial activity that is working and defining a new space of opportunity. Pioneers and opportunists who have preferential access to the newly released energy and resources will be the first to jump to entrepreneurial action and the overall diversity of the cluster will increase. As the system matures into the conservation phase, consolidation across firms will establish new system-​level standards that, in turn, enable more specialized innovative activity. The structure of the ecosystem will have become denser and more interconnected across different scales of economic activity. As the networks become denser and more embedded, the structure of the ecosystem becomes more rigid and, consequently, more vulnerable to large-​scale disturbances. In certain configurations,13 a strong stochastic shock, such as a regulatory change, can significantly disentangle many networks of the structure and release abundant energy into the environment. Entrepreneurs will once again seek opportunity in this disturbance and begin to establish a new order to the ecosystem in the reorganization phase. Figure 13.4 provides a generalized summary of the features of each of the four phases. Note that the two axes of the diagram indicate levels of realized entrepreneurial potential and the connectedness of the economic ecosystem. In the next subsections we discuss in more detail the two life-​cycle loops of ‘release and reorganization’, and ‘exploitation and conservation’ in more detail according to the processes that enable systemic phase shifts.

258   Auerswald and Dani 4. Reorganization – Release of stored nutrients introduces novelty – Phase characteristics: restructuring; highest uncertainty; new order Potential

2. Conservation – ‘Climax’ or equilibrium phase – Resource utilization shifts from growth to system maintenance – Phase characteristics: stability; rigidity; strong interdependencies; vulnerability to disturbance

Connectedness 1. Exploitation 3. Release – ‘Creative destruction’ phase – High competition for resources – System collapse releasing stored nutrients – New opportunities from increasing diversity – Phase characteristics; uncertainty; chaos; – Phase characteristics; pioneers; opportunity; disturbance; de-stability innovation; uncertainty

Figure 13.4   Phase Characteristics of the Entrepreneurial Ecosystem. Source: Author’s adaptation.

Release to Reorganization Release refers to the opportunity that fuels the creative destruction phase. Resulting from some external disturbance, the tightly knit connected structures of the ecosystem come undone and large amounts of stored capital and energy are released within the ecosystem. This initiates the undoing of old established networks from the prior period of succession. Networks established during the shift from exploitation to conservation mature over a long period of time; however, the structural shift from release to reorganization occurs over a much shorter time scale and is very disruptive. Although established networks deteriorate and the interconnectedness of the ecosystem declines, reorganization sets the stage for a new rapid phase of exploitation and entrepreneurial opportunity followed by a long period of innovation and economic succession.

Exploitation to Conservation Exploitation refers to the colonization of disturbed ecosystems where species capture easily accessible resources. It is the beginnings of establishing order to a chaotic system. The conservation phase, however, is the ‘climax’ phase of succession where stored nutrients and energy are at their peak and the system has achieved a high level of interconnectedness. It is a result of a long period of growth and reorder in the system, and refers to the phase of the adaptive cycle when the ecosystem has developed strong and complex interdependencies.

Economic Ecosystems   259 As the ecosystem embarks on a long period of succession, it develops broader networks and increases its connectivity. When no market niches are left unexploited, entrepreneurs will look for new opportunities through innovations in established technologies rather than opportunity identification or imitation. A key feature of succession is the macro-​level stability of the system relative to significant churn at the micro-​levels of the organization. Consequently, strong cross-​level synergies such as spin-​offs from large firms and more active mergers and acquisitions markets develop in the ecosystems. The networks that build on these interactions not only incentivize greater innovative activity, but also reinforce the structures of the ecosystem.

Health Diversity Jane Jacobs (1961, 1969)  made early and seminal advancements on the hypothesis that increasing economic diversity is key to the vitality of cities. She described the engines of growth for regions to be enabled by increasing connectivity to cities, as well as increasing economic diversity within the region itself. Glaeser et al. (1992) studied a cross-​section of US cities and found that ‘at the city-​industry level, specialization hurts, competition helps, and city diversity helps employment growth’. Work by Feldman and Audretsch (1999) similarly emphasized the importance of economic diversity in innovative activity. Saxenian (1994) documented the relative success of Silicon Valley in developing its technology sector in the 1980s as compared with Boston Route 128. She argued that inter-​firm and inter-​industry networks in Silicon Valley played a significant role in providing it with regional technological advantage as a result of knowledge externalities leading to greater innovative outputs. Hidalgo et al. (2007) studied the co-​occurrence of products in the export portfolios of countries to identify a relatedness measure across products, based on the expectation that countries are more likely to produce goods together that require ‘similar institutions, infrastructure, physical factors, technology, or some combination thereof ’ (Hidalgo et al., 2007, p.  484). The revealed network of products, called the ‘product space’, showed that more sophisticated products were located in denser regions of the network, while less sophisticated products were on the periphery. Furthermore, they also found that advanced countries tended to have more diverse and densely networked product spaces than the less developed countries. They explain the developmental constrains on countries in terms of the connectedness of their product space and the co-​evolutionary patterns of their firms. Hidalgo et al. (2007) apply a biological analogy similar in intent as the adaptive walks of firms along ‘holey adaptive landscapes’: Think of a product as a tree and the set of all products as a forest. A country is composed of a collection of firms, i.e., of monkeys that live on different trees and exploit those products. The process of growth implies moving from a poorer part of the forest, where trees have little fruit, to better parts of the forest. This implies that monkeys would have to jump distances, that is, redeploy (human, physical, and institutional) capital toward goods that are different from those currently under production. Traditional growth theory assumes there is always a tree

260   Auerswald and Dani within reach; hence, the structure of this forest is unimportant. However, if this forest is heterogeneous, with some dense areas and other more-​deserted ones, and if monkeys can jump only limited distances, then monkeys may be unable to move through the forest. If this is the case, the structure of this space and a country’s orientation within it become of great importance to the development of countries (Hidalgo et al., 2007, p. 482).

Regions develop comparative advantage by having diverse but related economic structures. Neffke and Henning (2013) studied the flows of labour across industries to identify an ‘industry space’. They define a skill-​relatedness measure based on the expectation that workers are more likely to move across jobs that have similar skill requirements. Consequently, industries that have similar skill requirements should show greater cross-​industry flows after controlling for other industry dynamics. Applying the industry space to study regional diversification, they find that firms are 100 times more likely to diversify into industries that are more skill-​related.

Ecosystem Resilience and Adaptive Capacity The notion of resilience is an ecological concept that in economic geography has most often been applied to a region’s capacity to resist and recover from disturbances, including natural disasters (Fingleton et al., 2012). Rose (2004) discusses the behavioural response of individuals and regional markets to large-​scale disruptive events such as earthquakes within a computation general equilibrium framework. He defines resilience as ‘the inherent and adaptive responses to hazards that enable individuals and communities to avoid some potential losses’ (Rose, 2004, p. 41). Notably, he distinguishes the ‘inherent’ response as that which allows for the substitution of inputs within the system for those that were affected by the disturbance, and the ‘adaptive’ response as that which actively reconfigures the network of relationships between suppliers and customers for better reallocation of resources. Both these responses reflect the expectation for the region’s social and economic structure returning to a pre-​disaster equilibrium as its ability to accommodate shocks (Rose, 2004). While initially appealing to traditional regional economists, economic geographers highlight that as a necessity of the response to the shock the economic structure of the region also changes and a new equilibrium state is reached through adjustment (Martin, 2012). Accordingly, an evolutionary approach has been recommended whereby following a disturbance, the ecosystem does not return to pre-​disaster equilibrium, but instead is set on a new equilibrium path (Holling, 2001). In our ecosystem perspective we adopt not only a similar conception of ecological resilience set in an evolutionary framework, but also emphasize the dynamically stable growth trajectories of regional economies (Martin and Sunley, 2007; Boschma, 2015). We consider ecosystem resilience to be the ability of the economy to return to, or continue on, its path of long-​term development following a disturbance, and also allow for the ecosystem to develop new pathways to development as a response to the disturbance. The trade-​off between these two structural responses is framed in the literature in terms of ecosystem adaptation versus adaptability (Hassnik, 2010; Pike et  al., 2010; Boschma, 2015). Where adaptation refers to a region’s ability to return to its pre-​disaster development trajectory, adaptability refers to the region’s ability to innovate new pathways for growth (Pike et al.,

Economic Ecosystems   261 2010)  through industry speciation, for instance. In both these responses an ecosystem’s diversity plays a critical role. In terms of adaptation, regions with more diversified industrial sectors have been shown to better accommodate sector-​specific shocks, provided other sectors in the economy are not strongly coupled with the affected sector in terms of supply-​chain relations, but are similar in their skill composition (Essletbichler, 2007; Frenken et al., 2007; Neffke and Henning, 2013). These circumstances limit the diffusion of the supply or demand shock across sectors while providing better-​matching opportunities for labour displaced by the shock in other local skill-​related sectors. In terms of adaptability, diversified ecosystems have improved capacity for algorithmic recombination (Auerswald, 2008) resulting from ‘Jacobs’ externalities’ (Jacobs, 1961), thus enabling opportunity for the region to innovate new development pathways. Industry speciation, in this case, is one potential development of ‘new pathways’ that can improve ecosystem outcomes as has been empirically identified by Neffke et al. (2011), but co-​evolutionary patterns in institutional change also facilitate or inhibit the ability of ecosystems to respond to large disturbances. Key to the evolutionary perspective worth iterating here is the path-​dependent and co-​ evolutionary nature of the ecosystem, because an ecosystem may respond to a disturbance in a maladaptive fashion that can reduce the overall welfare of the system, or set it upon a new evolutionary path that is less socially optimal than the initial development trajectory (Holling, 2001). Yet, in general, the potential for adaptation versus adaptability of the ecosystem is viewed as a trade-​off whereby the prior is the expected response of densely connected and interdependent structures, while adaptability is more plausible for systems that have looser, more malleable ties.

Entrepreneurial Dynamism Entrepreneurship is at the heart of economic change (Schumpeter, 1911), yet it takes an entire ecosystem to help entrepreneurs bring innovations to market. Novelty in an economic ecosystem is a result of coordinated activities both within firms and with agents external to the firm. Entrepreneurship is a collective achievement that resides not only within the parent organization of the innovation but also in the construction of an industrial infrastructure that facilitates and constrains innovation. This infrastructure includes (1) institutional arrangements to legitimize, regulate, and standardize a new technology; (2) public-​resource endowments of basic scientific knowledge, financing mechanisms, and a pool of competent labor; (3) development of markets, consumer education, and demand; and (4) proprietary research and development, manufacturing, production, and distribution functions by private entrepreneurial firms to commercialize the innovation for profit (Van de Ven et al., 1999, p. 149).

Furthermore, innovation is a dynamic, non-​linear process. Some innovations can have a radical impact on the market, while most others are incremental (Freeman and Soete, 2009), but it is far from a well-​defined phenomenon and requires the coming together of many factors to produce the right opportunity for commercialization. For instance, many inventions never make it to market because complementary factors or vital inputs for production and

262   Auerswald and Dani commercialization may not be available yet (Fagerberg, 2005). Feldman and Zoller (2011) found evidence that the presence of dealmakers in regional economies played an influential role in fostering regional entrepreneurship. Dealmakers acted as brokers who functioned to shape the network, manage structural holes, and ‘connect disparate actors to social networks’ (Feldman and Zoller, 2011, p. 27). Such coordinated activities across various organizations within institutional frameworks allude to multi-​level, networked, and interdependent relationships that define and enable the innovative process. Entrepreneurial action is essential to evolution in economic ecosystems as it provides a selection mechanism that impacts structural change and is thus an important indicator for assessing the health of economic ecosystems.

Conclusion At the time when Alfred Marshall stated that ‘the Mecca of the economist lies in economic biology rather than in economic dynamics’, the concept of the ecosystem had not yet been developed; the ‘modern synthesis’ in evolutionary biology had not yet occurred; and mathematical models of evolutionary processes did not yet exist. Nearly a century later, economists have access to powerful analytical tools developed by evolutionary biologists and ecologists that have the potential to be directed towards the study of economic ecosystems. In proposing that economic systems are, literally, ecosystems, and thus that they may be fruitfully be studied as such, we are mindful of the fact that human beings do differ along multiple dimensions from other biological entities. We expect that the path for research that we suggest in this chapter ultimately will lead to an understanding that the study of economic ecosystems requires different tools from those developed by ecologists and evolutionary biologists. Thus, our contention in this chapter is not that humans are the same as other biological entities, or that existing biologically inspired models ultimately will prove adequate in the study of social systems. Rather, it is that we will only understand the boundaries for the application of biologically inspired models if we begin by taking economic systems seriously as ecosystems, and studying their properties from that starting point. In such a process, we are only at the beginning.

Notes 1. Herbert Spencer’s (1857) early work on evolution, specifically his seminal essay ‘Progress: Its Law and Cause’ arguing that all structures in the universe evolved from simplicity towards ever-​increasing complexity, had a foundational impact on developing biologically informed theories in the study of physical and social systems. 2. Thorstein Veblen, influenced by Spencer’s early work, first used the term ‘evolutionary economics’ in his 1898 essay ‘Why is economics not an evolutionary science?’ (Veblen, 1898). More recently, pioneering work by Nelson and Winter (1982) has embraced biological analogies to advance the field of evolutionary economics. 3. Acs et al. (2014) introduced the related, but distinct, concept of a national entrepreneurial ecosystem.

Economic Ecosystems   263 4. Inbreeding also leads to greater risk of generically caused disease, as well as diminished resilience of the phenotype. 5. As Lucas (1993, p. 263) finds: ‘A growth miracle sustained for a period of decades clearly must thus involve the continual introduction of new goods, not merely continued learning on a fixed set of goods.’ 6. Syverson (2011) provides a survey. Productivity is a residual measure of how well firms convert their inputs to outputs after accounting for observed characteristics, that is, a measure for how well firms perform in given market structures (Syverson, 2004a), as well as their ability to imitate the practices of the most productive firms (Bloom and Van Reenen, 2010). Firm productivity is directly associated with firm-​specific attributes that, although unobserved, explain the wide dispersion in firm behaviour, even for firms in the same industry producing similar output. In the USA alone, considering manufacturing plants at the four-​digit level of industry, plants in the ninetieth percentile of the productivity distribution made twice as much output with the same inputs as plants in the tenth percentile of the distribution (Syverson, 2004b). Hsieh and Klenow (2009) found these productivity differences to be even larger for firms in China and India, where the ninetieth percentile made nearly five times the output of the firms in the tenth percentile. Foster et al. (2008) account for price changes (idiosyncratic demand shifts) that can affect the measure of productivity across firms and show that the differences in output persist even in the case of firms in industries that produce homogenous products. How firms organize their production activities to produce output has been a core consideration of economic theory. 7. For example, as pioneered by Romer (1986, 1990, 1994). An inspiration for these theories was Marshall’s observation, which was quoted at the outset, that ‘the mysteries of the trade become no mysteries; but are as it were in the air’. Either Marshall was mistaken on this point, or processes of production have become sufficiently more complex in the intervening century that his observation no longer holds. 8. The biological analogy grants these routines the evolutionary features of deterministic behaviour, heritable characteristics across generations of routines, and selection based on some routines being better suited to their markets than others (Nelson and Winter, 1982, pp. 51–​138). 9. Along similar lines Winter (1968) observes:  ‘ “Knowing how to bake a cake” is clearly not the same thing as “knowing how to bring together all of the ingredients for a cake.” Knowing how to bake a cake is knowing how to execute the sequence of operations that are specified, more or less closely, in a cake recipe.’ 10. Early foundational work on innovation as search by Evenson and Kislev (1976) modelled a simple trade-​off between the costs of search and the expected yield of success to determine that a research team will stop the search process when the costs exceed the expected returns from success. Weitzman (1979) iteratively added to this discussion by introducing a dynamic programming model of an optimal sequential search strategy. His model identified objective conditions under which a research team should engage in search, in what order should they pursue multiple research options, and when should they stop the search. In support of this observation, Kauffman et al. (1994) provided a simulation model to show that parallel processing algorithms can increase the efficiency of search for complex designs. 11. An interesting recent development along these lines is the application of satellite imaging and big data to study how economic systems grow or shrink over relatively short timeframes (Kearns, 2015). Such methods help delineate physical boundaries of economic ecosystems.

264   Auerswald and Dani 12. In his book Fitness Landscapes and the Origin of Species, Gavrilets (2004) makes a strong case for the modern synthesis of the evolutionary approach to studying speciation across academic disciplines outside of evolutionary biology. He outlines a series of models that can be used to study various types and rates of speciation of populations. He developes a series of models that can be applied to various studies beyond their immediate application in evolutionary biology to even the social sciences. 13. Stochastic events may affect ecological niches differently in an ecosystem (Holling, 2001). Similarly, a large disturbance to an entrepreneurial ecosystem may lead to collapse of some entrepreneurial niches over others.

References Acs, Z.J., Estrin, S., Mickziewicz, T., and Szerb, L. (2014). ‘The continued search for the solow residual: the role of national entrepreneurial ecosystems’. IZA Discussion Paper No. 8652, 2014 November (Bonn: Institute for the Study of Labor). Adams, J.D., Black, G.C., Clemmons, R., and Stephan, P. (2005). ‘Scientific teams and institutional collaborations: evidence from U.S. universities, 1981–​1999’. Research Policy 34: 259–​285. Aghion, P. and Howitt P. (1992). ‘A model of growth through creative destruction’. Econometrica 60: 323–​335. Appleyard, M.M. (2003). ‘The influence of knowledge accumulation on buyer-​ supplier co-​development projects’. The Journal of Product Innovation Management 20: 356–​373. Arrow, K.J. (1962). ‘The economic implications of learning by doing’. The Review of Economic Studies 29: 155–​173. Auerswald, P.E., Branscomb, L., Gorman, S., Kulkarni, R., and Schintler, L. (2007). ‘Placing innovation:  a Geographical Information Systems (GIS) Approach to identifying emergent technological activity’. Working Paper, U.S. Department of Commerce, Gaithersbur, MD: Advanced Technology Program. Auerswald, P.E. (2008). ‘Entrepreneurship in the theory of the firm’. Small Business Economics 30: 111–​126. Auerswald, P.E. (2010). ‘Entry and Schumpeterian profits:  how technological complexity affects industry evolution’. Journal of Evolutionary Economics 20: 553–​582. Auerswald, P.E. (2015). ‘Enabling Entrepreneurial Ecosystems’ in D.B. Audretsch, A.N. Link, and M.L. Walshok (eds) The Oxford Handbook of Local Competitiveness, pp. 54–​83 (New York, NY: Oxford University Press). Auerswald, P.E. and Branscomb, L.M. (2008). ‘Research and innovation in a networked world’. Technology in Society 30: 339–​347. Auerswald, P.E., Kauffman, S., Lobo, J., and Shell, K. (2000). ‘The production recipes approach to modeling technological innovation:  an application to learning by doing’. Journal of Economic Dynamics and Control 24: 389–​450. Bailey, R.G. (2009). ‘Scale of Ecosystem Units’ in R.G. Bailey (ed.) Ecosystem Geography: From Ecoregions to Sites, pp. 25–​28 (New York: Springer Science & Business Media). Banks-​Leite, C. and Ewers, R.M. (2009). Ecosystem Boundaries (Chichester: eLS, John Wiley and Sons). Bengtsson, J., Nilsson, S.G., Franc, A., and Menozzidi, P. (2000). ‘Biodiversity, disturbances, ecosystem function and management of European forests’. Forest Ecology and Management 132: 39–​50.

Economic Ecosystems   265 Bloom, N. and Van Reenen, J. (2010). ‘Why do management practices differ across firms and countries?’ Journal of Economic Perspectives 24: 203–​224. Boschma, R. (2015). ‘Towards an evolutionary perspective on regional resilience’. Regional Studies 49: 733–​751. Coase, R.H. (1937). ‘The nature of the firm’. Economica 4: 386–​405. Ericson, R. and Pakes, A. (1995). ‘Markov-​perfect industry dynamics: a framework for’. Review of Economic Studies 62: 53–​82. Essletbichler, J. (2007). ‘Diversity, Stability and Regional Growth in the United States, 1975–​2002’ in K. Frenken (ed.) Applied Evolutionary Economics and Economic Geography, pp. 203–​299 (Cheltenham: Edward Elgar). Evenson, R.E. and Kislev, Y. (1976). ‘A stochastic model of applied research’. Journal of Political Economy 84: 265–​281. Fagerberg, J. (2005). ‘Innovation: A Guide to the Literature’ in J. Fagerberg, D. Mowery, and R.R. Nelson (eds) Oxford Handbook of Innovation, pp. 1–​28 (New York: Oxford University Press). Feldman, M.P. and Audretsch, D.B. (1999). ‘Innovation in cities: science-​based diversity, specialization, and localized competition’. European Economic Review 43: 409–​429. Feldman, M. and Zoller, T.D. (2011). ‘Dealmakers in place:  social capital connections in regional entrepreneurial economies’. Regional Studies 46: 23–​37. Fingleton, B., Garretsen, H., and Martin, R. (2012). ‘Recessionary shocks and regional employment: evidence on the resilience of U.S. regions’. Journal of Regional Science 52: 109–​133. Foster, L., Haltiwanger, J., and Syverson, C. (2008). ‘Reallocation, firm turnover, and efficiency: selection on productivity or profitability?’ American Economic Review 98: 394–​425. Freeman, C. and Soete, L. (2009). ‘Developing science, technology and innovation indicators: what we can learn from the past’. Research Policy 38: 583–​589. Frenken, K., Van Oort, F.G., and Verburg, T. (2007). ‘Related variety, unrelated variety and regional economic growth’. Regional Studies 41: 685–​697. Gavrilets, S. (1997). ‘Evolution and speciation on holey adaptive landscapes’. Trends in Econolgy and Evolution 12: 307–​312. Gavrilets, S. (1999). ‘A dynamical theory of speciation on holey adaptive landscapes’. The American Naturalist 154: 1–​22. Gavrilets, S. (2004). Fitness Landscapes and the Origin of Species (Princeton, NJ: Princeton University Press). Gavrilets, S. and Gravner, J. (1997). ‘Percolation on the fitness hypercube and the evolution of reproductive isolation’. Journal of Theoretical Biology 184: 51–​64. Geels, F.W. (2002). ‘Technological transitions as evolutionary reconfiguration processes:  a multi-​level perspective and a case-​study’. Research Policy 31: 1257–​1274. Genus, A. and Coles, A.-​M. (2008). ‘Rethinking the multi-​level perspective of technological transitions’. Research Policy 37: 1436–​1445. Glaeser, E.H., Kallal, H., Scheinkman, J., and Schleifer, A. (1992). ‘Growth in cities’. Journal of Political Economy 100: 1126–​1152. Gunderson, L. and Holling, C.S. (2001). Panarchy: Understanding Transformations in Human and Natural Systems (Washington, DC: Island Press). Hassnik, R. (2010). ‘Regional resilience: a promising concept to explain differences in regional economic adaptability?’ Journal of Regions, Economy, and Society 3: 45–​58. Hidalgo, C.A., Klinger, B., Barabasi, A., and Hausman, R. (2007). ‘The product space conditions the development of nations’. Science 317: 482–​487.

266   Auerswald and Dani Holling, C.S. (2001). ‘Understanding the complexity of economic, ecological, and social systems’. Ecosystems 4: 390–​405. Holling, C.S., Schindler, D.W., Walker, B., and Roughgarden, J. (1995). ‘Biodiversity in the Functioning of Ecosystems: An Ecological Primer and Synthesis’ in C. Perrings, K.G. Mäler, C. Folke, C.S. Holling, and B.O. Jansson (eds) Biodiversity Loss: Ecological and Economic Issues (Cambridge: Cambridge University Press). Hopenhayn, H.A. (1992). ‘Entry, exit, and firm dynamics in long run equilibrium’. Econometrica 60: 1127–​1150. Hsieh, C.-​T. and Klenow, P.J. (2009). ‘Misallocation and manufacturing TFP in China and India’. Quarterly Journal of Economics 124: 1403–​1448. Isenberg, D.J. (2010). ‘How to start an entrepreneurial revolution’. Harvard Business Review 88: 40–​50. Jacobs, J. (1961). The Death and Life of Great American Cities (New York: Random House). Jacobs, J. (1969). The Economy of Cities (New York: Random House). Jovanovic, B. (1982). ‘Selection and the evolution of industry’. Econometrica 50: 649–​670. Kash, D.E. and Rycroft, R. (1999). The Complexity Challenge: Technological Innovation for the 21st Century (London: Pinter). Kauffman, S.A. (1989). ‘Adaptation on Rugged Fitness Landscapes’ in D. Stein (ed.) Lectures in the Sciences of Complexity: The Santa Fe Series, pp. 527–​618 (New York: Addison Wesley). Kauffman, S.A. and Johnsen, S. (1991). ‘Coevolution to the edge of chaos: coupled fitness landscapes, poised states, and coevolutionary avalanches’. Journal of Theoretical Biology 149: 467–​505. Kauffman, S. and Levin, S. (1987). ‘Toward a general theory of adaptive walks on rugged landscapes.’ Journal of Theoretical Biology 128: 11–​45. Kauffman, S.A., Macready, W.G., and Dickinson, E. (1994). Divide to Coordinate: Coevolutionary Problem Solving (Santa Fe, NM: Santa Fe Institute). Kearns, J. (2015). ‘Satellite Images Show Economies Growing and Shrinking in Real Time’. Bloombery, 8 July https://​www.bloomberg.com/​news/​features/​2015-​07-​08/​satellite-​images-​ show-​economies-​growing-​and-​shrinking-​in-​real-​time (last accessed 19 March 2017). Krugman, P. (1991). ‘Increasing returns and economic geography’. Journal of Political Economy 99: 483–​499. Lucas, R.E. (1993). ‘Making a miracle’. Econometrica 61: 251–​272. McNerney, J., Farmer, J.D., Redner, S., and Trancik, J. (2010). ‘Role of design complexity in technology improvement’. Proceedings of the National Academy of Sciences of the United States of America 108: 9008–​9013. Marshall, A. (1920). Principles of Economics: 8th edition (London: Macmillan). Martin, R. (2012). ‘Regional economic resilience, hysteresis and recessionary shocks’. Journal of Economic Geography 12: 1–​32. Martin, R. and Sunley, P. (2007). ‘Complexity thinking and evolutionary economic geography’. Journal of Economic Geography 7: 573–​601. Muneepeerakul, R., Lobo, J., Shutters, S., Gomez-​Lievano, A., and Qubbaj, M.R. (2013). ‘Urban economies and occupation space: can they get “there” from “here”?’ PLoS ONE 8: e73676. Murcia, C. (1995). ‘Edge effects in fragmented forests—​implications for conservation’. Trends in Ecology and Evolution 10: 58–​62. Muth, J.F. (1986). ‘Search theory and the manufacturing production function’. Management Science 32: 948–​962. Neffke, F. and Henning, M. (2013). ‘Skill-​ relatedness and firm diversification’. Strategic Management Journal 34: 297–​316.

Economic Ecosystems   267 Neffke, F., Henning, M., and Boschma, R. (2011). ‘How do regions diversify over time? Industry relatedness and the development of new growth paths in regions’. Economic Geography 87: 237–​266. Nelson, R.R. and Winter, S.G. (1982). An Evolutional Theory of Economic Change (Cambridge, MA: Harvard University Press). Office of Management and Budget (2013). Revised Delineations of Metropolitan Statistical Areas, Micropolitan Statistical Areas, and Combined Statistical Areas, and Guidance on Uses of the Delineations of These Areas (Washington, DC: Executive Office of the President of the United States). Pike, A., Dawley, S., and Tomaney, J. (2010). ‘Resilience, adaptation and adaptability’. Cambridge Journal of Regions, Economy and Society 3: 59–​70. Piore, M.J. and Sabel, C.E. (1984). The Second Industrial Divide:  Possibilities for Prosperity (New York, NY: Basic Books). Porter, M.B. (1990). The Competitive Advantage of Nations (New York: The Free Press). Romer, P. (1986). ‘Increasing returns and long-​run growth’. Journal of Political Economy 94: 1002–​1037. Romer, P. (1990). ‘Endogenous technological change’. Journal of Political Economy 98: 1002–​1037. Romer, P.M. (1994). ‘The origins of endogenous growth’. Journal of Economic Perspectives 8: 3–​22. Rose, A. (2004). ‘Defining and measuring economic resilience to disasters’. Disaster Prevention and Management 13: 307–​314. Rosenthal, S.S. and Strange, W.C. (2001). ‘The determinants of agglomeration’. Journal of Urban Economics 50: 191–​229. Rosenthal, S.S. and Strange, W.C. (2004). ‘Evidence on the nature and sources of agglomeration economies’ in G. Duranton, J.V. Henderson and W.C. Strange (eds) Handbook of Regional and Urban Economics, Volume 4, pp. 2119–​2171 (San Diego, CA: Elsevier). Saxenian, A. (1994). Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA: Harvard University Press). Schumpeter, J.A. (1911). The Theory of Economic Development. An Inquiry Into Profits, Capital, Credit, Interest and the Business Cycle (Cambridge, MA: Harvard University Press). Schumpeter, J.A. (1939). Business Cycles: A Theoretical, Historical and Statistical Analysis of the Capitalist Process (New York: McGraw-​Hill Book Company). Siggelkow, N. and Levinthal, D.A. (2003). ‘Temporarily divide to conquer:  centralized, decentralized, and reintegrated organizational approaches to exploration and adaptation’. Organization Science 14: 650–​669. Spencer, H. (1857). Progress: Its Law and Cause (New York: J. Fitzgerald). Stadler, B.M., Stadler, P.F., Wagner, G.P., and Fontana, W. (2001). ‘The topology of the possible:  formal spaces underlying patterns of evolutionary change’. Journal of Theoretical Biology 213: 241–​274. Stangler, D. and Bell-​ Masterson, J. (2015). Measuring an Entrepreneurial Ecosystem (Kansas City, MO: Kauffman Foundation Research Series on City, Metro, and Regional Entrepreneurship). Strumsky, D. and Lobo, J. (2003). ‘If it Isn’t Broken, Don’t Fix it:  Extremal Search on a Technology Landscape’. SFI Working Paper Series. Strumsky, D., Lobo, J., and van der Leeuw, S. (2012). ‘Using patent technology codes to study technological change’. Economics of Innovation and New Technology 21: 267–​286.

268   Auerswald and Dani Syverson, C. (2004a). ‘Market structure and productivity: a concrete example’. Journal of Political Economy 112: 1181–​1222. Syverson, C. (2004b). ‘Product substitutability and productivity dispersion’. Review of Economics and Statistics 86: 534–​550. Syverson, C. (2011). ‘What determines productivity?’ Journal of Economic Literature 49: 326–​365. Tansley, A.G. (1935). ‘The use and abuse of vegetational concepts and terms’. Ecology 16: 284–​307. Van de Ven, A., Polley, D.E., and Venkataraman, S. (1999). ‘Building an Infrastructure for the Innovation Journey’ in A. Van de Ven, D. E. Polley, and S. Venkataraman (eds) The Innovation Journey, pp. 149–​180 (New York: Oxford University Press). Veblen, T. (1898). ‘Why is economics not an evolutionary science?’ The Quarterly Journal of Economics 12: 373–​397. Viner, J. (1932). ‘Cost curves and supply curves’. Zeitschrift für Nationalökonomie 3: 23–​46. Wallsten, S.J. (2001). ‘An empirical test of geographic knowledge spillovers using geographic information systems and firm-​level data’. Regional Science and Urban Economics 31: 571–​599. Weitzman, M.L. (1979). ‘Optimal search for the best alternative’. Econometrica 47: 641–​654. Winter, S.G. (1968). Toward a Neo-​Schumpeterian Theory of the Firm (Santa Monica, CA: The RAND Corporation). Wright, S. (1932). ‘The Roles of Mutation, Inbreeding, Crossbreeding, and Selection in Evolution’. Proceedings of the Sixth International Congress on Genetics, Vol. I, pp. 356–​366.

Chapter 14

How Geo g ra ph y Shapes—and I s Sha pe d by —the Int e rnet Chris Forman, Avi Goldfarb, and Shane Greenstein Introduction In the mid-​1990s the inter-​networking infrastructure and protocols of the research community became available to private users. Known colloquially as the ‘Internet’, and newly enhanced with the additional functionality of the World Wide Web, the Internet became adopted widely and spread quickly. The new network grew at an astonishing rate, and transformed economic activity across a breadth of sectors. The Internet almost defied historical precedent for combining rapid diffusion with large economic impact, generating large and impatient investments by households and businesses, and catalysing major restructuring across a large range of industries (Greenstein, 2015). What consequences did the spread of the Internet have for the geographical location of economic activity? The precise economic effects defy simple description. Offline, distance creates a variety of economic frictions and, at first glance, the Internet reduces these frictions to be more similar across locations. However, because the strength of these frictions differs across places, the marginal impact of removing them will differ. To put it plainly, the Internet has different economic consequences in distinct locations because local factors shape the impact. Specifically, the Internet reduces the importance of three interrelated location-​ level frictions in economic transactions:  communication costs, transportation costs, and search costs. Communication costs are lower on the Internet because it is inexpensive to communicate with others, whether they are in the same building or across the world. The cost of sending and receiving email and other forms of digital communication are the same, irrespective of the distance between the sender and receiver. Transportation costs are lower on the Internet in two ways. Firstly, for goods that can be digitized, distribution is nearly free. Delivering physical newspapers to people’s doors

270   Forman et al. requires a costly operation to transport the physical object; once digital infrastructure is in place, delivering digital news has almost zero marginal cost. Music, books, video, and other information-​based goods also benefit from near-​zero distribution costs. Secondly, for items that cannot be digitized, online interactions reduce the need for travel. A consumer does not need to travel when ordering an item to be shipped. Digital conferencing can reduce the need for physical travel. The reduction in search costs on the Internet follows directly from lower transportation and communication costs. If communication costs and transportation costs fall, then it is easier to compare potential choices before making a selection. These choices can not only be physical goods, but they can also be business decisions such as outsourcing and hiring. Economic models show that lower search costs have distinct implications from the general models of lower communication and transportation costs. We begin with the hypothesis that the Internet had a variety of effects across locations because the impact of lower communication, transportation, and search costs varies by location, depending on three main factors:  local preferences, the availability of substitutes, and the availability of complements. The principal goal of this chapter is to show the manifest ways in which research has documented the scope of changes. We build on prior reviews of the interaction of online and offline, especially Forman (2014) and also Goldfarb (2012) and Lieber and Syverson (2012). Zook’s chapter in this book (Chapter 30) complements our approach and highlights how the Internet has affected the geography of global capital flows, reducing the importance of distance in some ways, and increasing it in others. We begin by discussing models about whether and how the Internet might decrease the importance of geographical factors in economic activity. This is followed by an exploration of how location has affected the incentives for firms and consumers to adopt Internet technology. The following sections then, in turn, discuss the consequences of Internet adoption for the geography of wealth and productivity, innovation, consumer purchasing, and globalization and trade. We conclude with some directions for future research.

Death of Distance? Perhaps the most commonly discussed framework is the ‘death of distance’ model popularized by Cairncross’s (1997) book of the same title and Friedman’s (2005) The World is Flat. This model assumes that the Internet is a substitute for other communication, transportation (or distribution), and search technologies. Electronic communication therefore implies that agents will rely less on offline communication channels (especially face-​to-​face), and substitute into digital communication. Similarly, Internet distribution replaces physical transportation of goods and Internet search replaces physical search. A key implication of these models is that cities become less important as the Internet diffuses: there is a decrease in the relative value of agglomeration benefits related to low communication, transportation, and search costs within the city. One implicit assumption for such models is that the costs of using the electronically enabled channel will fall for all types of communication, transportation, and search equally: they

How Geography Shapes—and is Shaped by—the Internet    271 do not highlight the comparative advantage that one channel may enjoy over others in certain types of communication. In search, some attributes may be easier to appreciate online than others (Lal and Sarvary, 1999; Brown and Goolsbee, 2002). In transportation, some goods may be easier to ship than others (Lal and Sarvary, 1999). Even in communication, a large body of literature has shown the unique advantages that face-​to-​face interaction has over other forms of communication,1 particularly in communicating certain kinds of tacit or ‘sticky’ knowledge (e.g. von Hippel, 1998). For example, while information technology (IT) may be an effective means of coordinating ongoing projects and collaborations, it may be less effective as a means of establishing new partnerships or collaborative relationships (e.g. Gaspar and Glaeser, 1998; Charlot and Duranton, 2006; Glaeser and Ponzetto, 2007). In this way, rich offline communication might make online communication more valuable, suggesting complementarity between face-​to-​ face and digital communication (Gaspar and Glaeser, 1998).

Adoption and Use of the Internet Geographical variance in the adoption of IT produces variance in reduction of search, distribution, and especially communication costs. That is because adoption of the Internet relies on complementary inputs such as broadband service or expertise in implementing IT systems, which varies across locations, creating variance in the net benefits to adoption. As a result, the diffusion of the Internet did not have the same impact on all locations.

Internet Adoption Among Firms In Forman et al. (2005), we demonstrated that the costs and gross benefits of technology adoption may vary significantly across geographical locations, often in quite different ways. For example, the highest-​value users of Internet technology may reside in rural areas and small cities, because of the potential for the Internet to reduce the costs of economic isolation. However, the costs of adoption may also be highest in such regions because of the absence of key complementary inputs such as broadband service, skilled labour, and third-​ party services. We showed that the net benefits of adoption for firms in 2000 were decreasing in location size for basic Internet technologies that reduced the costs of economic isolation. Further, because implementation is straightforward, the frictions associated with adoption costs do not vary much across locations. In contrast, the net benefits of adoption are increasing in location size for advanced technologies that require extensive adaptation and co-​ invention (Bresnahan and Greenstein, 1996) to be used successfully. This is particularly true for establishments in small, single-​establishment firms that are unable to rely on IT skills and other complementary inputs that may reside elsewhere in the organization and that rely on external markets to facilitate implementation of new IT (Forman et al., 2008). Thus, the net benefits of Internet-​enabled frontier services are frequently higher in urban areas. Using data on domain name registrations, Kolko (2000) and Moss and Townsend (1997) also document cases in which adoption of Internet technology increases location size.

272   Forman et al. Because domain name registrations are based on the business location of the registration rather than that of the Internet service provider (ISP), they can be a useful measure of the location of Internet adoption; however, they are limited because some firms have Internet but no domain name and because domain names are typically only associated with the headquarters of multi-​establishment firms, which are disproportionately in larger cities (Aarland et al., 2007) The availability of complements necessary for Internet adoption appear to be increasing in location size. For example, the supply of third-​party outsourcing firms that are used in IT implementations are increasing in location size, as are the use of outsourcing services (Arora and Forman, 2007; Ono, 2007). As outsourcing firms frequently play a key role in the implementation of frontier Internet services, such variance in local outsourcing supply is likely to influence adoption costs. Further, a rich literature has demonstrated variance in the supply of broadband providers and the quality of the ISPs across locations (e.g. Greenstein, 2000; Downes and Greenstein, 2002; Prieger, 2003; Flamm, 2005; Zook, 2005). Such variance may influence the costs of technical inputs such as broadband service and, more broadly, Internet connectivity.

IT Adoption Among Individuals A variety of studies has shown variance in the extent of individual Internet adoption across urban and rural (and small city) regions (e.g. Hindman, 2000; Wellman et al., 2001; Mills and Whitacre, 2003; Agarwal et al., 2009). Much of the urban–​rural variance appears as a result of other factors that may be correlated with location, such as income, age, and education (e.g. Hindman, 2000; Mills and Whitacre, 2003). Even after controlling for these demographic factors, location characteristics may play an independent role on adoption because of differences in the availability of complements such as the local supply of Internet service. Owing to low fixed costs of entry and the widespread availability of entrepreneurs from related industries, dial-​up ISPs quickly entered most small geographical markets relatively soon after the commercialization of the Internet, providing access at competitive rates in all but the smallest markets (Downes and Greenstein, 2002; Greenstein, 2015). In contrast, owing to differing economics of deployment, differences in the supply of broadband providers have persisted (Prieger, 2003; Flamm, 2005; Zook, 2005; Grubesic, 2012). Spillovers have also likely shaped the adoption of Internet technology among individuals. For example, Agarwal et al. (2009) have shown that adoption is correlated among users in the same metropolitan statistical area. Their results support the existence of network effects, particularly among those located in regions with high housing density and those within a dense network of social interactions. The approach of Agarwal et al. (2009) builds upon earlier work by Goolsbee and Klenow (2002), who found evidence that spillovers related to Internet use influenced the adoption of personal computers. Goldfarb (2006) also provides evidence that local spillovers influenced individual adoption of the Internet: the impact of prior university attendance on Internet use is much higher for people who attended university in the mid-​1990s (and people who live with them) than for others.

How Geography Shapes—and is Shaped by—the Internet    273

Consequences of the Internet for Wealth and Productivity A small literature has focused on the implications of IT investment for local economic outcomes, shedding light on whether the diffusion of the Internet has contributed to convergence or divergence in income. One view supporting convergence stresses that the diffusion of the Internet led to employment growth and wage gains in regions that have low populations and are economically isolated; that is, by lowering communication costs, the Internet contributes most to economic growth in regions that are not well off. An alternative view stresses divergence; that is, that the Internet disproportionately benefits regions that have large cities with highly skilled populations who already have high incomes. Research in this area is, in part, motivated by work on the broader literature on enterprise IT that has emphasized the value of complementary labour and information inputs to achieving value from IT systems (e.g. Bresnahan and Greenstein, 1996; Bresnahan et al., 2002; Bloom et al., 2012). As was highlighted in the previous section, the supply of some of these complements is distributed heterogeneously across locations. The returns to IT investment may be greater in large cities owing to Marshallian externalities. In addition, theories of skill-​biased technical change have suggested that IT investment is complementary with skilled labour (e.g. Katz and Autor, 1999; Autor et al., 2003), which is found in larger quantities within cities. Further, the productivity benefits of IT investment have been shown to be systematically higher for a subset of ‘IT-​intensive’ industries that have typically had long-​lived IT capital investments (e.g. Stiroh, 2002; Jorgenson et al., 2005), and the productivity benefits of IT investments among IT-​intensive firms have been found to be particularly strong in large cities (e.g. Henderson, 2003). The presence of IT-​intensive industries can indirectly increase the supply of other complements. As a result, firms’ response to new IT will be likely be non-​uniform across regions, leading to divergence. Forman et al. (2012) examined the relationship between investment in advanced Internet technology and local wage growth across 2743 counties in the USA. We identified a puzzle in the relationship between Internet investment and local wages. On average, the impact of advanced Internet investment on local wages is small and on employment is non-​existent. However, investment in the Internet is correlated with wage and employment growth in about 6 per cent of US counties, representing 42 per cent of the US population. These counties were already well-​off prior to 1995, with high income, large populations, high skills, and concentrated IT use. These well-​off counties averaged 28 per cent wage growth from 1995 to 2000 (unweighted by population), while all counties averaged just 20 per cent wage growth over this period. The Internet exacerbates regional wage inequality, explaining over half the additional wage growth experienced by the 6 per cent of counties that were already well-​off. This is the pay-​off puzzle: only a few counties experienced wage growth, despite widespread Internet investment. Dranove et al. (2014) demonstrate a similar result in exploring the effects of health information technology on hospital costs between 1996 and 2009. Hospitals in IT-​intensive

274   Forman et al. locations experienced declines in costs three years after adoption, while hospitals in other locations experienced an increase in costs even after six years. These differences appear to be driven by agglomeration of IT employment in other hospitals. There is less empirical evidence on convergence. One example is Kolko (2007), who finds that service industries that trade with one another are more likely to collocate within the same zip code if IT use is high, but less likely to collocate in the same state if IT use is high. One potential interpretation of this result is that IT reduces the costs of transporting services output over long distances. Formal theory has provided some insights into other ways that the Internet may reshape the geography of production through changes to communication patterns. For example, Gaspar and Glaeser (1998) present a model and some empirical evidence that show that Internet-​enabled communications may not only substitute for some face-​to-​ face interactions but also increase the demand for face-​to-​face interactions that cannot easily be performed electronically. This can make cities even more appealing, particularly for industries or activities where ideas are complex and difficult to communicate electronically. Lower communication costs may enable some industries like manufacturing to move out of cities. However, innovative ideas-​producing industries that involve frequent exchange of complex knowledge will likely continue to rely on face-​to-​face interactions in cities, and these industries will likely continue to be agglomerated (Glaeser and Ponzetto, 2007). Such industry-​level variance has been used to explain the resurgence of cities with high concentrations of innovative industries, like New  York and San Francisco, and for the decline of traditional manufacturing centres, such as Detroit and Cleveland. Similarly, the costs of performing headquarters and support functions have historically been lower in cities owing to the propensity of such functions to outsource business services (Aarland et  al., 2007), while some production activities can take place outside of big cities. If Internet-​enabled communications costs fall so may the costs of managing multiple firm locations. As a result, we may observe both an increasing number of multi-​unit organizations and a specialization of cities by functional area (Duranton and Puga, 2005). Much work has investigated the role of upgrades to Internet access in the form of broadband technologies. Several papers have also investigated the relationship between local broadband availability and local economic outcomes (e.g. Gillett et  al., 2006; Crandall et al., 2007; Van Gaasbeck et al., 2007; Kolko, 2012). These have consistently found a positive relationship on average between local broadband availability and local employment growth. However, as Kolko (2012) notes, the majority of these studies show correlation, not causality—​it is difficult to identify empirically whether broadband provision leads to employment growth or whether regions with high employment growth have more broadband providers. Further, none of these studies examine heterogeneity in outcomes in region; in particular, whether broadband availability disproportionately benefits large or small cities or urban or rural regions. Overall, the evidence suggests Internet technology has the biggest economic effect on larger cities, although there are particular situations in which small cities and rural areas do benefit.

How Geography Shapes—and is Shaped by—the Internet    275

Consequences of the Internet for Innovation Lower communications costs can influence the productivity of scientists, as well as shape the localization of innovation.2 This line of research is important because understanding what influences the productivity of individual scientists is inherently important for firms and the science of science policy. Further, researchers have long observed that innovative activity is localized (e.g. Jaffe et al., 1993), and there is an important public policy question about the extent to which recent declines to communication costs can enable more geographically dispersed innovation (e.g. Macher and Mowery, 2008). If the diffusion of the Internet has differing implications for the geographical concentration of innovation compared with the geographical concentration of production, this suggests important changes in the geographical concentration of the value chain over the very long term (Forman et al., 2015). Most research in this area has examined how IT investments have influenced research productivity and collaboration patterns in the academe. Research in this area provides ambiguous predictions, reflecting not only differences in research design, but also differences in setting, ranging from economics (Hammermesh and Oster, 2002; Rosenblatt and Mobious, 2004), life sciences (Winkler et  al., 2009; Ding et  al., 2010), and engineering (Agrawal and Goldfarb, 2008) to other disciplines (Jones et al., 2008). Several authors have examined time-​series trends in research collaborations. Such research generally finds that co-​authorship has risen over time across a wide variety of fields (Hammermesh and Oster, 2002; Jones et  al., 2008), and co-​authorship has increasingly spanned university boundaries (Jones et al., 2008) and metropolitan areas (Hammermesh and Oster, 2002). While research has found that co-​authorship has increased, it has also found that the trend has been increasingly towards segmentation or ‘balkanization’ of research collaboration. If agents have preferences to collaborate with those with similar preferences and if IT lowers the costs of collaborating across distance, than the diffusion of new IT will lead to a ‘balkanization’ of communities and that within-​group separation may increase even while the costs of distant collaborations decline (Rosenblatt and Mobius, 2004; Van Alstyne and Brynjolfsson, 2005). Research has found a direct empirical link between digital communication and the rise in research collaboration. Agrawal and Goldfarb (2008) examines how BITNET adoption influenced the likelihood of collaboration among engineering scientists in universities. They find that BITNET disproportionately increased collaboration between top-​tier and middle-​ tier universities, particularly among those that were collocated. They argue that this result may reflect gains from trade, perhaps through the increased use of underutilized research equipment or increased specialization. Research on life scientists has also shown that investments in the Internet are associated with an increase in researcher productivity (Winkler et al., 2009; Ding et al., 2010), and disproportionately aids researchers in lower tier institutions (Winkler et al., 2009) and female scientists (Ding et al., 2010). Forman and van Zeebroeck (2012) examine commercial innovation, demonstrating that Internet adoption leads to a significant increase in the likelihood of research collaborations among inventors within the same firm in geographically dispersed locations, but no

276   Forman et al. increase in collaborations among collocated inventors. That is, their data support the view that Internet adoption led to a decline in the costs of coordinating distant research. Forman et  al. (2015) examined whether the Internet increased or decreased the geographical concentration of invention. This can be framed much like the prior work on wages and productivity:  convergence or divergence. The results generally favour the view that the Internet worked against the concentration of invention. In particular, while invention became more geographically concentrated over this period, this is not true for the counties that were leaders in business Internet adoption.

Consequences of the Internet for Consumer Behaviour Electronic commerce can benefit consumers through lower prices (surveyed in Baye et  al., 2006), lower search costs and greater product selection among online merchants (Brynjolfsson et al., 2003), and better convenience (e.g. Sinai and Waldfogel, 2004; Forman et al., 2009). These benefits are likely to be particularly salient for consumers in small local markets, where the number of retailers is likely to be low and where consumers with idiosyncratic or minority preferences may be underserved (e.g. George and Waldfogel, 2003; Waldfogel, 2003). In addition, consumers with relatively unusual preferences may benefit, even if they live in densely populated areas (Sinai and Waldfogel, 2004; Choi and Bell, 2011). The Internet reduces distribution costs for a variety of consumer goods. If distribution is digital, as for music, news, and (increasingly) movies and books, then the marginal cost of distribution is near zero, across all locations. Even when distribution still requires shipping, these costs fall, and they likely fall more for relatively isolated individuals. They also fall more for some goods than other goods (Ellison and Ellison, 2009). For example, distribution costs remain high online for big-​ticket items such as automobiles (Overby and Forman, 2014). The fall in distribution costs is likely to improve the welfare of economically isolated consumers. Balasubramanian (1998) structured the main consequences of lower distribution costs in his model of consumer channel choice in commodity markets. In the model, consumers trade-​off between the fixed disutility costs of buying online (e.g. shipping time or the inability to view the product physically) and the transportation costs of buying online (in addition to the price differences between the two channels). A variety of empirical papers has documented the usefulness of this framework on understanding geographical patterns in online purchasing. Forman et al. (2009) show substitution between online and offline stores: when a Walmart, Barnes and Noble, or Borders store opens offline, local sales on Amazon change. Brynjolfsson et al. (2009) show substitution between online and offline channels in apparel. By lowering distribution costs, electronic commerce lowers the sum of purchase price plus transportation costs for consumers. Lower distribution costs online also enable a wider selection of products, particularly for geographically isolated customers. Online retailers have the potential to offer a much greater product selection than what can be carried by any physical store (Brynjolfsson et al., 2003; Anderson, 2006) and new search tools make it easier to find niche products. As a result, consumer propensity to purchase such ‘niche’ or ‘long tail’ products online is greater in

How Geography Shapes—and is Shaped by—the Internet    277 online than offline channels (Zentner et al., 2013) and is insensitive to local offline supply (Brynjolfsson et al., 2009; Forman et al., 2009). Electronic commerce can improve offline options if information about offline products is offered online. This can lead to either lower average prices or less price dispersion, or both. For example, the increasing use of electronic commerce in wholesale car auctions led to price convergence across geographically dispersed auction sites: buyers increasingly shifted from using high-​price to low-​price auction sites because they could more easily observe prices in other locations (Overby and Forman, 2014). Competition from the online channel also shifted the market structure of offline industries such as travel agencies, bookstores, and new-​car dealers towards larger lower-​cost establishments (Goldmanis et al., 2009). While growing evidence shows that consumers use the Internet to lower the costs of economic isolation, there also remains significant evidence that location significantly shapes how consumers behave online (see Bell, 2014 for a review). Consumers are more likely to visit websites that are hosted by firms geographically proximate to them. One reason is that a lot of online content is local (Sinai and Waldfogel, 2004); further, consumer tastes are spatially correlated and vary significantly across regions (e.g. Jank and Kannan, 2005). As a result, consumers may self-​select into websites that tailor to their preferences; this is particularly true for taste-​dependent digital products such as music, games, and pornography (Blum and Goldfarb, 2006). Consumers have also shown a preference to trade with others who are geographically proximate to them. Again, this is, in part, because many products are heavily taste dependent or consumed locally (Hortacsu et al., 2009). In addition to reductions in communication, distribution, and search costs, there are regulatory reasons why consumer use of the Internet varies by location. For example, Internet purchasing is higher in places with higher sales taxes (Goolsbee, 2000; Ellison and Ellison, 2009; Anderson et al., 2010; Einav et al., 2014). Internet advertising is more expensive and more effective in places with restrictions on offline advertising (Goldfarb and Tucker, 2011a). Copyright can add costs to international distribution (Aguiar and Waldfogel, 2014). Also, regulation of the Internet varies across locations. For example, privacy regulation and Net neutrality regulation vary across countries (Goldfarb and Tucker, 2011b; Lee and Wu, 2009). The interplay of consumer behaviour between the online and offline worlds is particularly important to understanding mobile Internet behaviour. Consumer behaviour on the mobile Internet differs from that on the traditional fixed-​line Internet: some types of search costs are higher and consumers are even more likely to browse on websites that are geographically proximate (Ghose et al., 2013). Research has emphasized how mobile advertising can influence demand for local products and services (Ghose et al., 2014; Luo et al., 2014).

Consequences of the Internet for Globalization and Trade A wealth of anecdotal evidence has been used to argue that IT investments, by reducing coordination costs, have facilitated the globalization of economic activity (e.g. Friedman, 2005). However, systematic empirical evidence on the link between Internet investment and globalization is relatively rare. One exception is the work of Freund and Weinhold (2002,

278   Forman et al. 2004), who examine the association between IT investment and trade in goods and services and find that increases in web hosts are associated with an increase in goods exports and an even greater climb in services exports. They argue that the Internet will have a greater impact on trade volume in services as many services can now be transported costlessly. In other words, trade in goods is likely to increase because of lower search and communication costs, while trade in services is likely to increase even more because transportation costs also fall through digital distribution. Recent work has sought to understand the implications of IT investments for global supply chains. Fort (2014) finds that IT investments lead to the fragmentation of production in industries where production specifications are most commonly formalized in electronic formats. However, Fort’s evidence also suggests that electronic communication lowers the coordination costs of production fragmentation disproportionately more for domestic than for foreign sourcing. Fort argues this latter result may reflect a need for skilled suppliers. Country-​level IT adoption is also associated with an increase in north–​south vertical mergers, particularly for industries with low ‘routineness’; this may suggest that IT investments will facilitate monitoring of distant enterprises (Basco and Mestieri, 2014). While systematic evidence linking IT investments to changes in trade or the structure of global supply chains are relatively rare, researchers have claimed that time series trends in each of these variables may be influenced, in part, by IT investments. For example, Blinder (2006) and others have argued that what can be traded has been steadily increasing over time. While traditionally any good that can be shipped and placed in a box has been considered tradable, IT has changed the sets of services that can be delivered at a distance (Blinder, 2006). Some services work cannot ever be done at a distance—​hairdressers need hair to cut and janitors cannot be in another continent to do their job (Blinder, 2006). However, there are a number of occupations at the margin, and recent research has attempted to identify which kinds of work presently represent tradable services or soon will. The methods used to answer this question vary widely. One approach is to examine the geographical concentration of an industry to identify whether its outputs are likely to be tradable (Jensen and Kletzer, 2005). A more common method has been to construct an index of whether work is routine or can be codified using descriptions of work from references like the US Dictionary of Occupational Titles. Work that is more routine or can be more easily codified can, for example, be more easily performed at a distance (e.g. Mithas and Whitaker, 2007; Autor and Dorn, 2013; Basco and Mestieri, 2014). Overall, research in this area demonstrates that many occupations and forms of work remain non-​tradable, although the margin between tradable and non-​tradable is likely changing over time, and IT investments are increasing the potential for transactions to span great distances and country boundaries.

Directions for Future Research This chapter has summarized the literature on the importance of geography in understanding the causes and consequences of the diffusion of the Internet. It has emphasized that the

How Geography Shapes—and is Shaped by—the Internet    279 Internet reduces three key interrelated economic frictions: communication costs, transportation costs, and search costs. The impact of reducing these frictions varies across locations. It depends on three factors that vary locally: preferences, the availability of substitutes, and the availability of complements. Thus, a reduction in communications cost benefits some locations more than others. Overall, the existing literature that stresses ‘the death of distance’ is too simplistic; offline differences across locations leads to heterogeneous impacts of Internet technology, and many offline factors mediate how the diffusion of the Internet shapes the location of economic activity. This literature is far from complete. A number of open questions remain, particularly with respect to the broad economic consequences of Internet adoption across geographies and with respect to how these consequences might change as the technology changes. In terms of the broad geographical consequences, one key open puzzle is why the Internet has only increased economic outcomes in a small number of locations. Despite the promise of a death of distance, the Internet has thus far exacerbated differences in income across locations. One promising avenue to study this puzzle comes from microdata on the wage profiles of individuals, either from US census data (Vilhuber and McKinney, 2014) or from online resume and career histories (e.g. Agrawal and Tambe, 2016). A second open puzzle is how the Internet has affected trade in services. The key challenge to study this question is that services often do not cross borders in easily measured ways. In the absence of systematic internationally representative data, one promising avenue might be to do case studies of particular firms (or even government agencies) that provide access to their email and web browsing. One source of such information is legal proceedings, as with the release of the Enron emails. A related approach identifies the role of strong and weak institutions for policing piracy across countries (Athey and Stern, 2015). Technological change will also continue to provide new questions. Much of the research in this chapter has focused on the causes and economic consequences of broadband adoption. Smartphones and the mobile Internet are already changing the geography of consumer behaviour and are likely to affect firms and non-​profits. Social media provide the opportunity to meet, and keep in touch with, friends from around the world, yet most social networks remain highly local (Agrawal et al., 2015). Zook’s chapter (Chapter 30) provides case studies of how new technologies such as Bitcoin can rework global capital flows. We foresee the prospects for research about how new applications that make use of broadband, smartphones, and social media affect economic activity. New Internet applications do not diffuse evenly across geography and such uneven diffusion provides quasi-​natural experiments for understanding the impact of the Internet on local economic activity. For example, Craigslist did not deploy across US geography all at once, and that enabled research to analyse its impact on newspaper advertising (Seamans and Zhu, 2014). In a similar spirit, ride-​sharing services such as Uber build off the use of smartphones, and have not entered every city at the same time, enabling research to examine how its deployment affects activities, such as drunk driving (Greenwood and Wattal, forthcoming). A variety of other technologies are on the horizon that may also someday influence the location of economic activity. These technologies will provide new puzzles to study and deeper understanding of the interaction between digital communication and economic geography.

280   Forman et al.

Notes 1. Starting with Polanyi (1966) but continuing on in diverse fields such as urban economics and regional studies (e.g. Moss, 1998; Leamer and Storper, 2001; Storper and Venables, 2004) and communications (e.g. Daft et al., 1987). 2. Lower search costs also have important implications for innovation, but less so for the geography of innovation. In particular, McCabe and Snyder (2015) show that lower online search costs through online referencing and JSTOR lead to increased citations to past work.

References Aarland, K., Davis, J.C., Vernon Henderson, J., and Ono, Y. (2007). ‘Spatial organization of firms:  the decision to split production and administration’. RAND Journal of Economics 38: 480–​494. Agrawal, A. and Goldfarb, A. (2008). ‘Restructuring research: communication costs and the democratization of university innovation’. American Economic Review 98: 1578–​1590. Agrawal, A.K. and Tambe, P. (2016). ‘Private equity and workers’ career paths:  the role of technological change’. The Review of Financial Studies 29: 2455–​2489. Agarwal, R., Animesh, A., and Prasad, K. (2009). ‘Social interactions and the “digital divide”: explaining variations in internet use’. Information Systems Research 20: 277–​294. Agrawal, A., Catalini, C., and Goldfarb, A. (2015). ‘Crowdfunding: geography, social networks, and the timing of investment’. Journal of Economics and Management Strategy 24: 253–​274. Aguiar, L. and Waldfogel, J. (2014). ‘Digitization, copyright, and the welfare effects of music trade.’ Working paper, University of Minnesota. Anderson, C. (2006). The Long Tail:  Why the Future of Business is Selling Less of More (New York: Hyperion). Anderson, E., Fong, N., Simester, D., and Tucker, C. (2010). ‘How sales taxes affect customer and firm behaviour:  the role of search on the Internet.’ Journal of Marketing Research 47: 229–​239. Arora, A. and Forman, C. (2007). ‘Proximity and information technology outsourcing: how local are IT services markets’. Journal of Management Information Systems 24: 73–​102. Athey, S. and Stern, S. (2015). ‘The Nature and Incidence of Software Piracy: Evidence from Windows’ in A. Goldfarb, S. Greenstein, and C. Tucker (eds) Economic Analysis of the Digital Economy, pp. 443–​481 (Chicago, IL: University of Chicago Press). Autor, D. and Dorn, D. (2013). ‘The growth of low-​skill service jobs and the polarization of the US labor market’. American Economic Review 103: 1553–​1597. Autor, D., Levy, F., and Murnane, R.J. (2003). ‘The skill content of recent technological change: an empirical exploration’. Quarterly Journal of Economics 118: 1279–​1334. Balasubramanian, S. (1998). ‘Mail versus mall: a strategic analysis of competition between direct marketers and conventional retailers’. Marketing Science 17: 181–​195. Basco, S. and Mestieri, M. (2014). ‘Mergers along the global supply chain: information technologies and routineness.’ Working Paper, Toulouse School of Economics. Baye, M., Morgan, J., and Scholten, P. (2006). ‘Information, Search, and Price Dispersion’ in T. Hendershott (ed.). Handbook of Economics and Information Systems, pp. 323–​376 (Amsterdam: North-​Holland).

How Geography Shapes—and is Shaped by—the Internet    281 Bell, D. (2014). Location is Still Everything: The Surprising Influence of the Real World on How We Search, Shop, and Sell in the Virtual One (Boston, MA: New Harvest). Blinder, A.S. (2006). ‘Offshoring: the next Industrial Revolution?’ Foreign Affairs 85: 113–​128. Bloom, N., Sadun, R., and Van Reenen, J. (2012). ‘Americans do IT better: US multinationals and the productivity miracle’. American Economic Review 102: 167–​201. Blum, B. and Goldfarb, A. (2006). ‘Does the internet defy the law of gravity?’ Journal of International Economics 70: 384–​405. Bresnahan, T. and Greenstein, S. (1996). ‘Technical progress in computing and in the uses of computers’. Brookings Papers on Economic Activity: Microeconomics 1996: 1–​78. Bresnahan, T.F., Brynjolfsson, E., and Hitt, L.M. (2002). ‘Information technology, workplace organization, and the demand for skilled labor: firm-​level evidence’. Quarterly Journal of Economics 117: 339–​376. Brown, J.R. and Goolsbee, A. (2002). ‘Does the internet make markets more competitive? Evidence from the life insurance industry’. Journal of Political Economy 110: 481–​507. Brynjolfsson, E., Hu, Y., and Smith, M. (2003). ‘Consumer surplus in the digital economy: estimating the value of increased product variety’. Management Science 49: 1580–​1596. Brynjolfsson, E., Hu, Y., and Rahman, M.S. (2009). ‘Battle of the retail channels: how product selection and geography drive cross-​channel competition’. Management Science 55: 1755–​1765. Cairncross, F. (1997). The Death of Distance (Cambridge, MA: Harvard University Press). Charlot, S. and Duranton, G. (2006). ‘Cities and workplace communication: some quantitative French evidence’. Urban Studies 43: 1365–​1394. Choi, J. and Bell, D. (2011) ‘Preference minorities and the internet’. Journal of Marketing Research 58: 670–​682. Crandall, R., Lehr, W., and Litan, R. (2007). ‘The effects of broadband deployment on output and employment:  a cross-​sectional analysis of U.S. data’. Issues in Economic Policy Discussion Paper No. 6, Brookings Institution. Daft, R.L., Lengel, R.H., and Trevino, L.K. (1987). ‘Message equivocality, media selection, and manager performance: implications for information systems’. MIS Quarterly 11: 354–​366. Ding, W.W., Levin, S.G., Stephan, P.E., and Winkler, A.E. (2010). ‘The impact of information technology on academic scientists’ productivity and collaboration patterns’. Working Paper, Haas School of Business, University of California, Berkeley. Downes, T. and Greenstein, S. (2002). ‘Universal access and local Internet markets in the U.S.’. Research Policy 31: 1035–​1052. Dranove, D., Forman, C., Goldfarb, A., and Greenstein, S. (2014). ‘The trillion dollar conundrum:  complementarities and health information technology’. American Economic Journal: Economic Policy 6: 239–​270. Duranton, G. and Puga, D. (2005). ‘From sectoral to functional urban specialization’. Journal of Urban Economics 57: 343–​370. Einav, L., Knoepfle, D., Levin, J., and Sundaresan, N. (2014). ‘Sales taxes and Internet commerce’. American Economic Review 104: 1–​26. Ellison, G. and Ellison, S.F. (2009). ‘Tax sensitivity and home state preferences in Internet purchasing’. American Economic Journal: Economic Policy 1: 53–​7 1. Flamm, K. (2005). ‘The role of economics, demographics, and state policy in broadband availability’. Working Paper, University of Texas, Austin. Forman, C. (2014). ‘How has Information Technology Use Shaped the Geography of Economic Activity?’ in F. Giarratani, G.J.D. Hewings, and P. McCann. (eds) Handbook of Industry Studies and Economic Geography, pp. 253–​270 (Cheltenham: Edward Elgar).

282   Forman et al. Forman, C. and van Zeebroeck, N. (2012). ‘From wires to partners: how the Internet has fostered R&D collaborations among firms’. Management Science 58: 1549–​1568. Forman, C., Ghose, A., and Goldfarb, A. (2009). ‘Competition between local and electronic markets: how the benefit of buying online depends on where you live’. Management Science 55: 47–​57. Forman, C., Goldfarb, A., and Greenstein, S. (2005). ‘How did location affect the adoption of the commercial Internet? Global village vs. urban density’. Journal of Urban Economics 58: 389–​420. Forman, C., Goldfarb, A., and Greenstein, S. (2008). ‘Understanding the inputs into innovation: do cities substitute for internal firm resources?’ Journal of Economics and Management Strategy 17: 295–​316. Forman, C., Goldfarb, A., and Greenstein, S. (2012). ‘The Internet and local wages: a puzzle’. American Economic Review 102: 556–​575. Forman, C., Goldfarb, A., and Greenstein, S. (2015). ‘Information Technology and the Distribution of Inventive Activity’ in A.B. Jaffe and B.F. Jones (eds) The Changing Frontier: Rethinking Science and Innovation Policy, pp. 169–​196 (Chicago, IL: University of Chicago Press). Fort, T.C. (2014). ‘Technology and production fragmentation: domestic versus foreign sourcing’. Working Paper, Tuck School of Business at Dartmouth College. Freund, C. and Weinhold, D. (2002). ‘The Internet and international trade in services’. The American Economic Review 92: 236–​240. Freund, C. and Weinhold, D. (2004). ‘The effect of the Internet on international trade’. Journal of International Economics 62: 171–​189. Friedman, T. (2005). The World is Flat:  A  Brief History of the Twenty-​ First Century (New York: Farrar, Straus, and Giroux). Gaspar, J. and Glaeser, E.L. (1998). ‘Information technology and the future of cities’. Journal of Urban Economics 43: 136–​156. George, L. and Waldfogel, J. (2003). ‘Who affects whom in daily newspaper markets?’ Journal of Political Economy 111: 765–​784. Ghose, A., Goldfarb, A., and Pil Han, S. (2013). ‘How is the mobile Internet different? Search costs and local activities’. Information Systems Research 24: 613–​631. Ghose, A., Li, B., and Liu, S. (2014). ‘Mobile trajectory-​based advertising:  evidence from a large-​scale randomized field experiment’. Working Paper, NYU. Gillett, S., Lehr, W., Osorio, C., and Sirbu, M. (2006). Measuring the Economic Impact of Broadband Deployment (Washington, DC: U.S. Department of Commerce, Economic Development Administration). Glaeser, E.L. and Ponzetto, G.A.M. (2007). ‘Did the death of distance hurt Detroit and help New York?’ NBER Working Paper 13710. Goldfarb, A. (2006). ‘The (teaching) role of universities in the diffusion of the Internet’. International Journal of Industrial Organization 24: 203–​225. Goldfarb, A. (2012). ‘Internet and the Offline World’ in S.N. Durlauf and L.E. Blume (eds) The New Palgrave Dictionary of Economics, online edition http://​www.dictionaryofeconomics. com/​article?id=pde2012_​I000313 (last accessed 7April 2017). Goldfarb, A. and Tucker, C. (2011a). ‘Search engine advertising: channel substitution when pricing ads to context’. Management Science 57: 458–​470. Goldfarb, A. and Tucker, C. (2011b). ‘Privacy regulation and online advertising’. Management Science 57: 57–​7 1.

How Geography Shapes—and is Shaped by—the Internet    283 Goldmanis, M., Hortacsu, A., Syverson, C., and Emre, O. (2009). ‘E-​commerce and the market structure of retail industries’. The Economic Journal 119: 1–​32. Goolsbee, A. (2000). ‘In a world without borders: the impact of taxes on Internet commerce.’ Quarterly Journal of Economics 115: 561–​576. Goolsbee, A. and Klenow, P.J. (2002). ‘Evidence on learning and network externalities in the diffusion of home computers’. Journal of Law and Economics 45: 317–​343. Greenstein, S. (2000). ‘Building and delivering the virtual world: commercializing services for Internet access’. Journal of Industrial Economics 48: 391–​411. Greenstein, S. (2015). How the Internet Became Commercial: Innovation, Privatization, and the Birth of a New Network (Princeton, NJ: Princeton University Press). Greenwood, B.N. and Wattal, S. (forthcoming). ‘Show me the way to go home: an empirical investigation of ride sharing and alcohol related motor vehicle homicide’. MIS Quarterly. Grubesic, T.H. (2012). ‘The national broadband map:  data limitations and implications for public policy evaluation’. Telecommunications Policy 36: 113–​126. Hammermesh, D.S. and Oster, S.M. (2002). ‘Tools or toys? The impact of high technology on scholarly productivity’. Economic Inquiry 40: 539–​555. Henderson, J.V. (2003). ‘Marshall’s scale economies’. Journal of Urban Economics 53: 1–​28. Hindman, D.B. (2000). ‘The urban–​rural digital divide’. Journalism and Mass Communication Quarterly 77: 549–​560. Hortacsu, A., Asis Martinez-​Jerez, F., and Douglas, J. (2009). ‘The geography of trade in online transactions:  evidence from eBay and MercadoLibre’. American Economic Journal: Microeconomics 1: 53–​74. Jaffe, A., Trajtenberg, M., and R. Henderson (1993). ‘Geographic localization of knowledge spillovers as evidenced by patent citations.’ The Quarterly Journal of Economics 108: 577–​598. Jank, W. and Kannan, P.K. (2005). ‘Understanding geographic markets of online firms using spatial models of customer choice’. Marketing Science 24: 623–​634. Jensen, J.B. and Kletzer, L.G. (2005). ‘Tradable services: understanding the scope and impact of services offshoring’. Brookings Trade Forum 2005: 75–​133. Jones, B.F., Wuchty, S., and Uzzi, B. (2008). ‘Multi-​university research teams: shifting impact, geography, and stratification of science’. Science 322: 1259–​1262. Jorgenson, D., Ho, M.S., and Stiroh, K. (2005). Productivity Volume 3: Information Technology and the American Growth Resurgence (Cambridge, MA: MIT Press). Katz, L.F. and Autor, D.H. (1999). ‘Changes in the Wage Structure and Earnings Inequality’ in O. Ashenfelter and D. Card (eds) Handbook of Labor Economics, Volume 3A, pp. 1463–​1558 (Amsterdam: Elsevier). Kolko, J. (2000). ‘The Death of Cities? The Death of Distance? Evidence from the Geography of Commercial Internet Usage’ in Vogelsang, I. and B.M. Compaine (eds) The Internet Upheaval, pp. 73–​98 (Cambridge, MA: MIT Press). Kolko, J. (2007). ‘Agglomeration and co-​ agglomeration of services industries.’ MPRA Paper 3362. Kolko, J. (2012). ‘Broadband and local growth’. Journal of Urban Economics 71: 100–​113. Lal, R. and Sarvary, M. (1999). ‘When and how is the Internet likely to decrease price competition?’ Marketing Science 18: 485–​503. Leamer, E.E. and Storper, M. (2001). ‘The economic geography of the Internet age’. Journal of International Business Studies 32: 641–​665. Lee, R. and Wu, T. (2009). ‘Subsidizing creativity through network design: zero pricing and network neutrality’. Journal of Economic Perspectives 23: 61–​76.

284   Forman et al. Lieber, E. and Syversson, C. (2012). ‘Online v. Offline Competition’ in M. Peitz and J. Waldfogel (eds) Oxford Handbook of the Digital Economy, pp. 189–​223 (Oxford:  Oxford University Press). Luo, X., Andrews, M., Fang, Z., and Wei Phang, C. (2014). ‘Mobile Targeting’. Management Science 60: 1738–​1756. McCabe, M. and Snyder, C.M. (2015). ‘Does online availability increase citations? Theory and evidence from a panel of economics and business journals’. The Review of Economics and Statistics 97: 144–​165. Macher, J. and Mowery, D. (2008). Innovation in Global Industries: U.S. Firms Competing in a New World (Washington, DC: National Academies Press). Mills, B.F. and Whitacre, B.E. (2003). ‘Understanding the non-​metropolitan-​metropolitan digital divide’. Growth and Change 34: 219–​243. Mithas, S. and Whitaker, J. (2007). ‘Is the world flat or spiky? Information intensity, skills, and global service disaggregation’. Information Systems Research 18: 237–​259. Moss, M.L. (1998). ‘Technology and cities’. Cityscape:  A  Journal of Policy Development and Research 3: 107–​127. Moss, M.L. and Townsend, A.M. (1997). ‘Tracking the net: using domain names to measure the growth of the Internet in U.S. cities’. Journal of Urban Technology 4: 47–​60. Ono, Y. (2007). ‘Market thickness and outsourcing services’. Regional Science and Urban Economics 37: 220–​238. Overby, E. and Forman, C. (2014). ‘The effect of electronic commerce on geographic purchasing patterns and price dispersion’ Management Science 61: 431–​453. Polanyi, M. (1966). ‘The logic of tacit inference’. Philosophy 41: 1–​18. Prieger, J.E. (2003). ‘The supply side of the digital divide: is there equal availability in the broadband Internet access market?’ Economic Inquiry 41: 346–​363. Rosenblatt, T. and Mobius, M. (2004). ‘Getting closer or drifting apart’. Quarterly Journal of Economics 119: 971–​1009. Seamans, R. and Zhu, F. (2014). ‘Responses to entry in multisided markets. The impact of Craigslist on newspapers’. Management Science 60: 476–​493. Sinai, T. and Waldfogel, J. (2004). ‘Geography and the Internet: is the Internet a substitute or complement for cities?’ Journal of Urban Economics 56: 1–​24. Stiroh, K.J. (2002). ‘Information technology and the U.S.  productivity revival:  what do the industry data say?’ American Economic Review 92: 1559–​1576. Storper, M. and Venables, A.J. (2004). ‘Buzz:  face-​to-​face contact and the urban economy’. Journal of Economic Geography 4: 351–​370. Van Alstyne, M. and Brynjolfsson, E. (2005). ‘Global village or cyber Balkans? Modeling and measuring the integration of electronic communities’. Management Science 51: 851–​868. Van Gaasbeck, K., Perez, S., Sharp, R., Schaubmeyer, H., Owens, A., and Cox, L. (2007). ‘Economic effects of increased broadband usage in California. summary report’. Sacramento Regional Research Institute. Vilhuber, L. and McKinney, K. (2014). ‘LEHD infrastructure files in the census RDC—​overview’. Center for Economic Studies Paper 14–26. von Hippel, E. (1998). ‘Economics of product development by users: the impact of “sticky” local information’. Management Science 44: 629–​644. Waldfogel, J. (2003). ‘Preference externalities: an empirical study of who benefits whom in differentiated-​product markets’. The RAND Journal of Economics 34: 557–​568.

How Geography Shapes—and is Shaped by—the Internet    285 Wellman, B., Haase, A.Q., Witte, J., and Hampton, K. (2001). ‘Does the Internet increase, decrease, or supplement social capital? Social networks, participation, and community commitment’. American Behavioural Scientist 45: 437–​456. Winkler, A.E., Levin, S.G., and Stephan, P.E. (2009). ‘The diffusion of IT in higher education:  publishing productivity of academic life scientists’. Working Paper, University of Missouri–​St. Louis. Zentner, A., Smith, M., and Kaya, C. (2013). ‘How video rental patterns change as consumers move online’. Management Science 59: 2622–​2634. Zook, M.A. (2005). The Geography of the Internet Industry: Venture Capital, Dot-​coms, and Local Knowledge (Malden, MA: Blackwell Publishing).

Chapter 15

Schumpet e ria n Custom er s ? H ow Active U sers C o - ​c re at e Innovat i ons Gernot Grabher and Oliver Ibert Co-​creation of Value: Open Firms and Active Customers ‘The customer is king’, of course. Despite this universal mantra, the role of the customer in economic geography up until recently seems to have been confined to a traditional role of a passive recipient of products at the end of the value chain. Innovation, in particular, has been conceived as an affair within and between firms. The customer, the alleged king, has largely been absent from the portrayals of the most influential economic geographical innovation models (for a detailed review see Grabher et al., 2008, pp. 258–​260). However, during the past two decades, in adjacent fields such as management studies and economic sociology a novel paradigm to conceptualize the role of consumers in innovation processes has been advanced:  value co-​creation (Prahalad and Ramaswamy, 2004; O’Hern and Rindfleisch, 2009; Banks and Potts, 2010; Bogers at al., 2010; Cova et al., 2011; Ramaswamy and Ozcan, 2014; Garcia Haro et al., 2014). In contrast to the traditional approach, the term ‘co-​creation refers to any activity in which the consumer participates in an active and direct way with the company to design and develop new products, services, or processes’ (Garcia Haro et al., 2014, p. 70). Communication between firms and customers is no longer understood as being unidirectional and short-​termed. Rather, co-​creation essentially implies a continuous collaboration between producers and customers to create joint value (Grönroos, 2008). For instance, customers and producers jointly engage in the development of connoisseurship that augments knowledgeability and prompts increasingly sophisticated demands (Jeannerat, 2013). Consumers, in fact, are no longer seen as mere buyers of commodities

Schumpeterian Customers?   287 but are more and more perceived (and perceive themselves) as competent users (von Hippel, 2005) who contribute valuable knowledge to innovation processes and who have the power and capacity to intervene at all stages in the value creation process (Garcia Haro et al., 2014; Ramaswamy and Ozcan, 2014). This empowerment transforms users into ever more entrepreneurial agents who, ultimately, might even switch sides and turn into producers themselves (Shah and Tripsas, 2007; Haeflinger et  al., 2010; Agarwal and Shah, 2014; Brinks and Ibert, 2015). The Schumpeterian momentum, then, shifts from the producer to the customer. The re-​apprehension of consumption, however, is not only confined to a change from passive customers into active users, but, in fact, is also shifting along a second dimension, from individual to collective action. The formation of tastes and preferences, the patterns of adoption, and the domestication of, as well as the resistance to, products, are, indeed, genuine social processes that are deeply enmeshed in a variety of networks (see e.g. Howells, 2004). The locus of these social processes, and therein lays the new quality, increasingly shifts to open arenas and public domains (Callon et al., 2002, p. 195). Moreover, beyond ephemeral forms of association, more enduring communities are mobilized for collective knowledge production (Mahr and Lievens, 2012). The emblematic case in point, of course, is open-​source communities like Linux (Weber, 2004). Community self-​organization and value co-​creation also pioneers novel modes of virtual co-​presence, both synchronous and asynchronous (Ritzer and Jurgenson, 2010; Grabher and Ibert, 2014). The Internet, however, does not induce co-​creation practices in a deterministic fashion; instead, the social practices of co-​creation and the technical affordances of Internet tools co-​evolve (Haythornthwaite, 2005). Even though the Internet has enabled co-​creation practices, and open-​source communities have contrived techniques and practices of virtual knowledge sharing, co-​creation is not at all confined to intangible content or software code. As we will demonstrate, the respective practices proliferate in a broad range of domains that heavily rely on embodied skills and close interaction between humans and materials, like, for instance, furniture making, medical practices, trend sports, or cooking. In each of these cases, virtual knowledge sharing and locally situated interaction between practitioners and their material workarounds interfuse and enact spatially distributed but locally anchored networks of more or less similar situated practices (Faulconbridge, 2010; Brinks and Ibert, 2015). The geographical engagement with innovation1 processes more recently has extended to the wide spectrum of spatial arrangements in innovation processes that are created and enacted by innovative users or that sustain intensified user–​producer interaction. Spatial innovation theories can no longer neglect the customer as an increasingly significant source of entrepreneurial dynamics. From the perspective of the geography of consumption the involvement of users into innovation processes has to be taken into account as these practices fundamentally shift the ways consumers use and acquire products and the ways consumers interact with other consumers and producers. In this chapter, we advance a conceptual framework for a more systematic economic geographical exploration of the contested terrain of co-​creation.2 More specifically, we propose a typology of different modes of co-​creation that seeks to capture the heterogeneity of approaches and to provide a conceptual template for further empirical research on user involvement in innovation.

288   Grabher AND IBERT

Situating Co-​creation What, then, is so different and challenging about co-​creation that it deserves the attention of economic geographers? From a historical perspective, learning from and with the customer is anything but a new practice for corporations. The more traditional repertoire for involving the knowledge of customers is composed of deductive, as well as inductive, approaches.

The deductive approach Market research epitomizes the chief logic of the deductive approach that primarily targets general needs and knowledge about the customer. It has developed an extensive register of tools, ranging from large-​scale quantitative surveys to more qualitative instruments, such as focus groups, usability tests, and field ethnographies. Although market research continues to be perceived as an indispensable input into innovation processes, the increasing volatility of preferences and the segmentation of demand into differentiated niches drastically reveal the limitations of this approach, however sophisticated the tools. The concept of ‘mass customization’ (Tseng and Piller, 2011) is an attempt to compensate for the limitations of the deductive approach by offering the modular flexibility to specify products according to individual customers’ needs or tastes: the Dell strategy. However, the customized product is not a result of the user’s direct involvement in the development process; rather, it is assembled from a given variety of modules after development is complete. The customer, thus, cannot exert a ‘voice’ to influence the innovation process directly, but is left with the ‘exit’ option—​simply to decline the offer—​which provides, at best, indirect feedback to the producer.

The inductive approach Whereas deductive strategies move top-​down from the aggregate market to the representative customer, the inductive approach works bottom-​up from the individual customers to market segments. The quintessential organizational form of inductive user–​producer interaction is the project that is built around the distinctive needs of individual customers. Projects, in fact, rather than for customers are performed with customers (Girard and Stark, 2002). As the project sponsor, the customer specifies the task to be achieved and contributes critical knowledge throughout the entire development process. The locus of control, in other words, rests primarily with the customer; his or her ‘voice’ shapes the course of the project. Although projects are built around the specific needs of an individual customer, the knowledge that is generated in the course of the project may be useful beyond its clearly specified temporal and technical boundaries. Projects are not isolated learning episodes; rather, the one-​off mission can feed into a cumulative process of learning from a series of related ventures (Brady and Davies, 2004).

Schumpeterian Customers?   289 Table 15.1 Situating Co-​creation Deductive approach

Co-​creation

Inductive approach

Target group

Representative customer (‘market research’)

Hybrid community

Individual customer (‘project’)

Type of knowledge

Knowledge about the customer

Knowledge of the community

Knowledge of the customer

Locus of control

Producer

Producer and customer

Customer

Mode of governance

Exit

Exit, voice, and loyalty

Voice

Mode of interaction

Absence

Virtual co-​presence

Physical co-​presence

Learning trajectory

Customization

Iteration

Modularization

Source: authors’ own design.

Co-​creation Deductive and inductive strategies signify the poles of a continuum of customer involvement in innovation. More recently, this continuum has been increasingly extended by a class of new approaches in between these two poles that pursue the logic of co-​creation (Table 15.1). While producer–​customer interaction is set up as a 1:n–​environment in deductive approaches and as a 1:1 relation in inductive approaches, co-​creation is built around ‘hybrid communities’ (Grabher and Ibert, 2014). These communities encompass not only experts and professionals, but are also composed of sophisticated customers with intimate knowledge about product architecture and heavy users who are familiar with products from everyday use (Ramaswamy and Ozcan, 2014). Owing to this heterogeneous composition hybrid communities unfold horizontal (expert-​to-​expert; user-​to-​user) and vertical (expert-​ to-​user, professional-​to-​layperson) dynamics and catalyse the generation of new knowledge in two ways. Firstly, in the vertical relation between users and producers, they combine the iterative and intensive interaction of inductive strategies with the broader representation of the market of deductive strategies; they are deeply involved and widely focused at the same time. Secondly, and more critically, the circulation of knowledge also unfolds laterally between customers. The lateral exchange of individual users’ experience, recommendations, and warnings, the revelation of individual workarounds and product modifications, makes explicit users’ tacit knowledge and it ‘unsticks’ von Hippel’s (1994) ‘sticky information’, at least partially (Grabher and Ibert, 2014). Hybrid communities thus exert a powerful ‘voice’. At the same time, ‘exit’ is a real option, as the termination of the relationship would incur only low sunk costs. Beyond this ‘exit’ or ‘voice’ calculus, however, hybrid communities are essentially governed by ‘loyalty’ (Wiertz and de Ruyter 2007; Brinks and Ibert, 2015). Loyalty pushes the dynamics of interaction

290   Grabher AND IBERT from the singular intervention to ongoing conversation and affords the social context for the circulation of knowledge that ranges from fairly personal experience to elaborate design proposals. Owing to these dynamics the locus of control is not fixed, but seems to shift back and forth between the customer and the producer in the course of development processes. Although control may eventually gravitate more strongly towards the producer or the customer, co-​creation essentially implies a redistribution of power, although an unstable and contested one. Communities may be instrumentalized in a straightforward fashion or may develop ‘a life of their own’ (Wiertz and de Ruyter, 2007, p. 370) that evades the control of the producer or even turns against him or her (Thrift, 2006, p. 290). While direct physical co-​presence is the prime constellation of inductive approaches and deductive strategies are limited to indirect interaction, co-​creation pioneers novel modes of virtual co-​presence, both synchronous and asynchronous. The Internet pushes the development process beyond the familiar organizational domains and transforms innovation into an activity that is spread across multiple locations and that mobilizes ever more heterogeneous sources of knowledge in real time. Yet, the Internet does not only afford a passive link between globally dispersed sites. Rather, it is a site of collaborative knowledge production (Grabher and Maintz, 2007; Hwang et al., 2015).

Towards a New Knowledge Ecology of Innovation: Mapping Co-​creation The historically new key features of co-​creation—​hybrid communities, the redistribution of power between the customer and producer, and virtuality—​denote a wide corridor within which producer–​customer interaction is reshaped. To capture the width and breadth of the geography of co-​creation, we propose a typology that unfolds along two dimensions (see Table 15.2). The horizontal dimension denotes the degree of involvement and stretches from consultation over participation to generation and thus elucidates the unequal distribution of power

Table 15.2 A Typology of Co-​creation Formats and Practices Producer-​driven

User-​driven

Consultation

Participation

Generation

Epistemic community

Expressive user E.g. Bicycle design Usage knowledge Physical co-​presence

Lead user E.g. Medical equipment Design knowledge Physical co-​presence

Professional user E.g. Open source Procedural knowledge Virtual co-​presence

Practising community

Consumer community E.g. Instant food Usage knowledge Virtual co-​presence

User community E.g. Computer games Design knowledge Blended co-​presence

Interest community E.g. Geocaching Procedural knowledge Blended co-​presence

Schumpeterian Customers?   291 among producers and users involved in co-​creation. Consultation designates a type of limited and producer-​driven interaction in which the customer primarily collaborates in the role of a layperson and control remains largely on the side of the producer. Participation entails a deeper, but still primarily producer-​driven, form of involvement in which the customer holds the status of a quasi-​expert. Although users and user communities often create their own forums for interaction and hence are more independent from producers than in consultation, the focal products and brands remain under the control of producers. Similarly, generation involves expert knowledge that is mainly accumulated by using and modifying a product. Generation, however, denotes a shift from producer-​to user-​driven development as it often occurs in situations in which communities evolve around novel practices or genres unknown to established producers. In the vertical dimension, the proposed typology draws a distinction according to the prevailing locus of knowledge production. The first type corresponds with a process of deliberate and goal-​oriented knowledge production. Knowledge is produced in the vertical producer–​customer relation that is strictly focused on the specific ‘epistemic object’ (Knorr Cetina, 2001, pp. 181–​184) and is governed by an accepted procedural authority (Amin and Roberts, 2008). The second type refers to a practice in which knowledge is produced as a by-​ product of socializing and situated learning. In this dimension, knowledge is not only produced in the orchestrated vertical producer–​customer relation but crucially unfolds in the horizontal exchange and evolving socializing among customers. The vertical dimension, put briefly, distinguishes between the epistemic community (Knorr Cetina, 1999) and the practising community (Neff and Stark, 2003; Müller and Ibert, 2015). By moving along these two axes, we differentiate six types of co-​creation that embody unique constellations of types of knowledge and spatial arrangements of interaction. Despite the essential differences among these six modes of co-​creation, they do not represent static categories demarcated by strict boundaries. Communities are prone to social dynamics that may eventually transform one type of co-​creation into another type (see e.g. Brinks and Ibert, 2015). In this sense, the typology provides a continuum along which we identify ideal–​ typical constellations that may move in the course of their life in one direction or the other.

The Expressive User An example of this type of producer–​user integration is the consultation of bike messengers and devoted car users in a project on the design of bicycles (Grabher et al., 2008). The enrolment of expressive users from the edges of the market, where one can ‘find “extreme” users who live differently, think differently, and consume differently’ (Brown and Katz, 2011) might be instrumental to approach challenges from a novel angle and to foster the proverbial thinking ‘out of the box’. The rich experience of expressive users cannot be seized through the traditional repertoire of market research, such as questionnaires. Rather, the critical information is often part of the ‘embodied’ (Blackler, 1995, p. 1024) knowledge that eludes the user’s full awareness. Following the paradigm of ‘design thinking’ (Brown, 2008), the collaborative production of knowledge unfolds largely through selecting and testing prototypes so the user can be observed instantaneously while interacting with the product. ‘Rarely will the everyday people […] be able to tell us what to do. The only way we can get to know them is to seek them

292   Grabher AND IBERT out where they live, work and play’ (Brown and Katz, 2011, p. 382). Critical information for product development can thus only be harnessed outside the R & D laboratory in the real-​life contexts of users and consumers. For a meaningful interpretation of the observed behaviour, however, additional interview data may provide helpful background information on the user’s irritations or emotions during the test situation (Grabher et al., 2008). Thus, expressive users are also partly valued for their capacity to explicate the experiences they have gained in everyday practice with a commodity and to convey motivations of adoption or consistent non-​adoption. The interaction between the producer and expressive user is orchestrated by the producer and focuses strictly on an object (e.g. a new bicycle); social dynamics among users are inconsequential for the knowledge dynamics. The involvement of expressive users may thus push stages of the development process outside the corporate boundaries, but in terms of geographical distance not too far from the producer’s site for pragmatic reasons. As the knowledge of the expressive user is rich but not necessarily specialized, expressive users typically can be identified in close vicinity of the producer. However, if the involvement of extreme users (like the ‘collector who owns 1400 Barbie dolls’ or the ‘professional car thief ’; Brown and Katz, 2011, p. 382) is imperative, producers are ready to cross larger distances.

The Consumer Community Consumer communities are initiated and maintained by professional and commercial producers and thus are also categorized as ‘firm-​hosted communities’ (Grabher and Ibert, 2014). Manufacturing firms set up the online forum of exchange, employ the community’s moderators, define and police the norms of interaction, and can, if deemed necessary, set the agenda by explicitly soliciting feedback on specific topics. Illustrative cases in point are the Consumer Channel of Kraft Foods, the Huggies Baby Network of Kimberly-​Clark (Grabher et al., 2008), or Dell’s Idea Storm Forum (Grabher and Ibert, 2014). Like expressive users, members of such communities are assumed to contribute mainly non-​expert knowledge. For instance, consumers of Kraft packet soups know little about the recipe or the natural and chemical ingredients. Yet, their vivid interaction offers several clues on the everyday situations in which such products are prepared, the social occasions on which they are served (or cannot be served), and how these products are appreciated (or not) by followers of different kinds of diet (representing target groups of the producer). The categorical difference between expressive user and consumer community is denoted by the locus of knowledge circulation and production. In contrast to the ‘intimacy averse’ (Mateos-​ Garcia and Steinmueller, 2008) product-​focused dialogue between the producer and the expressive user, the exchange of knowledge in consumer communities is, to a significant extent, a by-​product of ongoing conversation within a community that provides sociability, information, support, and a sense of belonging, however ephemeral (Wiertz and de Ruyter, 2007; Langner and Seidel, 2015). Consumer communities are particularly appreciated for the conversations that unfold not only around mutual advice in solving everyday problems, but may also crystallize around shared hobbies or current events that are not or only loosely related to the focal product or brand. However, these conversations are not derided as detracting noise by producers, but are valued as catalysts that foreground coping strategies in everyday life and as filters

Schumpeterian Customers?   293 and gauges of unmet needs and wider concerns. Knowledge revealed in lateral exchange is regarded as more authentic, reliable, and richer than information solicited through the traditional repertoire of market research, such as the standard questionnaire. Consumer communities are dispersed online communities, and interaction occurs almost exclusively in virtual co-​presence. If face-​to-​face events are arranged at all, they are staged as a forum for public gratification. They are regarded as instrumental neither for creating the social dynamics of community formation nor for accessing concealed layers of knowledge. On the contrary, more surprising insights may surface because the community members remain within their diverse local contexts while interacting. In cases of dissent and misunderstanding, for instance, the members have to enrich their contributions with additional contextual information to clarify their statements for the physically absent interaction partners (Grabher and Ibert, 2014). These parts of lateral conversation may be useful for producers because they unveil the multitude of locally situated strategies that users apply to integrate a commodity into their daily lives.

The Lead User Pioneering physicians who contribute to the specification of functions and the interface design of tomography scanners, for example, epitomize the principal features of lead users (Grabher et al., 2008). Producers seek to integrate lead users into development projects because they are at the frontend of the adoption curve and ahead of the market (von Hippel, 1986, p. 795; Jeppesen and Laursen, 2009). The lead user embodies close-​to-​expert knowledge about the architecture and modus operandi of the product. For instance, experienced medical surgeons do not only know the technical sequences of diagnostics performed by the scanner, but also have a vivid idea about the assumptions underlying the programming of the sequencing software by medical equipment engineers. This design knowledge defies any straightforward transfer from user to producer. Rather, it is co-​created through joint reflection in the context of application, such as the diagnostic department of a hospital, for example. Relevant knowledge is ‘sticky’ and place bound, as it is to an extent inscribed in the physical layout of workflows. Observation and socializing in practice, however short-​lived, is imperative because users are generally no longer aware of the problems they have already ‘solved’ through their own idiosyncratic workarounds. The circulation and generation of knowledge are focused on the product and do not evolve in lateral conversations in communities. Instead, the particular setting of producer–​user interaction may even preclude community building as peer dynamics (like status competition among physicians) may distract the development process from more general market needs (Grabher et al., 2008). Although frequent face-​to-​face communication is one of the emblematic characteristics of lead user–​producer interaction, the prerequisites to initiate physical co-​presence without any great effort are not the main drivers of the respective geography of knowledge creation. Rather, the situated nature of the lead user’s knowledge and the information that virtually ‘sticks’ (von Hippel, 1994) at the place of application induce time–​spatial practices that alternate between basic research in the producer’s laboratory and phases of collaborative knowledge production at the customer’s site (Ibert, 2010). Hence, in spatial terms, the integration of lead users unfolds a dispersed and temporary geography that follows the pattern of user

294   Grabher AND IBERT locations right to the diagnostic centres and hospitals in the case of medical equipment. As physical co-​presence at the site of usage is indispensable, lead users are most likely not located in close proximity of the producer.

The User Community Typically, user communities are launched by community members who also create and enforce the rules of interaction in a self-​organized process. However, these communities are not independent from professional producers. Rather, the object of common interest is associated with a distinct brand or even a specific product. Hence, they might be categorized as ‘firm-​related communities’ (Grabher and Ibert, 2014). The Command & Conquer 3 computer-​game community, (Grabher et al., 2008), IKEA fans, photographers using Nikon cameras, BMW motorcyclists (Grabher and Ibert, 2014), or the Adult Fans of Lego (Tapscott and Williams, 2006, pp. 130–​131) embody the principal features of a user community. Similar to lead users, members of a user community contribute intricate design knowledge to producer-​driven innovation that they have acquired through intense use and ongoing tinkering. The accumulation of this close-​to-​expert knowledge, however, does not resemble the goal-​oriented and systematic co-​creation with the lead user; rather, it unfolds in a more unfocused socialized process that evolves within the community. Nevertheless, user communities seize the opportunities of online interaction (see also Mahr and Lievens, 2012, p. 174). The asynchronicity of interaction allows for heedful interrelating and reflective reframing of problems and iterative prototyping (Hargadon and Bechky, 2006). Moreover, the ‘hypertextual’ nature of online conversations ‘encourages writerly, active reading rather than passive consumption of what has been produced by a conventional authorial author’ (Gulbrandson and Just, 2011, p. 1099). Despite playfulness, the practice of explicit cross-​ referencing in online interaction generates a collective memory (codified in databases with advanced navigation functions) and a certain sense of focus within user-​communities. Like the consumer community, user communities combine the vertical dimension of dialogue with the producer and the lateral dynamics of conversation among users. User communities regularly evolve into enduring and socially differentiated formations with status hierarchies, community norms, and conventions. At the periphery of the community, the vast majority of members lurk in the background and observe silently. Closer to the core of the community, a smaller group of temporary active members provide ‘focal feedback’ (Jeppesen, 2001, p. 17) by testing variants, revealing ‘bugs’, mutually solving problems, or simply spreading more or less relevant information to a responsive audience. Producers like Lego explicitly encourage inventive tinkering in their copyright management by publicly granting a ‘right to hack’ (Tapscott and Williams, 2006, p. 130; Ramaswamy and Ozcan, 2014). As they reach the top of the status hierarchy, members of the core group not only enjoy the highest esteem and professional authority of the community members, but also become formally acknowledged as community leaders or ‘helpful authorities’ (Jeppesen, 2001, p. 22) by the producer. Crucially, members of the core group provide also novel inputs into wider community and, thereby, propel processes of collective tinkering and knowledge accumulation. By absorbing knowledge from the outside and sharing it within the community, core members, phrased differently, take on the role of both lead adopters and gatekeepers (Jeppesen and Laursen, 2009, p. 1588).

Schumpeterian Customers?   295 The geography of collaborative knowledge production in user communities blends virtual exchange with transient physical encounters. Community members can share knowledge without physical co-​presence by purposefully creating similar material conditions at different sites. For instance, in the case of IKEA fans or Nikonians, community members frequently refer to product IDs and order codes to assemble almost identical material ‘constellations of practice’ (Faulconbridge, 2010) at geographically dispersed locations. Photographers who share knowledge in the Nikonian discussion forum seek to enrich the shared knowledge base by discussing to what extent their insights apply only for specific places and particular circumstances, or are also valid for more general challenges. In such cases the similarity and dissimilarity across the sites respectively creates a multi-​locally shared material–​virtual context that facilitates the exploration, variegation, and validation of the shared knowledge (Grabher and Ibert, 2014). Even in the case of computer gamers, who rely excessively on online interaction, interaction in user communities is not restricted to the virtual realm. Rather, the community occasionally gathers at particular events, like game conventions or ‘LAN’- parties, in the case of computer gamers, or the Lego World exhibits in the case of the Adult Fans of Lego. These events constitute temporary sites of knowledge performance (Thrift, 2000). ‘Being there’ offers opportunities to experience corporeally not only the ‘look’n feel’ of different products, but also to engage physically with their actual performance and to compare functionalities and features directly during collective and challenging application. During an event, participants experience the resonance of the community to modified features and new versions of products, and vaguely perceived user preferences become manifest through critique or encouragement. Furthermore, as a showcase, the event represents the achievements of a community. The choice of items that are deemed worthy of display and the proper ‘positioning’ (Power and Jansson, 2008) of brands or products at the event, in a sense, project the involved knowledge domains into the physical space and hint at relational structures within the field. Community gatherings, in fact, are also conducive for strengthening the social cohesion of the spatially dispersed community (Franke and Shah, 2003, pp. 160–​161); conventions are not only about shoptalk, but also partying.

The Professional User The emblematic cases of generation by professional users are, of course, open-​source projects. Open source, pioneered in the software field with Mozilla (a browser), with Apache (server software), and, most prominently, with Linux (an operating system) are founded on two chief principles: public ownership of the intellectual property and a production model of generating knowledge in a globally dispersed context (see e.g. Lakhani and von Hippel, 2003). More recently, initiatives that seek to exploit the principles of open source have emerged beyond software, in the development of drugs, for example. Open-​source projects have been launched particularly in the development of treatments for diseases that affect a comparatively small number of people (such as those with Parkinson’s disease) or that mainly affect poor countries (with diseases such as malaria or typhoid) (e.g. The Tropical Disease Initiative; The Synaptic Leap; see also Tapscott and Williams, 2006, pp. 169–​172; Yaqub and Nightingale, 2012).

296   Grabher AND IBERT These diverse projects share a strict focus on the ‘epistemic object’ (Knorr Cetina, 2001, pp. 181–​184) and a deliberate organization towards generating knowledge about the joint project, whether it is an operating system or a vaccine against typhoid. In the case of Linux, a committee, in charge of evaluating inputs from dispersed users, represents a procedural authority to guide the process of knowledge generation; lateral conversation and sociability are not typically part of the repertoire of these communities. The code of conduct emphasizes strict ‘on-​topic’ professionalism (Ren et al., 2007), distracting statements are explicitly precluded from the circulation of knowledge. Users not only contribute design knowledge, but, as this case amply illustrates, also generate knowledge of how to produce the solution collectively. As exemplified by open-​source projects, professional communities typically, but not exclusively, are online-only communities (Amin and Roberts, 2008). Software projects like Linux indicate that even complex tasks can be achieved over longer periods in Internet-​only global communities. Physical interaction may occur, but it does not seem imperative. On the one hand, face-​to-​face interaction hardly appears necessary for reasons related to knowledge exchange, at least for projects in which tasks can be parsed and modularized and the product is digital or can be transformed into a digital form (Mateos-​Garcia and Steinmueller, 2008). On the other hand, face-​to-​face encounters do not seem indispensable for strengthening the social coherence or reinforcing the relational ties within the community (Amin and Roberts, 2008). Rather, the identification with the goal of the project and a procedural authority that guides the production of knowledge afford the resilience of these communities and their tolerance of a high turnover of membership (Ren et al., 2007, p. 400). However, virtual interaction again does not necessarily lead to arbitrary geographies. Rather, analogous to user communities, the spatial patterns of the ongoing interaction among professional users are shaped by the unevenly distributed material and locally situated preconditions for knowledge production. Virtually transmitted data can be turned into productive use only if they will be put into practice in an adequate work setting. Developing a sequence of open-​source code requires little more than an up-​to-​date computer system with a fitting development environment; testing a modified formula of a vaccine, in contrast, can be achieved only at highly specific sites that provide, for instance, a well-​equipped laboratory, experimental animals, and technical staff. Shared practice across distance, then, is not a result of socialization in a context of co-​location, but rather of collective heedful engagement (Weick and Roberts, 1993) across similar but physically distanciated material contexts. Crucially, shared practices afford the preconditions for those interrelating activities that are critical for triggering moments of collective creativity (Hargadon and Betchky, 2006): help seeking, help giving, reflective reframing (in which each actor in turn attends to and builds upon the comments and actions of others), and affirmation (e.g. through organizational values that support individuals seeking and providing help and reflective reframing).

The Interest Community Interest communities had already evolved in the pre-​Internet area around sports like mountain biking, kayaking, and windsurfing (for an overview, see Grabher et al., 2008). However, they have spectacularly taken off with the Internet in the vast and rapidly expanding universe

Schumpeterian Customers?   297 dubbed Web 2.0. that boosted the generation of contents in ‘remix genres’ (Tapscott and Williams, 2006, p. 137) like music and video production or in pioneering new forms of amateur journalism or citizen science. Empowered by a new generation of digital tools, users in this expanding universe increasingly differentiate into niche communities that are organized around shared interests. The online feature-​film project The Swarm of Angels, sandboarding (Grabher and Ibert, 2014), geocaching, or fingerboarding (Brinks and Ibert, 2015) are examples of interest communities that are organized around a shared passion. An example of an interest community that has emerged around a shared (existential) concern is the dichloroacetic acid (DCA) forum, an online forum on which an informal global research network that seeks to advance the development of a new drug against cancer interacts with a community of committed members, mainly patients with cancer and their relatives (Grabher and Ibert, 2014). The evolution of interest communities is not the result of a goal-​oriented and systematic endeavour of a collective project to invent. Typically, they metamorphosed out of combinations of everyday problem solving, competitive performance, piecemeal improvements, and serendipitous encounters. Accidental discoveries resulted from using products and technologies in ways in which they were originally not conceived. For instance, geocaching emerged from a small group of technophiles who explored the possibilities of using Global Positioning System (GPS) devices for non-​military purposes after the deactivation of the distortion of the GPS signal in 2000. To test the accuracy of the undistorted GPS signal, one member had the idea of hiding a container in the landscape and posting the GPS coordinates on the Internet. What started as a technical test soon transformed into hybrid online/​offline hide-​and-​seek game (Brinks and Ibert, 2015). There is nothing particularly heroic or technologically spectacular about these reconfigurations. However, they elucidate how lifting products and technologies out of their prescribed context, reconfiguring them, and placing them in a new context opens up a development trajectory that might even engender a new genre that subsequently continues to be driven by users (see also Pinch, 2003) who eventually might turn their passion into profit by establishing new firms (Shah and Tripsas, 2007; Agarwal and Shah, 2014). Knowledge created here, as in the case of the professional user, extends beyond design knowledge and encompasses knowledge about how to produce it. In the infant stages of mountain biking, production was also performed by the bikers, who began to build bikes commercially for others and thus laid the foundation of a craft business that later turned into a mainstream industry (Lüthje et al., 2005, p. 954). Interest communities blend physical encounter and virtual co-​presence in various combinations. In the case of sports, in which competition and performance are at the essence of the activity, temporary physical co-​presence at events is indispensable (Brinks and Ibert, 2015). Particularly, competitions afford the key sites for display, comparison, and collective tinkering. In the case of fashion, innovations by interest communities are urban phenomena (Kawamura, 2006) as the dense co-​location of various overlapping subcultures is conducive to cross-​cultural re-​contextualization (Jacobs, 1969). The vast domain of the ‘remix cultures’ (Tapscott and Williams, 2006, p. 137), however, typically evolves online and, like user communities and interaction among professional users, unfolds a dispersed geography around materially similar ‘constellations of practice’ (Faulconbridge, 2010).

298   Grabher AND IBERT

Discussion The various practices and formats of co-​creation problematize some of the key notions of the economic that theorists (for very good reasons, of course) implicitly assume as given. For instance, the traditional clear-​cut separation between producers and consumers becomes more porous and sometimes even contested. Products, for instance, transform from a fixed and frozen thing into a ‘variable’ (Callon et al., 2002, p. 197) and increasingly lose their ‘object-​ivity’ (Knorr Cetina, 2001, p. 528). Through their lack of ‘ontological stability’ (Zwick and Dholakia, 2006, p. 21), products become focal entities of iterative, never-​ending processes of experimentation, negotiation, and modification: they remain in a state of ‘permanently beta’ (Neff and Stark, 2003). As epistemic objects, products are not given facts but reveal their look, content, shape, and ‘story’ in processes of interaction, observation, and examination. ‘Such objects demonstrate a propensity to change their “face in action” vis-​a-​ vis consumers through the continuous addition and subtraction of properties’ (Zwick and Dholakia, 2006, pp. 20–​21). Additionally, co-​creation practices also highlight an affective dimension of human–​object relations. In particular, in community-​based approaches of co-​creation most participants have developed strong emotional bonds to the objects of inquiry. Community members became initiated to co-​creation practices by experiencing moments of affective amazement while using (rather than consuming) a product in a particular practice (Jeannerat, 2013). The subsequent desire to stabilize such positive affective experiences has been interpreted as one of the main driving force for the formation of communities (Brinks, 2016). Co-​creation practices also put into question the notion of markets as simple means of selling products that are composed at the terminus of the value chain. The market becomes a forum for an ongoing dialogue between producers and consumer communities (Thrift, 2006, p. 287), a dialogue that is deeper than in the classical accounts on marketing or market research. The market is no longer outside the value chain, acting as the locus of interchange between the producer and the consumer. Greater interactivity means that the market, in a sense, ‘pervades the entire system’ (Prahalad and Ramaswamy, 2004, p. 125). Not far from the invisible hand that is ritually criticized as an unrealistic assumption of mainstream economics, markets are frequently absent from economic geographical analyses. A stronger appreciation of co-​creation practices in economic geography will most likely shift this absent assumption centre stage of the analysis. The notion of markets is brought down from the totalizing force ‘out there’ to the level of actual practices of negotiation between producers and consumers (Callon and Muniesa, 2005).

Conclusions Co-​creation practices pose new challenges to economic geographical inquiry and offer fascinating new possibilities to theorize the spatiality of learning and innovation. In particular, consumer communities and interest communities are narrow in terms of their focal topic but far-​reaching in terms of spatiality (see also Hwang et al., 2015). Apart from the joint passion, participants share rather little in common, most of them have never met personally,

Schumpeterian Customers?   299 and direct interaction normally has to be initiated and orchestrated across physical distance. The resulting spatiality of innovation is intriguing and at the same time challenging for geographical inquiry. The traditional approaches to study production by exploring either local and regional concentrations or global organizational networks are not sufficient to grasp the unfolding spatial logics of co-​creation. Co-creation practices exploit all available online/​ offline technologies and media to share ideas across physical distance (Bathelt and Turi, 2011; Bathelt and Henn, 2014; Maskell, 2014), enacting a much more diffuse and ephemeral geography. Firstly, collaborative knowledge production through producer–​customer interaction does not fit well into the registers of our more enduring geographies of knowledge creation among producers. Co-​creation practices rather leave ephemeral spatial imprints around temporary physical encounters: at the trade fair, the sports contest, or the hackathon. However, temporary co-​presence is by no means confined to these obvious and exceptional occasions. On the contrary, encounters at rather mundane sites of everyday use such as the home kitchen or the sandboarding track are less spectacular, but by no means less important. Collaborative knowledge production at temporary encounters is sustained through physical mobility, of experts, users, and prototypes. Co-​creation, in other words, is sustained by shifting physical geographies of circulation (Urry, 2003), a theme that has partly been picked up in debates on temporary clusters (Maskell et al., 2006), for example, but opens up rich opportunities for further exploration. Secondly, this ephemeral geography in a sense reiterates the importance of the specific physical site of encounter and interaction. User knowledge is inscribed in the physical layout of the workplace, in the temporal sequencing of everyday routines, in the improvised workarounds. Co-​creation, in other words, is not only about company representatives talking to customers, but also about interaction at these unique constellations of things and objects that make up the site of usage. Co-​creation thus shifts the locus of knowledge-​production from the R & D department right to the site of usage or, more generally, pushes knowledge production from the ‘context of discovery’ to the ‘context of application’ (Gibbons et al., 1994), at least temporarily. Thirdly, co-​creation dramatically re-​values the role of virtual co-​presence. The Internet, however, is not merely about speeding up, spreading out, and lowering costs of communication. Nor is it a simple (and deficient) substitute or artificial extension of face-​to-​face communication. The Internet is charged with social software that tracks, categorizes, and channels information, sediments memory, or automates word-​of-​mouth, aggregates idiosyncratic interests, induces connectivity, and sustains communities (Grabher and König, 2017). In a sense, social software turns networks ‘inside out’ (Riles, 2000): it turns networks from latent social embeddedness into a strategic practice to deliberately furnish knowledge ecologies. The increased importance of virtual platforms for interaction, however, does not result in a-​spatial or immaterial practices of interaction. For instance, when users share knowledge in an online environment, community members establish a common framework within which they negotiate which elements of the present material contexts are crucial to establish a shared practice and which of these elements, respectively, represent mere local idiosyncrasies. In other words, community members actively create identical material ‘constellations of practice’ (Faulconbridge, 2010) at geographically dispersed locations to allow a more meaningful in situ validation, exploration, and variation of the virtually shared knowledge. As

300   Grabher AND IBERT a result, co-​creation practices give rise to complex ‘topologies of learning’ (Faulconbridge, 2014) in which actors utilize similarities and differences between local contexts to create new insights. The highly dynamic geographies of co-​creation pose some exciting puzzles of wider relevance that seem worthwhile to be addressed by future research. On the righthand side of our typology of co-​creation practices (see Table 15.2, p. 290), innovating users who operate by and largely independently from firms or producers are located. We maintain that this particular type of user, in fact, appears of eminent importance for economic development. Interest communities share knowledge in order to refine designs, extend functionalities, or invent new fields of application—​and hence create value long before the first firms enter the field. In other words, the role of interest communities is not confined to the co-​creation of value; rather they are the originators of value and, indeed, create markets and invent genres. Although research on user–​entrepreneurs as company founders is already available (Shah and Tripsas, 2007; Haeflinger et al., 2010; Agrawal and Shah, 2014), knowledge on the spatialities of the respective economic activities remains patchy at best. User–​entrepreneurs, by transforming ideas born out of everyday practice into marketable products, thus combine local opportunities with globally shared knowledge (Brinks and Ibert, 2015). Hence, they truly signify the advent of the Schumpeterian customer.

Notes 1. We understand innovation in a broad Schumpeterian sense of ‘new combinations’ in the realm of products, processes, markets, inputs, or organization (Schumpeter, 1912/​2006). 2. Our geographical conceptualization of user-​driven innovation processes as presented in this chapter has been developed in close dialogue between theory and empirical fieldwork over the past ten years. Two successive research projects contributed substantially to it. The first research project, ‘Mobile Places, Virtual Networks: The Geographies of User-​Induced Innovation Processes’ (2006–​09; funded by the German Research Foundation GR 1913/​7), was an explorative study of user-​driven innovation processes across a broad range of sectors and organizational settings. As a result of this study we developed a typology of different ways of integrating users in innovation processes (Grabher et al., 2008). In this research context we also advanced a conceptualization of collaborative knowledge practices across physical distance, their peculiarities, strengths, and weaknesses (Grabher and Ibert, 2014). This research has been complemented with a second research project on ‘Sources and Paths of Innovation: A Spatial Perspective on the Dynamics of Knowledge Generation and Utilisation in the Economy’ (2012–​14; funded as Lead Project of the Leibniz Institute for Regional Development and Structural Planning Erkner) in which a particular focus was on the time-​spatial dynamics of selected innovation biographies that have strongly been driven by user entrepreneurs. In this project we developed further conceptualizations of different types of communities of practice, their impacts on innovation processes (Müller and Ibert, 2015), and an empirically grounded phase model of the process of firm foundation from within communities of interest (Brinks and Ibert, 2015). Warm thanks are owed to our colleagues Verena Brinks, Saskia Flohr, Felix C. Müller, and David Tamoschus, who contributed to our research projects.

Schumpeterian Customers?   301

References Agarwal, R. and Shah, S.K. (2014). ‘Knowledge sources of entrepreneurship: firm formation by academic, user and employee innovators’. Research Policy 43: 1109–​1133. Amin, A. and Roberts, J. (2008). ‘Knowing in action: beyond communities of practice’. Research Policy 37: 353–​369. Banks, J. and Potts, J. (2010). ‘Co-​creation games: a co-​evolutionary analysis’. New Media & Society 12: 253–​270. Bathelt, H. and Henn, S. (2014). ‘The geographies of knowledge creation over distance: toward a typology’. Environment and Planning A 46: 1403–​1424. Bathelt, H. and Turi, P. (2011). ‘Local, global and virtual buzz: the importance of face-​to-​face contact in economic interaction and possibilities to go beyond’. Geoforum 42: 520–​529. Blackler, F. (1995). ‘Knowledge, knowledge work and organization: an overview and interpretation’. Organization Studies 16: 1021–​1046. Bogers, M., Afuah, A., and Bastian, B. (2010). ‘Users as innovators: a review, critique and future research directions’. Journal of Management 36: 857–​875. Brady, T. and Davies, A. (2004). ‘Building project capabilities: from exploratory to exploitative learning’. Organization Studies 25: 1601–​1621. Brinks, V. (2016). ‘Situated affect and collective meaning: a community perspective on the creation and commercialization of value in enthusiast-​driven fields’. Environment and Planning A 48: 1152–​1169. Brinks, V. and Ibert, O. (2015). ‘Mushrooming entrepreneurship: the dynamic geography of enthusiast-​driven innovation’. Geoforum 65: 363–​373. Brown, T. (2008). ‘Design thinking’. Harvard Business Review 86: 84–​92. Brown, T. and Katz, B. (2011). ‘Change by design’. Journal of Product Innovation Management 28: 381–​383. Callon, M. and Muniesa, F. (2005). ‘Economic markets as collective devices’. Organization Studies 26: 1229–​1250. Callon, M., Meadel, C., and Rabeharisoa, V. (2002). ‘The economy of qualities’. Economy and Society 31: 194–​217. Cova, B., Dalli, D., and Zwick, D. (2011). ‘Critical perspectives on consumers’ role as “Producers”: broadening the debate on value co-​creation in marketing processes’. Marketing Theory 11: 231–​241. Faulconbridge, J. (2010). ‘Global architects: learning and innovation through communities and constellations of practice’. Environment and Planning A 42: 2842–​2858. Faulconbridge, J. (2014). ‘Putting the individual in context: paths, capitals and topologies of learning’. Prometheus 32: 75–​82. Franke, N. and Shah, S. (2003). ‘How communities support innovative activities: an exploration of assistance and sharing among end-​users’. Research Policy 32: 157–​178. Garcia Haro, M.A., Martinez Ruiz, M.P., and Martinez Canas, R. (2014). ‘The effects of the value co-​creation process on the consumer and the company’. Expert Journal of Marketing 2: 68–​81. Gibbons, M., Limoges, C., Nowotny, H., Schwartzman, S., Scott, P., and Trow, M. (1994). The New Production of Knowledge:  The Dynamics of Science and Research in Contemporary Societies (London: Sage). Girard, M. and Stark, D. (2002). ‘Distributing intelligence and organizing creativity in new media projects’. Environment and Planning A 34: 1927–​1949.

302   Grabher AND IBERT Grabher, G. and Maintz, J. (2007). ‘Learning in Personal Networks:  Collaborative Knowledge Production in Virtual Forums’ in H. Hof and U. Wengenroth (eds) Innovation Research: Concepts, Methods, and Perspectives, pp. 187–​202 (Münster: LIT). Grabher, G. and Ibert, O. (2014). ‘Distance as asset? Knowledge collaboration in hybrid virtual communities’. Journal of Economic Geography 14: 97–​123. Grabher, G. and König, J. (2017). ‘Performing Network Theory? Reflexive Relationship Management on Social Network Sites’ in B. Hollstein, W. Matiakse, and K.-​U. Schnapp (eds) Networked Governance. Governing Networks and Governance via Networks, pp. 121–​140 (Berlin: Springer Publishers). Grabher, G., Ibert, O., and Flohr, S. (2008). ‘The neglected king: the customer in the new knowledge ecology of innovation’. Economic Geography 84: 253–​280. Grönroos, C. (2008). ‘Service logic revisited:  Who creates value? And who co-​creates?’ European Business Review 20: 298–​314. Gulbrandson, I.T. and Just, S.N. (2011). ‘The collaborative paradigm: towards an invitational and participatory concept of online communication’. Media, Culture & Society 33: 1095–​1108. Haefliger, S., Jäger, P., and von Krogh, G., (2010). ‘Under the radar: industry entry by user entrepreneurs’. Research Policy 39: 1198–​1213. Hargadon, A.B. and Bechky, B.A. (2006). ‘When collections of creatives become creative collectives: a field study of problem solving at work’. Organization Science 17: 484–​500. Haythornthwaite, C. (2005). ‘Introduction:  computer-​ mediated collaborative practices’. Journal of Computer-​Mediated Communication 10: doi: 10.1111/​j.1083-​6101.2005. Howells, J. (2004). ‘Innovation, consumption and services: encapsulation and the combinatorial role of service’. The Service Industries Journal 24: 19–​36. Hwang, E. L., Sing, P. V., and Argote, L. (2015). ‘Knowledge sharing in online communities: learning to cross geographic and hierarchical boundaries’. Organization Science 26: 1593–​1611. Ibert, O. (2010). ‘Relational distance:  sociocultural and time-​spatial tension in innovation practices’. Environment and Planning A 42: 187–​204. Jacobs, J. (1969). The Economy of Cities (New York: Random House). Jeannerat, H. (2013). ‘Staging experience, valuing authenticity: towards a market perspective on territorial development’. European Urban and Regional Studies 20: 370–​384. Jeppesen, L.B. (2001). ‘Making consumer knowledge available and useful: the case of computer games’. Danish Research Unit for Industrial Dynamics (DRUID). Working paper No. 01-​10 http://​ideas.repec.org/​s/​aal/​abbswp.html (last accessed 28 April 2017). Jeppesen, L.B. and Laursen, K. (2009). ‘The role of lead users in knowledge sharing’. Research Policy 38: 1582–​1589. Kawamura, Y. (2006). ‘Japanese teens as producers of street fashion’. Current Sociology 54: 784–​801. Knorr Cetina, K. (1999). Epistemic Cultures: How the Sciences Make Knowledge (Cambridge, MA: Harvard University Press). Knorr Cetina, K. (2001). ‘Objectual practice’ in T.R. Schatzki, K. Knorr Cetina, and E. von Savigny (eds) The Practice Turn in Contemporary Theory, pp. 175–​188 (London: Routledge). Lakhani, K.R. and von Hippel, E. (2003). ‘How open-​source software works: “free” user-​to-​user assistance’. Research Policy 32: 923–​943. Langner, B. and Seidel, V.P. (2015). ‘Sustaining the flow of external ideas: the role of dual social identity across communities and organization’. Journal of Product Innovation Management 32: 522–​538. Lüthje, C., Herstatt, C., and von Hippel, E. (2005). ‘User-​innovators and “local” information: the case of mountain biking’. Research Policy 34: 951–​965.

Schumpeterian Customers?   303 Mahr, D. and Lievens, A. (2012). ‘Virtual lead user communities: drivers of knowledge creation for innovation’. Research Policy 41: 167–​177. Maskell, P. (2014). ‘Accessing remote knowledge—​the roles of trade fairs, pipelines, crowdsourcing and listening posts’. Journal of Economic Geography 14: 883–​902. Maskell, P., Bathelt, H., and Malmberg, A. (2006). ‘Building global knowledge pipelines: the role of temporary clusters’. European Planning Studies 14: 997–​1013. Mateos-​Garcia, J. and Steinmueller, W.E. (2008). ‘Open, but how much? Growth, conflict and institutional evolution in Wikipedia and Debian’ in A. Amin and J. Roberts (eds) Community, Economic Creativity, and Organization, pp. 254–​281 (Oxford: Oxford University Press). Müller, F.C. and Ibert, O. (2015). ‘(Re-​)sources of innovation: Understanding and comparing time-​spatial innovation dynamics through the lens of communities of practice’. Geoforum 65: 338–​350. Neff, G. and Stark, D. (2003). ‘Permanently beta: responsive organization in the Internet-​era’ in P. Howard and S. Jones (eds) Society Online: The Internet in Context, pp. 173–​188 (Thousand Oaks, CA: SAGE). O’Hern, M. and Rindfleisch, A. (2009). ‘Customer co-​creation:  a typology and research agenda’. Review of Marketing Research 6: 84–​106. Pinch, T. (2003). ‘Giving birth to new users: how the minimoog was sold to rock and roll’ in N. Oudshoorn, and T. Pinch (eds) How Users Matter. The Co-​construction of Users and Technologies, pp. 247–​270 (Cambridge, MA: MIT Press). Power, D. and Jansson, J. (2008). ‘Cyclical clusters in global circuits: overlapping spaces in furniture trade fairs’. Economic Geography 84: 423–​448. Prahalad, C.K. and Ramaswamy, V. (2004). The Future of Competition:  Co-​creating Unique Value with Customers (Boston, MA: Harvard Business School Press). Ramaswamy, V. and Oczan, K. (2014). The Co-​Creation Paradigm (Stanford, CA:  Stanford University Press). Ren, Y., Kraut, R., and Kiesler, S. (2007). ‘Applying common identity and bond theory to design of online communities’. Organization Studies 28: 377–​408. Riles, A. (2000). The Network Inside Out (Ann Arbor, MI: University of Michigan Press). Ritzer, G. and Jurgenson, N. (2010). ‘Production, consumption, prosumption. the nature of capitalism in the age of the digital “prosumer” ’. Journal of Consumer Culture 10: 13–​36. Schumpeter, J.A. (1912/​2006). Theorie der wirtschaftlichen Entwicklung (Berlin:  Duncker & Humblot). Shah, S.K. and Tripsas, M. (2007). ‘The accidental entrepreneur: the emergent and collective process of user entrepreneurship’. Strategic Entrepreneurship Journal 1: 123–​140. Tapscott, D. and Williams, A.D. (2006). Wikinomics. How Mass Collaboration Changes Everything (New York: Penguin). Thrift, N. (2000). ‘Performing cultures in the new economy’. Transactions of the Association of American Geographers 90: 674–​691. Thrift, N. (2006). ‘Re-​inventing invention:  new tendencies in capitalist commodification’. Economy and Society 35: 279–​306. Tseng, M.M. and Piller, F. (2011). The Customer-​ centric Enterprise:  Advances in Mass Customization and Personalization (Berlin: Springer) Urry, J. (2003). ‘Social networks, travel and talk’. British Journal of Sociology 54: 155–​175. von Hippel, E. (1986). ‘Lead users: a source of novel product concepts’. Management Science 32: 791–​805.

304   Grabher AND IBERT von Hippel, E. (1994). ‘“Sticky information” and the locus of problem-​solving: implications for innovation’. Management Science 40: 429–​439. von Hippel, E. (2005). Democratizing Innovation (Cambridge, MA: MIT Press). Weber, S. (2004). The Success of Open Source (Cambridge, MA: Harvard University Press). Weick, K.E. and Roberts, K.H. (1993). ‘Collective mind in organizations: heedful interrelating on flight decks’. Administrative Science Quarterly 38: 357–​381. Wiertz, C. and de Ruyter, K. (2007). ‘Beyond the call of duty: why customers contribute to firm-​hosted commercial online communities’. Organization Studies 28: 347–​376. Yaqub, O. and Nightingale, P. (2012). ‘Vaccine innovation, translational research and the management of knowledge accumulation’. Social Science & Medicine 75: 2143–​2150. Zwick, D. and Dholakia, N. (2006). ‘The epistemic consumption object and postsocial consumption: expanding consumer–​ object theory in consumer research’. Consumption, Markets and Culture 9: 17–​43.

Chapter 16

The Geo g ra ph y of the Cre at i v e Indu stries:  Th e oret i c a l Sto cktaki ng a nd E m pirical Illu st rat i on Mark Lorenzen Introduction Early on, social sciences paid attention to the economics of arts and culture under the heading of ‘cultural’ industries (Horkheimer and Adorno, 1944; Hirsch, 2000). More recently, attention has turned to a broader economy comprised by the so-​called creative industries. The creative industries can be defined as those that are organized principally to take advantage of human creativity and capture its market value, and include ‘cultural’ industries such as visual and performing arts, and also media, software, architecture, and design (some definitions also include events, sports, heritage, and tourism). ‘Creativity’ is a term often used about invention of something new by combining elements that already exist (Boden, 1990; Sternberg, 1999). Such processes hinge upon individuals capable of and willing to engage in open-​ended and uncertain activities of search and experimentation. Thus, most companies in the creative industries are skill-​intensive, leveraging expert labour and specialized technologies for value creation through producing new experience product ‘content’. Creative industries are important for dissemination of cultural expressions and values (Giddens, 2000; Anheier and Isar, 2008). Furthermore, they have comparatively high impact on economic development owing to their prospects for social and economic growth and spillovers to other sectors in the economy (Cunningham and Potts, 2015). For these reasons, creative industries attract interest from policymakers (e.g. European Commission, 2001; OECD, 2006; UNESCO, 2006; DCMS, 2007) and from scholars across the social sciences (e.g. Caves, 2000; Towse, 2003; Hartley, 2004; Ginsburgh and Throsby, 2006; Flew, 2013; Hesmondhalgh, 2013; Jones et al., 2015). In economic geography, there is also growing

306   Lorenzen attention paid to creative industries (e.g. Scott 2000, 2005; Power and Scott, 2004; Cooke and Lazzeretti, 2008; Lorenzen et al., 2008; Vorley et al., 2008; Pratt and Jeffcut, 2009; Lazzeretti, 2013). For geographers, creative industries are particularly interesting owing to their complex interrelationships with cultural, social, and economic societal factors and profound ongoing changes of their local, as well as global geography. The latter development makes the creative industries a useful vehicle for employing and examining theory on the geographical organization of economic activity. This chapter will stylize recent developments of the geography of the creative industries and summarize the main theoretical arguments made for explaining these developments. To illustrate, the chapter will then apply some of these arguments in a brief juxtaposition of the recent developments of the film industries in Copenhagen, Denmark, and Mumbai, India (‘Bollywood’).

The Geography of the Creative Industries Clustered Production Since antiquity, art and culture has been centred in cities (Jennings, 2011), and a key feature of the geography of the creative industries is pronounced clustering of production activities. Even after the emergence of technologies that enable mass production and the separation of production and consumption (e.g. book printing and audiovisual recording and broadcasting), the tendency of creative production to cluster in larger cities has intensified (Scott, 2000; Lazzeretti, 2013). During the second half of the twentieth century, well-​ studied creative industry clusters have grown larger and more diversified (Langdale, 1997; Power and Scott 2004; Scott, 2005; Cooke and Lazzeretti, 2008). Recently, creative industry clusters in emerging economies have also attracted scholarly attention (Barrowclough and Kozul-​Wright, 2006).

Local Consumption Some creative industries, such as advertising, design, and architecture, are service providers. Their production processes encompass customization in direct interaction with clients, and long-​lasting client relationships are valuable. Consequently, these industries mostly serve purely local, typically urban, markets. Other creative industries, such as publishing, audiovisuals and media, games, and performing arts, serve consumer markets. In some of these, such as concerts and stage plays, a significant part of production entails physical presence of audiences and hence takes place at the same time and locality as consumption—​again, typically in larger cities. Other creative consumer industries are able to reproduce (replicate) products at low cost, and, consequently, separate production from consumption (Caves, 2000). However, the latter (as measured by, for instance, product distribution and sales) often exhibit distinct national patterns:  most cultural consumer industries produce for the home market. Throughout the twentieth century, only the USA and a handful of Western European countries managed to export on a large

The Geography of the Creative Industries    307 scale. Over the last decades, however, this geography of consumption is changing, and more countries, such as Japan, India, and Brazil, now have substantial exports of creative products (UNCTAD, 2008; Lorenzen, 2009; Hirsch and Gruber, 2015; Khaire, 2015; Pratt, 2015).

Spatial Division of Labour During the second half of the twentieth century, the creative industries saw the emergence of a spatial division of labour between increasingly specialized clusters. Like in many other industries, this process was been primarily cost-​driven, with multinational enterprises (MNEs) in some creative industry clusters offshoring labour-​intensive production activities to other clusters with lower factor costs. Hollywood’s ‘runaway’ productions in Toronto and Wellington are well-​known cases (Wasko, 2003; Coe, 2015). This offshoring continues today, but during the last decades, it has been complemented by a knowledge-​ based spatial division of labour (Mudambi, 2008). One facet of this is increased product content sourcing by MNEs from local producers (e.g. record labels or book publishers) with the purpose of distributing this content globally. Another facet of the increased importance of local specialized knowledge is international co-​productions between production companies, suppliers, and artists in different creative industry clusters (Morawetz et al., 2007).

Complex and Interconnected Global Geography The result of the aforementioned developments is a shift towards higher complexity of globally competing and complementary creative industry clusters. For half a century, the creative industries had a relatively stable hierarchy of clusters with Hollywood, New York, Berlin, Rome, Milan, London, and Paris as the top performers, exporters, and offshorers. Recently, however, this hierarchy is being challenged by the catch-​up by globally operating clusters such as Beijing, Hong Kong, Bollywood (Mumbai), Tokyo, Istanbul, Mexico City, Bangkok, and Singapore. Furthermore, emerging market clusters such as Nollywood (Lagos) are creating regional markets for ‘frugal’ cultural product innovation (Barnard and Tuomi, 2008). The geography of the creative industries is becoming more complex and interconnected in (at least) two dimensions: (i) exports and cultural hegemony; and (ii) value creation and value capture.

Exports and Cultural Hegemony From a configuration with only a handful of clusters exporting creative products (and, in the process, disseminating cultural expressions and values worldwide), the creative industries are seeing a shift towards a diverse population of clusters in a variety of countries that serve local, regional, and global markets. Not only does this create notable economic development in new localities, but it also challenges the thesis of homogenization and Western cultural hegemony (Appadurai, 1996; Tomlinson, 1999) by making cultural expressions and values from, for example, Japan, India, and Brazil globally known.

308   Lorenzen

Value Creation and Value Capture From a hierarchy with dominant creative industry clusters capturing the bulk of value through focusing on intellectual property (IP)-​generating activities and other clusters undertaking offshored less attractive activities, the creative industries are seeing a shift towards global production networks of specialized, knowledge-​intensive clusters (Coe, 2015; Pratt, 2015). These networks have more complex and fluid distribution of activities, generation of IP rights, and value capture (Hirsch and Gruber, 2015). The next section takes stock of what we know about the mechanisms that shape the above geography of the creative industries, and attempt to explain theoretically the current developments.

Theoretical Stocktaking The following combines arguments from economic geography and neighbouring disciplines, primarily economics, and management.1 The theories are presented in a thematical rather than chronological order. In turn, the section will outline explanations of the clustering of creative production, localization of creative consumption, spatial division of labour among creative industry clusters, and the emergence of an increasingly complex and interconnected global geography in the creative industries.

External Economies Two well-​established theoretical arguments in economic geography, the ‘Marshallian’ idea of external economies of agglomeration and the ‘Jacobian’ perspective of external economies arising from urbanization, are decisive in explaining clustering of creative production. Creative industries need to develop product content through recombination of aesthetic and/​or narrative elements (Caves, 2000; Hesmondhalgh, 2013). For that reason, production in the creative industries is organized in temporary projects that bring together diverse sets of creative skills and other resources on a temporary basis (Grabher, 2002; Pratt, 2002). Some creative industries have relatively long product cycles and/​or a focus on product brand and quality. Hence, they use relatively long-​lasting projects and reuse resources, such as labour, between projects. This entails internal scale economies and leads to integrated ‘production studios’, as in Hollywood during the first half of the twentieth century, or in some contemporary computer games companies and advertising agencies. Most creative industries, however, have short product life cycles and need to develop highly varied new product content. Thus, they use projects with non-​standardized, flexible tasks and highly diverse skills, people, and motivations. This entails no internal scale economies and projects are typically market based (Lorenzen and Frederiksen, 2005), organized as temporary networks of collaborating freelancers (e.g. artists or writers) and specialized skill containers (e.g. technical services), orchestrated by project coordinators (e.g. record labels or film producers) (Whitley, 2006). During the last decades, this form of organization of production has become dominant in many creative industries, including filmed entertainment and

The Geography of the Creative Industries    309 recorded music: the production studios in Hollywood, London, and even Bollywood have long since disintegrated. Market-​based projects hinge on external economies made possible by spatial agglomeration. Geographical proximity, local institutions, and the presence of specialized local project coordinators and boundary spanners in creative industry clusters allow labour and companies to collaborate and specialize (Storper, 1989; Storper and Christopherson, 1987; Scott, 2000). For example, Grabher (2002), Pratt (2002), and Lorenzen and Frederiksen (2005) demonstrate how proximity lowers transaction costs and allows for external economies in production projects among clustered advertising, media, and music companies. Another external economy of agglomeration relates to the local labour market, where training, exchange, and formal education of skilled labour is made possible by the large size of the local creative industries (Marshall, 1890). Creative industry clusters are unequivocally urban (Scott, 2000; Power and Scott, 2004:  Scott, 2008; Lazzeretti, 2013). External economies related to urbanization (Jacobs, 1961) include the unique ability of large cities to attract creative talent (Glaeser et al., 2001; Florida et al., 2008). They also include the presence of urban venture capital and of clusters in other industries providing synergies in reusing IP and brands (e.g. across films, books, soundtracks, and merchandise) and inspiring experimentation and new forms of product content generation (Lorenzen and Frederiksen, 2008).

Localized Learning The argument of localized learning (in its several guises) has been prominent in economic geography since the late 1990s. It adds to the explanation of clustering of creative production, and is also useful for explaining the local nature of creative consumption. The argument’s fundamental tenet is that while transaction costs of operating globally (across distance) have generally declined owing to improved information, digitization, logistics and transportation technologies, spatial costs related to knowledge and learning remain significant for a range of industries (Morgan, 2004; McCann, 2007; Christopherson et al., 2008). The reason is that the production, appropriation, and sharing of knowledge is context specific (Gertler, 2003): it depends on local ‘codebooks’ (Maskell and Malmberg, 1999; Maskell, 2001) and embeddedness in high-​trust local networks (Maskell and Lorenzen, 2004; Gordon and MacCann, 2005). Furthermore, knowledge may be ‘in the air’ locally (Marshall, 1890) through local informal information spillovers (Storper and Venables, 2004). The creative industries are a set of industries with high spatial costs of knowledge and learning, and localized learning is an important element of external economies in cultural industry clusters. Cultural expressions and understanding of cultural values are dependent upon local context, so access to local networks and information spillovers is crucial in the production of creative content. For example, Bathelt (2005) outlines how local information ‘buzz’ can propagate sharing of knowledge among clustered media production companies. Furthermore, consumer trends and preferences change so fast that creative companies often need to be physically present in a market: they need to access local knowledge in order to design effective marketing campaigns and distribution activities. By contrast, on export markets, creative companies suffer from notable liabilities of foreignness (Hoskins and Mirus, 1988; Oh, 2001): they know little of cultural expressions, trends, and preferences, and

310   Lorenzen have difficulties tapping into local networks in order to obtain pertinent knowledge. For that reason, most cultural consumer industries focus on their home market.

Market Scale The localized learning argument is not helpful in explaining why some cultural industry clusters have been more successful than others in exporting beyond their home market. In order to explain the early export dominance of the USA and a few Western European countries, we may turn to an argument originating in marketing theory in management: that of market-​scale advantages. As creative production is labour-​intensive and uncertain, in order to be profitable, creative products need either be sold at premium price (on niche markets) or in large quantities (on mass ‘mainstream’ markets (Caves, 2000)). For reasons outlined earlier, creative industries first and foremost focus on the home markets in order to capture value. For most of the twentieth century, only a few countries were able to build notable exports, and they all shared one feature: they had sizeable home markets. On mainstream markets for creative products, there are substantial scale advantages in marketing, as the marginal costs of, for example, promotion campaigns decrease rapidly (Vogel, 1998; Lee and Waterman, 2007). Creative companies with large home markets are able to build mass marketing that can be used not just on home markets, but also for exports. Even if companies with mass marketing still suffer from liabilities of foreignness on export markets, they will be powerful compared with companies with less home market scale and marketing muscle (Elberse and Eliashberg, 2003; Bakker, 2005). On mainstream markets, efficient distribution is necessary to take advantage of mass marketing (Donahue, 1987; Vogel, 1998). During the twentieth century, when many creative consumer products were distributed in physical format (e.g. CDs, books, prints), there were significant scale advantages owing to decreasing marginal costs of transport and contact to retailers. Today, even with digitization and online distribution, distribution on mainstream markets still entails large investments (and online distribution sites hinge on efficient marketing for visibility). Consequently, creative companies with large home markets stand at an advantage when building mass distribution infrastructures that can be used for exports: the most prominent digital distribution platforms are owned by companies in the USA, Japan, China, UK, and France. Home market-scale advantages not only account for why countries with large home markets have been successful exporters, but they also explain why such countries are virtually impenetrable to imports: competitors from smaller countries have smaller home markets, less marketing and distribution muscle, and stand at a comparative disadvantage when trying to enter larger countries with strong incumbents (Bakker, 2005).

Sunk Cost While the market-​scale argument explains the successful exports by a few countries, it does not explain why a range of countries with very sizeable home markets, such as China, India, and Brazil, have not been able to export until late in the twentieth century. Furthermore,

The Geography of the Creative Industries    311 market scale is not helpful for understanding the spatial division of labour and hierarchy among creative industry clusters in the twentieth century. To understand these phenomena, we turn to an argument derived from industrial economies: that of sunk cost. As outlined earlier, creative companies with large home markets have leveraged home market scale to invest in mass marketing and distribution. Such investments were made, in particular, by US film companies (since the 1910s and accelerating since World War II), US and UK recorded music companies (since World War II), and by US, French, German, and Japanese cross-​media conglomerates (during the latter half of the twentieth century) (Wildman, 1995; Hoskins et  al., 1997; Vogel, 1998). These investments represented sunk costs (i.e. irretrievable investments),2 changing the overall nature of competition in creative industries by setting the minimum competitive scale of marketing and distribution higher (Caves, 2000; Bakker, 2005, 2015). Early-​mover companies having sunk costs into marketing and distribution are able to reinforce continuously their competitive advantage: capturing revenues at home, as well as abroad, they have the capacity to sink further costs in factors that further influence the nature of competition (e.g. production quality values and technological formats). Thus, the sunk-​cost argument explains the export success of early movers, such as the USA and a few Western European countries, as well as why latecomers, such as China, India, and Brazil, were not able to build exports in spite of being strong in their home markets: they faced entry barriers, in the guise of minimum competitive scale of marketing, distribution, production quality values, and so forth, that made them unable to compete on export markets. Sunk cost also helped early-​mover companies in the USA and Europe to build ‘cultural competitiveness’: Preferences of consumers for particular types of product content, such as cultural expression, or, simply, language. By investing in marketing and distribution on export markets over decades, US and European countries (in particular, the UK) have influenced consumer preferences on export markets (Wildman, 1995; Vogel, 1998). For example, the preferences in many countries for Hollywood-​type productions and for English-​ language films over other foreign-​language films have been created by Hollywood itself. The greater differences in cultural expression there are among the world’s other exporting film clusters, the greater the export advantages of Hollywood (Hoskins et al., 1997; Papandrea, 1998; Oh, 2001; Elberse and Eliashberg, 2003). The sunk-​cost argument explains the dominance of Hollywood and a range of other creative industry clusters in the USA and Western Europe on export markets everywhere, in terms of exports, cultural hegemony, value creation, and value capture: for the latter half of the twentieth century, these early-​moving and scale-​based clusters became the ones to produce, develop IPs, and export, and later, the ones to offshore the most costly production activities in, for example, media and computer games to later-​moving clusters that were slower to build exports. To a high extent, early sunk cost influenced the later spatial division of labour in the creative industries.

Connectedness and Connectivity During the last decades, the global geography of the creative industries has been changing rapidly, with more and more countries exporting, and with a more knowledge-​based, less hierarchical, and more fluid spatial division of labour. In order to understand the emergence

312   Lorenzen of this more complex and interconnected global geography, we turn to a theoretical argument developed in the intersection of the international business studies in management, economic geography, and innovation studies: global connectedness and connectivity. During the last decade, scholars have paid increasing attention to connections between clusters (Amin and Thrift, 1992; Dicken et al., 2001; Bathelt et al., 2004; Hudson, 2005; Coe et al., 2008; Lorenzen and Mudambi, 2013, 2015). One type of such connection is personal relationships, such as family relations, friendships, and acquaintanceships between individuals living in different clusters (diasporas) (Lorenzen and Mudambi, 2013). This type of connection has been on the rise during the last decades as mobility of labour has increased due to political (de-​) regulation, emerging international employment opportunities, and developments in communication and transportation technologies that enable personal relationships to be maintained across geographical distance. Personal relationships enable flows of knowledge and finance and facilitate entrepreneurial opportunities across clusters (Saxenian, 2006). A second type of connection is organizational pipelines (Bathelt et al., 2004), such as strategic alliances, joint ventures, or ownership-​spanning organizations or organizational units located in different clusters (the most prominent example being MNE branch plants). Organizational pipelines have increased dramatically over the last half of the twentieth century, owing to declining spatial costs of transportation, communication, and governance across geographical distance. This type of connections facilitates flows of goods (trade), as well as more substantial flows of knowledge and finance than personal relationships allow for. Through personal relationships and organizational pipelines, different clusters may be connected with varying power and network centrality of the partaking individuals and organizations. A cluster’s particular configuration of connections, that is, its ‘connectivity’, is likely to influence its ability to innovate and capture value (Lorenzen and Mudambi, 2013, 2015). Creative industry clusters are increasingly connected by personal relationships between diasporas of industry professionals, artists, and other skill-​holders. For instance, Currid-​ Halkett and Ravid (2012) report on a global network of celebrities from different clusters who meet and interact at various industry events. The rise of personal relationships between creative industry clusters may explain the rise in international co-​productions, that is, project networks that cross the boundaries of clusters, bringing labour and companies from different clusters together temporarily. Organizational pipelines between creative industry clusters, in the guise of local subsidiaries of MNEs (e.g. record labels or publishers) have been present for more than half a century. However, across the creative consumer industries, MNEs are now changing the mandate of local branch plants. From being mostly about local marketing and distribution of imported products, the mandate is changing towards being more ‘competence-​creating’ (Cantwell and Mudambi, 2005), sourcing local content for local, and sometimes also export, markets. Furthermore, MNEs are increasingly changing their offshoring strategies towards knowledge-​based global production networks, using suppliers in different clusters on account of their skills rather than their low labour costs. The search for local specialized knowledge also leads MNEs to enter into creative-​industry clusters by joint ventures and acquisitions rather than branch plants. For instance, Disney’s rather unsuccessful local branch plant in Bollywood has been supplanted by an acquisition of major local production company in order to access its localized knowledge and develop product content for both the Indian and export markets (Lorenzen and Täube, 2008).

The Geography of the Creative Industries    313 The rise of connections between clusters, and between clusters and their export markets for creative products, helps to explain how clusters in small countries and late-​coming clusters are increasingly able to overcome their disadvantages and entry barriers. Firstly, global production networks in the guise of co-​productions increasingly develop exportable products, by including labour and companies from different clusters, contributing with their specific knowledge of different markets. Some co-​productions that target particular export markets may be able to take advantage of diasporic connections to these markets in obtaining knowledge of trends and preferences. Secondly, creative products are increasingly digitized, and, as a result, global niche markets,3 such as those for Manga cartoons or Kung Fu films, are emerging. On such markets, audiences have high consumption capital (Caves, 2000) and are actively searching for content, and small, late-​coming companies can leverage their personal relationships and forge strategic alliances in order to market and distribute. While revenues on global niche markets for ‘world’ and ‘arthouse’ creative products are still modest, they are distributed across a broad range of companies and clusters globally, by contrast to revenues on mainstream markets that are still captured largely by the hitherto dominant clusters in Hollywood, New York, Berlin, and so forth.

Empirical Illustration: The Filmed Entertainment Clusters in Copenhagen And Bollywood This section illustrates the chapter’s theoretical arguments by way of examples from the filmed entertainment industry.4 This remains the largest of the creative industries measured by revenues. As a paradigmatic case, it has attracted scholarly attention from a range of social sciences, including not the least a range of empirical studies by geographers (e.g. Christopherson and Storper, 1986; Storper and Christopherson, 1987; Storper, 1989; Brito Henriques and Thiel, 2000; Scott, 2000, 2005; Coe, 2001, 2015; Turok, 2003; Coe and Johns, 2004; Christopherson, 2006; Lorenzen, 2007, 2009; Currah, 2007; Kaiser and Liecke, 2007; Vang and Chaminade, 2007; Mossig, 2008; Zademach, 2009; Johns, 2010). Not surprisingly, the bulk of studies of the filmed entertainment industry focuses on Hollywood. In the following, we will address two less studied empirical examples, those of the Danish cluster in Copenhagen and the Indian cluster in Mumbai (‘Bollywood’).

Copenhagen Producing 20–​25 feature films annually and with a turnover of roughly US$135 million in 2014, the Copenhagen filmed entertainment cluster is small. With a small home market, significant public subsidies, and organized as a swarm of small producers relying on the distribution channels of a few larger corporations, Copenhagen resembles many other small filmed entertainment clusters in Europe. However, with 4.5 films produced per Danish citizen (Hollywood produces around two), a home market share of 30 per cent, and a strong

314   Lorenzen presence in global arthouse markets, Copenhagen performs notably better than comparative small filmed entertainment clusters. This aligns well with the theoretical arguments outlined in this chapter. The cluster was early to build external economies, as well as early-​mover sunk cost advantages on export markets, but owing to lack of home market scale, lost its exports. Later, through localized learning and the emergence of connectedness to clusters in other European filmed entertainment clusters, Copenhagen rebuilt its performance, recapturing a large share of its home market and establishing a presence on global niche markets. The filmed entertainment cluster in Copenhagen originated with the world’s oldest continuously operating film production company, Nordisk Film (established 1906). For the first decades, the cluster operated without notable external economies: in a model not unlike the early Hollywood studios, Nordisk Film integrated much of the production in the cluster. In the silent film era, the localized knowledge of cultural expressions and preferences was not acutely important and liabilities of foreignness on export markets not significant. Consequently, Nordisk Film, moving early in sinking cost into production quality, as well as mass marketing and distribution, came to dominate not only the Danish home market, but also European export markets: until World War I, Copenhagen was Europe’s most exporting film cluster. Hollywood’s rise to dominance, along with the spread of sound films in the 1930s and the resulting importance of localized knowledge, meant a collapse of Copenhagen’s film exports. After World War II, Hollywood increased investments in Europe, aided by favourable export conditions for American products. With a very small home market, the Copenhagen cluster had no scale to counter the competition from Hollywood, and like all other European filmed entertainment clusters, even if purchase power was growing, lost the majority share of its home market to Hollywood imports. From the late 1960s, the Danish state invested massively in upgrading of the Copenhagen cluster. From 1964, the Danish Film Act channelled tax revenues from cinemas back into production subsidies. The Act was expanded notably in 1972 with the establishment of the Danish Film Institute, channelling subsidies though independent consultants, advising film production projects, and promoting skill development. During the first decades, subsidies were handed out mostly to niche films, but since a revision of the Act in 1989, it also promotes mainstream films with up to 60 per cent of their production budget. This created incentives for incumbent companies to invest in marketing, distribution, and exhibition infrastructures (cinemas). It also attracted new entries to the cluster, creating external economies: with new and more specialized local labour and companies, projects became increasingly market-​based and relied on shared knowledge of expressions, styles, and market trends (as well as of successful ways of obtaining public subsidies). External economies also related to urbanization, as traded and untraded interdependencies with the Copenhagen TV, theatre, and software games industries grew in importance. As a result of these developments, Copenhagen recaptured significant shares of its home market from Hollywood since the 1960s. Another important public investment in the Copenhagen cluster was the establishment in 1966 of the National Film School of Denmark. After a restructuration and increased public funding in the 1980s, the school became successful in training filmmakers to manoeuver between niche and mainstream markets. A generation of new Danish directors trained at the School was able to target successfully the emerging global niche markets and win awards at high-​profile international festivals, such as the Academy (Bille August; Susanne Bier) and Cannes (Lars von Trier; Thomas Vinterberg). Coined by the latter two directors, the Danish

The Geography of the Creative Industries    315 ‘Dogme 95’ manifesto of avant-​garde film-​making was successful in adding to the international brand of Danish films and promote niche exports. Inspired by the growing exports to global niche markets, the largest Copenhagen companies began investing in infrastructures to promote exports of mainstream films, particularly to Scandinavia and Western Europe. Since 2000, the export rate of Danish films has grown to above 10 per cent. The post-​1980 generations of film-​makers who increasingly collaborated inside the cluster also began to establish global connectedness in the guise of international co-​productions with companies in clusters in, for example, Sweden, Norway, the UK, and Germany. Such connections were successful in obtaining funding (regional production subsidies), incorporating new talent, and establishing new channels to European markets. The majority of the Danish films winning awards at international festivals are international co-​productions, but in spite of the positive impact on exports of global connectedness, the Copenhagen cluster remains predominantly locally embedded and focused on the home market. Of the twenty-three feature films produced in 2015, eight were international co-​productions.

Bollywood By contrast to Copenhagen, Bollywood is big business. Bollywood produces roughly the same number of films as Hollywood, 200–​250 films annually, and sells more than a billion tickets annually.5 Even if the low Indian purchase power means that Bollywood’s earnings are around a tenth of Hollywood’s, less than US$2 billion in 2014, Bollywood enjoys no public subsidies and, like its American cousin, is unabashedly profit-​seeking in its pursuit of mass markets for filmed entertainment, increasingly on export markets. While these two clusters in some ways resemble each other, their historical contingencies have been very different, resulting in very different industry structures and geographies. Bollywood’s history is as long as Hollywood’s (and Copenhagen’s), but the cluster is currently undergoing comparatively dramatic restructuring and geographical change. The theoretical arguments outlined in this chapter are very relevant to understanding this development. Significant external economies have enabled Bollywood to become dominant on the very diverse Indian home market, but given the low purchase power here, the cluster did not earn sufficient revenues to sink early investments into marketing or distribution capacity. Thus, it did not develop notable exports before Hollywood’s rise to dominance. However, during the last decades, Bollywood’s global connectedness now facilitates rapid catch-​up to Hollywood in terms of exports, value capture in global networks, and dissemination of cultural expressions and values. The first Indian film was produced in 1913 in Mumbai (formerly Bombay), and filmed entertainment clusters emerged here and in a handful of other major Indian cities. During the first decades of the twentieth century, Bollywood had to carve out its existence confronted with British censorship and Hollywood imports, but with the presence of large urbanization economies in Mumbai, where talent and venture capital flocked, Bollywood grew fastest of all Indian filmed entertainment clusters. After two decades, more than half of India’s film production took place here. With the advent of sound films, film imports to India from Hollywood collapsed and the home markets segregated along regional languages. The other Indian filmed entertainment clusters focused on each their regional market and developed integrated studios and star actors with durable regional audience brands. Bollywood, by contrast, focused on Hindi,

316   Lorenzen a language widely used in Mumbai owing to centuries of inflow of Hindi-​speaking immigrants. Hindi became India’s national language with independence in 1947 and this created a substantial market across India, in the Indian diaspora, and in India’s neighbouring countries. The Hindi film market remained very segregated in terms of social class, regional culture, and religion. Bollywood developed product content able to appeal across these divides, recombining expressions, narratives, and styles from Indian regional art forms, as well as international trends (this formula was later coined masala, ‘spice mix’). As a result, Bollywood captured the major share of the Indian national market. Today, the cluster represents 15 per cent of India’s film output but captures 40 per cent of film revenues. The reason for Bollywood’s success in developing such product content was that the cluster shifted from an integrated studio system, dominant in its first three decades, to market-​based project networks of freelancing actors, directors, technicians, specialized stages, and studios, and small, typically family owned production companies. The reason for Bollywood’s shift from internal to external economies was a high entry of small independent film-​makers from other Indian cities and from the North (e.g. the film cluster in Lahore) following the division of India and Pakistan in 1947. While Bollywood’s horizontal disintegration made the cluster able to develop competitive product content for the Indian market, it also created high transaction costs in film production and prevented the cluster from sinking costs into marketing and distribution infrastructures. Consequently, India’s film exhibition remained fragmented, and Bollywood had difficulties in collecting from the thousands of stand-​alone cinemas across India. Furthermore, for most of the twentieth century, Bollywood collected virtually no export revenues. The cluster had no export infrastructure to compete with Hollywood, and Bollywood consumption outside India largely took place as imports by small-​scale and inefficient independent agents from the Indian diaspora. During the last decades of the twentieth century, Bollywood began to build a new type of global connectedness with a tremendous impact on the cluster’s performance and geography. Bollywood’s early connectedness—​the personal relationships between Bollywood producers and directors and European filmmakers—​disappeared with the studio system, and for almost fifty years, the cluster remained largely isolated from global markets and other filmed entertainment clusters. However, Indian emigration for education and work burgeoned after the 1950s, and by the end of the century, more than ten million non-​resident Indians and persons of Indian origin lived in North America, the UK, and the Middle and Far East. Direct interaction between the diaspora and their family and friends in India has been propagated by growing purchase power of the diasporas and the Indian middle class, as well as cheap air travel and communication technology. Bollywood is now highly connected by personal relationships between Bollywood producers, directors, investors, and other professionals, and thousands of individual consumers, investors, and skill-​holders of Indian origin living abroad, and there is also growing immigration into Bollywood by specialized skill-​holders (actors, scriptwriters, directors, etc.) from the Indian diasporas. As a result, knowledge flows into Bollywood of Western and global trends and preferences, allowing the cluster to change its product content to become more exportable. Hence, Bollywood challenges Hollywood as an exporter. At first, the cluster targeted the very attractive market of the Indian diasporas in the UK and North America. In the process of developing product content and export infrastructures to serve the diaspora, Bollywood also became able to reach other export markets. Bollywood is now a major player on global markets for filmed entertainment products. During the first decade of the twenty-​first

The Geography of the Creative Industries    317 century, the cluster’s export revenues grew by 450 per cent, and since then, export rates have been between 10 and 15 per cent. Bollywood has become the largest foreign exporter of filmed entertainment to the US market, and every year several Bollywood films are in the top-​twenty box-​office charts in North America and the UK. The export boom means that Bollywood also challenges Hollywood with respect to dissemination of cultural expressions and values: Bollywood films now appear at most international film festivals, as well as on national TV across the globe, and new generations of cinemagoers are increasingly familiar with Indian film stars and styles. By setting up its own global production networks, Bollywood also challenges Hollywood’s dominance in the spatial division of labour in the filmed entertainment industry. Focusing on generating IP rights, Bollywood is now a major offshorer of less value-​capturing production tasks to specialized suppliers in other countries (both low-​cost, such as Pakistan and the Philippines, and knowledge-​intensive, such as the UK, Australia, and the USA). The cluster routinely shoots on location, typically in iconic locations such as New York, London, Dubai, or Sydney that add to the films’ export potential, and also places outward foreign direct investment in filmed entertainment companies in the UK and North America (e.g. Bollywood company Reliance Entertainment owns the majority share of Hollywood’s Dreamworks). With annual growth rates between 10 and 20 per cent, Bollywood currently outperforms most other filmed entertainment clusters in emerging economies. At home, the cluster is expanding urbanization economies as it interacts with complementary industries in Mumbai like TV, pop music, computer games, and advertising. Abroad, the cluster continues to sink costs into marketing and distribution infrastructures, and now own more than 250 cinemas and several TV distribution platforms in North America alone. It is largely owing to Bollywood that entertainment is currently India’s second biggest growth sector. Bollywood’s expansion is fuelled partly by growing Indian purchase power and more efficient collection at home. However, given the significantly higher purchase power on export markets, Bollywood’s global connectedness lies at the heart of the cluster’s rapid growth and the reconfiguration of its geography.

Conclusion The chapter has provided an introduction to the geography of the creative industries, as well as the set of theories regularly used to understand them. The creative industries are diverse and dynamic, and the chapter’s overview of their recent geographical development has been highly stylized. Likewise, the chapter has only focused on a narrow set of theories from economic geography and its neighbouring disciplines, providing a very condensed overview. Nevertheless, the chapter has attempted to show that while well-​established theories of external economies, market size, and sunk cost are still valid in understanding the fundamental organization of the creative industries, we need to engage recent and less-​established theories of localized learning, connectedness, and connectivity to understand the creative industries’ wide-​ranging global changes. Thus, the creative industries remain not just important in their own right, but also as an excellent showcase for the current shortcomings and potentials of theory development in economic geography.

318   Lorenzen

Notes 1. By necessity, what follows is very schematic, omitting much of the richness of insight presented in the source material. 2. Such sunk costs are referred to as ‘endogenous’ as they are made by early-​mover companies as a competitive strategy, not determined by exogenous industry factors such as technologies (Sutton, 1991). 3. By contrast to an international market, which is usually defined as the sum of several distinct national markets with different institutions and preferences, a global market is integrated across countries. Sharing institutions and preferences, a global niche market may offer significant scale (Daly, 1999). 4. This section draws upon empirical material also reported in Lorenzen (2007, 2009), Lorenzen and Taübe (2008), Andersen (2013), and Lorenzen and Mudambi (2013), as well as figures and overviews provided by Federation of Indian Chambers of Commerce and Industry/​KPMG (2015), the Danish Film Institute (dfi.dk, accessed 1 March 2016) and the Danish Producers’ Association (pro-​f.dk, accessed 1 March 2016). 5. Bollywood is the best known, but by no means the only, filmed entertainment cluster in India. Clusters in other Indian major cities all produce hundreds of regional-​language films annually, making India the world’s largest film producer with well over 1000 films annually.

References Amin, A. and Thrift, N. (1992). ‘Neo-​Marshallian nodes in global networks’. International Journal of Urban and Regional Research 16: 571–​578. Andersen, K.V. (2013). ‘The problem of embeddedness revisited: collaboration and market types’. Research Policy 42: 139–​148. Anheier, H.K. and Isar, Y.R. (eds) (2008). Cultures and Globalization: The Cultural Economy (London: SAGE). Appadurai, A. (1996). Modernity at Large: Cultural Dimensions of Globalisation (Minneapolis, MN: University of Minnesota Press). Bakker, G. (2005). ‘The decline and fall of the European film industry: sunk costs, market size, and market structure, 1890–​1927’. Economic History Review 58: 310–​351. Bakker, G. (2015). ‘Sunk Costs and the Dynamics of Creative Industries’ in C. Jones, M. Lorenzen, and J. Sapsed (eds) The Oxford Handbook of the Creative Industries, pp. 351–​386 (Oxford: Oxford University Press). Barnard, H. and Tuomi, T. (2008). ‘How demand sophistication (de-​)limits economic upgrading: comparing the film industries of South Africa and Nigeria’. Industry and Innovation 15: 647–​668. Barrowclough, D. and Kozul-​Wright, Z. (eds) (2006). Creative Industries and Developing Countries: Voice, Choice and Economic Growth (London: Routledge). Bathelt, H. (2005). ‘Cluster relations in the media industry: exploring the ‘distanced neighbour’ paradox in Leipzig’. Regional Studies 39: 105–​127. Bathelt, H., Malmberg, A., and P. Maskell (2004). ‘Clusters and knowledge: local buzz, global pipelines and the process of knowledge creation’. Progress in Human Geography 28: 31–​56.

The Geography of the Creative Industries    319 Boden, M.A. (1990). The Creative Mind:  Myths and Mechanisms (London:  Weidenfeld and Nicholson). Brito Henriques, E. and Thiel, J. (2000). ‘The cultural economy of cities: a comparative study of the audiovisional sector in Hamburg and Lisbon’. European Urban and Regional Studies 7: 253–​268. Cantwell, J.A. and Mudambi, R. (2005). ‘MNE competence-​creating subsidiary mandates’. Strategic Management Journal 26: 1109–​1128. Caves, R.E. (2000). Creative Industries:  Contracts Between Art and Commerce (Cambridge, MA: Harvard University Press). Christopherson, S. (2006). ‘Behind the scenes: how transnational firms are constructing a new international division of labour in media’. Geoforum 37: 739–​751. Christopherson, S., Garretsen, H., and Martin, R. (2008). ‘The world is not flat: putting globalization in its place’. Cambridge Journal of Regions, Economy and Society 1: 343–​349. Chistopherson, S. and Storper, M. (1986). ‘The city as studio, the world as backlot: the impact of vertical disintegration on the location of the motion picture industry’. Environment and Planning D: Society and Space 4: 305–​320. Coe, N.M. (2001). ‘A hybrid agglomeration? The development of a satellite Marshallian industrial district in Vancouver’s film industry’. Urban Studies 38: 1753–​1775. Coe, N.M (2015). ‘Global Production:  Networks in the Creative Industries’ in C. Jones, M. Lorenzen, and J. Sapsed (eds) The Oxford Handbook of the Creative Industries, pp. 486–​501 (Oxford: Oxford University Press). Coe, N., Dicken, P., and Hess, M. (2008). ‘Global production networks: realizing the potential’. Journal of Economic Geography 8: 271–​295. Coe, N.M. and Johns, J. (2004). ‘Beyond Production Clusters:  Towards a Critical Political Economy of Networks in the Film and Television Industries’ in D. Power and A. Scott (eds) The Cultural Industries and the Production of Culture, pp. 188–​204 (London: Routledge). Cooke, P. and Lazzeretti, R. (eds) (2008). Creative Cities, Cultural Industry Clusters, and Local Economic Development (Cheltenham: Edward Elgar). Cunningham, S. and Potts, J. (2015). ‘Creative Industries and the Wider Economy’ in C. Jones, M. Lorenzen, and J. Sapsed (eds) The Oxford Handbook of the Creative Industries, pp. 387–​ 404 (Oxford: Oxford University Press). Currah, A. (2007). ‘Hollywood, the Internet and the world: a geography of disruptive innovation’. Industry and Innovation 14: 359–​383. Currid-​Halkett, E. and Ravid, G. (2012). ‘Stars and the connectivity of cultural industry world cities: an empirical social network analysis of human capital mobility and its implications for economic development’. Environment and Planning A 44: 2646–​2663. Daly, H.E. (1999). ‘Globalization versus internationalization:  some implications’. Ecological Economics 31: 31–​37. DCMS (2007). Staying Ahead:  The Economic Performance of the UK’s Creative Industries (London: The Work Foundation and UK Department for Culture, Media and Sport). Dicken, P., Kelly, P.F., Olds, K., and Yeung, H.W. (2001). ‘Chains and networks, territories and scales: towards a relational framework for analyzing the global economy’. Global Networks 1: 89–​112. Donahue, S.M. (1987). American Film Distribution: The Changing Marketplace (Ann Arbor, MI: University of Michigan Press). Elberse, A. and Eliashberg, J. (2003). ‘Demand and supply dynamics for sequentially released products in international markets: the case of motion pictures’. Marketing Science 22: 329–​354.

320   Lorenzen European Commission (2001). ‘Employment trends and sectors of growth in the cultural economy: final report’ (Brussels: European Commission DG Employment and Social Affairs). Federation of Indian Chambers of Commerce and Industry/​ KPMG (2015). ‘Shooting for the stars:  FICCI-​ KPMG Indian media and entertainment industry report 2015’ (Mumbai: Federation of Indian Chambers of Commerce and Industry). Flew, T. (2013). Global Creative Industries (Cambridge: Polity Press). Florida, R., Mellander, C., and Stolarick, K. (2008). ‘Inside the black box of regional development:  human capital, the creative class and tolerance’. Journal of Economic Geography 8: 615–​649. Gertler, M.S. (2003). ‘Tacit knowledge and the economic geography of context, or the undefinable tacitness of being (there)’. Journal of Economic Geography 3: 75–​99. Giddens, A. (2000). Runaway World:  How Globalisation is Reshaping Our Lives (New York: Routledge). Ginsburgh, V.A. and Throsby, D. (eds) (2006). Handbook of the Economics of Art and Culture (Amsterdam: Elsevier). Glaeser, E.L., Kolko, J., and Saiz, A. (2001). ‘Consumer city’. Journal of Economic Geography 1: 27–​50. Gordon, I. and McCann, P. (2005). ‘Innovation, agglomeration, and regional development’. Journal of Economic Geography 5: 523–​543. Grabher, G. (2002). ‘The project ecology of advertising:  tasks, talents and teams’ Regional Studies 36: 245–​262. Hartley, J. (ed.) (2004). Creative Industries (Oxford: Wiley-​Blackwell). Hesmondhalgh, D. (2013). The Cultural Industries (3rd edition) (London: SAGE). Hirsch, P.M. (2000). ‘Cultural industries revisited’. Organization Science 11: 356–​361. Hirsch, P.M. and Gruber, D.A. (2015). ‘Digitizing Fads and fashions: disintermediation and glocalized markets in creative industries’ in C. Jones, M. Lorenzen, and J. Sapsed (eds) The Oxford Handbook of the Creative Industries, pp. 421–​435 (Oxford: Oxford University Press). Horkheimer, M. and Adorno. T. (1944). Dialectic of Enlightenment (2002 translation by Edmund Jephcott) (Stanford, CA: Stanford University Press). Hoskins, C. and Mirus, R. (1988). ‘Reasons for U.S. dominance in the international trade in television programmes’. Media, Culture and Society 70: 499–​515. Hoskins, C., McFadyen, S., and Finn, A. (1997). Global Television and Film: An Introduction to the Economics of the Business (Oxford: Clarendon). Hudson, R. (2005). Economic Geographies: Circuits, Flows and Spaces (London: SAGE). Jacobs, J. (1961). The Death and Life of Great American Cities (New York: Random House). Jennings, J. (2011). Globalizations and the Ancient World (Cambridge: Cambridge University Press). Johns, J. (2010). ‘Manchester’s film and television industry: project ecologies and network hierarchies’. Urban Studies 47: 1059–​1077. Jones, C., Lorenzen, M., and Sapsed, J. (eds) (2015). The Oxford Handbook of the Creative Industries (Oxford: Oxford University Press). Kaiser, R. and Liecke, M. (2007). ‘The Munich feature film cluster: its degree of global integration and explanations for its relative success’. Industry and Innovation 14: 385–​399. Khaire, M. (2015). ‘Entrepreneurship in Creative Industries and Cultural Change: Art, Fashion and Modernity in India’ in C. Jones, M. Lorenzen, and J. Sapsed (eds) The Oxford Handbook of the Creative Industries, pp. 200–​218 (Oxford: Oxford University Press). Langdale, J.V. (1997). ‘East Asian broadcasting industries: global, regional, and national perspectives’. Economic Geography 73: 305–​321.

The Geography of the Creative Industries    321 Lazzeretti, L. (ed.) (2013). Creative Industries and Innovation in Europe: Concepts, Measures and Comparative Case Studies (Abingdon: Routledge). Lee, S.-​W. and Waterman, D. (2007). ‘Theatrical feature film trade in the United States, Europe, and Japan since the 1950s: an empirical study of the home market effect’. Journal of Media Economics 20: 167–​188. Lorenzen, M. (2007). ‘Internationalization vs. globalization of the film industry.’ Industry and Innovation 14: 349–​357. Lorenzen, M. (2009). ‘Creativity in Context:  Content, Cost, Chance, and Collection in the Organization of the Film Industry’ in A. Pratt and P. Jeffcut (eds) Creativity and Innovation in the Cultural Economy, pp. 93–​118 (London: Routledge). Lorenzen, M. and Frederiksen, L. (2005). ‘The management of projects and product experimentation: examples from the music industry’. European Management Review 2: 198–​211. Lorenzen, M. and Frederiksen, L. (2008). ‘Why do Cultural Industries Cluster? Localization, Urbanization, Products and Projects’ in Cooke, P. and Lazzeretti, R. (eds) Creative Cities, Cultural Industry Clusters, and Local Economic Development, pp. 155–​179 (Cheltenham: Edward Elgar). Lorenzen, M. and Mudambi, R. (2013). ‘Clusters, Connectivity and Catch-​up: Bollywood and Bangalore in the Global Economy’. Journal of Economic Geography 13: 501–​534. Lorenzen, M. and Mudambi, R. (2015). ‘Clusters and Global Innovation:  The Role of Connectedness and Connectivity’ in D. Archibugi and A. Filipetti (eds) Handbook of Global Science, Technology, and Innovation, pp. 212–​227 (New York: Wiley-​Blackwell). Lorenzen, M. and Taäbe, F.A. (2008). ‘Breakout from Bollywood? The roles of social networks and regulation in the evolution of Indian film industry’. Journal of International Management 14: 286–​299. Lorenzen, M., Scott, A.J., and Vang, J. (2008). ‘Geography and the cultural economy’. Journal of Economic Geography 8: 589–​592. McCann, P. (2007). ‘Globalization and economic geography:  the world is curved, not flat’. Cambridge Journal of Regions, Economy and Society 1: 351–​370. Marshall, A. (1890). Principles of Economics (London: Macmillan). Maskell, P. (2001). ‘Towards a knowledge-​based theory of the geographical cluster’. Industrial and Corporate Change 10: 921–​943. Maskell, P. and Lorenzen, M. (2004). ‘The cluster as market organization’. Urban Studies 41: 975–​993. Maskell, P. and Malmberg, A. (1999). ‘Localised learning and industrial competitiveness’. Cambridge Journal of Economics 23: 167–​186. Morawetz, N.J., Hardy, C., Haslam, C., and Randle, K. (2007). ‘Finance, policy and industrial dynamics: the rise of co-​productions in the film industry’. Industry and Innovation 14: 421–​443. Morgan, K. (2004). ‘The exaggerated death of geography: learning, proximity and territorial innovation systems’. Journal of Economic Geography 4: 3–​21. Mossig, I. (2008). ‘Global networks of the motion picture industry in Los Angeles/​Hollywood using the example of their connections to the German market’. European Planning Studies 16: 43–​59. Mudambi, R. (2008). ‘Location, control and innovation in knowledge-​intensive industries’. Journal of Economic Geography 8: 699–​725. OECD (2006). ‘International measurement of the economic and social importance of culture’ (Paris: Organisation for Economic Co-​operation and Development, Statistics Directorate). Oh, J. (2001). ‘International trade in film and the self-​sufficiency ratio’. Journal of Media Economics 14: 31–​44.

322   Lorenzen Papandrea, F. (1998). ‘Protection of domestic TV programming’. Journal of Media Economics 11: 3–​15. Power, D. and Scott, A.J. (eds) (2004). Cultural Production and Production of Culture (London: Routledge). Pratt, A.C. (2002). ‘Hot jobs in cool places: the material cultures of new media product spaces; the case of the South of the Market, San Francisco’. Information, Communication and Society 5: 27–​50. Pratt, A.C. (2015). ‘Creative Industries and Development:  Culture in Development, or the Cultures of Development?’ in C. Jones, M. Lorenzen, and J. Sapsed (eds) The Oxford Handbook of the Creative Industries, pp. 502–​512 (Oxford: Oxford University Press). Pratt, A.C. and Jeffcutt, P. (eds) (2009). Creativity, Innovation in the Cultural Economy (London: Routledge). Saxenian, A. (2006). The New Argonauts:  Regional Advantage in the Global Economy (Cambridge, MA: Harvard University Press). Scott, A.J. (2000). The Cultural Economy of Cities: Essays on the Geography of Image-​Producing Industries (London: SAGE). Scott, A.J. (2005). On Hollywood: The Place, the Industry (Princeton, NJ: Princeton University Press). Scott, A.J. (2008). Social Economy of the Metropolis:  Cognitive–​Cultural Capitalism and the Global Resurgence of Cities (Oxford: Oxford University Press). Sternberg, R.J. (ed.) (1999). Handbook of Creativity (Cambridge: Cambridge University Press). Storper, M. (1989). ‘The transition to flexible specialization in the US film industry: external economies, the division of labour, and the crossing of industrial divides’. Cambridge Journal of Economics 3: 273–​305. Storper, M. and Christopherson, S. (1987). ‘Flexible specialization and regional industrial agglomerations: the case of the US motion-​picture industry’. Annals of the Association of American Geographers 77: 260–​282. Storper, M. and Venables, A.J. (2004). ‘Buzz:  face-​to-​face contact and the urban economy’. Journal of Economic Geography 4: 351–​370. Sutton, J. (1991). Sunk Cost and Market Structure:  Price Competition, Advertising, and the Evolution of Concentration (Cambridge, MA: MIT Press). Tomlinson, J.B. (1999). Globalisation and Culture (Chicago, IL: University of Chicago Press). Towse, R. (ed.) (2003). A Handbook of Cultural Economics (Cheltenham: Edward Elgar). Turok, I. (2003). ‘Cities, clusters and creative industries:  the case of film and television in Scotland’. European Planning Studies 11: 549–​565. UNCTAD (2008). ‘Creative economy report 2008’ (Geneva: United Nations Conference on Trade and Development). UNESCO (2006). Understanding Creative Industries: Cultural Statistics for Public Policy-​making (Montreal: UNESCO). Vang, J. and Chaminade, C. (2007). ‘Cultural clusters, global-​local linkages and spillovers: theoretical and empirical insights from an exploratory study of Toronto’s film cluster’. Industry and Innovation 14: 401–​419. Vogel, H.L. (1998). Entertainment Industry Economics: A Guide for Financial Analysis (4th edition) (New York: Cambridge University Press). Vorley, T., Mould, O., and Lawton Smith, H. (2008). ‘Introduction to geographical economies of creativity, enterprise and the creative industries’. Geografiska Annaler, Series B, Human Geography 90: 101–​106.

The Geography of the Creative Industries    323 Wasko, J. (2003). How Hollywood Works (London: SAGE). Whitley, R. (2006). ‘Project-​based firms: new organizational form or variations on a theme?’ Industrial and Corporate Change 15: 77–​99. Wildman, S.S (1995). ‘Trade liberalization and policy for media industries: a theoretical examination of media flows’. Canadian Journal of Communication 20: 367–​388. Zademach, H.-​M. (2009). ‘Global finance and the development of regional clusters: tracing paths in Munich’s film & TV industry’. Journal of Economic Geography 9: 697–​722.

Chapter 17

Firms in C ont e xt : Internal and E xt e rna l Drivers of Su c c e s s Mercedes Delgado Defining The Boundaries of Clusters Locational attributes are at the core of formulating and implementing firm and regional strategies. In the Competitive Advantage of Nations, Porter (1990) developed the microeconomic business environment framework to help firms, regions, and countries assess the sources of locational advantages and disadvantages, and decide how to compete. This influential framework has four interrelated components, graphically depicted in a diamond: factor conditions (the costs, quantity, and quality of a variety of inputs); the context for firm strategy and rivalry (incentives to invest and compete); demand conditions (the sophistication of local customers); and related and supporting industries (the presence of suppliers and complementary industries). In this diamond framework, clusters play an important role on firm and regional competitiveness. Clusters are geographical concentrations of related industries and firms that are connected through various types of linkages (skill, technology, knowledge, supply, demand, and others) and supporting institutions (training, trade promotion, educational, or financial). These clusters emerge in the conditions that a specific microeconomic business environment provides in a region. Within clusters, firms compete but also cooperate (Porter, 1998). The agglomeration of related economic activity in a small number of locations is a central feature of economic geography (Marshall, 1920; Porter, 1990, 1998; Krugman, 1991; Feldman, 1994, 1999; Ellison and Glaeser, 1997). A number of distinct drivers of agglomerations have been identified in the literature. The early work by Marshall (1920) highlighted three drivers associated with improved firm performance: supplier–​buyer linkages; labour market pooling; and knowledge spillovers. Over time, an extensive literature broadened the set of agglomeration drivers. In Porter’s (1990, 1998) cluster framework many agglomeration forces can be at work: not only Marshallian agglomerations, but also local demand

Firms in Context    325 conditions and specialized institutions. Other studies have focused on the organizational structure of businesses (e.g. the presence of specialized firms) and socio-​economic networks of firms and individuals as important agglomeration drivers (Saxenian, 1994; Storper, 1995; Markusen, 1996; Sorenson and Audia, 2000; Becattini et al., 2003; Dahl and Sorenson, 2012; among others). Clusters are everywhere. In the USA some common examples of clusters are information technology in Silicon Valley, biopharmaceuticals in Boston, and financial services in New York city, but there are many others. Each region has some comparative advantages manifested in its clusters. The types of strategies, organizational practices, operations, and, ultimately, performance of firms are shaped by the attributes of the locations and clusters they participate in. But, how best to define and measure the boundaries of a cluster? In order to implement cluster research and offer implications for managers and policymakers, we need an operational definition of clusters that measures their boundaries (i.e. the set of related economic activities that constitutes a cluster). Because there are many ways to group related firms and industries, we need to assess and score groupings. That is, are they the best groups possible? Do they capture the types of linkages we are interested in? In Defining Clusters of Related Industries (Delgado et al., 2016), we tackle some of these questions. Over the last twenty years, two main approaches to defining clusters have developed: definitions based on inter-​industry linkages inferred from multiple regions (referred to as benchmark or comparable cluster definitions); and definitions based on observed linkages in a single region (referred to as region-​specific cluster definitions).1 There are advantages and disadvantages to each approach.

Benchmark Cluster Definitions A set of benchmark cluster definitions allocates individual industries (or technology classes) into specific clusters based on the inter-​industry linkages that are inferred from multiple regions. By defining clusters as a fixed set of related industries, we can compare different regions in terms of that cluster definition. Some studies use national-​level data to capture particular inter-​industry linkages, including knowledge links based on patenting (see e.g. Glaeser and Kerr, 2009); input and output links (see e.g. Feser, 2005); labour occupation links (see e.g. Glaeser and Kerr, 2009; Neffke and Henning, 2013); and product similarity as defined by the industry classification system (Frenken et al., 2007). Other studies define measures based on the co-​location patterns of industries across many regions to capture various types of linkages (Ellison and Glaeser, 1997; Porter, 2003). A few studies then use these measures of inter-​industry relatedness to define benchmark clusters. Some studies define input–​output clusters based on supplier–​buyer links between industries (Feser, 2005). Others define innovation clusters by measuring knowledge linkages among industries or technology classes. These studies focus on manufacturing industries, and group them based on patent citation patterns (see e.g. Koo 2005), or based on their science and technological base (Feldman and Audretsch, 1999). Finally, in groundbreaking work, Porter (2003) defined clusters based on the co-​location patterns of traded industries across regions, following the principle that co-​location reveals the presence of many linkages

326   Delgado between industries. Using a comprehensive set of service and manufacturing industries, the methodology first distinguishes traded versus local industries. Traded industries are geographically concentrated and produce goods and services that are sold across regions and countries (e.g. surgical instruments). Local industries sell primarily in the local market and are geographically dispersed (e.g. retail). Because local industries do not cluster, they are excluded. To measure the relatedness between a pair of traded industries, Porter (2003) computes the pairwise correlation of industry employment across regions and then uses an iterative approach to define clusters. Using particular benchmark cluster definitions, a growing body of research has shown that the presence of related economic activity matters for regional and industry performance, including job creation, innovation, and entrepreneurship (see, among others, Feldman and Audretsch, 1999; Porter, 2003; Feser et al., 2008; Glaeser and Kerr, 2009; Delgado et al., 2010, 2014; Neffke et al., 2011).2 These findings are based on different definitions of clusters (which often use the subset of manufacturing industries), and without a methodology to generate and compare alternative sets of cluster definitions it is difficult to reconcile them. In ‘Defining clusters of related industries’ (Delgado et al., 2016), we address this issue by developing a novel clustering algorithm to generate and assess alternative sets of cluster definitions—​groups of industries closely related by skill, technology, supply, demand, and/​ or other linkages. Our approach to defining clusters accounts for multiple types of inter-​ industry linkages, and allows us to compare alternative sets of cluster definitions by computing scores for each set. To derive clusters of traded industries, we use cluster analysis: numerical methods to classify similar objects (cities, firms, genes, industries, technology classes, people, etc.) into groups (Everitt et al., 2011). Cluster analysis measures how each object (e.g. an industry) is related to any other object, and then creates groups (termed ‘clusters’) so that objects in the same group are more similar to each other than to objects in other groups (clusters). We generated many cluster configurations (i.e. groupings of industries) based on alternative clustering functions, inter-​industry relatedness measures, and parameter choices. Our method develops a score function to choose a candidate configuration that will capture the broadest range of inter-​industry linkages. The methodology concludes with an expert assessment and adjustment of individual clusters. The result is the US benchmark cluster definitions (BCD), which groups all traded industries covering services and manufacturing into fifty-one clusters based on the strength of input–​output links, shared labour occupations, and co-​location patterns of industries across regions. These clusters and their linkages are illustrated in Figure 17.1.3 The benchmark cluster definitions can be mapped into any administrative region (e.g. metropolitan statistical areas, economic areas, and states) and into spatial units based on density of businesses.4 Region-​industry data from the County Business Patterns and ZIP Code Business Patterns datasets are coded with the BCD to map the specialization in clusters for all regions in the USA. Using these and many other data sources, the U.S. Cluster Mapping Project (USCMP) has created a detailed regional cluster dataset and interactive tool that facilitates comparisons across regions and across clusters on numerous dimensions.5

BCR ≥ 95 percentile & RI ≥ 20% Jewelry & Precious Metals

BCR 90 -94 percentile

& RI ≥ 20%

Agriculture

Next closest clusters

Livestock Processing

Video Production

Performing Arts

Music & Sound Recording

Hospitality & Tourism

Marketing Services

Biopharma

Upstream Metals

Production

Plastics

Downstream Chemicals Upstream Chemicals

Metalworking

Automotive

Trailers & Appliances

Printing Services

Food Processing

Leather

Oil & Gas Apparel

Textiles

Footwear

Electricity

Coal Mining

Metal Mining

Figure 17.1   Portfolio of Fifty-​one Traded Clusters and their Connections. Note: Clusters with solid connection lines are highly related: the between cluster relatedness (BCR) score is above the ninetieth percentile value and the percentage of related industries (RI) is above 20%. Source: Delgado et al. (2016) and the U.S. Cluster Mapping Project.

Firms in Context    327

Financial Services Paper & Packaging

Tobacco

Downstream Metals

Distribution & eComm.

Business Services Environmental Services

Construction

Transport & Logistics

Insurance

Forestry

Medical Devices

IT & Analytical Instruments

Education & Knowledge Creation

Wood Products

Furniture Recreation & Electric Goods

Water Transport

Communications Equip. & Services

Vulcanized Materials

Lighting

Aerospace Fishing & Fishing Products

Nonmetal Mining

328   Delgado For example, consider the Information Technology (IT) and Analytical Instruments cluster, which contains twenty-​seven narrowly defined industries (six-​digit North American Industry Classification System code). Using this definition, we can evaluate how different regions compare in terms of their size and specialization in that cluster. Figure 17.2 shows the US regions (economic areas) with high employment specialization in IT and Analytical Instruments. While it is common knowledge that the San Jose–​San Francisco, CA Economic Area (which includes Silicon Valley) is highly specialized in IT, cluster mapping allows us to identify other regions that also are specialized in the cluster (e.g. Austin–​Round Rock, TX; Boston, MA; Seattle, WA; and Portland, OR). Our clustering method can be applied to other countries using their specific data. The BCD are a good starting point for many countries that may not have the appropriate data (i.e. defining clusters is best undertaken using data from large and diverse economies with integrated regions). Mapping the BCD in multiple countries has the additional advantage of allowing comparison of clusters across countries and can inform firm global location and investment decisions.6 What constitutes a good set of benchmark cluster definitions ultimately depends on the particular research and policy question. Clusters based on a single linkage (e.g. labour occupations) could inform policies in support of a particular link. Clusters based on multiple linkages (like the BCD) could facilitate externalities of various types. Multi-​linkage clusters could be most meaningful for policies supporting multiple complementarities across industries. For example, promoting the skills needed by the suppliers in the cluster and thus facilitating supplier–​buyer and knowledge linkages in the cluster and related clusters. They could also be more informative for firm location strategies, which are not based on a single link but rather on a complementary set of locational attributes (skills, suppliers, buyers, supporting institutions, etc.).

Region-​specific Cluster Definitions Benchmark cluster definitions can capture most economic activities and are necessary for studies that aim to examine clusters across regions. However, a limitation of the multi-​region cluster approach is that it overlooks specific inter-​industry linkages that may exist in particular regions. The region-​specific cluster approach instead focuses on a single region to measure industry and firm interdependencies and define the region’s clusters. Studies vary greatly in their industry coverage, type of economic unit (industry, technology class, or firm), regional unit, and method. The majority of studies focus on particular clusters and rely on existing organizations, industry directories, and other primary data collection and bottom-​ up approaches to identify clusters (e.g. Allen and Potiowsky’s (2008) work on Portland’s Green Building cluster). They can offer rich detail on the firms and institutions within particular defined clusters, but are less appropriate for comparing clusters across regions. One conceptual limitation to using region-​specific approaches to defining clusters is that they are based on observed linkages in a single region and therefore can be myopic (Bathelt et al., 2004; Maskell and Malmberg, 2007). Activities that are absent or weak in the region (e.g. industries and technology classes) are classified as unrelated to the other activities in the region. Thus, region-​specific cluster definitions may exclude complementary industries that would be useful to the cluster. Furthermore, some region-​specific clusters can be ‘opportunistic’ and ad hoc without reflecting the comparative advantages of the location.

High Employment Specialization and Share High Employment Specialization High Employment Share

Notes: Economic areas (EAs) with a red star are clusters with dual specialization in terms of employment and patenting (Delgado, 2016). An interactive map is available at the U.S. Cluster Mapping Project. EAs with high employment specialization in a cluster meet these criteria: location quotient (LQ) of cluster employment must be greater than the seventy-​fifth percentile (measured across all EAs with non-​zero employment in the cluster). Secondary criteria to differentiate marginal cases: LQ > 1 and share of national cluster employment and establishments greater than the twenty-​fifth percentile. EAs with high employment share in a cluster meet this criterion: share of national cluster employment must be greater than the ninetieth percentile. EAs with high employment specialization and share meet all of the above criteria.

Firms in Context    329

Figure 17.2   Strong IT and Analytical Instruments Clusters in the U.S., 2011.

330   Delgado

Complementarities between Benchmark and Region-​specific Cluster Definitions Benchmark and region-​specific clusters can, however, be complementary. Using cluster analysis, we could develop new tools for assessing and scoring region-​specific clusters against benchmark clusters. We could, for example, assess how different the region-​specific cluster is from the benchmark cluster, and why. Such differences may exist because the cluster is defined based on a particular technology versus a broader set of linkages, or because it has emerged in the intersection of various clusters. Understanding these differences is an important area for future research. Similarly, we should be assessing cluster organizations and other institutions for collaboration. The European Cluster Observatory lists thousands of cluster organizations—​that is, collective efforts by firms, public entities, and other institutions to improve the competitiveness of specific regional clusters (Sölvell et al., 2003). One recent example is the proliferation of water-​related cluster initiatives. They use different terminologies and ‘water cluster’ definitions: the ‘water economy’, the ‘blue economy’, the ‘ocean economy’, ‘water security’, ‘water technology’, and so on. The lack of proper methods for assessing cluster organizations can result in terminological and policy confusion and to less effective inter-​ firm collaborations. A cluster organization will be successful only if it is embedded in an actual regional cluster (Ketels and Protsiv, 2013). Thus, it is important to define and map clusters properly in order to be able to shape them. Future research could examine what types of governance, network structures, and incentives (e.g. competitions and prizes) in cluster organizations might be more effective in helping member firms to identify innovation and collaboration opportunities (Boudreau et al., 2011; Murray et al., 2012; Ketels and Protsiv, 2013).

Clusters and Regional Performance Cluster theory suggests that agglomeration economies arise across related industries and firms (Marshall, 1920; Porter, 1990, 1998). The bulk of the cluster literature uses detailed case studies for particular regional clusters to examine this theory (see e.g. Marshall, 1920; Porter, 1990, 1998; Swann, 1992; Saxenian, 1994; Bresnahan and Gambardella, 2004). In Delgado et al. (2014), we developed a novel empirical approach to evaluate cluster theory more systematically using a comprehensive set of narrowly defined industries that are mapped into clusters for all regions. If cluster agglomerations matter, then regional industries located within strong clusters will grow faster. This work moves beyond the traditional dichotomy of agglomeration forces:  localization (increasing returns to activities within a single industry) and urbanization (increasing returns to diversity at the overall regional level). Instead, we examine agglomeration economies that arise among related industries co-​located in a region. Using Porter’s (2003) benchmark cluster definitions, we find that agglomeration economies exist: industries participating in strong clusters (i.e. those with high relative employment presence in related industries) grew faster in terms of employment, entrepreneurship,

Firms in Context    331 and innovation (Delgado et al., 2010, 2012, 2014). Importantly, these positive effects are not driven by random groupings of industries (unrelated diversity), but rather by the presence of a set of co-​located and related industries.7 These cluster-​driven agglomerations matter for the growth of existing industries and for the creation of new industries, and they arise in different channels: across related industries within a cluster; across related clusters (e.g. between financial services and insurance services); and across the same cluster in neighbouring regions. These agglomeration benefits are not confined to the subset of manufacturing industries or high-​technology industries, which are the focus of many relevant studies (see e.g. Feldman and Audretsch, 1999; Glaeser and Kerr, 2009). In more recent work, I use the newly available set of cluster definitions (BCD) to evaluate the role of clusters in two additional dimensions of performance: resilience to economic shocks and inclusive prosperity in cities. Most policy prescriptions designed to mitigate shocks stress the need for diversification across industries to avoid that shocks propagate across related firms. However, economies of agglomeration can arise in strong clusters, and they could be at work during a recession and alleviate the effects of shocks. Whether clusters are vulnerable or resilient to shocks is an important empirical question (Delgado and Porter, 2015). We find that industries participating in strong clusters with other related (upstream and downstream) industries registered more resilient employment growth during the Great Recession (2007–​09); and business-​ to-​business linkages within clusters were a relevant channel of resilience. The economic resilience of firms and regions thus remains a fertile area for future research. In particular, understanding resilience in innovation and entrepreneurship; the types of inter-​firm networks that matter more for mitigating the effects of shocks (e.g. suppliers with many buyers versus suppliers with a long-​term interaction with a single buyer); and the long-​term evolution of clusters and regions (Boschma, 2015). Finally, there is increasing concern about the uneven distribution of economic growth and wealth within cities. More than 10 per cent of the US population lives in inner cities (i.e. parts of the city with concentrated poverty and high unemployment rates). To examine the role of clusters in inner city development, Delgado and Zeuli (2016) use unique data sets from the Initiative for a Competitive Inner City and the USCMP. We map clusters into the inner city and the surrounding city (e.g. the relative employment presence of the performing arts cluster inside and outside the Detroit inner city). We find that agglomeration economies do occur in inner cities and improve job creation. These externalities arise both within an inner city cluster and across the same cluster in the inner city and its nearby city. Thus, inclusive prosperity could be facilitated with policies to connect inner city businesses into city clusters. Unfortunately, how to best connect inner cities (and rural areas) to city clusters remains poorly understood.

What Attributes of Clusters Matter More for Performance? Prior studies have found that various types of economies of agglomeration may arise within clusters; these include input–​output linkages, access to demand, labour occupation linkages, and knowledge linkages (Marshall, 1920; Porter, 1990, 1998; Glaeser and Kerr, 2009; Ellison et al., 2010; Delgado et al., 2014; among others). Which types of

332   Delgado agglomerations matter more, and the specific mechanisms involved, are important topics for research. The answer to these questions might depend on the performance dimension under consideration (job growth, innovation, entrepreneurship, resilience to shocks, inclusive prosperity). Two locational attributes that have been the focus of an active debate are discussed next: the co-​location of innovation and production capacity and the presence of suppliers.

Co-​location of Innovation and Production Capacity in Clusters Researchers, policymakers, and business leaders are increasingly concerned that the geographical separation of innovation and manufacturing might limit the subsequent capacity of a location to innovate, and thus increase the vulnerability of firms to shocks (e.g. import competition and economic crises). This concern relates to the active debate in the economic and strategy literature on the role of co-​location of innovation and production on the performance of regions and their firms (see e.g. Dertouzos et al., 1989; Pisano and Shih, 2012; Ketokivi and Ali-​Yrkkö, 2009; Porter and Rivkin, 2012; Berger, 2013; Delgado et al., 2014; Delgado, 2016). Prior studies have found that manufacturing and (especially) innovation are each geographically concentrated (Feldman, 1994; Audretsch and Feldman, 1996; Audretsch, 1998; Alcacer, 2006). However, the geographically bounded complementarities between these two activities are not well understood. Theory predicts benefits from co-​locating innovation and production in industries where the manufacturing process is not standardized (Vernon, 1966; Duranton and Puga, 2001). A few industry studies yield findings consistent with this theory (e.g. Pisano (1997) and Alcacer and Delgado (2016) for biopharmaceuticals, and Fuchs and Kirchain (2010) for optoelectronics). In a study that uses a comprehensive set of industries (Delgado et al., 2014), we find that industries participating in clusters with dual strength in innovation (patenting) and production (employment) grew faster in terms of innovation. This suggests that co-​location of innovation and production matters for subsequent innovation and facilitates a broad set of linkages, including input–​output linkages and knowledge spillovers. Thus, innovation in a region is better understood in the context of co-​located industries and firms that are connected by many types of linkages (Marshall, 1920; Porter, 1998; Delgado et al., 2016). This finding is in contrast to studies that focus on knowledge (patenting) linkages as the only driver of innovation. The importance of co-​locating innovation and production may vary across clusters and industries depending on many factors, such as their maturity (Duranton and Puga, 2001), the modularity of the value chain (Baldwin and von Hippel, 2011), and the existence of technologies and managerial practices that facilitate the separation of R & D, design, and manufacturing (Fuchs and Kirchain, 2010; Fort, 2011; Tripathy and Eppinger, 2013). In a first step towards understanding these differences, in recent work I examine what US clusters have greater dual specialization in innovation (patenting) and production (employment), and how these co-​location patterns have changed over time (Delgado, 2016).8 Using the USCMP database, I compute the correlation between regional cluster strength in employment and patenting across regions for many cluster categories to build a new measure of the extent of co-​location of production and innovation in clusters (referred to as

Firms in Context    333 Dual Specialization Correlation (DSC)). The US clusters with high DSC include Information Technology and Analytical Instruments; Communications Equipment and Services; Aerospace Vehicles and Defense; Automotive; Medical Devices; Oil and Gas Production and Transportation; and Marketing, Design, and Publishing. While meaningful co-​location of innovation and production in regional clusters exist, and it holds for many cluster categories, I find important changes over time in co-​location patterns for specific clusters. One of the clusters with a large decline in dual specialization in the 2000s is Information Technology and Analytical Instruments. This cluster category is the top one by number of patents in the US (more than 30% of all patents), and had one of the largest DSC scores in 2011. The large DSC of this cluster is illustrated in Figure 17.2. For example, the top-​five regional clusters by employment specialization also have high patenting specialization (i.e. the Boise City–​Nampa, ID; Seattle–​Tacoma–​ Olympia, WA; San Jose–​San Francisco–​Oakland, CA; Burlington–​South Burlington, VT; and Austin–​Round Rock, TX economic areas). Although the DSC is high, it has declined greatly relative to 1998, in part, because of the increase in the number of clusters with production strength but weak patenting (e.g. Madison–​Baraboo, WI, or Dayton–​ Springfield–​Greenville, OH). If strong clusters in terms of production fail to improve their innovation strength in the future, they may experience low growth. Similarly, IT and Analytical Instruments clusters with single innovation strength (e.g. New  York–​ Newark–​Bridgeport, NY–​NJ–​CT–​PA) may find their ability to grow limited (Pisano and Shih, 2012; Delgado et al., 2014). We need to understand why the co-​location of innovation and production can weaken over time, and what the implications are for firm and regional performance. An optimistic answer is that some production and innovation linkages become less geographically bounded and can be implemented successfully across distant locations. A worrisome answer is that regional clusters are losing key suppliers of high-​tech inputs, reducing their production and innovation capacity. I discuss the crucial role of suppliers in clusters in the next section.

Business-​to-​Business Interactions in Clusters Suppliers can be important for agglomeration economies to arise in a location, and, in particular, for fostering innovation and entrepreneurship (Chinitz, 1961; Pisano, 1997; Porter, 1998; Helper et al., 2000; Glaeser and Kerr, 2009; Agrawal et al., 2014). When suppliers are clustered together and near their buyers, they can create agglomeration benefits through shared pools of skills, technologies, knowledge, and specialized inputs. But, who are the suppliers? Most prior work defines suppliers as manufacturers of intermediate goods. Instead, Delgado and Mills (2016) offer a more appropriate definition of suppliers that includes producers of both intermediate goods (e.g. microprocessors) and services (e.g. software). We introduce a new categorization that separates supply-​chain industries (i.e. those that sell their goods and services primarily to other businesses or governments) from business-​to-​consumer (B2C) industries (i.e. those that sell primarily to personal consumers). For example, engineering services and biological product manufacturing are supply-​chain industries. Most of the firms operating in these industries will be suppliers (i.e. sell mainly to other businesses or to the government).9

334   Delgado We find that supply chain industries compose a surprisingly large segment of the economy. Even more importantly, there are many suppliers of traded services (e.g. accounting, finance, marketing, design, logistics, and R & D services): four times as many traded supplier firms in services than in manufacturing. Thus, studies of the role of suppliers in regional and firm performance should take into account suppliers of both goods and services. There are significant differences between supply-​chain industries and B2C industries in terms of wages, labour pools, and innovative activity. Supply-​chain industries pay higher wages, have a larger relative presence of Science, Technology, Engineering, and Mathematics (STEM) occupations (especially suppliers of traded services), and account for most of the patenting activity in the USA. They also benefit more than B2C industries from being located in a strong cluster (Delgado and Porter, 2015). This further suggests that proximate business-​to-​business interactions are a particularly important mechanism for agglomeration benefits.

Which Firms Benefit More Within a Cluster? To answer this question properly, we need to take into account the interaction between firm and regional attributes. The mere existence of cluster agglomerations does not mean that all firms within the cluster will benefit. In fact, unless firms differentiate themselves in their competitive strategies, they will be more likely to fail in clusters (Porter, 1990, 1998; Sorenson and Audia, 2000). Clusters could facilitate the adoption of operational effectiveness (i.e. dissemination of best practices in the industries) setting the productivity bar higher and making the need for strategic positioning even more important (Porter, 1996, 1998). Some firms choose to locate and collaborate in clusters (see e.g. Saxenian, 1994; Porter, 1998; Helper et al., 2000), but others may avoid clusters to prevent their knowledge to spill over to competitors (see e.g. Shaver and Flyer, 2000; Alcacer and Chung, 2007). Few papers have examined the role of external agglomerations on the performance of a firm and its establishments systematically (e.g. Jaffe et al., 1993; Feldman, 1999; Henderson, 2003; Moretti, 2004; Bloom et al., 2012). Many papers consider a firm as a single unit and identify the location of its headquarters as the relevant region. This assumption can be very limiting as many firms have multiple domestic (and global) locations and their spatial organization is linked to their strategy and performance. A small but growing literature in economics, international business, and strategy considers how the benefits of agglomeration depend on attributes of the firm, as well as on attributes of their locations (Saxenian, 1994; Feldman, 1999; Mariani, 2002; Henderson, 2003; Moretti, 2004; Rosenthal and Strange, 2010; Alcacer and Chung, 2007, 2014; Alcacer and Delgado, 2016; Lessard et al., 2016, among others). There can be key interactions between the organization of the firm (start-​up, multi-​location, multinational, small or large, specialized versus vertically integrated, corporate organization, etc.) and its ability to extract agglomeration benefits from a location. One dimension of this work that has received more attention is the role of small firms in extracting and generating economies of agglomeration. For example, Henderson (2003) finds that the extent of localization economies is larger for single-​plant firms than for corporate plants, because the former are more specialized and dependent on the external environment.

Firms in Context    335

The Spatial Organization of Firms: Internal and External Agglomerations The location decision and performance of firms will depend on their portfolio of establishments and locations (see e.g. Dunne et al., 1988; Dunning, 1998; Alcacer and Delgado, 2016); therefore, multi-​unit firms’ location strategies may differ from those of start-​ups. Clusters matter for the creation of start-​ups because they generate externalities that reduce barriers to new business creation (e.g. among others, Glaeser and Kerr, 2009; Delgado et al., 2010). The attributes of the cluster (e.g. the presence of suppliers, buyers, skills, and supporting institutions) where a start-​up is born can play a particularly important role in shaping its strategy and survival. As firms grow and expand into new establishments, their location choices may be driven not only by external agglomerations in clusters, but also by internal agglomerations (intra-​ firm linkages that are facilitated by proximity and improve firm performance). Firms may choose to diversify geographically in multiple clusters to benefit from the competitive resources of each location (Enright, 2000; Bresnahan and Gambardella, 2004; Delgado et al., 2010). Or, firms may choose to co-​locate new establishments with existing ones to facilitate intra-​firm value-​chain linkages. Examples of firm location decisions consistent with internal agglomerations include Samsung’s co-​location of R & D and manufacturing in its Giheung Site in South Korea, Facebook’s new open-​floor building in Menlo Park, California, which holds thousands of people, and electric carmaker Tesla Motors’ plan to build a ‘gigafactory’ in Nevada to produce all of its batteries. As discussed earlier, a growing literature shows that geographical proximity between related firms generates external agglomerations that improve performance, thus creating incentives for co-​location. However, less is known about the spatial organization of firms and the role of same-​firm geographical proximity as a driver of location choices and performance. The strategy literature emphasizes that links between activities in the value chain are important for developing competitive advantage (Porter, 1996)  and for innovating (Cohen and Levinthal, 1990). To the extent that these interdependencies are weakened by distance, firms are likely to consider their existing geographical footprint as they decide where to grow. In Alcacer and Delgado (2016), we offer a novel framework that compares the role of internal and external agglomerations in firm domestic location choices across different activities in the value chain (R & D, manufacturing, and sales). External agglomerations are centrifugal forces that drive firms to disperse activities geographically in search of the best clusters; and internal agglomerations are centripetal forces that drive within-​firm co-​ location. These two forces can complement or oppose each other depending on firms’ prior locations. They may work in the same direction when a firm is already located in a strong cluster (e.g. Facebook in Silicon Valley), creating potential complementarities between internal and external agglomerations (Cohen and Levinthal, 1990). Yet, these forces also can present firms with a trade-​off: stay in a weak cluster to preserve intra-​firm links, or move some activities to a strong cluster. We discuss five sources of internal agglomerations—​geographically bounded, intra-​ firm linkages that positively impact performance. Three of them are similar to those identified by Marshall (1920) for external agglomeration economies but taking place within

336   Delgado firms: improved access to internal knowledge spillovers (Cohen and Levinthal, 1990; Adams and Jaffe, 1996; Azoulay, 2004); specialized labour (e.g. R & D personnel that can be shared across nearby research projects); and specialized inputs that could be accessed more efficiently if the upstream and downstream facilities of a firm are co-​located. Internal agglomerations also arise because geographical proximity enhances firms’ ability to control (Giroud, 2013; Kalnins and Lafontaine, 2013) and coordinate (Chandler, 1977; Henderson and Ono, 2008) activities across the value chain. These internal agglomerations may occur within activity and across-​activities. For example, within-​firm knowledge flows can be facilitated by same-​firm R & D co-​location (Van den Bulte and Moenaert, 1998; Chacar and Lieberman, 2003) and by R & D and manufacturing co-​location (Adams and Jaffe, 1996; Pisano, 1997; Mariani, 2002). Thus, to examine the role of internal and external agglomerations, we take into account the spatial organization of all activities in the value-​chain of firms. We find that both internal and external agglomerations positively affect the US location choices of new establishments opened by biopharmaceutical firms. Internal agglomerations vary by activity in the value chain. They are larger for R & D and manufacturing than for sales. Same-​firm co-​location occurs both within an activity (e.g. among plants) and across activities (e.g. between R & D and manufacturing). Internal agglomerations also vary by type of firm expansion: they matter less for the location of acquisitions than for new establishments. Overall, the review of the literature suggests that both internal and external agglomerations can influence the location choices and performance of firms. Thus, firm performance models that fail to incorporate both intra-​firm and inter-​firm linkages in a location can produce biased results.

Future Research to Integrate Economic Geography and Strategic Management Porter (2000) strongly suggested that ‘geography and location must become one of the core disciplines in management’. To date, the interaction between firm performance and the attributes of locations has not been properly examined. Such work might help firms to find the best strategic positioning in their locations in order to improve their comparative advantage. This merger of economic geography and management represents a broad research opportunity.

Improved Characterization of Locations and their Clusters To help firms and regions decide how to compete and grow, one must properly characterize locations and their clusters. While much progress has been made in defining and mapping clusters, there are still important areas for improvement. Firstly, to better assess why clusters matter, future work should examine the composition of firms and institutions for collaboration within clusters (the presence of suppliers, start-​ups, multi-​location firms, cluster organizations, etc.) and the nature of firm linkages (e.g. supplier–​buyer links; start-​up–​incumbent

Firms in Context    337 links; firm–​university links; ownership links; labour flows; co-​patenting and co-​publishing links). Firms’ and individuals’ socio-​economic networks also can play an important role in encouraging agglomeration economies to arise in a location (Saxenian, 1994; Storper, 1995; Feldman et al., 2005; Dahl and Sorenson, 2012; Bell et al., 2016). But these socio-​economic links are difficult to observe, so very few studies account for them. Secondly, we need to take into account the presence in a region (and its nearby regions) of related clusters that could foster between-​cluster agglomerations (e.g. dual strength in Information Technology and Medical Devices clusters). That would help us to better understand the evolution of regions and their firms. Thirdly, we need measures that capture innovation performance for all economic activities in a region, including those with low patent intensity (e.g. many services). The percentage of a firm’s sales owing to new product introductions captures innovativeness (e.g. Feldman and Audretsch, 1999; Arora et al., 2014), but this indicator is not broadly available across firms, locations, and over time. Recent studies have used attributes of firms at birth (e.g. name, filing for a trademark, or place of incorporation) to predict the real-​time innovation potential of start-​ups (Guzman and Stern, 2015). Another approach is to use innovation inputs to proxy for the innovation potential of regions and firms. In particular, the presence of STEM occupations in an industry can be used to evaluate its technology intensity (Hecker, 2005). Regions specialized in clusters with higher STEM content will likely be better able to innovate. More broadly, the science of innovation, which includes the development of analytical tools for assessing the innovation potential of a location and its firms, is an important area for future research (Feldman et al., 2012; Guzman and Stern, 2015).

The Interaction Between Location and Firm Strategy and Performance A firm’s context includes geographically bounded intra-​ firm and inter-​ firm linkages (internal and external agglomerations). Both types of linkages can play a role in firms’ location choices and performance (Alcacer and Delgado, 2016). The location decisions of firm activities are interrelated: for example, the location of firm R & D may be influenced by the presence of same-​firm R & D, manufacturing, sales, and support activities. In order to help firms choose where to grow, we need comprehensive theoretical and empirical frameworks that take into account the relationships between activities in the value chain, the relevance of internal and external agglomerations, and the fact that agglomerations may vary by the type of firm expansion (growth within existing establishments, opening new establishments, and acquisitions). Research that isolates internal and external agglomeration mechanisms is another desirable direction for future work. Shocks to firms and their locations (e.g. economic crisis, natural disasters, or re-​allocations of anchor firms) and policy changes might provide an opportunity for researchers to identify these mechanisms. The interaction between location and firm performance also depends on industry attributes. The maturity of an industry can influence firms’ ability to exploit agglomerations (e.g. Duranton and Puga, 2001; Dumais et al., 2002). Industries also can differ in their degree of

338   Delgado modularity (Baldwin and von Hippel, 2011), and affect the extent and types of agglomerations. Firms in industries with higher modularity (e.g. automotive) can more easily break the value chain across locations. Future work should examine these industry dynamics. The life cycle of firms may influence location choices too. Firm locational needs can be different at birth than in later growth stages (e.g. when firms diversify into new products). The place of birth of start-​ups is not random. Strong clusters are important for the birth and growth of startups (Delgado et al., 2010). As firms get larger, the extent of internal and external agglomerations may vary. Research that explores the evolution of the spatial organization of firms since birth would improve our understanding of firm performance. In particular, we could examine how the strategy and performance of start-​ups relate to location attributes. A start-​up born in a cluster with innovation strength but lacking production strength may need to define a different strategy than a start-​up born in a cluster with dual strength (Gans et al., 2016). The successful strategy would take into account the potential trade-​offs between breaking the value chain in different locations for R & D and production in order to exploit external agglomerations versus co-​locating same-​firm value-​chain activities to exploit internal agglomerations (Alcacer and Delgado, 2016). The definition of the location of birth of a start-​up is also an open research question. Some studies identify the location where the start-​up has the first paid employee (e.g. Delgado et al., 2010). Other studies focus on earlier stages and consider the place of registration of the start-​up (e.g. Guzman and Stern, 2015). Recent work shows that the birthplace of entrepreneurs and inventors also can influence their subsequent choices in the technology and geographical space (Dahl and Sorenson, 2012; Bell et al., 2016). Understanding the life cycle of start-​ups and inventors could inform policies for fostering innovation and entrepreneurship. Finally, firms’ spatial organization will depend on management practices that facilitate intra-​firm and inter-​firm interactions, even with geographical separation. These include outsourcing practices, labour mobility practices, monitoring and control practices, and information and communications technology investments (see e.g. Baldwin and von Hippel, 2011; Fort, 2011; Tripathy and Eppinger, 2013; Lessard et al., 2016). The role of location on performance can change as firms expand across regions and globally. Multi-​location firms are more likely to have the managerial capabilities and communication technology to break their value chain across locations; thus, they may benefit more from participating in multiple clusters. As the value chains of firms and their clusters become global, the consequences for the performance of particular clusters and their firms are unknown, and will depend on the complementarities or substitution among the intra-​regional, inter-​regional, and transnational linkages (Chacar and Lieberman, 2003; Bathelt et al., 2004; Ketels and Memedovic, 2008; Beugelsdijk et al., 2010). Future work should examine the interaction among the spatial organization of firms (within and across nations), management practices, and firm performance.

Acknowledgements Special thanks to Maryann Feldman, and to Gordon Clark, Meric Gertler, and Dariusz Wójcik, for the invitation to contribute to the New Oxford Handbook of Economic Geography and for their great feedback. I am grateful for insightful comments by Juan Alcacer,

Firms in Context    339 Rich Bryden, Christian Ketels, Myriam Mariani, Scott Stern, Michael Porter, Jane Wu, Samantha Zyontz, and participants in the The 4th Global Conference on Economic Geography, School of Geography and the Environment and the Smith School of Enterprise and the Environment, at the University of Oxford.

Notes 1. Many cluster definitions have been generated by researchers and practitioners based on both approaches and using alternative economic units (industry, technology class, or firms). See Cortright (2006) and Delgado et al. (2016) for a review. 2. See Rosenthal and Strange (2004), Cortright (2006), and Carlino and Kerr (2015) for a review of economies of agglomeration studies. 3. A detailed explanation (and Stata codes) to derive the BCD is offered in Delgado et al. (2016). 4. Some studies identify the geographic boundaries of a given cluster by examining the spatial density of businesses for particular industries (Duranton and Overman, 2005) or the density of patents for particular technology classes (e.g. Kerr and Kominers, 2010; Alcacer and Zhao, 2013). The goal is to identify locations with a high density of economic activity in a particular field that will facilitate inter-​firm connections and externalities. 5. The USCMP is an initiative supported by the Economic Development Administration of the U.S. Department of Commerce. The BCD, underlying data, and additional resources can be accessed at the USCMP portal (http://​clustermapping.us/​). 6. The BCD have been mapped into European regions by the European Cluster Observatory (http://​ec.europa.eu/​growth/​smes/​cluster/​observatory/​) and into the regions in Canada and Mexico. 7. Cluster strength is conceptually similar to the notion of ‘related variety’ introduced by Frenken et al. (2007) as strong clusters often include an array of related industries versus specialization in a single narrowly defined industry. An important empirical difference is that Frenken et al. (2007) classify industries in manufacturing and services as unrelated, but clusters can include both types of industries (Porter, 2003; Delgado et al., 2016). 8. As the measure of innovativeness used is patenting activity, the analysis is focused on thirty-​five US cluster categories with meaningful number of patents. 9. Using the 2002 U.S. Benchmark Input–​Output Account of the Bureau of Economic Analysis, they identify as supply-​chain industries those that sell less than one-​third of their output to personal consumption expenditures.

References Adams, J. and Jaffe, A. (1996). ‘Bounding the effects of R&D: an investigation using matched establishment-​firm data’. RAND Journal of Economics 27: 700–​721. Alcacer, J. (2006). ‘Location choices across the value chain: how activity and capability influence collocation’. Management Science 52: 1457–​1471. Alcacer, J. and Chung, W. (2007). ‘Location strategies for knowledge spillovers’. Management Science 53: 760–​776. Alcacer, J. and Chung, W. (2014). ‘Location strategies for agglomeration economies.’ Strategic Management Journal 35: 1749–​1761.

340   Delgado Alcacer, J. and Delgado, M. (2016). ‘Spatial organization of firms and location choices through the value chain.’ Management Science 62: 3213–​3243. Alcacer, J. and Zhao, M. (2016). ‘Zooming in: a practical manual for identifying geographic clusters’. Strategic Management Journal 37: 10–​21. Agrawal A., Cockburn, I., Galasso, A., and Oettl, A. (2014). ‘Why are some regions more innovative than others? The role of small firms in the presence of large labs.’ Journal of Urban Economics 81: 149–​165. Allen, J. and Potiowsky, T. (2008). ‘Portland’s green building cluster:  economic trends and impacts’. Economic Development Quarterly 22: 303–​315. Arora A., Cohen, W., and Walsh, J. (2014). ‘The acquisition and commercialization of invention in American manufacturing: incidence and impact’. NBER Working Paper 20264. Audretsch, D.B. (1998). ‘Agglomeration and the location of innovative activity.’ Oxford Review of Economic Policy 14: 18–​29. Audretsch, D.B. and Feldman, M.P. (1996). ‘R&D spillovers and the geography of innovation and production’. American Economic Review 86: 253–​273. Azoulay, P. (2004). ‘Capturing knowledge within and across firm boundaries: evidence from clinical development’. American Economic Review 94: 1591–​1612. Baldwin, C. and von Hippel, E. (2011). ‘Modeling a paradigm shift: from producer innovation to user and open collaborative innovation’. Organization Science 22: 1399–​1417. Bathelt, H., Malmberg, A., and Maskell, P. (2004). ‘Clusters and knowledge: local buzz, global pipelines, and the process of knowledge creation’. Progress in Human Geography 28: 31–​56. Becattini, G., Bellandi, M., Ottati, G.D., and Sforzi, F. (2003). From Industrial Districts to Local Development (Cheltenham: Edward Elgar). Bell, A.M., Chetty, R., Jaravel, X.L., Petkova, N., and Van Reenen, J. (2016). ‘The lifecycle of inventors’. Centre for Economic Performance, LSE Working Paper. Berger, S. (ed.) (2013). Making in America: From Innovation to Market (Cambridge, MA: MIT Press). Beugelsdijk, S., McCann, P., and Mudambi, R. (2010). ‘Place, space, and organization: economic geography and the multinational enterprise’. Journal of Economic Geography 10: 485–​493. Bloom, N., Schankerman, M., and Van Reenen, J. (2012). ‘Identifying technology spillovers and product market rivalry’. Econometrica 81: 1347–​1393. Boschma, R. (2015). ‘Towards an evolutionary perspective on regional resilience’. Regional Studies 49: 733–​751. Boudreau, K., Lacetera, N., and Lakhani, K. (2011). ‘Incentives and problem uncertainty in innovation contests: an empirical analysis’. Management Science 57: 843–​886. Bresnahan, T. and Gambardella, A. (eds) (2004). Building High-​Tech Clusters. Silicon Valley and Beyond (New York: Cambridge University Press). Carlino, G. and Kerr, W. (2015). ‘Agglomeration and Innovation’ in G. Duranton, J.V. Henderson, and W.C. Strange (eds.) Handbook of Regional and Urban Economics, Vol. 5, pp. 349–​404 (Oxford: Elsevier). Chacar, A.S. and Lieberman, M. (2003). ‘Organizing for technological innovation in the U.S. pharmaceutical industry’. Geography and Strategy—​Advances in Strategic Management 20: 299–​322. Chandler, A. (1977). The Visible Hand (Cambridge, MA: Belknap Press). Chinitz, B. (1961). ‘Contrasts in agglomeration: New York and Pittsburgh’. American Economic Review 51: 279–​289. Cohen, W. and Levinthal, D. (1990). ‘Absorptive capacity: a new perspective on learning and innovation’. Administrative Science Quarterly 35: 128–​152.

Firms in Context    341 Cortright, J. (2006). ‘Making sense of clusters:  regional competitiveness and economic development’. The Brookings Institution Metropolitan Policy Program https://​www. brookings.edu/​ research/​ m aking-​ s ense-​ o f-​ c lusters- ​ regional- ​ c ompetitiveness- ​ a nd-​ economic-​development/​ (last accessed 21 March 2017). Dahl, M.S. and Sorenson, O. (2012). ‘Home sweet home: entrepreneurs’ location choices and the performance of their ventures’. Management Science 58: 1059–​1071. Delgado, M. (2016). ‘The colocation of innovation and production in clusters’. DRUID 2016 Working Paper. Delgado, M. and Mills, K. (2016). ‘A new categorization of the U.S. economy: the role of supply chain industries in innovation and economic performance’. MIT Working Paper. Delgado, M. and Porter, M.E. (2015). ‘Clusters and the Great Recession’. DRUID 2015 Working Paper. Delgado, M. and Zeuli, K. (2016). ‘Clusters and regional performance: implications for inner cities’. Economic Development Quarterly 30: 117–​136. Delgado, M., Porter, M.E., and Stern, S. (2010). ‘Clusters and entrepreneurship’. Journal of Economic Geography 10: 495–​518. Delgado, M., Porter, M.E., and Stern, S. (2012). ‘Clusters, convergence, and economic performance’. NBER Working Paper 18250. Delgado, M., Porter, M.E., and Stern, S. (2014). ‘Clusters, convergence, and economic performance’. Research Policy 43: 1785–​1799. Delgado, M., Porter, M.E., and Stern, S. (2016). ‘Defining clusters of related industries’. Journal of Economic Geography 16: 1–​38. Dertouzos, M.L., Lester, R.K., Solow, R.M., and the MIT Productivity Commission (1989). Made in America: Regaining the Productive Edge (Cambridge, MA: MIT Press). Dumais, G., Ellison, G., and Glaeser, E.L. (2002). ‘Geographic concentration as a dynamic process’. Review of Economics and Statistics 84: 193–​204. Dunne, T., Roberts, M., and Samuelson, L. (1988). ‘Patterns of firm entry and exit in U.S. manufacturing industries’. RAND Journal of Economics 19: 495–​515. Dunning, J.H. (1998). ‘Location and the multinational enterprise: a neglected factor?’ Journal of International Business Studies 29: 45–​66. Duranton, G. and Overman, H.G. (2005). ‘Testing for localization using micro-​geographic data’. Review of Economic Studies 72: 1077–​1106. Duranton, G. and Puga, D. (2001). ‘Nursery cities: urban diversity, process innovation, and the life cycle of products’. American Economic Review 91: 1454–​1477. Ellison, G. and Glaeser, E. (1997). ‘Geographic concentration in U.S. manufacturing industries: a dartboard approach’. Journal of Political Economy 105: 889–​927. Ellison, G., Glaeser, E., and Kerr, W. (2010). ‘What causes industry agglomeration? Evidence from coagglomeration patterns’. The American Economic Review 100: 1195–​1213. Enright, M. (2000). ‘Regional clusters and multinational enterprises:  independence, dependence or interdependence?’ International Studies of Management and Organization 30: 114–​138. Everitt, B.S., Landau, S., Leese, M., and Stahl, D. (2011). Cluster Analysis (5th edition) (Chichester: John Wiley). Feldman, M.P. (1994). The Geography of Innovation (Boston, MA:  Kluwer Academic Publishers). Feldman, M.P. (1999). ‘The new economics of innovation, spillovers and agglomeration:  a review of empirical studies’. Economics of Innovation and New Technology 8: 5–​25.

342   Delgado Feldman, M.P. and Audretsch, D. (1999). ‘Innovation in cities: science-​based diversity, specialization, and localized competition’. European Economic Review 43: 409–​429. Feldman, M.P., Francis, J., and Bercovitz, J. (2005). ‘Creating a cluster while building a firm: entrepreneurs and the formation of industrial clusters’. Regional Studies 39: 129–​141. Feldman, M.P., Reed, A.G., Lanahan, L., McLaurin, G., Nelson, K., and Reamer, A. (2012). ‘Innovative Data Sources for Economic Analysis’ https://​maryannfeldman.web.unc.edu/​ files/​2012/​01/​Innovative-​Data-​Sources-​for-​Regional-​Economic-​Analysis.pdf (last accessed 21 March 2017). Feser, E.J. (2005). Benchmark Value Chain Industry Clusters for Applied Regional Research (Champaign, IL:  Regional Economics Applications Laboratory, University of Illinois at Urbana-​Champaign). Feser, E.J., Renski, H., and Goldstein, H. (2008). ‘Clusters and economic development outcomes’. Economic Development Quarterly 22: 324–​344. Frenken, K., van Oort, F.G., and Verburg, T. (2007). ‘Related variety, unrelated variety, and regional economic growth’. Regional Studies 41: 685–​697. Fort, T. (2011). ‘Breaking up is hard to do: why firms fragment production across locations’. US Census Bureau Center for Economic Studies Paper No. CES-​WP-​13–​35. Fuchs, E. and Kirchain, R. (2010). ‘Design for location? The impact of manufacturing offshore on technology competitiveness in the optoelectronics industry’. Management Science 56: 2323–​2349. Gans, J., Stern, S., and Wu, J. (2016). ‘The foundations of entrepreneurial strategy’. MIT Working Paper. Giroud, X. (2013). ‘Proximity and investment:  evidence from plant-​level data’. Quarterly Journal of Economics 128: 861–​915. Glaeser, E.L. and Kerr, W.R. (2009). ‘Local industrial conditions and entrepreneurship: how much of the spatial distribution can we explain?’ Journal of Economics and Management Strategy 18: 623–​663. Guzman, J. and Stern, S. (2015). ‘Where is Silicon Valley’. Science 347: 606–​609. Hecker, D. (2005). ‘High-​technology employments: a NAICS-​based updated’. Monthly Labor Review 128: 57–​72. Helper, S., MacDuffie, J.P., and Sabel, C.F. (2000). ‘Pragmatic collaborations: advancing knowledge while controlling opportunism’. Industrial and Corporate Change 9: 443–​483. Henderson, J.V. (2003). ‘Marshall’s scale economies’. Journal of Urban Economics 53: 1–​28. Henderson, J.V. and Ono, Y. (2008). ‘Where do manufacturing firms locate their headquarters?’ Journal of Urban Economics 63: 431–​450. Jaffe, A., Trajtenberg, M., and Henderson, R. (1993). ‘Geographic localization of knowledge spillovers as evidenced by patent citations’. Quarterly Journal of Economics 108: 577–​598. Kalnins, A. and Lafontaine, F. (2013). ‘Too far away? The effect of distance to headquarters on business establishment performance’. American Economic Journal: Microeconomics 5: 157–​179. Ketels, C. and Memedovic, O. (2008). ‘From clusters to cluster-​based economic development’. International Journal of Technological Learning, Innovation and Development 1: 375–​391. Ketels, C. and Protsiv, S. (2013). ‘Clusters and the New Growth Path for Europe’. WWW for Europe Working Paper No. 14. Ketokivi, M. and Ali-​Yrkkö, J. (2009). ‘Unbundling R&D and manufacturing: postindustrial myth or economic reality?’ Review of Policy Research 26: 35–​54. Kerr, W. and Kominers, S. (2010). ‘Agglomerative forces and cluster shapes’. NBER Working Paper 16639.

Firms in Context    343 Koo, J. (2005). ‘Knowledge-​based industry clusters: evidenced by geographical patterns of patents in manufacturing’. Urban Studies 42: 1487–​1505. Krugman, P. (1991). ‘Increasing returns and economic geography’. Journal of Political Economy 99: 483–​499. Lessard, D., Teece, D., and Leih, S. (2016). ‘The dynamic capabilities of meta-​multinationals’. Global Strategy Journal 6: 211–​224. Mariani, M. (2002). ‘Next to production or to technological clusters? The economics and management of R&D location.’ Journal of Management and Governance 6: 131–​152. Markusen, A. (1996). ‘Sticky places in slippery space:  a typology of industrial districts’. Economic Geography 72: 293–​313. Marshall, A. (1920). Principles of Economics; An Introductory Volume (London: Macmillan). Maskell, P. and Malmberg, A. (2007). ‘Myopia, knowledge development, and cluster evolution’. Journal of Economic Geography 7: 603–​618. Moretti, E. (2004). ‘Workers’ education, spillovers, and productivity: evidence from plant-​level production functions’. American Economic Review 94: 656–​690. Murray, F., Stern, S., Campbell, G., and MacCormack, A. (2012). ‘Grand innovation prizes: a theoretical, normative, and empirical evaluation’. Research Policy 41: 1779–​1792. Neffke, F. and Henning, M. (2013). ‘Skill-​ relatedness and firm diversification’. Strategic Management Journal 34: 297–​316. Neffke, F., Henning, M., and Boschma, R. (2011). ‘How do regions diversify over time? Industry relatedness and the development of new growth paths in regions’. Economic Geography 87: 237–​265. Pisano, G.P. (1997). The Development Factory:  Unlocking the Potential of Process Innovation (Boston, MA: Harvard Business School Press). Pisano, G.P. and Shih, W.C. (2012). Producing Prosperity: Why America Needs a Manufacturing Renaissance (Cambridge, MA: Harvard Business Review Press). Porter, M.E. (1990). The Competitive Advantage of Nations (New York: Free Press). Porter, M.E. (1996). ‘What is Strategy?’ Harvard Business Review November: 61–​78. Porter, M.E. (1998). ‘Clusters and Competition: New Agendas for Companies, Governments, and Institutions’ in M.E. Porter (ed.) On Competition, pp. 197–​299 (Boston, MA: Harvard Business School Press). Porter, M.E. (2000). ‘Location, Clusters, and Company Strategy’ in G.L. Clark, M. Feldman, and M. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 253–​ 274 (Oxford: Oxford University Press). Porter, M.E. (2003). ‘The economic performance of regions’. Regional Studies 37: 549–​578. Porter, M.E. and Rivkin, J.W. (2012). ‘Choosing the United States’. Harvard Business Review 90: 80–​91. Rosenthal, S.S. and Strange, W.C. (2004). ‘Evidence on the Nature and Sources of Agglomeration Economies’ in J.V. Henderson and J.F. Thisse (eds) Handbook of Regional and Urban Economics, Vol. 4, pp. 2119–​2143 (Amsterdam: Elsevier North). Rosenthal, S.S. and Strange, W.C. (2010). ‘Small establishments/​big effects:  agglomeration, industrial organization, and entrepreneurship’ in E. Glaeser (ed.) Agglomeration Economics, pp. 277–​302 (Cambridge, MA: National Bureau of Economic Research). Saxenian, A. (1994). Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA: Harvard University Press). Shaver, J.M. and Flyer, F. (2000). ‘Agglomeration economies, firm heterogeneity, and foreign direct investment in the United States’. Strategic Management Journal 21: 1175–​1193.

344   Delgado Sölvell, Ö., Lindqvist, G., and Ketels, C. (2003). The Cluster Initiative Greenbook (Stockholm: Ivory Tower AB). Sorenson, O. and Audia, P.G. (2000). ‘The social structure of entrepreneurial activity: geographic concentration of footwear production in the United States, 1940–​1989’. American Journal of Sociology 106: 424–​462. Storper, M. (1995). ‘The resurgence of regional economies, ten years later: the region as a nexus of untraded interdependencies’. European Urban and Regional Studies 2: 191–​221. Swann, P. (1992). The Dynamics of Industrial Clusters (Oxford: Oxford University Press). Tripathy, A. and Eppinger, S.D. (2013). ‘Structuring work distribution for global product development organizations’. Production and Operations Management 22: 1557–​1575. Van den Bulte, C. and Moenaert, R.K. (1998). ‘The effects of R&D team co-​location on communication patterns among R&D, marketing, and manufacturing’. Management Science 44: 1–​18. Vernon, R. (1966). ‘International investment and international trade in the product cycle’. Quarterly Journal of Economics 80: 190–​207.

Pa rt  I V

T H E  F I R M

Chapter 18

The L o g i c of Aggl omerat i on Gilles Duranton and William R. Kerr Introduction A core topic in economic geography is agglomeration economies, where cities and clusters of activity boost the productivity of firms located within them. Conceptual rationales date back to Marshall (1890), and theorists have done a remarkable job of formalizing and codifying these concepts, as reviewed by Duranton and Puga (2004) and Behrens and Robert-​ Nicoud (2015). The empirical literature has been slower to develop, however, especially as it relates to the role of firms in agglomeration forces. Most early studies on agglomeration economies focused on measuring the wage premiums paid to urban workers, which was quite natural given the ready availability of person-​level data collected through population censuses. While these data and studies could speak to the existence of agglomeration economies, they were inadequate for identifying channel(s) through which agglomeration economies percolate and how they shape behaviour. Said differently, the higher wages paid in larger cities could have descended equally from several different forms of firm and worker interactions. This was not just a loss for our academic description of the world—​ without knowing, for example, whether agglomeration forces are arising as a result of knowledge flows or better-​matching externalities, practical advice for policy and business leaders is limited. The most important advance for empirical research on agglomeration economies over the last two decades has been the development of firm-​and establishment-​level data sets of economic activity. These data have opened up the possibility of quantifying the role of firms in agglomeration economies and the productivity boost that large cities and clusters provide. Moreover, they have allowed the advent of new lenses for studying these questions: for example, continuous distance measures of geographical concentration (e.g. Duranton and Overman, 2005), estimations of productivity spillover decays within cities (e.g. Rosenthal and Strange, 2004, Arzaghi and Henderson, 2008), firm selection mechanism and productivity (e.g. Combes et al., 2012), and dynamic perspectives that include firm entry and exit (e.g. Dumais et al., 2002; Klepper, 2010; Glaeser et al., 2015). Alongside has been the

348   Duranton and Kerr development of complementary data sets for observing the actual interactions of workers and firms (e.g. employer–​employee data, patents, and citations). These data have also encouraged the measurement of the exchanges that we believe underlie the agglomeration economies—​for example the many studies using patent citations to measure local knowledge spillovers following Jaffe et al. (1993)—​and recent efforts have started to unite these with detailed firm location data. This chapter reviews two conceptual frameworks and proposes some interesting ways in which new micro-​level data can advance our understanding of the models. Our first framework, in the next section, considers the formation of cities as the equilibrium outcome of benefits to firm agglomeration in cities against the growing cost of living that larger cities endure (e.g. land scarcity, congestion, pollution). This work pulls from Duranton (2008), and we use this lens to describe useful ways that firm-​level data in rich and developing economies can advance our insights into the differences across cities in these economic forces. Our second framework, the section ‘Structures of Interactions within Cities’ is a conceptual model of interactions within a given cluster that pulls from Kerr and Kominers (2015). We describe through this lens the very nascent work on local interactions of firms and workers and how they give rise to agglomerative clusters. This is an area where we anticipate large advances will occur over the next two decades as we break open the black box of internal cluster dynamics and the relationships that define agglomeration economies. There are two large boundary conditions for this piece. Firstly, we only cover these two topics of interest out of a sea of opportunity. For example, firm-​level studies of how multi-​unit firms interact with local agglomeration economies versus internally sourced resources are woefully few in number (e.g. Tecu, 2012; Alcacer and Delgado, 2013). Likewise, a better understanding of how clusters interact with each other across countries and the role of multinational firms is critical for today’s global interconnectivity (e.g. Saxenian et al., 2002; Alfaro and Chen, 2014). Other examples include the extensive margin of employer–​employee relationship (e.g. spinouts and entrepreneurship), the implications of firm-​level market frictions like financing constraints for agglomeration, and so on. Secondly, our review does not contain many references, and space constraints require that we be equally stingy towards the classics and towards recent contributions. We focus here just on ideas related to firms and these two frameworks. Previous and contemporaneous reviews contain much more extensive documentation (e.g. Feldman, 2000; Audretsch and Feldman, 2004; Duranton and Puga, 2004, 2014; Rosenthal and Strange, 2004; Feldman and Kogler, 2010; Carlino and Kerr, 2015; Combes and Gobillon, 2015). Finally, and most importantly, while we conclude in this piece that new data and fresh thinking have opened up many exciting research opportunities regarding agglomeration with firm-​level data, it is important to emphasize that a mountain of data is never a substitute for a convincing research design. Agglomeration is a very complex process that involves trade-​offs and equilibrium outcomes, and the best empirical progress in this field comes when researchers identify a razor-​sharp way to cut through this complexity and identify causal relationships. We mention this upfront, so that we don’t have to endlessly repeat the warning throughout the review! We will make the most progress when these big opportunities are met with the right methodologies.

The Logic of Agglomeration    349

Differences in Agglomeration Across Economies We begin with a framework that depicts the formation of cities as an outcome of productivity benefits, cost factors, and labour supply/​migration decisions. Duranton (2008, 2014) fully develops this framework with implications and examples that we do not repeat here, and earlier developments of this conceptual framework are contained in Combes et al. (2005). This framework visually describes the key elements of a literature dating back to Henderson (1974) and even earlier on the fundamental factors that influence city size and composition. Figure 18.1 shows the three curves that are central to this framework. In both parts of Figure 18.1, the horizontal axis measures city size or possible labour supply as given by a population, N. The upper part of Figure 18.1 first shows the wage/​productivity curve w(N). Models of agglomeration feature an upward-​sloping wage curve where increases in a location’s economic activity boost the productivity of firms in the city or cluster and the wages that they pay. As noted in the ‘Introduction’ section, this relationship has been empirically shown in many different settings and time periods. In the original formulation of Marshall (1890), agglomeration forces were expressed in terms of customer–​supplier interactions, labour pooling, or knowledge exchange. Duranton and Puga (2004) provide a more theoretically amenable framework that emphasizes the sharing, matching, and learning processes of cities. Many classic studies have isolated features of these agglomeration economies, such as Helsley and Strange (1990), Porter (1990), Saxenian (1994), and Audretsch and Feldman (1996). Underlying these collocation benefits is the ability of firms to ship and sell their products on larger markets. An intuitive example is Hollywood. The spatial concentration of the movie production industry allows many productivity gains—​for example better matching of actors and actresses to specific parts, the emergence of specialized law firms that support the entertainment industry, the development of schools to train employees in primary and supporting roles—​with all production studios ultimately serving and competing in a global market for the best movies and largest global audiences. The top panel also shows a cost curve, H(N). Costs are also rising with city size. This aspect of agglomeration can be forgotten at times by academics—​to the extent that they adopt a ‘more is always better’ view of clustering—​but is very evident to those living in large cities. These costs come with higher rents and transportation costs, greater congestion, and perhaps environmental degradation (e.g. air pollution). Simply put, prices on Wall Street in New York or Sand Hill Road in Silicon Valley are extraordinary. In addition, the price of consumption goods may rise or fall with city size, depending upon the outcome of higher input prices versus potentially greater competition and diversity in large cities. As reviewed in greater detail by Duranton (2014), the cost curve is significantly less studied than the wage curve. Subtracting the cost curve from the wage curve, one determines a net benefit for w(N)–​ H(N), as shown in the lower half of Figure 18.1. This benefit is initially rising in city size as the early productivity benefits from larger city size dominate cost increases. After point B, these net benefits decline. To determine the stable size of a city, one must further specify how

350   Duranton and Kerr w (N ) Wage Curve

w* (a)

wB

N N*

HB H*

Cost of Living Curve

(b)

H (N )

w (N ) – H (N ) B

wB – HB

Labour Supply Curve

C

w * – H*

(c) Net Wage Curve

A

N 5

NB

N*

Figure 18.1   Agglomeration and City Formation. Notes: Figure represents the conceptual model of Duranton (2008) for the formation of cities. The horizontal axis measures city size as given by a population N. The upper part shows a wage/​productivity curve w(N) that is increasing a location’s economic activity as a result of agglomerative forces. The top panel also shows a cost curve H(N) that is similarly growing more acute in city size. Subtracting the cost curve from the wage curve, one determines a net benefit for w(N)–​ H(N), as shown in the lower half of the figure. This benefit is initially rising in city size as the early productivity benefits from larger city size dominate the cost increases. After point B, these net benefits decline. Through migration, labour supply increases to a city with rising net benefits. The stable equilibrium that forms will be at point C.

labour supply responds to the net benefits that a city provides (taken to be through migration to the city for simplicity). The figure presents a case where labour responds with some elasticity to higher net benefits, but that labour is not perfectly mobile (which, absent amenities, would be a horizontal line at a given net benefit level owing to the spatial equilibrium).

The Logic of Agglomeration    351 The stable equilibrium that forms will be at point C. The fact that the city is larger than what maximizes net benefits is commonly observed. There are many possible avenues for comparative analyses across cities using firm-​level data that we can illustrate with this model. (There are also many interesting questions that go beyond firm-​level studies, but we focus on these opportunities exclusively given the emphasis of this review.) As a starting point, we may investigate how the wage/​productivity curves of cities are influenced by micro-​interactions within clusters. The framework makes a reduced-​form connection of the upward-​sloping wage curve owing to productivity benefits, which is certainly observed in wage relationships. Yet we have collected very little evidence that quantifies the different agglomerative mechanisms in the background, especially in a comparative form across cities. Likewise, we have few studies that attempt to discern the quantitative roles of candidate explanations and assign importance. The explosion in firm-​level data offers great promise here. Take, for example, the rise of employer–​employee datasets that follow both workers and firms over time. In principle, one should be able to use these datasets to describe how well workers and firms are matched to each other. That is, controlling for the cost curve, does one observe improvements in employer–​employee matching that are commensurate with differences in city sizes? As a second example, many studies look at knowledge flows in local areas as an agglomerative rationale. Work in this line can turn towards tracing out how knowledge-​intensive firms accrue different benefits with larger cities that make them more productive. Alternatively, the mobility of workers of workers may also act as a vehicle for local spillovers (Serafinelli, 2014). What part of the wage curve slope comes from a greater volume of ideas, more timely access to the latest ideas, the diversity of ideas circulated, and similar features? Taken a step further, empirical work can also assess how firm production functions change with city size. Within the same industry, for example, some production models have stronger dependency on the complementary matching of team members, similar to the O-​Ring model of Kremer (1993). The theory posits that the success of a group depends deeply of the strength of its weakest member. Such models may be more likely to emerge in the largest cities, given that a deeper labour pool can allow for better choices across all of a team’s needs, and thereby provide added curvature to the wage/​productivity relationship. Anecdotally, for example, joint recruiting of complete teams on Wall Street across investment banks seems quite common, and Silicon Valley now talks about ‘acqui-​hiring’, where start-​ups are acquired mostly for their start-​up teams rather than developed technologies. In a similar spirit, multi-​unit firms place their headquarters frequently in larger cities, which could be, in part, due to this complementarity across top management team functions. Greater city size can also shape the industrial organization of the cluster, perhaps giving steepness to the wage curve. Fallick et al. (2006) describe how high-​velocity labour markets in clusters can facilitate a modular production structure. At the start, many small firms compete to identify the best idea or design, and then in the second stage the winning firm rapidly scales by hiring employees from the losing firms. This can speed innovation and create design improvements, if the labour market conditions and modular production design features are evident. Fallick et al. (2006) provide some evidence for this industrial organization in the computer industry in California, which echoes the anecdotal accounts of Saxenian

352   Duranton and Kerr (1994). More systematic assessments of how city size changes industrial organization is most warranted and feasible with firm-​level records. Greater availability of new data, including data from previously under-​represented regions, allows us to take a wider view and examine variations in clusters worldwide. Most empirical work on the subject considers advanced economies, but there are many firm-​level datasets for developing and emerging economies that are coming into widespread use. Duranton (2014) reviews the higher elasticities for the wage curve in China and India compared with advanced economies, and there is heterogeneity in outcomes among advanced countries. Further refining these cross-​country comparisons will help identify how economic development gaps emerge from an agglomeration perspective. These may be connected, for example, to the limited scaling of productive plants in India and Mexico compared with the USA that is identified by Hsieh and Klenow (2014). Moreover, developing and emerging economies may provide interesting insight into the model overall to the extent that their settings and institutions isolate some features from others. Local infrastructure, for example, is quite deficient in many poor countries and the subject of massive investments by national governments and international organizations. Many efforts are being made to upgrade these facilities, which can allow for real-​time analysis of firm behaviour as investments are being made, so long as a suitable control group can be identified. Many studies are pursuing such angles in Brazil, China, and India, and so on for highways and railroad access (Redding and Turner, 2015). As the bulk of these investments for the USA came before the collection of firm-​level datasets, the delayed start in other countries provides a unique advantage for ‘before and after’ views of the role of infrastructure for individual firm choices (e.g. input sourcing) and for overall distributions of firm outcomes. Similarly, developing economies face many challenges that are mostly absent in advanced countries. Three examples include dual-​housing markets (e.g. squatter housing without legal property rights), inefficient migration, and city favouritism, and Duranton (2008, 2014) traces out some of the conceptual implications from these three factors. To the extent that these factors isolate some features—​for example, city favouritism can be thought of as raising the wage curve of the primal city without impact on actual firm productivity—​we can learn more about the function of agglomeration and city formation generally in a specialized laboratory. At the same time, this framework has applications to related areas of economic research, including questions of interest to policymakers and business leaders. As one example, it can help to illuminate the role that migration plays in the development of clusters. The labour supply curve plays an important role in governing the degree of equilibrium net benefits in cities. Many urban models feature a spatial equilibrium with full mobility that pins down the same net benefits across all cities, described, for example, in Glaeser (2008). Deviations from this framework can provide an interesting range of labour supply responses. Constraints on internal migration in some countries (e.g. the Hukou system in China) may severely limit the ability of locations to expand, or push this expansion into non-​optimal forms like secondary labour markets. The extra-​legal nature of these migrations may place constraints on how much firms can take advantage of agglomeration economies that would otherwise emerge. Related, progress has been made in India for how organized sector firms tap into the unorganized sector that surrounds them through subcontracting relationships (e.g. Mukim, 2015).

The Logic of Agglomeration    353 At the other end of the spectrum, many immigration systems allow firms to choose locations for workers who are legally tied to the employer. In the USA, for example, most skilled immigrants hired for employment come through the H-​1B temporary visa programme or are an intra-​company transfer under the L-​1 visa programme. In both cases, employees are effectively assigned geographical locations while they are on the visa. This has the interesting implication of allowing firms to overcome some of the limits imposed by the labour-​ supply curve in this conceptual framework. Indeed, Kerr and Lincoln (2010) and Ruiz et al. (2012) describe how US immigration visas are used by firms in locations where they struggle to acquire labour resources, in addition to the better-​known cases of high-​tech clusters. Linking firm-​level data with immigrant employee traits may afford opportunities to learn about the limits locations place on firms by evaluating the workarounds that firms seek. As another example, firm-​level data sets have opened up new possibilities for the study of clusters and entrepreneurship, which is often viewed as a central driver for city growth. Recent examples of this work include Chinitz (1961), Michelacci and Silva (2007), Glaeser et al. (2010), Delgado et al. (2010), Behrens et al. (2014), and Glaeser et al. (2015), and yet, some big open areas remain for study. Firstly, there are two views of entrepreneurship supply across regions. One view is that entrepreneurs are very mobile and move across cities to opportunities. This is surely the right perspective for very high-​growth entrepreneurship like that observed in Silicon Valley. A second view is that entrepreneurs are very sticky and local in nature, such as the Chinitz (1961) depiction of differences in entrepreneurial supply across US cities due to their industrial legacies. This is most likely the right depiction for the lowest end of work, as the ‘local bias of entrepreneurship’ studies observe. Behind these two models are some very natural and intuitive economic forces like effective market size, the importance of local connections for business sales or financing, winner-​take-​all dynamics, and similar. Silicon Valley is home to a special cluster of consumer–​Internet firms given the very high agglomeration benefits in this industry, while cement manufacturing sits at the opposite end of the spectrum. What we have yet to discern is the relevant ranges of industries over which these two models operate—​ where do we tip from a local dynamic and sticky places to a world of specialized clusters? Secondly, and somewhat building on the modularity point described above, Glaeser et al. (2015) identify how industrial dynamics differ in cities and clusters with a strong entrepreneurial base. Entrepreneurship does not yield city growth through an endless replication of small firms, but instead links to local employment growth through an up-​or-​out scaling process that scales up the best new entrants to become the largest employers in a city. This line of work is very young, however, and needs much greater empirical attention with firm-​level data. This scaling relationship and the entrepreneurial supply functions depicted earlier will have first-​order implications for whether the wage/​productivity curve takes a common shape across cities or is very heterogeneous as a result of differences in industrial structures. A final important feature of this literature is the depiction of why industries move across locations and the implications for the cities that house them. Duranton and Puga (2001) describe how product and technology maturity leads industries to move out of expensive nursery cities towards cheaper locations. In other work, Duranton (2007) formalizes a model of city evolution that has industry movements across cities related to breakthrough inventions at its core, and Kerr (2010) provides some empirical evidence along the lines

354   Duranton and Kerr of this model in terms of patenting behaviour. Panel datasets on firms are starting to be of sufficient time dimension for firm-​level analyses to provide additional insights about these dynamics. Likewise, faster product cycles are allowing more rapid observation of these movements. Obviously, this section can only provide a partial list of possible topics for future research, as the abundance of new kinds of data has opened up countless ways to extend this model of city formation. The view of cities as a consequence of the competing benefits and costs that arise from clustering behaviour has existed for decades in one form or another, and we now have the chance to flesh it out further by illuminating some of the underlying mechanisms or extending the model to new settings.

Structures of Interactions Within Cities We next turn to the internal structure of agglomeration economies for cities and the implications for firms. For this exercise, we use the theoretical framework of Kerr and Kominers (2015). This model and its empirical applications have roots in the observation that most micro-​level studies of how workers and firms interact (e.g. commuting patterns, patent citations) show a much shorter geographical distance of interaction than the actual footprint of cities and clusters. Said differently, commuting patterns tend to twenty miles or less, but agglomerations for labour pooling appear to stretch much farther. While perhaps obvious, a second line of work—​such as Rosenthal and Strange (2001, 2003), Duranton and Overman (2005, 2008), and Ellison et al. (2010)—​finds that regional-​based approaches for measuring agglomeration forces yield seemingly quite reasonable and informative depictions for how agglomeration forces influence cluster size. For example, these studies tend to find that technology-​or knowledge-​based clusters appear smaller and more tightly knit than agglomerations building upon labour pooling or customer–​supplier connections, despite examining data that are orders of magnitude larger than the underlying forces believed to cause the cluster. This is true when using political boundaries (e.g. counties vs. states) or continuous distance (e.g. 100 vs. 500 miles). The Kerr and Kominers (2015) model conceptualizes how these two empirical pieces can be reconciled through a theory of small, overlapping regions of interaction visualized in Figure 18.2. There is a set of sites for businesses that is shown by the dots. Firms enter in sequence and without foresight, and the set of potential sites is fixed. Sites with black dots have already been chosen by firms, and the light dots are sites available to the next entrant. Each firm participates in the cluster by interacting with neighbours that fall within a maximal spillover radius, indicated by the dashed circles. The network to which the firm is connected by its neighbours can have arbitrary amounts of benefit transmission for the baseline model. The maximum spillover radius has a limited and defined boundary due to fixed costs of interaction that exceed a decaying benefit to interaction at some point. This model pictures large-​area clustering that arises as a result of small, contained interaction effects that overlap each other. Two important points follow. Firstly, the introduction of fixed costs and defined effective spillover boundaries yields interesting and testable

The Logic of Agglomeration    355 Unoccupied site D Maximal spillover radius

C B X A

Y Z

Marginal entrant is currently indifferent across available sites as none are within the maximal radius of interaction with the existing cluster

Figure 18.2   Internal Structure of Clusters. Note: Image illustrates the Kerr and Kominers (2015) theoretical framework for the formation of an agglomeration cluster due to small, contained interaction effects among firms that overlap each other.

predictions regarding the internal structure of clusters that we discuss later. This framework makes clear why the short-​distance interactions that we measure with commuting flows or patent citations are different from the distances that we consider with region-​based data. As we trace out, this model can start to identify many useful lines of inquiry regarding firms and agglomeration going forward. Secondly, the model provides a rationale for why multiple clusters form. In our simple illustration, the next entrant is indifferent among available sites, including sites X, Y, and Z, because none of the remaining sites is within reach of the populated cluster. In other words, clusters fill up, and this marginal entrant will choose randomly among the remaining available sites and start a new cluster. This foundation provides a basis for comparative statics of spillover radius size and cluster structure. Consider, for example, a second agglomerative force that has a longer maximal spillover radius than what is shown in Figure 18.2 because the second force has a slower rate of benefit decay (or a lower fixed cost of interaction). Such a model would increase the size of the circles in Figure 18.2 and transform site X into the strictly preferred next location because it can participate in the existing cluster with a longer radius. The full model formally traces out how a longer maximal radius results in fewer, larger, and less-​dense clusters. Thus, we can use the shapes and sizes of clusters to back-​out the micro-​forces beneath them. From this launching point, Kerr and Kominers (2015) test the broader model predictions by measuring how far apart patents in thirty-​six technologies tend to be from each other when they cite each other in their patent filings, a measure of the spillover radius for each technology. Some technologies like semiconductors use very localized networking, while others exhibit much longer spatial horizons. Using estimates from the USA and the UK, they show that technologies with shorter effective interaction distances exhibit smaller and denser clusters.

356   Duranton and Kerr Kerr and Kominers (2010) contains an extensive analysis of technology and worker flows for Silicon Valley, an iconic agglomeration that is composed of overlapping tech spaces. We select a couple of these pieces to illustrate the model in greater depth and provide some new research ideas about the internal structure of clusters. Walking through a couple of detailed maps will help us visualize the concepts that follow. Figure 18.3 describes the construction of the technology core for Silicon Valley. Looking across the entire San Francisco Bay Area, the core of Silicon Valley includes the top-​ten zip codes in terms of patent filings and eighteen of the top twenty-​five. Figure 18.3(A) shows the three most important zip codes—​for each, the zip code is indicated with the star, and the other three points on the connected shape are the three zip codes that the focal zip code cites the most in its patent filings (Hall et al., 2002). On average, these three external zips contain 41 per cent of local external citations for a zip code. Figure 18.3(B) does the same for the top-​ ten zip codes. The primary technology sourcing zones for these zip codes in the core are also fully contained in the core, even though we have made no restrictions in design and these sourcing zones could have been included anywhere in the San Francisco Bay Area. These zones are small, overlapping regions, and in the next map, we represent the core as a shaded triangle whose longest side is twenty-​five miles in length. Figure 18.4 shows the seven largest zip codes for patenting that are not contained in the core itself (#12, 13, 17, 19, 22, 24, 25). The Silicon Valley core depicted in Figure 18.3 is represented on this larger map as the shaded triangle. Similar to Figure 18.3, the shape of each technology sourcing zone is determined by the three zip codes that firms in the focal zip code cite most in their work. For visual ease, San Ramon and Santa Clara are indicated on the edge of the map at the location of their primary transportation route. Downtown San Francisco and Oakland, CA, are to the north and off the map. As observed for the core, these technology zones are characterized by small, overlapping regions. The three zip codes that are labelled with numbers are three of the four largest zip codes for patenting in the San Francisco Bay Area that are outside of the Silicon Valley core. Zone 1, which covers Menlo Park, extends deepest into the core. Zone 2, for Redwood City, CA, shifts up and encompasses Menlo Park and Palo Alto but has less of the core. Zone 3, which covers South San Francisco, further shifts out and brushes the core. None of the technology sourcing zones traverses the whole core, much less the whole cluster, and only the closest zip code (Menlo Park) even reaches far enough into the core to include the area of Silicon Valley where the greatest number of patents are issued. While technology sourcing for individual firms is localized, the resulting cluster extends over a larger expanse of land. There are many possible avenues for further research that we can illustrate with this model. One line of investigation addresses structural variations within a cluster. A comparison of Figures 18.3 and 18.4 suggests that the cluster’s structure in the Silicon Valley core may be different from what exists at the periphery. Both areas show localized interactions, but the core exhibits great overlap among these regions in Figure 18.3 that resembles the density around site C in the illustration in Figure 18.2. By contrast, the periphery in Figure 18.4 shows overlapping zones that resemble the A–​B–​C structure of sites in Figure 18.2. The baseline model is mostly agnostic about these features and to whether each location only derives benefits from the locations that it directly touches versus all members of the cluster benefiting equally. Nonetheless, stronger empirical and theoretical depictions of how this networking exists, how it is priced into wages and locations, and so on, is first order. Arzaghi and Henderson (2008) provide some characterization with advertising agencies in Manhattan,

(a)

(b)

Notes: Figure describes the construction of the technology core for Silicon Valley. The core includes the top-​ten zip codes in the San Francisco Bay Area for patenting and eighteen of the top twenty-​five. For each zip code, we present a technology sourcing zone that depicts the three zip codes that the firm in the focal zip code cites the most in their patent filings. (a) The three largest patenting zip codes and their sourcing. (b) Top-​ten zip codes. While unrestricted in design, the primary technology sourcing zones are all contained in the core. These zones are small, overlapping regions that often exhibit directional transmission.

The Logic of Agglomeration    357

Figure 18.3   Tech Sourcing in Silicon Valley Core.

358   Duranton and Kerr 3

San Ramon

2

1

Santa Clara

Figure 18.4   Tech Sourcing Around Silicon Valley. Notes: Figure continues to characterize technology flows for the San Francisco area. The Silicon Valley core depicted in Figure 18.3 is represented on this larger map as the shaded triangle. The Silicon Valley core contains eighteen of the top twenty-​five zip codes for patenting in the San Francisco area. This figure includes the seven largest zip codes for patenting that are not contained in the core itself. Similar to Figure 18.3, the shape of each technology sourcing zone is determined by the three zip codes that firms in the focal zip code cite most in their work. For visual ease, San Ramon and Santa Clara are indicated on the edge of the map at the location of their primary transportation route. As observed for the core, these technology zones are characterized by small, overlapping regions that exhibit directional transmission.

and Rosenthal and Strange (2003) comparatively estimate production function decays in several industries. Kemeny et al. (2015) further describe the central role of dealmakers and social capital in local areas. The availability of firm-​level data with detailed geographical coding provides many opportunities here. On a similar note, it may also be possible to determine a relationship between the types of interactions within clusters and the shapes of the zones that arise. Technology spillover zones are directional in nature. Our depictions of sourcing zones are unrestricted in the sense that the three most important zip codes could lie in any direction from the focal zip code. In Figures 18.3 and 18.4, however, the technology sourcing zone almost always lies within a ninety-​degree arc from the focal zip code. This pattern is very strong at peripheral locations, as there is a general flow of information or knowledge from the Silicon Valley core to Redwood City and Palo Alto, and then further to South San Francisco, and so on. Even within the tightly knit core itself, the zones are remarkably lop-​sided.

The Logic of Agglomeration    359 Kerr and Kominers (2015) also analyse the commuting patterns of scientists and engineers using IPUMS data. Commuting patterns tend to be more diffuse, and as a consequence the zones of interaction around labour inputs into firms appear significantly less directional. Visually, one is more likely to see workers for a firm commuting in equal measure from all sides, compared with technology sourcing that tends to be concentrated in one direction. These issues need to be traced out further, both in terms of how they reflect the nature of agglomeration forces (e.g. the pooling nature of labour inputs) and how they then affect the overall structure or operations of resulting clusters. The interactions between incumbents and entrants also deserve greater attention (e.g. Combes and Duranton, 2006). Expanding on this theme, there is a surprising gap in our knowledge about how skyscrapers affect the structure of interactions. Naturally, tall buildings allow greater density for a given land area, but do they do more by adding a third dimension to local structures? This is unlikely to affect patenting and technology firms, which tend not to locate in the core of urban areas, but it could be central to the functioning and organization of very high-​end financial and professional service firms like Wall Street. Observing heavy levels of agglomeration that occur within a single large building, versus across nearby skyscrapers, may signal the relative importance of different types of agglomerative forces (e.g. knowledge sharing versus labour pooling). Many cities have imposed and later removed maximum building heights, allowing models of these forms to be tested empirically with firm-​level data. Shifting focus a bit, this model can also be extended to include the impact of physical features on clusters and their constituent firms. For example, the background to Figure 18.4 demonstrates the roles of geographical features (e.g. mountains, protected land) and transportation routes (e.g. highways, bridges). These forces substantially govern how the peripheral zones access the core of cluster. While not shown for visual reasons, these same features also play an important role in the technology flows evident in the core of Silicon Valley. These connections between local infrastructure and firm-​level interactions are quite understudied and yet important for local policy and business choices. Agrawal et al. (2014) provide a recent contribution on roads and knowledge flows within cities that signal this importance. Moreover, new tools now exist for measuring travel time and associated costs that can translate spatial distances into real terms and account for congestion. For example, Google Maps reports that the expected lengths of time needed to drive across the three labeled technology zones in Figure 18.4 are comparable, which may indicate the length of the sourcing zones is determined more by interaction costs rather than true spatial distance. Similarly, the model has straightforward extensions that allow for fixed natural advantages that firms also want to access. These could be traditional natural advantages (e.g. harbours, coal mines) or ‘manmade’ advantages like universities, military bases, or state capitols. We have the capacity now to understand better how MIT’s location affects the biotech community that surrounds it in Kendall Square and, in turn, the broader Cambridge and Boston communities. We can also examine more closely the detailed ‘inner workings’ of clusters and the concerns of individual firms. An important theme throughout this work is the breaking up of a city or cluster to recognize that all firms do not automatically participate equally in benefits. At one level this is obvious, but it is remarkable how little our existing work factors in where

360   Duranton and Kerr in a cluster a firm is located. Moreover, recent studies emphasize the differences in locations within clusters for women-​owned firms and their networking benefits (e.g. Rosenthal and Strange, 2012, Ghani et al., 2013), and an extensive literature describes similar or worse segmentation on racial lines. This segmentation means excluded groups receive unequal benefits from the cluster. Chatterji et al. (2014) describe policy issues related to this and other micro-​cluster perspectives related to entrepreneurship and innovation. New firm-​level data also allow the study of entry and exit, with one major theme of recent empirical work being the high degree to which we observe entry and exit coinciding with each other in local areas. Said differently, while rapid long-​term growth may favour entry over exit in a cluster, the bigger take away from the data is that some areas are very dynamic, with lots of entry and exit, while others are less so. These features deserve greater attention. Moreover, entry and exit might allow more advanced statements about the attractiveness of places. Our simple model highlights the important information contained in the location decisions of marginal entrants, which, in principle, could be observed through panel data. Pricing theory emphasizes the valuable information contained at the margin, and some of these insights could be applied to the study of clusters if researchers fully learn how to harness firm dynamics in local areas. On the management front, scholars can use new data to study how the location choices of individual firms affect their performance. This is challenging, of course, given that location choices are intimately connected with business models and strategies, as reviewed in the MBA course material of Kerr and Brownell (2011) and Kerr et al. (2011). A starting point is to note that (i) benefits differ across firm types for each location; (ii) costs are mostly generic as landlords tend to rent property for the rental prices regardless of tenant’s industry; and (iii) managers have limited knowledge about all of these features. Thus, one should find differences in performance depending upon the quality of site chosen for a particular business. More speculative, the presence of directional flows like the technology zones in Figure 18.4 open up the possibility for asymmetry over short distances—​whether a firm locates in the path of flows from the cluster core towards an important periphery point may make a meaningful difference in the frequency and quality of interactions the firm makes, even holding fixed the distance from the core and the overall density/​prices of the area selected. Finally, while much of this discussion in this chapter focuses on clusters formed by firms within the same industry, a fruitful area for future investigation is the clustering behaviour of firms in multiple industries. Recent research emphasizes the important degree to which firms in related industries interact (e.g. Ellison et al., 2010; Faggio et al., 2014) and theoretical constraints on the degree to which these joint location decisions are well aligned (e.g. Helsley and Strange, 2014). There seems to be an unlimited scope for empirical advancement in this regard, and we note three pieces here. Firstly, Jacobs (1970) and Duranton and Puga (2001) describe how local industrial diversity can give rise to new industries, and micro-​data on firms provide deep scope for advancing our understanding of nursery cities and this cross-​fertilization process. Secondly, several studies describe the particular importance of supplier industries for entry choices following Chinitz (1961), including Helsley and Strange (2002) and Glaeser and Kerr (2009). With new data, one should be able to follow the material flows along these lines. Finally, models of entrepreneurship often depict founders as picking up a varied, ‘jack of all trades’ background (e.g. Lazear, 2005), which can be studied in local areas using employer–​employee data and movements of future founders across firms.

The Logic of Agglomeration    361

Conclusions The advent of micro-​level data is a boon to our understanding of agglomeration economies. In the short time since data like the Census Bureau’s Longitudinal Research Database first became available, researchers have fleshed out a much richer portrait of economic geography. Likewise, many European countries now have tremendous data resources for studying these issues given the depth of personal and firm-​level information available. As a consequence, firms are no longer anonymous and in the background, but are instead playing a central role in our modern depictions of the internal workings of cities and the differences over locations. The future promises to be as bright with the development of new employer–​ employee datasets, the linking of micro-​data like patents to the establishment-​level records in administrative datasets, and the known and unknown tools that big data may shortly provide to researchers. The two conceptual lenses that we developed in this review highlight the massive opportunity for empirical research that lies ahead. We won’t repeat here the tactical ideas provided within this chapter, but instead circle back to two broader themes. Firstly, agglomeration is fundamentally an equilibrium outcome where firms and workers weigh benefits against costs. Theoretical models in this regard are quite well developed, and there is a valuable tradition in this field for a heavy interplay of theory with empirics. Thus, empiricists can and should use theory to identify interesting conceptual topics that exist but have little empirical foundation. In some cases (e.g. matching externalities, infrastructure impacts), new and forthcoming data are already opening up exciting research opportunities. In these cases, our main hope is that researchers craft empirical work into frames that match theory (e.g. measuring commuting time costs vs. simple distances). We also hope that researchers make more use of powerful computing resources and estimation procedures to shed light on higher-​ order and non-​linear effects, beyond the standard linear models. This curvature in treatment effects is central for our understanding of agglomeration, as Figure 18.1 shows, but poorly measured to date. In other cases, the development of new data remains a priority and one that should be valued by the profession. To us, three issues are paramount. Firstly, we need a better understanding of how agglomeration operates in large cities that are full non-​manufacturing firms. This requires getting beyond patent data to measure knowledge spillovers, and it demands greater attention to the economics of skyscrapers. Secondly, we need to think harder about global interconnections and how this impacts clusters and firms in different countries. Agglomeration fundamentally connects into trade of outputs, and this does not stop at national borders for goods or services. In the theme of this chapter, the natural starting point here is richer work with multinational firms and the operations of their many establishments. Finally, the urbanization of the developing nations is one of the biggest issues facing the world over the next decade. We are woefully behind on understanding how agglomeration is similar or different in these places, which is a first-​order concern given that our existing insights are being used to define policies here and now! We are very hopeful that in twenty years, we will look back on many studies that developed new data for clusters outside of the developed world, harnessed real-​time variation in the emergence and growth of these economies to garner insights that advanced economies cannot reveal given their long-​ standing development, and delivered a deep and beneficial policy impact.

362   Duranton and Kerr Finally, we reiterate that a mountain of data does not equal insight. For insight to be realized, these new data sets must be paired with strong and convincing research designs. At this point, the usefulness of the clever ‘natural experiment’ to study these issues is well understood (e.g. Bleakley and Lin, 2012), and we hope that more of these empirical gems are unearthed. We also think the great research promise in many developing nations is to design interventions to facilitate future rigorous programme evaluation. Yet, the real trade-​off is not between unfounded correlations and the perfectly random shock. For important topics, we need to figure out the best ways to continually raise the empirical bar, recognizing the dual need to identify causal relationships and also acknowledge the equilibrium nature of these forces makes it very hard. We also need to learn to take better advantage of the insights that global data and research can provide. For example, we have good data for countries at many points along the spectrum of income inequality (e.g. very compressed Nordic structures to the USA and beyond). There appears to be a good opportunity to learn through these global similarities and differences by embracing more meta work across places.

References Agrawal, A., Galasso, A., and Oettl, A. (2014). ‘Roads and innovation’. CEPR Discussion Paper 10113. Alcacer, J. and Delgado, M. (2013). ‘Spatial organization of firms and location choices through the value chain’. Harvard Business School Working Paper 13–​025. Alfaro, L. and Chen., M. (2014). ‘The global agglomeration of multinational firms’. Journal of International Economics 94: 263–​276. Arzaghi, M. and Henderson, J.V. (2008). ‘Networking off Madison Avenue’. Review of Economic Studies 75: 1011–​1038. Audretsch, D. and Feldman, M. (1996). ‘R&D spillovers and the geography of innovation and production’. American Economic Review 86: 630–​640. Audretsch, D. and Feldman, M. (2004). ‘Knowledge Spillovers and the Geography of Innovation’ in Henderson, J.V. and Thisse, J-​F. (eds) Handbook of Urban and Regional Economics, Vol. 4, pp. 2713–​2739 (Amsterdam: North-​Holland). Behrens, K. and Robert-​Nicoud, F. (2015). ‘Agglomeration Theory with Heterogeneous Agents’ in Duranton, G., Henderson, J.V., and Strange, W. (eds) Handbook of Regional and Urban Economics, Vol. 5, pp. 175–​245 (Amsterdam: North-​Holland). Behrens, K., Duranton, G., and Robert-​Nicoud, F. (2014). ‘Productive cities: sorting, selection, and agglomeration’. Journal of Political Economy 122: 507–​553. Bleakley, H. and Lin, J. (2012). ‘Portage: path dependence and increasing returns in U.S. history’. Quarterly Journal of Economics 127: 587–​644. Carlino, G. and Kerr, W. (2015). ‘Agglomeration and Innovation’ in Duranton, G., Henderson, J.V., and Strange, W. (eds) Handbook of Regional and Urban Economics, Vol. 5, pp. 349–​404 (Amsterdam: North-​Holland). Chatterji, A., Glaeser, E., and Kerr, W. (2014). ‘Clusters of Entrepreneurship and Innovation’ in Lerner, J. and Stern, S. (eds) Innovation Policy and the Economy, Vol. 14, pp. 129–​166 (Chicago, IL: University of Chicago Press). Chinitz, B. (1961). ‘Contrasts in agglomeration: New York and Pittsburgh’. American Economic Review 51: 279–​289.

The Logic of Agglomeration    363 Combes, P. and Duranton, G. (2006). ‘Labour pooling, labour poaching and spatial clustering’. Regional Science and Urban Economics 36: 1–​28. Combes, P. and Gobillon, L. (2015). ‘The Empirics of Agglomeration Economies’ in Henderson, J.V., Duranton, G., and Strange, W. (eds) Handbook of Regional and Urban Economics, Vol. 5, pp. 247–​348 (Amsterdam: North-Holland). Combes, P., Duranton, G., Gobillon, L., Puga, D., and Roux, S. (2012). ‘The productivity advantages of large cities: distinguishing agglomeration from firm selection’. Econometrica 80: 2543–​2594. Combes, P., Duranton, G., and Overman, H. (2005). ‘Agglomeration and the adjustment of the spatial economy’. Papers in Regional Science 84: 311–​349. Delgado, M., Porter, M., and Stern, S. (2010). ‘Clusters and entrepreneurship’. Journal of Economic Geography 10: 495–​518. Dumais, G., Ellison, G., and Glaeser, E. (2002). ‘Geographic concentration as a dynamic process’. Review of Economics and Statistics 84: 193–​204. Duranton, G. (2007). ‘Urban evolutions: the fast, the slow, and the still’. American Economic Review 97: 197–​221. Duranton, G. (2008). ‘Viewpoint: from cities to productivity and growth in developing countries’. Canadian Journal of Economics 41: 689–​736. Duranton, G. (2014). ‘Growing through cities in developing countries’. World Bank Research Observer 30: 39–​73. Duranton, G. and Overman, H. (2005). ‘Testing for localization using micro-​geographic data’. Review of Economic Studies 72: 1077–​1106. Duranton, G. and Overman, H. (2008). ‘Exploring the detailed location patterns of U.K. manufacturing industries using micro-​geographic data’. Journal of Regional Science 48: 213–​243. Duranton, G. and Puga, D. (2001). ‘Nursery cities: urban diversity, process innovation, and the life cycle of products’. American Economic Review 91: 1454–​1477. Duranton, G. and Puga, D. (2004). ‘Micro-​foundations of Urban Agglomeration Economies’ in Henderson, J.V. and Thisse, J.-​F. (eds) Handbook of Urban and Regional Economics, Vol. 4, pp. 2063–​2117 (Amsterdam: North-​Holland). Duranton, G. and Puga, D. (2014). ‘The Growth of Cities’ in Aghion, P. and Durlauf, S. (eds) Handbook of Economic Growth, Vol. 2, pp. 751–​843 (Amsterdam: North-​Holland). Ellison, G., Glaeser, E., and Kerr, W. (2010). ‘What causes industry agglomeration? Evidence from coagglomeration patterns’. American Economic Review 100: 1195–​1213. Faggio, G., Silva, O., and Strange, W. (2014). ‘Heterogeneous agglomeration’. SERC Working Paper 0152. Fallick, B., Fleischman, C., and Rebitzer, J. (2006). ‘Job-​hopping in Silicon Valley: some evidence concerning the microfoundations of a high-​technology cluster’. Review of Economics and Statistics 88: 472–​481. Feldman, M. (2000). ‘Location and Innovation: The New Economic Geography of Innovation, Spillovers, and Agglomeration’ in Clark, G., Feldman, M., and Gertler, M. (eds) The Oxford Handbook of Economic Geography, pp. 373–​394 (Oxford: Oxford University Press). Feldman, M. and Kogler, D. (2010). ‘Stylized Facts in the Geography of Innovation’ in Hall, B., and Rosenberg, N. (eds) Handbook of the Economics of Innovation, Vol. 1, pp. 381–​410 (Oxford: Elsevier). Ghani, E., Kerr, W., and O’Connell, S. (2013). ‘Local industrial structures and female entrepreneurship in India’. Journal of Economic Geography 13: 929–​964. Glaeser, E. (2008). Cities, Agglomeration and Spatial Equilibrium (Oxford: Oxford University Press).

364   Duranton and Kerr Glaeser, E. and Kerr, W. (2009). ‘Local industrial conditions and entrepreneurship: how much of the spatial distribution can we explain?’ Journal of Economics and Management Strategy 18: 623–​663. Glaeser, E., Kerr, S., and Kerr, W. (2015). ‘Entrepreneurship and urban growth: an empirical assessment with historical mines’. Review of Economics and Statistics 97: 498–​520. Glaeser, E., Kerr, W., and Ponzetto, G. (2010). ‘Clusters of entrepreneurship’. Journal of Urban Economics 67: 150–​168. Hall, B., Jaffe, A., and Trajtenberg, M. (2002). ‘The NBER Patent Citation Data File: Lessons, Insights and Methodological Tools’ in Jaffe, A. and Trajtenberg, M. (eds) Patents, Citations, and Innovations: A Window on the Knowledge Economy, pp. 403–​460 (Cambridge, MA: MIT Press). Helsley, R. and Strange W. (1990). ‘Matching and agglomeration economies in a system of cities’. Regional Science and Urban Economics 20: 189–​212. Helsley, R. and Strange, W. (2002). ‘Innovation and input sharing’. Journal of Urban Economics 51: 25–​45. Helsley, R. and Strange, W. (2014). ‘Coagglomeration, clusters, and the scale and composition of cities’. Journal of Political Economy 122: 1064–​1093. Henderson, J.V. (1974). ‘The size and types of cities’. American Economic Review 61: 640–​656. Hsieh, C.T. and Klenow, P. (2014). ‘The lifecycle of plants in India and Mexico’. Quarterly Journal of Economics 129: 1035–​1084. Jacobs, J. (1970). The Economy of Cities (New York: Vintage Books). Jaffe, A., Trajtenberg, M., and Henderson, R. (1993). ‘Geographic localization of knowledge spillovers as evidenced by patent citations’. Quarterly Journal of Economics 108: 577–​598. Kemeny, T., Feldman, M., Ethridge, F., and Zoller, T. (2015). ‘The economic value of local social networks’. Working paper. Kerr, W. (2010). ‘Breakthrough inventions and migrating clusters of innovation’. Journal of Urban Economics 67: 46–​60. Kerr, W. and Brownell, A. (2011). ‘Location choice for new ventures:  choices within cities’. Harvard Business School Background Note 812–​036 (Boston, MA: Harvard Business School). Kerr, W. and Kominers, S. (2010). ‘Agglomerative forces and cluster shapes’. NBER Working Paper 16639. Kerr, W. and Kominers, S. (2015). ‘Agglomerative forces and cluster shapes’. Review of Economics and Statistics97: 877–​899. Kerr, W. and Lincoln, W. (2010). ‘The supply side of innovation: H-​1B visa reforms and U.S. ethnic invention’. Journal of Labor Economics 28: 473–​508. Kerr, W., Nanda, R., and Brownell, A. (2011). ‘Location choice for new ventures: cities’. Harvard Business School Background Note 811–​106 (Boston, MA: Harvard Business School). Klepper, S. (2010). ‘The origin and growth of industry clusters: the making of Silicon Valley and Detroit’. Journal of Urban Economics 67: 15–​32. Kremer, M. (1993). ‘The O-​Ring theory of economic development’. Quarterly Journal of Economics 108: 551–​575. Lazear, E. (2005). ‘Entrepreneurship’. Journal of Labor Economics 23: 649–​680. Marshall, A. (1890). Principles of Economics (London: Macmillan). Michelacci, C. and Silva, O. (2007). ‘Why so many local entrepreneurs?’ Review of Economics and Statistics 89: 615–​633.

The Logic of Agglomeration    365 Mukim, M. (2015). ‘Coagglomeration of formal and informal industry: evidence from India’. Journal of Economic Geography 15: 329–​351. Porter, M. (1990). The Competitive Advantage of Nations (New York: The Free Press). Redding, S. and Turner, M. (2015). ‘Transportation Costs and the Spatial Organization of Economic Activity’ in Henderson, J.V., Duranton, G., and Strange, W. (eds) Handbook of Regional and Urban Economics, Vol. 5, pp. 1339–​1398 (Amsterdam: North-Holland). Rosenthal, S. and Strange, W. (2001). ‘The determinants of agglomeration’. Journal of Urban Economics 50: 191–​229. Rosenthal, S. and Strange, W. (2003). ‘Geography, industrial organization, and agglomeration’. Review of Economics and Statistics 85: 377–​393. Rosenthal, S. and Strange, W. (2004). ‘Evidence on the Nature and Sources of Agglomeration Economies’ in Henderson, J.V. and Thisse, J.F. (eds) Handbook of Regional and Urban Economics, Vol. 4, pp. 2119–​2171 (Amsterdam: North-​Holland). Rosenthal, S. and Strange, W. (2012). ‘Female entrepreneurship, agglomeration, and a new spatial mismatch’. Review of Economics and Statistics 94: 764–​788. Ruiz, N., Wilson, J., and Choudhury, S. (2012). ‘Geography of H-​1B workers: demand for high-​ skilled foreign labor in U.S. metropolitan areas’. Brookings Institute Report. Saxenian, A. (1994). Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA: Harvard University Press). Saxenian, A., Motoyama, Y., and Quan, X. (2002). Local and Global Networks of Immigrant Professionals in Silicon Valley (San Francisco, CA: Public Policy Institute of California). Serafinelli, M. (2014). ‘“Good” firms, worker flows and local productivity’. Working paper, University of Toronto, Ontario, Canada. Tecu, I. (2012). ‘The location of industrial innovation: does manufacturing matter?’ PhD Thesis, Brown University.

Chapter 19

N et work Geo g ra ph i e s and Geo gra ph i c a l Net works: C o-​de pe nde nc e and C o-​e volu t i on of M u ltinat i ona l Enterprises a nd  Spac e Simona Iammarino and Philip McCann Introduction This chapter discusses some of the dramatic changes taking place in the relationship between multinational enterprises (MNEs) and geographical space, highlighting implications for both theory and empirical research. Over the last few decades, geographical specificity at the sub-​national and subregional level has become increasingly important for the strategy, organization, and performance of MNEs, and, in turn, MNEs have become progressively more significant for local and regional economic development (Iammarino and McCann, 2013). Such a bilateral and mediated relationship, in principle valid for any kind of business firm, is particularly important and effective in the case of multinational firms. In the modern phase of globalization MNEs have experienced much faster and deeper transformations than other firm types (i.e. small-​and medium-​sized or large multi-​plant uni-​ national firms) owing to their bridging role across diverse geographical, technological, and institutional systems. The growing interdependence of geographical specificity and MNE behaviour and organization has rendered problematic some of the theoretical constructs traditionally applied in international economics and business studies of MNEs, and at the same time poses serious challenges for empirical investigation. In order to understand such difficulties and to better grasp the contemporary geographical features of multinational firms, it is necessary both to reflect on the traditional models and to update them in the light of the current realities. This

Network Geographies and Geographical Networks    367 is the aim of this chapter, which is divided into six sections. The next section discusses the substantial absence of geographically specific characteristics in the main analytical multidisciplinary framework of MNE studies. ‘The Dichotomies Host versus Home and Determinant versus Impact’ highlights the limitations that have arisen with respect to the use of typical dichotomies such as home versus host and determinant versus impact in the investigation of MNE operations and their interactions with different economic actors and contexts. ‘The Dichotomy Horizontal versus Vertical Integration’ revises the distinction between horizontal versus vertical integration of corporate activities across geographical and institutional boundaries, and argues that different geographies have emerged in relation to new types of firm integration. As a result, the dichotomy hierarchy versus network has lost most of its contrasting power, as both are forms of governance of production and innovation simultaneously applicable at the corporate and spatial level. This is discussed in ‘The Dichotomy Hierarchies versus Networks and the Interdependence Between MNEs and Geographical Space’, which highlights the increasing co-​dependence and co-​evolution of corporate and geographical networks and hierarchies in the international division of labour. The final section offers brief concluding remarks.

MNEs and Space: The Theory and the Real World For now almost four decades the eclectic Ownership–​Location–​Internalization (OLI) paradigm—​originally formulated by John Dunning (e.g. 1977, 1981)  and subsequently updated and adapted by Dunning himself (e.g. 2001, 2009) and a number of other scholars—​ has provided a broad analytical framework for examining both the growth of multinational activity and its changing patterns over time. The OLI paradigm has had the capacity to accommodate and compare different major economic, business, and managerial theories aimed at explaining ownership (O) advantages, that is, why firms become multinational; location (L) advantages, that is, where firms go to internationalize their activities; and internalization (I) advantages, that is, how firms carry out their multinational experience. In spite of the criticisms that the OLI has encountered regarding its explanatory power, it has consistently been demonstrated that the approach is still suitable for interpreting the complexity of MNE expansion, particularly in a historical perspective (e.g. Dunning 1995, 2001; Cantwell and Narula, 2001). In contrast, the OLI paradigm’s ability to integrate effectively both micro-​and macro-​ perspectives of the MNE phenomenon does not translate well when we look at the issue of geographical space. The reason for this failure does not reside only in the OLI itself, but also in the limitations of the main theories that are subsumed into the paradigm. In effect, although strongly micro-​founded, both traditional economic theory, with its focus on international production and product/​market dimensions, and the international business and management literature, which concentrates on the business firm, have mostly treated geography at a highly stylized and unspecific macro-​level. This is typically that of the country, if not of the macro-​region or in some cases even of the continent. Geography has therefore been intended as an international geography rather than as a subnational space (McCann and

368   Iammarino and McCann Mudambi, 2004) in which geographical issues are treated primarily in terms of MNE affiliates or subsidiaries being located either in a home or parent country versus being in a foreign or overseas country. Moreover, the definition of what is understood as being home or foreign is largely treated as being independent of scale, such that a foreign country could be either small or large. While some papers and some research programmes within international business sought to deal with these issues in a more detailed manner and to provide more nuance on these issues than others, the basic fact remains that the traditional notion of geographical space within economics and international business studies tends to be vague and undefined, and it does not go beyond simple locational definitions of being located on either one side or the other of a national border. The substantial absence of any actual geography in the international business literature ironically parallels the criticisms made by international business scholars of both the so-​ called new trade theories and the new economic geography (NEG) with respect to their largely abstract treatment of space (see e.g. Dunning, 1998; Buckley and Ghauri, 2004; Ietto-​ Gillies, 2012). Yet, one of the main merits of the basic NEG framework has been the demonstration of the critical tensions which exist between cores and peripheries, reviving academic interest in the atavistic contrast between central and more marginal regions, and the role that firm location plays in these tensions. Some seminal work in economic geography (e.g. Audretsch and Feldman, 1996; Feldman, 1996), which partly builds on and also extends the issues raised by NEG models, addresses the crucial role of knowledge flows across space, and the extent to which the location of firms also shapes how these flows are themselves subject to core and periphery patterns. These issues have subsequently become central themes in the research agendas of urban economics, regional science, and economic geography, and international business has recently been striving to make an impact on these research progresses. Part of the explanation for the surprising delay in acknowledging the role of space in MNE behaviour lies in the fact that most of the international economics and business emphasis has been on either the macroeconomics of international production or on the characteristics of the internal processes of the firm. Prior to the advent of the modern era of globalization in the early 1990s, both of these lines of enquiry were motivated primarily by historical contingency (Dunning, 2001), in the sense that the priorities of foreign investment decisions made by MNEs were perceived as being far more related to national and international differences than to regional (sub-​national) heterogeneity. Even more fundamentally, the theory of the MNE has developed along the lines of a sharp distinction between firms, as organizations subject to centralized control, and markets, treated as environments characterized by independent actors engaged in full arm’s length transactions. Although both of these lines have deeply evolved over time and remain theoretically valid and useful in today’s economic analysis, they also still represent the major constraints to the full integration of a real spatial dimension within the study of MNEs. Indeed, it is not accidental that the criticisms to the traditional organizational dichotomy market versus hierarchy, and the elaboration of more relevant heuristic models incorporating the network as the prevailing form of economic coordination in contemporary economic systems, have used geographical specificity to illustrate the power of the network governance (e.g. Powell, 1990). Among the first generation of scholars involved in developing the early classical model of the MNE, both Stephen Hymer and Raymond Vernon are crucial contributors, being the first two commentators implicitly or indirectly aware of the long-​run evolution of the relationships between multinational firm organization and economic geography. Vernon’s

Network Geographies and Geographical Networks    369 notion of the product life cycle (Vernon, 1960, 1966), which has since played such a key part in the international business literature, was originally understood as a phenomenon operating at the sub-​national level, in which core cities and regions played different roles from peripheral regions (Vernon, 1959). Vernon understood that a close mapping between technological and geographical structures was likely to be a natural outcome of the investment choices that a multi-​plant firm must make, and that this logic applied even more to the MNE as a type of multi-​plant firm. Meanwhile, Hymer’s (1970, 1972) visionary perspective was that the hierarchical geography of multinational headquarters and knowledge activities will map closely on to the future global urban hierarchy, with higher-​order functions taking place in higher-​order cities. Hymer’s long-​run insight that there must be a close mapping between organizational hierarchies and geographical hierarchies is all the more remarkable given that, at the time of his writing, modern globalization as we understand it today seemed a distant and far-​fetched notion. These insights of Hymer and Vernon, linking technological, organizational, and geographical hierarchies were largely overlooked in the international business literature during the following three or four decades, with instead other intuitions by these two scholars being afforded much greater significance. Given the historical context this is partly understandable. Yet, even now after two decades of rethinking the role played by multinationals in economic geography, these key insights, and how they link to the current world still remain largely underexplored (Iammarino and McCann, 2015). Modern globalization, which began to emerge in the final years of the 1980s and the early 1990s, has been characterized by various key features (Iammarino and McCann, 2013). Firstly, the share of developing and emerging economies on global foreign direct investment (FDI) inflows has grown steadily and, for the first time in history, accounted for more than half of the world total in 2012, confirming a massive transformation in the global geography of foreign investment (UNCTAD, 2013). Secondly, much of these flows are characterized by cross-​border mergers and acquisitions, as well as by greenfield investments, and again some two-​thirds of FDI inflows are in services, with the remaining one-​third involving manufacturing sectors. Thirdly, the majority of these cross-​border flows span neighbouring or even adjacent countries, rather than being genuinely global transactions. The result is that groups of bordering economies are becoming ever more integrated economically into what are known as global regions (Guy, 2015). Global regionalism is also characterized by a slicing up and recombination of global value chains, in which establishments and groups of activities are unbundled (Baldwin, 2011), primarily across groups of neighbouring economic systems. At the core of these global regions are global cities, which house most of the power functions of large corporations (Iammarino and McCann, 2015). The nature and scale of modern globalization has been far beyond what anyone in the 1960s and 1970s could have imagined, but the ideas of Hymer and Vernon already touched on the interdependence of global organizational, technological, and geographical hierarchies in the long-​run firm-​geography trajectories. Much of the available real-​world evidence points to a growing importance of new types of structures and forms of economic coordination which go well beyond the traditional distinction between the firm and the market. These structures include networks, value chains, and quasi-​market relationships of various kinds that have grown exponentially in importance (Guy, 2009). Such massive, profound, and ongoing changes have obviously affected both the nature and configuration of the OLI advantages and their interactions. Although the latter are very difficult to disentangle

370   Iammarino and McCann bilaterally, we argue that the changes in the global institutional and technological environment have had important repercussions for the balance of the ‘three-​legged stool’ of the OLI (Dunning, 1998, 2009, p. 5), affecting, in particular, the centrality of location and its interaction with both ownership and internalization advantages.

The Dichotomies Host versus Home and Determinant versus Impact Ownership advantages (O-​advantages) have been historically exploited largely via internalization and vertical integration (I-​advantages). Growing multinationality generates new O-​advantages through experience and capability accumulation, which can then be exploited by both internal and external means (Castellani and Zanfei, 2004, 2006), giving rise to positive cumulative causation mechanisms that reinforce such advantages. The growing internationalization of one of the major, if not the primary, MNE O-​advantages—​its technological competence and innovative capacity—​has resulted in a renewed questioning of the rather narrow role conferred upon MNE units (i.e. affiliates and subsidiaries) by Vernon’s highly influential product life cycle (PLC) model. The aim of redefining MNE units as key creators of innovation and technological knowledge was originally suggested by Dunning (1970) and later developed by Cantwell (1989) and Fors (1998) among others, building on the seminal work of Edith Penrose (1956, 1959). The MNE is then defined as a bundle of productive resources and competencies—​physical, human, and technological—​which are idiosyncratic to each specific enterprise and represent the firm’s major competitive advantage. Two of the major criticisms of the PLC explanation of international production have important implications in geographical terms. Firstly, on the basis of the evolution of the contemporary world economy, O-​advantages are to be attributed to firms, rather than countries, thus making the geographical origin of MNEs much less predetermined by the national level. This point, seemingly captured in Vernon’s original framework, has been highlighted by both regional economists (e.g. Taylor, 1986, 1987) and international business scholars (e.g. Cantwell, 1995), who attribute the limitations of the PLC model to an inadequate conceptualization of both the firm and technological progress. Secondly, observation suggests that agglomeration forces have attracted MNE activities—​especially high-​value-added ones—​to particular locations in both advanced and emerging economies, thus making the geographical destination of MNEs progressively less dependent on purely cost-​based and relative endowment considerations. Indeed, this latter observation points precisely to the geographical specificity of knowledge O-​advantages acquisition: capability and innovation accumulation processes are ever more reliant on sources that are external to any single firm (however large and multinational it may be) and are highly spatially situated (Storper, 1997, 2013). Evolutionary views of technological change applied to MNE behaviour and strategy have contemplated the interactions between O and L, providing grounds for some significant advances in the field. O-​advantages are increasingly dependent on the ability to explore and select among a wide range of knowledge and innovation sources (e.g. Cantwell and Iammarino, 1998, 2003; Cantwell and Piscitello, 1999). Mostly intangible L-​advantages are highly concentrated within specific regions, cities, and local systems, and contribute to

Network Geographies and Geographical Networks    371 enhancing firm-​specific O-​advantages, which, in turn, strengthen those of the many locations where the MNE is present. Thus, when competitive advantages are seen through the lens of a fine-​grained economic geography and perceived as simultaneously firm-​specific and place-​specific, the host–​home categories—​mainly based on the direction of FDI (Ietto-​ Gillies, 2012)—​cease to be analytically useful. Instead, the sources of knowledge, both intra-​ and extra-​firm, and the overall openness, or connectivity, of the firm with its surrounding environment become far more relevant issues. It is worthwhile to stress the difference between simple connectedness, defined in terms of the architecture of transport and communications infrastructure, and the much broader concept of connectivity, which is a behavioural concept incorporating the capability of individuals, organizations, and institutions to interact and engage across geographical space and within networks. Connectivity emphasizes the degree of two-​way (inward and outward) openness of regions, and also of firms and actors there located, in terms of many behavioural and organizational dimensions of knowledge connectedness. The identification of the MNE’s spatial behaviour is thus a result of complex interactions between firm(s), industry, organizational, and knowledge characteristics. The simple nation-​based host–​home dichotomy therefore becomes largely meaningless, particularly in relation to knowledge flows. Indeed, core regions are those places where host and home actually overlap to a great extent, and the direction of such flows is eminently bi-​or multilateral. The blurring of home/​lending origin versus host/​borrowing destination is even more relevant when looking at the economic impact of MNEs. As seen above, the major research questions have focused on the determinants of MNE behaviour—​that is, why, where, and how firms become multinationals—​and the OLI is by construction primarily a paradigm for understanding the causes of multinationality. However, the effects of MNE operations, although considered within the OLI, are undoubtedly not treated with the same systematic approach. In the OLI framework, each particular activity performed across geographical boundaries is seen as a consequence of the specific advantages of the MNE itself. When it comes to the evaluation of the effects of MNEs, the centrality of the micro-​level as a unit of analysis is obviously less clear cut. The impact of MNE activity can operate simultaneously at different levels: that of the firm, of the industry, of the region, of the country and also of the global level. Trying to disentangle them has proved to be extremely tricky, if not impossible. As Ietto-​Gillies (2012) points out, the assessment of MNE effects is intrinsically associated with the explanations of why, how, and where the multinational enterprise operates. The critical relevance of geographical specificity (the where exactly?) has made it virtually impossible to separate the questions on the determinants from those on the impacts of MNEs in relation to different actual places. For instance, the examination of the growth effect of an MNE’s location choice on the firm itself is seen to be a function of the number and variety of knowledge sources and accumulation that the MNE derives from that particular region among many others. Similarly, the development impact on the region in which an MNE locates depends on the number and variety of knowledge exchanges between the MNE and the surrounding territory and its economic actors, and particularly local firms (including other MNEs). Mutually reinforcing MNE–​environment knowledge exchange is to be two-​way in order to foster sustained local learning, innovation rates, externalities, and knowledge spillovers (Crescenzi et al., 2015). The major focus of the conceptual and empirical literature on the effects of MNEs has been on localized externalities. The widespread difficulty of directly observing externalities,

372   Iammarino and McCann allied with a fuzzy notion of the L (e.g. Dunning, 1981, 1988; Blömstrom, 1989; Kokko, 1992; Lall, 1993), have largely limited the consideration of externalities to one particular type and to only one direction, namely that of spillovers transmitted from MNEs to the host location (e.g. Blomström and Persson, 1983; Blomström and Kokko, 1998; Javorcik, 2004; Javorcik and Spatareanu, 2008), disregarding the critical link that goes from region-​specific L-​advantages to the growth of the MNE itself. Recent empirical literature has suggested the need for a more comprehensive approach in modelling the emergence of positive externalities as a two-​way relationship, rather than as a unilateral pipeline. In this vein, a rediscovery of the role of domestic firms as more than simply passive recipients of foreign capabilities and technologies has gained momentum. The obvious problem arising here, which is particularly serious in empirical research, is the inherent endogeneity of the causal relationship between MNE investment and local firms’ innovation and growth. Some of these contributions have tried to deal more efficiently with such concerns, either by taking advantage of the novel availability of panel data or by controlling for the endogeneity of the regressor of interest (e.g. Benfratello and Sembenelli, 2006; Driffield, 2006; Haskel et al., 2007; Crespo et al., 2009; Crescenzi et al., 2015). Conflicting results on the impact of MNEs may also stem from unobserved firm heterogeneity on both sides; however, this dimension has so far been qualified mainly with respect to MNE characteristics, while scant attention has been devoted to domestic and localized firms’ features (Crescenzi et al., 2015). Finally, the search for often unspecified spillovers in the economic literature has somewhat hidden the effects of MNEs on capabilities development, particularly in developing economies. One effective application of the OLI paradigm has been on development issues, through the concept of investment development path (IDP), always applied at national or broader geographical scale (e.g. Dunning, 1981, 1988, 2001; Dunning and Narula, 1996; Narula, 1996). The main IDP tenet is that as a country develops, the configuration of the OLI advantages facing both MNEs and domestic firms changes, as do their interactions, eventually reversing the directionality of foreign investment. This is clear in the recent impressive surge of outward FDI from developing and emerging locations. All in all, the circularity unsolved by the traditional analytical categories—​home versus host, determinant versus impact—​need to be addressed, both theoretically and empirically, by future research efforts.

The Dichotomy Horizontal versus Vertical Integration What is also rather surprising is how the reciprocal interactions between internalization and location have so far been almost entirely ignored. Instead, a major focus of theoretical and empirical research has been on firm growth as a consequence of the increasing intensity of ownership and internalization. This is particularly interesting if we take into account the recent major transformation in corporate integration and organization. Caves’s work (1971) was the first to introduce the groundbreaking distinction between horizontal and vertical integration of MNE operations across national boundaries. Horizontal

Network Geographies and Geographical Networks    373 FDI (HFDI) implies the production of the same good or service produced at home in a new foreign location, thus replicating identical production processes across countries. Vertical FDI (VFDI) instead involves the shift abroad of some stages of the production process, either backward (upstream), forward (downstream), or both, thus fragmenting the MNE production process vertically across countries. Caves also allowed for the possibility of MNEs carrying out foreign production that is neither horizontally nor vertically integrated, but of ‘conglomerate diversification’ (Caves, 1971, p. 3). In the case of HFDI the ownership advantages are possessed by MNEs operating in industries characterized by oligopolistic structure and substantial product differentiation, and particularly those with the most considerable knowledge-​or research-​intensive activities. The major advantages of VFDI relate to the structural features of the markets in which MNEs are active: in particular, to where there are incentives to eliminate oligopolistic uncertainty for input supplies and to raise barriers to entry by vertically integrating the stages of production across national boundaries. The subsequent Knowledge Capital Model (KCM) (e.g. Markusen, 1984, 2002; Markusen and Venables, 1998; Venables, 1999; Carr et al., 2001) rests on the main idea that MNEs are intensive in the use of knowledge-​based assets. The approach combines both horizontal integration associated with the proximity to demand, and vertical integration associated with the search for lower costs, as determinants of MNE location and investment activities. Drawing extensively upon Caves’ work, although curiously seldom acknowledging it, the KCM splits multinational firms into two types:  horizontally integrated firms known as type-​h firms, and vertically integrated firms known as type-​v firms (Markusen, 2002). One of the main assumptions of the KCM model is that the firm’s knowledge assets are basically a public good within the firm, whose costs of supply to the firm’s foreign plants are very low. Whether the firm decides to supply foreign and overseas markets directly via exports or via local supply from foreign affiliates depends on the balance between domestic production economies of scale and international trade or transport costs (e.g. Markusen, 2002; but also Caves, 1971, 1982). In general, high trade and transport costs encourage FDI as firms seek to gain easier access to a foreign market, while low trade costs encourage domestic production and exporting. Similarly, high economies of scale encourage single-​site production and exporting, whereas low economies of scale encourage the establishment of different facilities in different locations. Regarding the patterns of FDI, type-​h MNEs tend to dominate when the markets in both the origin and host locations are large and similar in terms of their local labour skills’ endowments, whereas type-​v MNEs tend to dominate when the markets differ substantially in terms of their size and the endowment of local labour skills. Trade theorists have recently noted an apparent conflict between the KCM explanation of MNE activity and the trends observed in current globalization. As Neary (2009) argues, if transport and trade costs fall, which has, indeed, been the case in recent decades, then according to the KCM one would expect that HFDI will decrease, as exporting from a domestic location should become more attractive. However, HFDI and multinationalism have increased dramatically over recent decades, thereby producing outcomes that appear to be counterintuitive to the KCM main tenet. One way to reconcile these observations is to assume that the set-​up costs of individual foreign establishments have fallen over time. Indeed, building plants and establishing new turnkey production facilities is becoming increasingly sophisticated, thereby pointing to the conclusion that set-​up costs of overseas establishments are falling. However, plant set-​ up costs may well also involve issues related to labour knowledge, skills, and training, and

374   Iammarino and McCann there is very little evidence to suggest that these costs have decreased. Alternatively, we could assume that the location-​specific economies of scale have actually become less important over recent years, and that their decline has been even greater than that in transport and trade costs. Yet, there is growing evidence that the geography of many production systems and input–​output chains is becoming more spatially fragmented (e.g. Parr et al., 2002; Klier and Rubenstein, 2008). In addition, the suggestion that location-​specific economies of scale have fallen over time appears to be at odds with the fundamental assumptions of the NEG and with the wealth of evidence on the increasing worldwide importance of agglomeration effects (e.g. OECD, 2006; World Bank, 2009; McKinsey Global Institute, 2013). A  third attempt at reconciling theory and observation is to suggest that, as transport costs fall, the potential profits of foreign acquisitions systematically favour the acquiring firms, thereby promoting outward FDI based on mergers and acquisitions (Neary, 2009): this insight is, in fact, consistent with the experience of the European Union. Thus, the dichotomy horizontal versus vertical integration seems to be no longer suitable for reflecting the main organizational forms of MNE international operations. This has been acknowledged by KCM scholars, who recognize that MNEs follow complex integration strategies rather than those in one or either category (e.g. Grossman, et al., 2003; Yeaple, 2003; Neary, 2009). MNEs are mostly both horizontally and vertically integrated, mixing up different strategies including that of the international diversification across products and space firstly described by Caves (1971, 1982). In the words of Neary (2009, p. 215) ‘the distinction between horizontal and vertical FDI is useful for pedagogic purposes but otherwise not very helpful’. More research efforts to better grasp the increasing complexity of MNE strategies are becoming urgent.

The Dichotomy Hierarchies versus Networks and the Interdependence Between MNEs and Geographical Space Globalization, in the context of the creation and expansion of strategically planned worldwide networks of investment, production, sales of goods, and services underpinned by worldwide movements of people, money, knowledge, and ideas, dates back to the fifteenth and sixteenth centuries. More than four centuries of global expansion were interrupted by the two world wars, the 1930s Depression, and the Bretton Woods systems, each of which contributed to a global retrenchment of economic activities into primarily national or transatlantic networks of relations (Iammarino and McCann, 2013). This longstanding retrenchment was overturned again in the last decade of the twentieth century, driven by technological, institutional, and organizational changes. The advent of the modern Internet provided for a common platform on which communications, management control, and analytical technologies could be integrated. At the same time, institutional transformations were leading to falls in international trade and investment barriers associated with the movement towards macro-​areas of free trade and economic integration such as the European Single Market, North American Free Trade Agreement (NAFTA), MERCOSUR, Association of Southeast Asian Nations (ASEAN), and Closer Economic Relations (CER). These trends

Network Geographies and Geographical Networks    375 were overwhelmingly driven and spearheaded by MNEs, building on both the new communications technologies and institutional changes, that seized the new opportunities for outsourcing and offshoring in order to reconfigure and redesign the spatial and organizational logic of their global activities. International MNE networks have represented the strategic integration of geographically distinct paths of both production and innovation activity (Cantwell, 1989). The stable relationship between ownership and control, which has long been understood as problematic when looked at from the divide between investor and executives, has been disturbed both along the supply chain and within the corporation (Ietto-​Gillies, 2012). In outsourcing strategies, ownership may change, but control of value-​chain activities is largely retained by MNEs through various means of pressure on the suppliers and their competitive bidding (e.g. narrow transfers of technology, strict product specifications, tight supplying schedules, etc.) (UNCTAD, 2013). Conversely, in integration and offshoring strategies, ownership is not altered, but the distribution of control within the MNE can vary greatly. Different degrees of autonomy of affiliates and subsidiaries can lead to intra-​firm competition and even to various degrees of restraint in the control of the central MNE headquarters (e.g. Birkinshaw and Hood, 1998, 2000; Birkinshaw et al., 2005). In these network structures, international trade, production, and knowledge creation occur both within the individual MNE and also within networks, some of which are highly spatially concentrated and some of which largely aspatial. The actual geographical and organizational configurations depend on the specific patterns of unbundling (Baldwin, 2011) of enterprises and activities, which, in turn, depend on how the global value chains are being sliced up, reconfigured, and recombined (Gereffi and Korzeniewicz, 1994). This implies a revision of the role and activities played by individual establishments within the overall MNE structure. Indeed, the location of corporate headquarters of large MNEs has nowadays little geographical connection with the location base for specific business units and operations. Rather, different configurations are employed by different MNEs depending on their industries, technologies, knowledge assets, and also the nature of the interactions they undertake with potential suppliers and customers. For many MNEs, the role of subsidiaries and affiliates has transformed from the previously largely passive recipients of knowledge and instructions from parent country-​based headquarters to highly autonomous localized centres of knowledge creation and exploitation that feed back into the global MNE knowledge network system. This role of knowledge generation often implies the affiliate becoming increasingly independent and heavily embedded in its local context by building on local human capital and knowledge networks, as well via the more traditional buyer–​supplier pecuniary linkages. Such organizational changes have had major implications for the location choices and advantages of MNEs, which are increasingly dependent on the balance between technological competencies and capabilities within and outside the firm, or on the integration of various sources of knowledge that are both internal and external to the firm. Different geographies have emerged in relation to different types of MNE integration, (de-​)centralization of firm control, unbundling of headquarters and core functions, and smoothing of organizational structures (McCann and Mudambi, 2005; Iammarino and McCann, 2013, 2015). Such changes have highlighted the fuzziness of the notion of hierarchy as a corporate structure versus networks as a form of spatial coordination. Increasingly, the empirical

376   Iammarino and McCann evidence points to the intersubstitutability of space and organizations into the two typical forms of governance, and to the prevailing heterogeneity of both MNEs and specific geographical contexts. Nowadays, the multiple locations of MNEs are best understood by referring to specific sub-​ national areas—​such as regions, cities, or industrial clusters—​where a firm locates its main functions, including strategic decision-​making, R & D, and other core production activities. Each different MNE function tends to favour different spatial characteristics, thus pushing towards the dispersion of functions across various (sub-​national) locations. Opposing this dispersion force are linkages between different stages in the production chain which can encourage firms to co-​locate different activities in the same location (Defever, 2006). Indeed, recent evidence on the economic geography of MNEs indicates that, in the cross-​ border co-​location of the different stages of the value chain of MNE affiliates in the context of the European Union, MNE headquarters do not display any pull effect over the location of any other corporate function (Defever, 2006; Ascani et al., 2016). Goerzen et al. (2013) have shown that ‘competence-​exploiting’ and ‘competence-​creating’ (Cantwell and Mudambi, 2005)  activities of MNEs follow very different spatial patterns:  while the former tend to agglomerate in global or core locations, the latter—​far more valuable for local economic development—​tend to concentrate in other metropolitan (or rather less core) regions, giving rise to geographical hierarchies based on functions rather than on firm counts or industry. These differing patterns also suggest that the relationships between MNE affiliates and the geography of knowledge networks and spillovers are likely to be far more varied and nuanced than the simple stylized linkages popularized in the traditional regional and urban economics and economic geography literature. Standard arguments in urban economics assume that firms locating in larger, or global, cities benefit the most from the learning, sharing, and matching opportunities locally available (e.g. Duranton and Puga, 2004). However, in the light of the issues raised here, the potential advantages for MNEs associated with learning and sharing in large cities are not so obvious. Firstly, the MNE generates advantages via the sharing of place-​specific assets, knowledge, and know-​how within the corporate geographical network—​in the revisited OLI this would reflect the interaction between O, L, and I. The rationale for the MNE in leveraging these internalized assets and organizational advantages to some extent precludes the potential benefits available from locating in global cities. Secondly, while the potential knowledge inflows gained by locating in cities due to spillovers may appear prima facie to be attractive to MNEs, the danger of experiencing knowledge outflows via unintended knowledge leakages outside the firm may be at least as significant as any possible inflow benefits (McCann and Mudambi, 2005; Iammarino and McCann, 2013). The fact that MNEs place so much emphasis on knowledge internalization—​even more so in their present network structures—​suggests that knowledge-​related functions will often be located somewhat away from large cities. This may be particularly relevant where high levels of secrecy are required. Following the same line of argument but from a different perspective, in the ‘world city hypothesis’ (Friedmann, 1986) and much of the ‘global city network’ (GCN) literature, the bulk of the connections between global cities or core cities in core regions—​where large MNEs are mostly headquartered—​and more peripheral regions of the global economy are argued to take place through regional articulators, or core cities in peripheral regions (e.g. Beaverstock et al., 1999). While this may be true for some large MNEs approaching peripheral regions from the top down and with certain market-​seeking strategies, the same may

Network Geographies and Geographical Networks    377 not necessarily apply to businesses in peripheral cities of peripheral regions interacting with the global economy from the bottom up. Recent research suggests that firms in peripheral regions often bypass these regional articulators, instead seeking to plug into business networks, either in global cities directly or in peripheral cities within core regions (Datu, 2014; Crescenzi et al., 2017). In other words, those firms form what we may call inter-​peripheral networks, thus bypassing the core cities within their respective periphery more often than implied by the literature. This suggests that, while the global city–​regional articulator-​global periphery model implied in much of the GCN literature may be appropriate for the most global MNEs, other models may be required to understand the behaviour of emerging MNEs or MNEs from emerging places (Datu, 2014; Crescenzi et al., 2017), prompting new theoretical and empirical challenges for future research agendas.

Conclusions As Dunning (1977) observed within the eclectic OLI framework, MNE location behaviour—​ the L—​has become increasingly intertwined with both ownership and internalization. The links between technological, organizational, and geographical hierarchies first outlined by Vernon (1959) and Hymer (1972) have often evolved into flatter networks and are much more complex, richer, and heterogeneous than their representation in theoretical models, which have not being able to keep up with actual changes. Corporate geographical networks of functions determine spatial hierarchies, and network geographies between core and peripheral locations in both core and peripheral regions influence the integration patterns of MNEs, particularly from emerging economies. This implies that when a multinational relocates or invests in new subsidiaries, this geographical choice itself changes the internal organizational and internalization logic of the MNE. If the (re)location involves important functions, then the additional O-​and I-​advantages may be quite significant, thereby altering the relative positioning and roles of units within the corporate network. Similarly, the location of an MNE function in a locality, by definition, re-​shapes both the technological and connectivity features of the region. O, I, and L ought all to be seen as interacting at every stage of the MNE location decision, and hierarchies and networks across space are to be seen as being continually reshaped along with the technological and organizational changes within corporations. As such, while many aspects of the traditional location theory toolkit have proved to be still relevant for analysing the spatial dimension of MNEs, there are additional crucial issues which need to be considered that draw from literatures other than economic geography or urban economics. We are still some way off from a comprehensive and integrated spatial theory framework for the modern MNE, and tackling the analytical and empirical challenges associated with specific geography remains the main arena in which progress ought to be made. This is important because, while concepts such as pipelines—​or, as seen in another chapter of this Handbook, highways—​are currently popular in geography, the overwhelming majority of international movements of tangible and intangible resources are undertaken within MNEs and their internal and external networks, implying a multi-​lateral directionality of flows. Economic geography, international business and management, and regional and urban economics need to make far more joint progress in understanding the co-​dependence

378   Iammarino and McCann and co-​evolution of MNEs and geographical space in order to develop a more sophisticated narrative of modern globalization.

References Ascani A., Crescenzi R., and Iammarino S. (2016). ‘Economic Institutions and the Location Strategies of European Multinationals in their Geographical Neighbourhood’. Economic Geography 92(4): 401–429. Audretsch, D.B. and Feldman, M.P. (1996). ‘Knowledge spillovers and the geography of innovation and production’. American Economic Review 86: 630–​640. Baldwin, R. (2011). ‘Trade and industrialisation after globalisation’s 2nd unbundling: how building and joining a supply chain are different and why it matters’. NBER Working Paper No. 17716, December. Beaverstock, J.V., Taylor, P.J., and Smith, R.G. (1999). ‘A roster of world cities’. Cities 16: 445–​458. Benfratello, L. and Sembenelli, A. (2006). ‘Foreign ownership and productivity: is the direction of causality so obvious?’ International Journal of Industrial Organization 24: 733–​751. Birkinshaw, J. and Hood, N. (1998). ‘Multinational subsidiary evolution: capability and charter change in foreign-​owned subsidiary companies’. Academy of Management Review 23: 773–​795. Birkinshaw, J. and Hood, N. (2000). ‘Characteristics of foreign subsidiaries in industrial clusters’. Journal of International Business Studies 31: 141–​154. Birkinshaw, J., Hood, N., and Young, S. (2005). ‘Subsidiary entrepreneurship, internal and external competitive forces, and subsidiary performance’. International Business Review 14: 227‒248. Blomström, M. (1989). Foreign Investment and Spillovers (London: Routledge). Blomström, M. and Kokko, A. (1998). ‘Multinational corporations and spillovers’. Journal of Economic Surveys 12: 247‒277. Blomström, M. and Persson, H. (1983). ‘Foreign investment and spillover efficiency in an underdeveloped economy:  evidence from the Mexican manufacturing industry’. World Development 11: 493‒501. Buckley, P. and Ghauri, P.N. (2004). ‘Globalisation, economic geography and the strategy of multinational enterprises’. Journal of International Business Studies 35: 81‒98. Cantwell, J. (1989). Technological Innovation and Multinational Corporations (Oxford: Basil Blackwell). Cantwell, J. (1995). ‘The globalisation of technology: what remains of the product cycle model?’ Cambridge Journal of Economics 19: 155‒174. Cantwell, J. and Iammarino, S. (1998). ‘MNCs, technological innovation and regional systems in the EU: some evidence in the Italian case’. International Journal of the Economics of Business, Special Issue 5: 383–​408. Cantwell, J. and Iammarino, S. (2003). Multinational Corporations and European Regional Systems of Innovation (London: Routledge). Cantwell, J. and Mudambi, R. (2005). ‘MNE competence-​creating subsidiary mandates’. Strategic Management Journal 26: 1109‒1128. Cantwell, J. and Narula, R. (2001). ‘The eclectic paradigm in the global economy’. International Journal of the Economics of Business 8: 155‒172.

Network Geographies and Geographical Networks    379 Cantwell, J. and Piscitello, L. (1999). ‘The emergence of corporate international networks for the accumulation of dispersed technological capabilities’. Management International Review 39 (Special Issue 1): 123–​147. Carr, D.L., Markusen, J.R., and Maskus, K.E. (2001). ‘Estimating the knowledge-​capital model of the multinational enterprise’. American Economic Review 91: 693‒708. Castellani, D. and Zanfei, A. (2004). ‘Choosing international linkage strategies in the electronics industry:  the role of multinational experience’. Journal of Economic Behavior & Organization 53: 447‒475. Castellani, D. and Zanfei, A. (2006). ‘Multinational firms, innovation and productivity’. Economia Politica, Le edizioni del Mulino 2: 351‒353. Caves, R.E. (1971). ‘International corporations:  the industrial economics of foreign direct investment’. Economica 38: 1–​27. Caves, R.E. (1982). Multinational Enterprise And Economic Analysis (Cambridge: Cambridge University Press). Crescenzi, R., Datu, K., and Iammarino, S. (2016). ‘European cities and foreign investment networks’. Papers in Evolutionary Economic Geography (PEEG) 16.16, Utrecht University, Section of Economic Geography. Crescenzi, R., Datu, K., and Iammarino, S. (2017). ‘European Cities and Foreign Investment Networks’. Scienze Regionali - Italian Journal of Regional Sciences 16(2): 79–​110. Crespo, N., Fontoura, M.P., and Proenca, I. (2009). ‘FDI spillovers at regional level: evidence from Portugal’. Papers in Regional Science 88: 591–​607. Datu, K. (2014). ‘The role of the global network of cities in the economies of developing cities and regions:  Lagos and West Africa’. PhD Dissertation, Department of Geography & Environment, London School of Economics, London. Defever, F. (2006). ‘Functional fragmentation and the location of multinational firms in the enlarged Europe’. Regional Science and Urban Economics 36: 658‒677. Driffield, N. (2006). ‘On the search for spillovers from foreign direct investment (FDI) with spatial dependency’. Regional Studies 40: 107–​119. Dunning, J.H. (1970). Studies in International Investment (London: Allen & Unwin). Dunning, J.H. (1977). ‘Trade, location of economic activity’. Proceedings of a Nobel Symposium Held at Stockholm, pp. 395–​418 (London: Macmillan). Dunning, J.H. (1981). International Production and the Multinational Enterprise (London: Allen & Unwin). Dunning, J.H. (1988). ‘The eclectic paradigm of international production: a restatement and some possible extensions’. Journal of International Business Studies 19: 1–​31. Dunning, J.H. (1995). ‘Reappraising the eclectic paradigm in an age of alliance capitalism’. Journal of International Business Studies 26: 461‒491. Dunning, J.H. (1998). ‘Location and the multinational enterprise: a neglected factor?’ Journal of International Business Studies 29: 45‒66. Dunning, J.H. (2001). ‘The eclectic (OLI) paradigm of international production: past, present and future’. International Journal of the Economics of Business 8: 173‒190. Dunning, J.H. (2009). ‘Location and the multinational enterprise: John Dunning’s thoughts on receiving the Journal of International Business Studies 2008 Decade Award’. Journal of International Business Studies 40: 20‒34. Dunning, J.H. and Narula, R. (eds) (1996). Foreign Direct Investment and Governments: Catalysts for Economic Restructuring (London and New York: Routledge).

380   Iammarino and McCann Duranton, G. and Puga, D. (2004). ‘Micro-​foundations of Urban Agglomeration Economies’ in J.V. Henderson and J.F. Thisse (eds) Handbook of Regional and Urban Economics (1st edition), pp. 2064–​2117 (Amsterdam: Elsevier). Feldman, M.P. (1996). The Geography of Innovation (Dordrecht: Kluwer Academic Publishers). Fors, G. (1998). ‘Locating R&D abroad: the role of adaptation and knowledge-​seeking’ in P. Braunerhjelm and K. Ekholm (eds) The Geography of Multinational Firms, pp. 117–​134 (Boston, Dordrecht, and London: Kluwer Academic Publishers). Friedmann, J. (1986). ‘The world city hypothesis’. Development and Change 17: 69–​83. Gereffi, G. and Korzeniewicz, M. (eds) (1994). Commodity Chains and Global Capitalism (Westport, CT: Praeger). Goerzen, A., Asmussen, C.G., and Nielsen, B.B. (2013). ‘Global cities and multinational enterprise location strategy’. Journal of International Business Studies 44: 427–​450. Grossman, G., Helpman, E., and Szeidl, A. (2003). ‘Optimal integration strategies for the multinational firm’. NBER Working Paper 10189. Guy, F. (2009). The Global Environment of Business (Oxford: Oxford University Press). Guy, F. (2015). ‘Globalisation, regionalization and technological change’ in D. Archibugi and A. Filippetti (eds) The Handbook of Global Science, Technology and Innovation, pp. 575–​596 (Oxford: Wiley-​Blackwell). Haskell, J.E., Pereira, S.C., and Slaughter, M.J. (2007). ‘Does inward foreign direct investment boost the productivity of domestic firms?’ Review of Economics and Statistics 89: 482–​496. Hymer, S. (1970). ‘The efficiency contradictions of multinational corporations’. American Economic Review 60: 441–​448. Hymer, S. (1972). ‘The multinational corporation and the law of uneven development’ in J.N. Bhagwati (ed.) Economics and World Order: From the 1970s to the 1990s, pp. 113–​140 (New York: The Free Press). Iammarino, S. and McCann, P. (2013). Multinationals and Economic Geography: Location and Technology, Innovation (Cheltenham: Edward Elgar). Iammarino, S. and McCann, P. (2015). ‘MNE innovation networks and the role of cities’ in D. Archibugi and A. Filippetti (eds) The Handbook of Global Science, Technology and Innovation, pp. 290–​312 (Oxford: Wiley-​Blackwell). Ietto-​Gillies, G. (2012). Transnational Corporations and International Production. Concepts, Theories and Effects (2nd edition) (Cheltenham: Edward Elgar). Javorcik, B.S. (2004). ‘Does foreign direct investment increase the productivity of domestic firms? In search of spillovers through backward linkages’. The American Economic Review 94: 605‒627. Javorcik, B.S. and Spatareanu, M. (2008). ‘To share or not to share: does local participation matter for spillovers from foreign direct investment?’ Journal of Development Economics 85: 194–​217. Klier, T.H. and Rubenstein, J.M. (2008). ‘Who really made your car?’ Chicago Fed Letter, Federal Reserve Bank of Chicago, October. Kokko, A. (1992). Foreign Direct Investment, Host Country Characteristics, and Spillovers (Stockholm: Stockholm School of Economics, Economic Research Institute). Lall, S. (1993). The Interrelationship Between Investment Flows and Technology Transfer: An Overview of the Main Issues (Geneva: UNCTAD Secretariat). McCann, P. and Mudambi, R. (2004). ‘The location behaviour of the multinational enterprise: some analytical issues’. Growth and Change 35: 491‒524. McCann, P. and Mudambi, R. (2005). ‘Analytical differences in the economics of geography: the case of the multinational firm’. Environment and Planning A 37: 1857–​1876.

Network Geographies and Geographical Networks    381 McKinsey Global Institute (2013). ‘Urban world: the shifting of the global business landscape’. McKinsey & Company, October 2013. Markusen, J.R. (1984). ‘Multinationals, multi-​plant economies and the gains from trade’. Journal of International Economics 16: 205–​226. Markusen, J.R. (2002). Multinational Firms and the Theory of International Trade (Cambridge, MA: The MIT Press). Markusen, J.R. and Venables, A.J. (1998). ‘Multinational firms and the new trade theory’. Journal of International Economics 46: 183‒203. Narula, R. (1996). Multinational Investment and Economic Structure (London: Routledge). Neary, P.J. (2009). ‘Trade costs and foreign direct investment’. International Review of Economics and Finance 18: 207‒218. OECD (2006). Competitive Cities in the Global Economy (Paris: Organisation for Economic Co-​operation and Development). Parr, J.B., Hewings, G.D., Sohn, J., and Nazara, S. (2002). ‘Agglomeration and trade: some additional perspectives’. Regional Studies 36: 675‒684. Penrose, E. (1956). ‘Foreign investment and the growth of the firm’. Economic Journal 66: 220–​235. Penrose, E. (1959). The Theory of the Growth of the Firm (New York: John Wiley). Powell, W.W. (1990). ‘Neither market nor hierarchy—​network forms of organization’. Research in Organizational Behavior 12: 295–​336. Storper, M. (1997). The Regional World:  Territorial Development in a Global Economy (New York: Guilford Press). Storper, M. (2013). Keys to the City: How Economics, Institutions, Social Interaction, and Politics Shape Development (Princeton, NJ: Princeton University Press). Taylor, R.S. (1986). Value-​Added Processes in Information Systems (Norwood, NJ: Ablex). Taylor, R.S. (1987). ‘New thought in the twenties: the case of Springfield, Illinois’. Historian 49: 329–​347. UNCTAD (2013). ‘World Investment Report 2013: global value chains: investment and trade for development’. (New York and Geneva: United Nations). Venables, A.J. (1999). ‘The international division of industries:  clustering and comparative advantage in a multi-​industry model’. Scandinavian Journal of Economics 101: 495‒513. Vernon, R. (1959). The Changing Economic Function of the Central City (New York: Area Development Committee of CED). Vernon, R. (1960). Metropolis (Cambridge, MA: Harvard University Press). Vernon, R. (1966). The Myth and Reality of our Urban Problems (Cambridge, MA: Harvard University Press). World Bank (2009). World Development Report 2009:  Reshaping Economic Geography (Washington, DC: The World Bank). Yeaple, S.R. (2003). ‘The role of skill endowments in the structure of U.S. outward foreign direct investment’. The Review of Economics and Statistics 85: 726‒734.

Chapter 20

T he L o gic of Produ c t i on Net work s Henry Wai-​c hung Yeung Introduction In its World Investment Report 2013, UNCTAD (2013) estimated that some 80 per cent of international trade was organized through global production networks (GPNs) coordinated by lead firms investing in cross-​border productive assets and trading inputs and outputs with partners, suppliers, and customers worldwide. Empirically, there is no doubt that GPN and global value chains (GVCs) are the most critical organizational platforms through which production in primary, manufacturing, and service sectors is coordinated and organized on a global basis. A 2010 World Bank report on the post-​2008 world economy claims that ‘given that production processes in many industries have been fragmented and moved around on a global scale, GVCs have become the world economy’s backbone and central nervous system’ (Cattaneo et al., 2010a, p. 7). To analysts in many international organizations, GPNs are now recognized as the new long-​term structural feature of today’s global economy. These production networks are invariably the de facto organizational mechanism through which local economies at the local, regional, and national scales are integrated into the ever-​more interdependent global economy (see OECD, 2011; WTO and IDE-​JETRO, 2011; Elms and Low, 2013; OECD–​WTO–​UNCTAD, 2013; UNCTAD, 2013; World Bank, 2013). Over fifteen years ago, Peter Dicken (2000, p. 274) opened his chapter for the Oxford Handbook of Economic Geography with a compelling statement that ‘Without doubt, one of the most significant economic–​geographical developments of the twentieth century was the growth and spread of international direct investment, and of other forms of international economic involvement (such as collaborative ventures between firms), through the medium of the transnational corporation (TNC)’. Into the second decade of the new millennium, this ‘overwhelming and indisputable’ spread of international economic activity has become much better understood as a complex geographical phenomenon involving powerful transnational corporations coordinating their production networks on a worldwide basis (Dicken, 2015). While some of these networks operate through intra-​ firm trade among subsidiaries and affiliates within each transnational corporation, a large

The Logic of Production Networks    383 proportion of these cross-​border production networks involve inter-​firm and extra-​firm relations. In this chapter, I highlight and evaluate the most significant economic–​geographical research that examines the logic and role of production networks in facilitating global–​local economic integration. In particular, theoretical advancement and empirical studies of GPNs have emerged since the late 2000s as one of the most influential research foci in economic geography. This literature draws upon the earlier theoretical debates on business and industrial networks in economic geography during the 1990s (Dicken and Thrift, 1992; Cooke and Morgan, 1993; Grabher, 1993; Yeung, 1994, 2000, 2008). At the same time, it also develops close connections with the initial conceptual advancement on global commodity chains and GPNs in the cognate social science fields of economic sociology, development studies, and regional studies (Gereffi, 1994, 1999, 2014; Humphrey, 1995; Ernst and Kim, 2002; Humphrey and Schmitz, 2002; Sturgeon, 2002), and their manifestations in economic geography (Leslie and Reimer, 1999; Hughes, 2001; Smith et al., 2002; Hughes and Reimer, 2003).1 Emerging in the early 2000s, the Manchester school of GPN studies in economic geography—​first named as such in Bathelt’s (2006, p. 225) review of geographies of production—​is primarily defined in relation to several programmatic publications on GPNs (Dicken et al., 2001; Henderson et al., 2002; Coe et al., 2004, 2008a; Hess, 2004; Yeung, 2005a, 2009). Moving beyond the influential analysis of governance structures in the GVC literature associated with sociologist Gary Gereffi (1994, 1999; Gereffi et al., 2005), these studies in economic geography focus on relational network configurations and the strategic coupling of local and regional economies with GPNs (see reviews in Hess and Yeung, 2006a; Bair, 2008, 2009; Gibbon et al., 2008; Coe et al., 2008b; Coe, 2012; Parrilli et al., 2013; Neilson et al., 2014; 2015; Coe and Yeung, 2015; Smith, 2015).2 As a synopsis of what’s to come in the chapter, the GPN literature in economic geography is primarily concerned with two main analytical issues. Firstly, it addresses the critical question of geographical specialization in value activity, a common phenomenon in today’s global industries, rather than in finished goods or services as in conventional economic analysis. Grounded in sophisticated conceptual frameworks and empirical studies, this literature accounts better for the spatial (re)organization of production and value distribution on a global scale. It contributes to the academic and policy debates on why we should move from a ‘Made in X’ conception of economic activity to a ‘Made by the World’ notion of global production. Be it the USA, Germany, or China, ‘X’ here is often misconstrued conceptually as a national economy in conventional trade theory, and mismeasured statistically as the container of all value-​added (e.g. in trade data). Secondly, the GPN literature tackles one central problem in geographical analysis—​how are economic activities at different spatial scales integrated that, in turn, (re)produce uneven geographical political economy? This global–​local tension has been a longstanding concern in economic geography (Swyngedouw, 1992, 1997; Dicken, 1994; Peck and Tickell, 1994; Tickell and Peck, 1995; Yeung, 1998, 2002). The GPN approach unpacks critically the organizational mechanisms underlying the ruthless globalization of production and accentuates concomitantly the fundamental importance of local articulation into these GPN for economic development. The following sections first explain how the approach describes and explains the logic of this global–​local economic integration. It then considers the significance of this GPN literature, its key controversies, and the prospects and future directions for research in what might be termed GPN 2.0 research in economic geography.

384   Yeung

Production Networks and Global–​L ocal Economic Integration By definition, a production network necessarily involves more than one actor. These actors can be capitalist firms of different sizes, ownership structures, industrial specialization, national origins, and so on; they can also be non-​firm entities, such as the state, international organizations, labour groups, consumers, civil society organizations, and so on. As a significant economic actor in contemporary capitalism, the firm is responsible for the creation of economic value through production, a process of transforming material and intangible inputs into outputs such as intermediate or finished goods and services. Production is therefore a value-​transformative process incorporating all economic sectors, from extractive industries in the primary sector to manufacturing activities in the secondary sector and service industries in the tertiary sector. OECD-​WTO-​UNCTAD (2013, p. 16) estimated that as intermediate inputs to global production, services contributes directly and indirectly to over 30 per cent of the total value added in manufactured goods. In turn, several of these service activities are themselves organized and delivered through GPNs, as evident, for example, in finance, advertising, logistics, or retailing. In its linear form, this process of transformation is commonly understood as taking place through the value chain. If a firm internalizes much of this value-​chain activity through vertical integration, value transformation is deemed intra-​firm in nature. If a firm engages with other firms in the production of value through outsourcing, subcontracting, or strategic alliances, inter-​firm organizational relations are developed. Much of the current GVC literature focuses on the key actors, particularly the lead firms, who govern through inter-​firm relations this complex process of value transformation and capture (Gereffi et al., 2005; Bair, 2009; Ponte and Sturgeon, 2014). This (global) value-​chain approach to inter-​firm production relations, however, suffers from two major problems: (i) its linear interpretations of how production systems operate and how value is generated and distributed; and (ii) its narrow focus on lead firms in inter-​firm governance at the expense of other non-​firm actors and the wider institutional contexts. Building on and going beyond this chain-​based conception of global–​local production, the GPN approach advocates a network understanding of the economic–​geographical process of value transformation in a global mosaic of local and regional economies (Coe and Yeung, 2015, Chapter 2). In this approach, value transformation is conceptualized as taking place not just through the horizontal or inter-​firm movement of materials and/​or intangibles along the value chain in a particular industry, but it can also occur through the simultaneously vertical or extra-​firm relationships between firms in interconnected production processes and non-​firm actors. In inter-​firm relations, a lead firm can engage horizontally with a variety of other firms, such as strategic partners, specialized suppliers, and generic suppliers, in its value transformation. While a lead firm takes charge of product or market definition based on its firm-​specific competitive advantage and capabilities, its network members provide the partial or complete material supplies (e.g. the manufacture of intermediate or finished goods) and business solutions (e.g. the provision of critical or advanced services). In each segment of these horizontal inter-​firm organizational relations (e.g. lead firm–​supplier or strategic partner–​supplier), a firm can be subject to vertical influences from ‘above’

The Logic of Production Networks    385 (e.g. the state and the wider regulatory or industry context) and from ‘below’ (e.g. labour unions or customer pressures). In this network organizational form, coordinating production is conceptually much more than the existing GVC focus on the inter-​firm governance of value transformation—​often only in manufacturing industries. Production networks inevitably bring together both firms and non-​firm actors and create economic value through a combination of intra-​, inter-​, and extra-​firm relations. To conceptualize these different configurations of network relations, the GPN approach has deployed or developed three central concepts—​power relations, embeddedness, and strategic coupling. Power relations refer to the dynamic processes through which key actors in these production networks can compel or influence others to act in these actors’ interest. Embeddedness is a state of mutual dependency between actors in the same network. Strategic coupling takes place when actors in different networks work together to produce a common strategic outcome. Whereas power relations and embeddedness were, respectively, developed in Dicken et al. (2001) and Henderson et al. (2002), the last of these was original to the analytical framework first developed in Coe et al. (2004) and further refined and extended in Yeung (2009, 2015, 2016). A GPN is characterized by the central role of a globally significant lead firm in its coordination, control, and organizational involvement of a large number of intra-​firm affiliates, strategic partners, key customers, and non-​firm institutions. These complex interrelationships between the lead firm and other firms and non-​ firm actors should be conceptualized as both structural and relational in nature. A relational view of GPNs requires us to go beyond just a structural analysis of the global economy. It is also insufficient to focus exclusively on domestic organizations and institutions in order to understand global economic change. Such a relational analysis of GPNs focuses on the existence of differential power relations within a network (Dicken et al., 2001; Henderson et al., 2002; Yeung, 2005a; Yeung and Coe, 2015). A GPN reflects relational processes and structures in which, and through which, corporate power is exercised. As noted by Coe et al. (2008a, p. 272), production networks ‘reflect the fundamental structural and relational nature of how production, distribution and consumption of goods and services are—​indeed always have been—​organised. Although they have undoubtedly become far more complex organisationally, as well as far more extensive geographically, production networks are a generic form of economic organisation’. Powerful lead firms are those that drive GPNs and make things happen. Their ability to do so depends on their control of key resources (physical, political, economic, social, and technological). The GPN approach views power as the capacity to exercise that can only be realized through the process of exercising. The control of resources does not inadvertently imply that an actor is powerful until that power is exercised—​such control is only a necessary, but not sufficient, condition for the allocation of power to any actor. In other words, power should be conceived as a relational practice rather than a structural position within a network. With this broader understanding of the logic of (global) production networks and power relations, we can now explore their operationalization at the global and the local spatial scales and their crucial role in spatial economic integration. To begin, many production networks are local in nature because of territorial dynamics well explored in the leading economic-​geographical literature on agglomeration economies, clusters, and industrial districts. As a firm starts its initial operation in value transformation, it is more likely to develop localized linkages with other firms and actors in the same locality or region through territorialized economies such as traded and untraded interdependencies (Storper, 1997,

386   Yeung 2013). Defined as ‘locally serving partially or non-​tradable goods and services’ in Storper (2000, p. 160), this localized dimension of production networks is well known in economic geography and will not be repeated much further here. Suffice to say that the value transformation in these local production networks is strongly influenced by their territorial embeddedness, which refers to the significant role played by places and regions in shaping the unique characteristics of economic actors, such as firms and institutions, in these networks. This territorially based influence reflects the institutional structuring of firms in different business systems and industrial structures (e.g. Silicon Valley, Third Italy, and Baden-​ Württemberg). Institutional structures are defined as enduring and organized sets of relations among ‘prevailing institutions dealing with the constitution and control of key resources such as skills, capital, and legitimacy’ (Whitley, 1999, p.  5). These institutional structures form established systems of economic coordination and control in specific capitalist economies. They inherently shape the logics governing economic decision-​making, actions, and the market processes through establishing and enforcing conventions, values, views, norms, practices, and the so-​called rules of the game. While this shaping is not structurally deterministic, it does leave visible imprints on the nature and organization of firms and their local production networks (e.g. cooperative vs. adversarial relations). Once embedded territorially in these localized institutional structures, firms should be better able to mobilize these endowments to enhance their firm-​specific corporate power and value transformation. More significantly in a world of GPNs, we need to explore how and why these local production relations become globalized over time. In the existing GPN approach, network embeddedness is the main conduit through which local firms and institutions become articulated into GPNs. This network embeddedness can take place through outward or inward flows of direct investment and international trade in any of the following four forms: 1. The internationalization of a local firm to other localities or regions outside its home economy: this inside-​out process of foreign direct investment enables a local firm to embed into GPNs coordinated by lead firms based elsewhere, or to develop into a global lead firm in itself by incorporating other firms and non-​firm actors into its expanding production network. 2. The export of intermediate goods or services by a local firm to other firms outside its home economy: this inside-​out form of international trade in industrial supplies allows a local firm to remain in situ geographically and yet participate in GPNs coordinated by lead firms based elsewhere. 3. The inward investment of foreign firms in the locality or the home region of a local firm: this outside-​in process by either lead firms or suppliers based elsewhere facilitates the localized development of global linkages mediated through these foreign firms. With the participation of these foreign firms, local production networks become progressively globalized over time through outsourcing arrangements with the local subsidiaries of global lead firms and/​or global suppliers. 4. The import of intermediate goods or services by a local firm from dedicated suppliers outside its home economy: this outside-​in mode of network embeddedness takes place when foreign suppliers become articulated into the globalizing production network governed by a locally bounded lead firm.

The Logic of Production Networks    387 Through their embedding in GPNs, firms and non-​firm actors from a particular local or regional economy are connected with other network members irrespective of their geographical origin or local anchoring in particular places. As argued in Henderson et al. (2002, p. 453), ‘It is most notably the “architecture”, durability and stability of these relations, both formal and informal, which determines the agents’ individual network embeddedness (actor–​network embeddedness) as well as the structure and evolution of the GPN as a whole’ (see also Hess, 2004). Economic actors and their embedded relations with other actors in the same network are therefore crucial in determining their collective power and the precise configuration and coordination of GPNs. In this dynamic configuration, these actors draw upon divergent forms of power in order to take on an advantageous position in GPNs that favours their creation, retention, and capture of value. Those economic actors occupying a leading role in their GPNs tend to benefit disproportionately from the value processes associated with the market success of their products and services. Other non-​local partners incorporated into the same network can also benefit from enhanced opportunities to upgrade their operations and to improve their collective returns. In the GPN literature, strategic coupling is arguably the most widely known concept commonly deployed to explain this global–​local articulation of production networks (see Coe et al., 2004; Yeung, 2009, 2015, 2016; Coe and Hess, 2011). In its essence, strategic coupling with GPNs refers to the intentional convergence and articulation of actors in both local economies and GPNs for mutual gains and benefits. It should not be viewed as a static concept resulting in an end-​state articulation of a local economy into GPNs. The very strategic nature of this local–​global articulation, mediated through different actors such as firms and non-​firm institutions, necessitates continual interactions among these actors at different spatial scales. Instead of inherently localized assets and advantages, these interactive effects are the raison d’être of local and regional development in an era of economic globalization (Coe et al., 2004, p. 469). These effects necessitate a dynamic conception of the strategic coupling of local economies with GPNs. In this sense, the concept’s critics have unfortunately misunderstood its changing and multidimensional nature, and misattributed to it a one-​way flow of articulation (i.e. coupling does not allow for decoupling or recoupling). The concept allows us to connect and bridge two critical and yet relatively independent sets of economic dynamics—​territorial dynamics at the local scale and network dynamics at the global scale. By local dynamics, I mean the pre-​existing political and social institutions and economically productive assets that give rise to the unique character and composition of a local economy. As illustrated in the right-​hand side of Figure 20.1, these local institutions and assets are historically and geographically specific such that they cannot be easily reproduced and transformed within a relative short period of time. In other words, we expect a certain degree of mutual path dependency in this unique combination of local institutions and assets that define, albeit not determine, local and regional development trajectories. Meanwhile, network dynamics are much less governed by pre-​existing institutions at the local or even the national level. Instead, they are primarily constituted by economic actors, such as global lead firms, strategic partners, specialized suppliers, industrial and final customers, and so on. Some of these are large transnational corporations, whereas others are national or local firms. While embedded in specific national or regional economies, these economic actors are mostly driven by the competitive logics of seeking cost efficiency, market access and development, financialization and capital gain, and risk minimization through configuring their GPNs (Yeung and Coe, 2015). These GPN logics are therefore

388   Yeung Global Production Networks • Lead firms • Subsidiaries and suppliers • Customers and markets

Dynamic strategic coupling process

Bargaining and cooperation

Local Development • Value capture • Industrial and social upgrading • Sustainable growth trajectory

‘Local’ Institutions • State agencies • Labour organizations • Business associations

Dependency and transformations

Territorialized Assets • Technology and know-how • Industrial organization • Territorialized politics and social relations

Figure  20.1  Strategic Coupling and Global–​ Local Economic Integration Through Production Networks. Source: Adapted from Coe et al. (2004, Figure 1, p. 470).

firm-​and industry-​specific, and do not necessarily align with those in their home origin of local or national economies. In short, GPN dynamics are qualitatively different from localized territorial dynamics. Operating on its own in today’s global economy, each of these two sets of economic dynamics—​territorial and network—​does not produce sufficient thrust to propel local growth and regional change. While local dynamics are necessary for territorialized development to take place, their cumulative effects on local and regional economies can be greatly enhanced and sustained if they interact positively with broader network dynamics at the global scale. In this sense, a theoretical approach focusing on either set of dynamics is inadequate in explaining local growth and developmental change in an interdependent global economy. Most importantly, the positive outcome of these twin motors for local and regional development hinges on their mutual complementarity and dynamic articulation. This is where the concept of strategic coupling provides the most useful analytical purchase by bringing together both dynamics in accounting for local and regional development (see the case of East Asian development in Yeung, 2016). For a local economy to benefit from evolving network dynamics in different global industries, its localized assets must be integrated into complementary GPNs through the dynamic process of strategic coupling. In Figure 20.1, territorialized assets are embodied in local firms and technology institutes that cooperate with global lead firms and their partners and suppliers in different GPNs. This mutual articulation provides the underlying strategic platform for local and regional growth to occur. Strategic coupling is therefore a mutually dependent and constitutive process involving shared interests and cooperation between two or more groups of actors who

The Logic of Production Networks    389 otherwise might not act in tandem for a common strategic objective. It is a dynamic process through which actors in regional economies coordinate, mediate, and arbitrage strategic interests between local actors and their counterparts in GPNs. These trans-​local processes involve both material flows in transactional terms (e.g. equity investment and movement of intermediate or final goods) and non-​material flows (e.g. information, intelligence, and practices). By now, it should be clear that strategic coupling is not a static equilibrium concept, as is sometimes misunderstood by its critics (Bair and Werner, 2011; MacKinnon, 2012; Bair et al., 2013; Horner, 2014). As argued in Coe et al. (2004), strategic coupling process is neither automatic nor always successful; it needs to be unpacked and analysed because it changes over time, and its modus operandi varies in different geographical contexts. Access to the enabling mechanisms and technologies for this coupling may also be highly uneven geographically. In local and regional development, the concept certainly includes analytical possibilities for coupling/​ articulation, decoupling/​ disarticulation, and recoupling/​ re-​articulation (MacKinnon, 2012). The dynamics of the strategic coupling process depend critically on the bargaining and cooperation relationships between local institutions and key actors in GPNs, and on the continual success of local institutions in transforming territorialized assets favourable for sustaining ongoing developmental trajectories. In Table 20.1, I have summarized three major modes of strategic coupling that enable localities and regions in diverse East Asian economies to articulate into GPNs (see Yeung, 2010, 2016; and also Pickles et al., 2015 on East and Central Europe). These development trajectories are fairly variegated, ranging from Japan’s active pursuit of regional equality policies during post-​ war development to the strong focus in South Korea and Taiwan on building up national institutional capacity between the 1970s and the 1990s, the growth pole strategy pursued by Malaysia and Thailand since the early 1980s, and the more recent experimentation of China with regional devolution since the late 1980s. Each of these economies has experimented with one or more modes of strategic coupling in GPNs. Their developmental outcomes are defined by the aggregation of their network positions and value trajectories of local firms and foreign firms in different global industries.

Significance, Controversies, and Prospects for Future Research As a heuristic conceptual framework, the initial GPN approach developed by the Manchester school of economic geographers since the early 2000s has certainly advanced the debate on local and regional development in an era of economic globalization. Building on the earlier conceptions of business networks and commodity chains, this approach has stimulated substantial empirical studies of local and regional development in different national economies and global industries. It has also built significant theoretical bridges with other major research topics in economic geography (e.g. clusters and innovation, and local and regional development) and the wider social sciences (e.g. industrial upgrading, capital–​ labour relations, and developmental trajectories). Still, the GPN approach is inadvertently an unfinished attempt towards developing a truly explanatory theory—​it is more a conceptual apparatus than a coherent theory of the causal mechanisms of global–​local economic

390   Yeung Table 20.1 Strategic Coupling, Global Production Networks, and Local Development Trajectories Modes of strategic coupling International partnership (functional)

Indigenous innovation (organic)

Production platforms (structural)

• Spatial fix

Cost-​capability efficiency

Public subsidies

Lower production costs

• Organizational fix

Vertical specialization New competition and the rise of new lead firms

International outsourcing and subcontracting

• Technological fix

Faster time to market New product and process technologies

Enabling transport technologies

Transactional links, Reverse ‘brain drain’ and business intelligence, technological returnees and market knowledge

Managerial competence and intermediaries

Dynamics of global production networks

Coupling mechanisms • Transnational communities

• Industrial organization Rise of strategic partners and global localization of TNCs

Rise of national champions and new lead firms

SMEs and new industrial spaces

• States and institutions Explicit role and Implicit and explicit policy-​led: upgrading role: strategic industrial of labour, technology, policies and infrastructure

Explicit role but limited influence through fiscal and financial incentives

Local trajectories

Generic local assets and external dependency

Distinctive local Distinctive local assets assets and some local and strong autonomy autonomy

Likelihood of decoupling Medium

Low

High

East Asian examples

Singapore and Taipei-​ Seoul Metropolitan Hsinchu (Taiwan) Area (South Korea), Taipei-​Hsinchu (Taiwan), Singapore, and Yangtze and Pearl River Deltas (China)

Penang and Selangor (Malaysia), Greater Bangkok (Thailand), and Yangtze and Pearl River Deltas (China)

Relevant sectors

Electronics, petrochemicals, finance, transport, and logistics

Electronics, automobiles, apparel, and toys

Electronics, automobiles, transport, and communications

TNC, transnational corporation; SME, small-​and medium-​sized enterprise Sources: Adapted from Yeung (2009, Table 2, p. 338), MacKinnon (2012, Table 1, p. 240), and Horner (2014, Table 1).

The Logic of Production Networks    391 change. This section evaluates the significance and legacy of this initial GPN approach—​ dubbed GPN 1.0—​delves into some of its major controversies, and assesses its prospects and directions for future research. These reflections and discussion should provide a manifesto for the further development of this substantial genre of research in economic geography in the next ten to fifteen years.

Significance of GPN 1.0 When the GPN approach was first proposed in the early 2000s (Dicken et al., 2001; Henderson et al., 2002; Coe et al., 2004), economic geography was dominated primarily by the post-​Fordism debate in understanding urban and regional development (Peck, 2000; Scott, 2000) and what was then termed the cultural turn (Thrift and Olds, 1996; Lee and Wills, 1997; Thrift, 2000). At that time, most conceptions of production networks in economic geography were pitched at the local scale, as the flexible organizational response to the dismantling of vertically integrated Fordist production systems in what Scott (2000, p. 29) called ‘a world of regions’. This highly territorialized conception of production networks was clearly visible in Clark et al.’s (2000, p. 11) introductory chapter to the Oxford Handbook: ‘Notions such as clusters, networks, enclaves, and localised chains of value began with geographers, being a new vocabulary for describing the economic geography of post-​ Fordism and flexible accumulation. Whether deliberately or not, these notions have become incorporated into national policy-​making (witness the recent competitiveness manifesto announced by the UK government) and the development objectives of sub-​national regions’. Meanwhile, most conceptions of business networks under the guise of the cultural turn paid overwhelming analytical attention to their social and cultural determination in places and localities, ignoring the wider power relations and structural conditions underpinning these networks (Yeung, 2003; 2005a). After over a decade of conceptual work and empirical studies under the broad umbrella of GPN 1.0, a significant legacy of the GPN approach can be established in relation to the abovementioned two dominant frontiers of economic–​geographical research. First and foremost, the significance of the GPN literature is predicated on its analytical capacity to connect the geographical dots—​places and regions—​in a world mosaic of capitalist economies. Production networks are no longer conceptualized merely as place-​dependent territorial ensembles; they are now viewed as the key organizational platform bringing together the developmental trajectories of firms and non-​firm actors in different places and regions across the world economy. This approach not only provides an analytical window for economic geographers to examine GPNs as the core backbone of today’s interconnected world economy, but also eschews the narrow and, perhaps, excessive focus on endogeneity and intra-​territorial dynamics in much of the literature on local and regional development. In today’s world of regions, this GPN approach helps reconcile the analytical tension between globalization tendencies and localization economies in leading economic geography research. The conception of globalizing local and regional development through strategic coupling with GPNs (Coe et al., 2004; Yeung, 2009, 2015, 2016; MacKinnon, 2012), for example, offers a concrete approach to studying how such global–​local (dis)articulation can take place. It contributes much to the debate on local and regional development that remains enduring and significant in economic geography through to the 2010s (Pike et al., 2011; Scott, 2012; Storper, 2013; Dixon, 2014).

392   Yeung Secondly, the GPN framework advances an actor-​specific approach to understanding the organizational dynamics and geographical impacts of production networks. While it does not specifically offer a theory of the firm, GPN 1.0 and its associated literature have eschewed the neoclassical conception of the representative firm often adopted in earlier generations of economic-​geographical research prior to the 2000s (see Taylor and Asheim, 2001; Yeung, 2005b). The firm—​be it a lead firm, a strategic partner, or a supplier—​is never conceived as a blackbox transforming factor inputs into economic outputs. As a critical actor constituting and coordinating GPNs through asymmetrical power and embeddedness, the firm in these studies is a relational construct engaging in value activity under the influence and/​or co-​determination of other firms and non-​firm actors (e.g. state institutions, labour unions, standards organizations). These inter-​firm and extra-​firm relations can be cooperative and relational on the one hand, and competitive and captive on the other. Differing quite substantially from the cultural turn in economic geography, this agency of the firm in GPN thinking takes into better account both firm-​specific considerations (e.g. resources and capabilities, corporate decision-​making, and organizational practices and locational strategies) and broader institutional and territorial influences (e.g. financial systems, industrial relations, regulatory constraints, and so on). This relational understanding of firms and their value activity in GPNs represents a more nuanced appreciation of the industrial and economic dynamics embodied in and transmitted through these production networks. Thirdly, GPN research is significant in economic geography because of its integration of several other key research issues, such as the state, labour, finance, regulation, and the environment, that have now populated the subdiscipline. While the initial GPN 1.0 framework already emphasized non-​firm actors in the constitution of production networks, it has taken almost a decade for scholars to integrate these non-​firm actors and institutions more centrally into the diverse analyses of production networks and their geographical repercussions. In the original conceptions of Henderson et al. (2002) and Coe et al. (2004), non-​firm actors such as state institutions, labour organizations, and financial systems were featured, but their empirical research and substantiation was relatively underdeveloped until more recent years. These recent studies of GPNs focus on state and structural inequalities (Yeung, 2014, 2016; Selwyn, 2015; Smith, 2015), labour (Coe and Jordhus-​Lier, 2011; Rainnie et al., 2011; Selwyn, 2012; Coe and Hess, 2013; Herod et al., 2014), finance (Coe et al., 2014; Dörry, 2015), market (Murphy and Schindler, 2011; Murphy, 2012; Ouma, 2015), and the environment (Gregson et al., 2012; Herod et al., 2014; Lepawsky, 2015). This expansive integration of key geographical issues into the GPN approach validates its analytical robustness in tackling the three world referencing issues identified in this New Oxford Handbook: globalization, financial crisis, and the environment. GPNs and its economic–​geographical analysis represent one of the most fruitful avenues for understanding the complex interrelationships between lead firm-​coordinated production activity and global inequalities, financial integration, and environmental change.

Controversies While the legacy of GPN 1.0 is likely to present an enduring and productive theme for future research in economic geography, there are certainly major challenges and

The Logic of Production Networks    393 unresolved tensions confronting this genre of research. The most critical challenge has been the relative absence of a coherent theory of GPNs. Because of its initial grounding partially in the cultural turn of economic geography (e.g. embeddedness and actor–​network theory), the original GPN thinking and framework developed in Dicken et al. (2001), Henderson et al. (2002), and Coe et al. (2004) did not place explanatory causality squarely in the centre of its conceptualization. In doing so, GPN 1.0 in many ways remains an inadequately developed theory of GPNs. Although the initial framework has specified three interrelated conceptual categories of value, embeddedness, and power, it has not explicitly developed and specified the causal mechanisms linking these elements to the dynamic configurations of GPNs. This state of inadequate theory development in GPN theorization has led Hudson (2008), Sunley (2008), and Starosta (2010) to argue critically that existing conceptual approaches are not explanatory and causal enough to provide a coherent theory of GPNs. These epistemological challenges are grounded respectively in their perspectives of cultural political economy, institutionalism, and Marxism. To Sunley (2008, p. 8), ‘Networks are defined in such an elastic manner that they can include virtually anything. One cannot escape the conclusion that such a loose and ubiquitous idea explains everything and nothing’. Such an all-​embracing approach, he adds, is dangerous because ‘The end result is a push toward an economic geography that is immersed in managerial networks and uncritical descriptions of business elites and omniscient firms’ (Sunley, 2008, p. 10). Other economic geographers sympathetic to the GPN approach also point to its relative neglect of the state as a regulator and driver (Smith, 2015; see also Yeung, 2016), and labour as a key condition for the social reproduction of these production networks (Kelly, 2009, 2013; Selwyn, 2012). With hindsight, I believe this critique of weak analytical causality in GPN 1.0 is quite valid. Indeed, the Manchester school of GPN studies was not originally conceived as a school of thought in an epistemological or Kuhnian sense. The initial framework (Dicken et al., 2001; Henderson et al., 2002) or perspective (Coe et al., 2004) was developed as a heuristic device to broaden the then analysis of economic integration in the world economy by such approaches as the global commodity chains (subsequently morphed into GVCs) within and outside economic geography. Its subsequent uptake and noteworthy influence in economic geography and the adjacent social sciences was entirely unexpected by its key proponents. In short, the early conceptualization efforts in developing GPN 1.0 were not intended to be a coherent and comprehensive theory of global economic change. The intended aims of GPN 1.0 to ‘propose a relational view of ’ (Dicken et al., 2001, p. 91), to ‘outline an analytic framework … to understand’ (Henderson et al., 2002, p. 438), and to ‘conceptualise the connections between’ (Coe et al., 2004, p. 468) key network processes at the local and the global scales were rather symptomatic of the prevailing cultural turn in thinking in economic geography that was committed to avoiding the advancement of ‘a totalising framework capable of grasping the myriad complexities of economic globalisation’ (Henderson et al., 2002, p. 438). In fact, this domain of theory development has been captured by the massively growing influence of the GVC framework since Gereffi et al.’s (2005) theory of GVC governance. In their desire for a parsimonious GVC theory ‘to be useful to policymakers’, Gereffi et al. (2005, p. 82) prefer to ‘to create the simplest framework that generates results relevant to real-​world outcomes’. In doing so, they have knowingly and invariably underplayed the role of ‘history, institutions, geographical and social contexts, the evolving rules of the game, and path dependence matter; and many factors [that]

394   Yeung will influence how firms and groups of firms are linked in the global economy’. Causality and theory development, in this simplest and parsimonious form, have invariably been achieved at the expense of the critical connections and interrelationships of key economic–​ geographical issues discussed in the previous subsection. In the international economics literature, this quest for parsimonious models is even more pronounced. These economic models attempt to modify or extend the conventional trade-​in-​finished-​goods theory in international economics. Their explanatory emphasis is generally placed on decreasing communications and transport costs and trade-​related transaction costs that, in turn, facili­ tate the trading of tasks across borders through international production fragmentation and outsourcing arrangements. In addition to this major controversy over causality and explanation, GPN 1.0 faces another significant challenge from within the GPN and the GVC community, which is related to the core idea of strategic coupling in the GPN framework. This concern comes from what Bair and Marion (2011) call a disarticulations perspective that follows on from Bair’s (2008, 2009) earlier world systems-​inspired work on global commodity chains. This critical perspective on commodity chain studies (Bair et  al., 2013)  argues that research should focus as much on periods when links between GPNs and local economies are broken and remade, and the actors and events shaping such transitions, as on periods when the links are positive and productive. In particular, proponents of this disarticulations perspective are critical of ‘the tendency of researchers to pursue the newest production frontier of a particular commodity in order to analyse how a region becomes linked into a chain and how this incorporation impacts local actors’ (Bair and Marion, 2011, p. 989). To them, this inclusionary bias has led to the significant downplay of moments of exclusion in these processes of global–​local economic integration, such as production volatility, precipitous booms and busts, and historical patterns of dis/​investment and dispossession, that are crucial to understanding global inequalities and uneven development (see also MacKinnon, 2012; Horner, 2014). In my view, however, while this internal challenge has some validity, it does not differ fundamentally from the original conception of strategic coupling in GPN 1.0. As clearly argued in Coe et al. (2004) and Yeung (2009, 2015), the strategic coupling of local actors (firms and institutions) with lead firms in GPNs should not be construed as a functionalist argument because this coupling process is never automatic and always successful. Pre-​ existing patterns of local development and their coupling dynamics may conceal important power asymmetries in the bargaining and cooperation relationships between local institutions and actors in GPNs. These asymmetries may lead to significant negative consequences for regional economies, termed variously as the dark side of strategic coupling. As argued by Coe and Hess (2011, p. 134), ‘the embedding of GPNs into regional economies is of course no guarantee of positive developmental outcomes, even if it results in new or enhanced opportunities for value capture at the local level … In other words, although the articulation of regions in GPNs can produce significant economic gains on an aggregate level, in many cases it also causes intra-​regional disarticulations, for instance, through uneven resource allocation and the breakup of existing cultural, social and economic networks and systems’. In this more dynamic form of the GPN approach, local and regional development can be fruitfully thought of in evolutionary terms as being shaped by periods of strategic coupling in sequence with phases of decoupling and subsequent recoupling.

The Logic of Production Networks    395

Towards GPN 2.0: Prospects and Future Directions It should be clear by now that the GPN approach in economic geography has produced a significant stream of fruitful and exciting research for the last fifteen years. But the initial GPN 1.0 is facing some serious and valid challenges that render it increasingly obsolete. The prospect of this arena of economic–​geographical research can only be positive and strong if a new version in the form of GPN 2.0 can be developed. This section outlines some current efforts at building such a new direction in GPN research. An overall assessment is that the GPN 2.0 approach in economic geography is likely to advance theoretical development and empirical richness much more significantly than in the earlier phase. The future prospects of such an analytical approach are bright and productive in four critical dimensions: theory, methodology, empirical research, and policy influence. In terms of theory, GPN 2.0 will represent a major leap in conceptual robustness and advancement over the initial analytical frameworks discussed in this chapter. Such theory development work has now been underway. In particular, Coe and Yeung (2015) have devoted an entire monograph to developing such a coherent theory of GPNs, and Yeung (2016) has put in practice its empirical analysis of East Asian industrial transformation. In this theory, the primary objective is to explain uneven economic development, the ultimate dependent variable, in an interconnected global economy. It provides robust answers to a fundamental research question: how does development take place in different national and regional economies through their participation in value activity organized on the basis of GPNs? These answers can also offer new theoretical insights into why the organization and coordination of GPNs varies significantly across different industries, sectors, and economies, and their implications for economic development. In principle, the GPN theory focuses on the organization, dynamics, and strategies of GPNs and their causal relations with economic development. Specifically, Coe and Yeung (2015) conceptualize the organization of GPNs in terms of their delimitation, origins, intermediation, and territoriality. The key competitive drivers of GPN dynamics, such as optimizing cost-​capability ratios, sustaining market access and development, working with financial discipline, and managing risks, are theorized as the causal dynamics of empirical outcomes. These capitalist dynamics are critical independent variables that explain the different strategies adopted by actors in GPNs. Theorizing these actor-​specific strategies is an indispensable part of the theory because these strategies translate the causal power of the four competitive dynamics into differential value activities that, in turn, produce diverse developmental outcomes. While the theorization of competitive dynamics provides answers to the question of why GPNs form, the conceptual exposition on actor-​level strategies offers answers to the question of how these networks work and operate in different organizational fields and industrial sectors. In a theoretical sense, competitive dynamics in the global economy provide the structural properties of causality and emergence, whereas actor-​specific strategies serve as the corresponding mechanisms for organizing these networks. Taken together in the GPN theory, these dynamics and strategies co-​constitute the causal mechanisms of GPNs, explaining empirical outcomes in terms of economic development, for example firm growth, technological acquisition and innovation, industrial upgrading and sectoral transformation, local and regional development, and so on.

396   Yeung This theory development is a timely response both to the call for more causal and explanatory conceptualization of GPNs in economic geography and the theoretical deficit in the related literature on GVCs. In this latter literature, despite two well-​known edited volumes explicating global commodity chains (Gereffi and Korzeniewicz, 1994; Bair, 2009), none of its key proponents has advanced further theory development in a single book-​length publication. Indeed, in the GVC literature more broadly, the leading conceptual frameworks have been developed through several book chapters (e.g. Gereffi, 1994, 2005) and journal articles (e.g. Humphrey and Schmitz, 2002; Gereffi et al., 2005). In Gereffi et al. (2005, pp. 84–​88), arguably the most influential theory paper, theory development actually takes up no more than five pages. The rest of the paper is devoted to several antecedents of their GVC theory (e.g. global commodity chain work and transaction cost theory of industrial organization) and four sectorial case studies (i.e. bicycles, apparel, fresh vegetables, and electronics). Meanwhile, recent book-​length monographs on GVCs and GPNs have a strong empirical component (Gibbon and Ponte, 2005; Lane and Probert, 2009; Neilson and Pritchard, 2009; Posthuma and Nathan, 2011; Milberg and Winkler, 2013; Murphy and Carmody, 2015; Ouma, 2015; Pickles et al., 2015). But they do not have theory development as the central goal. Methodologically, GPN research must and will go beyond individual case studies and qualitative analysis. While it is true that a vast number of qualitative studies on GPNs have been conducted in economic geography during the last fifteen years, the future genre of such studies is likely to be both qualitative and quantitative. On the one hand, international organizations such as the Organisation for Economic Co-​operation and Development (OECD), World Trade Organization (WTO), and United Nations Conference on Trade and Development (UNCTAD), and Eurostat have compiled more sophisticated quantitative data on value-​added trade and investment. In May 2013, OECD–​WTO released for the first time its Trade in Value Added (TiVA) database.3 At around the same time, UNCTAD launched its UNCTAD–​Eora GVC database (1990–​2010). These quantitative databases will provide very exciting opportunities for breakthrough research into the central role and impact of GPNs in organizing the world economy. While for decades economic geographers have been quite content using empirical data from qualitative interviews and corporate case studies (Schoenberger, 1991; Clark, 1998; Yeung, 2003; Tokatli, 2015), the availability of these quantitative databases is likely to engender the parallel development of more sophisticated techniques for the spatial analysis of GPNs. However, this is not a blanket advocacy for returning to the heydays of the quantitative revolution in economic geography (see Scott, 2000; Barnes, 2001). Rather, this chapter envisages a more purpose-​specific deployment of both quantitative and qualitative methods, in combination with relevant geographical information systems techniques, to illustrate and explain the complex patterns of these production networks and their effects on the global mosaic of regions and places. This recombination of geographical methods will not only bridge the quantitative–​qualitative divide in economic geography, but also build on and further advance the already strong foundation of GPN research grounded in qualitative case studies (see also Hess and Yeung, 2006b; Coe et al., 2010). Empirically, this emerging GPN 2.0 has identified three particular areas in which significant new contributions to the existing empirical knowledge base can be made (Coe and Yeung, 2015). Firstly, more focused empirical studies can be conducted on the organizational dynamics of GPNs and industries. Avoiding the tendency to characterize GVC configurations

The Logic of Production Networks    397 at the industry level, there is a need to understand individual GPNs as lead firm-​based configurations. While they will exhibit industry traits, at the same time there will be considerable variation between GPNs even in the same industry or product category, depending on the nature of the lead firm in terms of its ownership mode, nationality, corporate culture, and strategic disposition. In turn, industries need to be understood as aggregations of these variable GPNs that intersect with each other through overlapping lead firms or key partners/​ suppliers. More empirical work is also needed to connect GPNs with their end users and markets, broadly defined. Many empirical studies tend to identify a lead firm and then work backwards or upstream to explore inter-​firm relationships. Future research should give as much, if not more, consideration to forward or downstream linkages, to reveal the networks of distributors, resellers, retailers, and so on, that connect lead firms to their markets, and to enable consideration of the important recent literature highlighting recycling and the post-​ consumption after-​lives of products and services. Equally, more empirical work is needed to examine the intersections of GPNs in different industries (e.g. the role of finance and logistics firms in the electronics industry or the role of information and communications technology products and services in automobiles, aviation, and banking). These cross-​cutting connections are not peripheral to the operation of GPNs, but rather are fundamental to the value creation, enhancement, and capture dynamics within all of the interconnecting industries. Secondly, the wider political economy of GPNs remains under-​theorized and the empirical research is inadequate (see Yeung, 2016). Yet GPNs are as much political formations as they are economic entities in the countries in which they operate. They necessarily involve control of and struggles over power, resources, and value on an ongoing basis. More empirical work should seek to understand better the extra-​firm or institutional dimensions to GPNs. This will involve detailed consideration of the relationships between the firm and non-​firm actors enrolled into, and/​or shaping from the outside, GPNs in both home and host contexts. In addition to the multi-​level state (e.g. international organizations and local authorities), the other key categories of non-​firm actor are labour, civil-​society organizations, industry associations, consumers, and standards-​setting agencies (the latter may overlap with some or all of the former categories). By foregrounding notions of control, power, struggle, rivalry, and contestation over value and its distribution, future empirical studies will endeavour to take seriously the constitutive role of such actors in GPNs and also move towards more rigorous theorization of domains which are currently all-​too-​often simply bracketed off as institutional context. Thirdly, many empirical studies in the existing literature have taken for granted the initial origin and formation of GPNs, preferring to investigate their internal dynamics only in the post-​establishment phase. There is significant analytical merit in developing a truly evolutionary approach to GPNs that foregrounds the factors underpinning their initial formation and their subsequent reconfigurations (both dimensions will vary considerably across firms, sectors, geographies, and time periods). The empirical aim is to chart the shifting intersections of or episodic shifts in the dynamics of technological change, market development, financialization, and risk mitigation in shaping the formation and development of GPNs. Some of these dynamics will be economy-​wide, some will be sector specific, and some will be distinctive to particular GPNs; untangling these complexities remains an important empirical challenge.

398   Yeung Finally, the bright prospects of future GPN research are underscored by the tremendous interests in public policies at all levels—​from the international policy community to the national and local organizations—​to plug into and benefit from GPNs. The dynamics of global production have recently received very significant policy attention in major international organizations, such as UNCTAD, OECD, WTO, International Labor Organization, World Bank, Eurostat, and so on. Each of these has produced a range of widely circulated official reports between 2011 and 2014.4 These organizations are primarily interested in the impact of GPNs on such critically important topics as economic development, technological innovation, economic and social upgrading, national competitiveness, industrial and corporate change, entrepreneurship and business strategy, investment patterns, global governance, shifting consumption patterns, global environmental change, and so on. For example, UNCTAD’s (2013, pp. 175–​176) World Investment Report 2013 contains the most comprehensive set of policy frameworks for promoting strategic coupling with GPNs (but not using that exact terminology, of course). In particular, it has identified the following key policy challenges for economic development in an interconnected world economy organized through the extensive presence of GPNs: 1. How to gain access and connect local firms to GPNs. 2. How to maximize the development benefits from participation in GPNs. 3. How to ensure that opportunities to industrial and social upgrading in GPNs are realized. 4. How to mitigate the risks involved in participation in GPNs. 5. How to align and synergize trade and investment policies in a world in which the two are inextricably intertwined. As one might imagine, these fairly major policy challenges are presented mostly at the national level as if the entire country can be plugged into GPNs, and the existing development policies can be reworked to stimulate such strategic coupling. To this policy effect, UNCTAD (2013, p. 175) recommends that ‘active promotion of GVCs and GVC-​led development strategies imply the encouragement and provision of support to economic activities aimed at generating exports in fragmented and geographically dispersed industry value chains, based on a narrower set of endowments and competitive advantages. And they imply active policies to encourage learning from GVC activities in which a country is present, to support the process of upgrading towards higher value-added activities and diversifying into higher value added chains’. In principle, all of these policy issues can be explained and understood from the perspective of GPN 2.0. As economic geographers, we can and should engage critically with the policy recommendations by various international organizations in relation to increasing participation in GPNs. As a critique of this favourable policy uptake at the international level, Fernández (2015, p. 212) argues that ‘Capitalising on the loss of prestige of the neoliberal tools and policies inspired by the WC [Washington Consensus] dogma, the GVC approach has gained increasing presence in the supranational agenda, providing new theoretical inputs for many assistance programs, financial projects, institutional advisers, and institutional workshops that a few years ago were outstandingly committed to self-​regulative market theory’. Such a critical view of the mobilization of GVC studies by international organizations and national governments is also found in recent work by geographers, such as Starosta (2010), Glassman (2011), and Neilson (2014).

The Logic of Production Networks    399

Conclusion Taking an embedded chronological approach, this chapter has outlined and assessed the evolving trajectory of GPN research in economic geography over the last fifteen years. While GPN 1.0 represents a significant epistemological shift towards a relational view of production networks and global–​local economic change, this shift is not yet paradigmatic in the Kuhnian sense owing to insufficient micro-​foundation and theory development. Nevertheless, the field has matured and evolved into an exciting arena for future research. It is well poised to develop new and innovative theory, richer empirical analysis, and higher policy relevance in the next decade or so. Indeed, we can now witness the early success in its business model of exporting to outside geography as a discipline. Avoiding intra-​disciplinary navel-​gazing and always looking out for inter-​disciplinary possibilities, GPN 1.0 in economic geography has already developed emerging integration and cross-​ fertilization with other social science disciplines, such as economic and development sociology, development studies, international political economy, regional studies, international business studies, and international economics. Looking forward, the next generation of GPN 2.0 research will likely capitalize on the unexpected success of its predecessor and set an ambitious agenda for economic geography to tackle the key challenges in a deeply interconnected world economy.

Notes 1. There is another important strand of literature in international economics that examines production fragmentation, outsourcing and offshoring, and trading tasks in tradable goods (Feenstra, 1998; Arndt and Kierzkowski, 2001; Antràs and Helpman, 2004; Grossman and Rossi-​Hansberg, 2008; Antràs and Chor, 2013; Baldwin and Venables, 2013; Milberg and Winkler, 2013; Koopman et al., 2014). But given my emphasis here on the GPN literature in economic geography, I will not attempt to engage with this body of work in international economics. Such an engagement can be found in Coe and Yeung (2015, Chapter 6). 2. Interestingly, some of the key topics in these statements on GPNs were mentioned in several chapters in the first edition of the Oxford Handbook—​inter-​firm business networks and firm-​place relationships (Dicken, 2000), clusters, and new inter-​organizational forms (Audretsch, 2000; Porter, 2000), and global commodity chains (Storper, 2000) and global innovation networks (Hotz-​Hart, 2000). But they are not developed in-​depth in any of these chapters. 3. http://​stats.oecd.org (last accessed 5 July 2017). 4. For a selection of these policy documents, see Cattaneo et  al. (2010b), OECD (2011), WTO and IDE-​JETRO (2011), Elms and Low (2013), UNCTAD (2013), and Gereffi and Luo (2014). In this policy discussion, I also draw upon personal experience in conducting conceptual training in global production networks and capacity-​building sessions for government policymakers and regulatory practitioners from a large range of developing economies in East, South, and South East Asia. These seminars and workshops were organized under the auspices of macro-​regional development organizations (e.g. the Asian Development Bank, ASEAN Secretariat, World Free Zones Organization, or Pacific

400   Yeung Economic Cooperation Council) or national governments (e.g. the Malaysian Institute for Economic Research).

References Antràs, P. and Chor, D. (2013). ‘Organizing the global value chain’. Econometrica 81: 2127–​2204. Antràs, P. and Helpman, E. (2004). ‘Global sourcing’. Journal of Political Economy 112: 552–​580. Arndt, S.W. and Kierzkowski, H. (eds) (2001). Fragmentation: New Production Patterns in the World Economy (Oxford: Oxford University Press). Audretsch, D.B. (2000). ‘Corporate Form and Spatial Form’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 333–​352 (Oxford: Oxford University Press). Bair, J. (2008). ‘Analysing global economic organization:  embedded networks and global chains compared’. Economy and Society 37: 339–​364. Bair, J. (ed.) (2009). Frontiers of Commodity Chain Research (Stanford, CA: Stanford University Press). Bair, J. and Werner, M. (2011). ‘Commodity chains and the uneven geographies of global capitalism: a disarticulations perspective’. Environment and Planning A 43: 988–​997. Bair, J., Berndt, C., Boeckler, M., and Werner, M. (2013). ‘Dis/​articulating producers, markets, and regions: new directions in critical studies of commodity chains’. Environment and Planning A 45: 2544–​2552. Baldwin, R. and Venables, A.J. (2013). ‘Spiders and snakes: offshoring and agglomeration in the global economy’. Journal of International Economics 90: 245–​254. Barnes, T.J. (2001). ‘Retheorizing economic geography: from the quantitative revolution to the “cultural turn” ’. Annals of the Association of American Geographers 91: 546–​565. Bathelt, H. (2006). ‘Geographies of production: growth regimes in spatial perspective 3—​towards a relational view of economic action and policy’. Progress in Human Geography 30: 223–​236. Cattaneo, O., Gereffi, G., and Staritz, C. (2010a). ‘Global Value Chains in a Postcrisis World: Resilience, Consolidation, and Shifting End Markets’ in O. Cattaneo, G. Gereffi, and C. Staritz (eds) Global Value Chains in a Postcrisis World: A Development Perspective, pp. 3–​20 (Washington, DC: World Bank). Cattaneo, O., Gereffi, G., and Staritz, C. (eds) (2010b). Global Value Chains in a Postcrisis World: A Development Perspective (Washington, DC: World Bank). Clark, G.L. (1998). ‘Stylized facts and close dialogue: methodology in economic geography’. Annals of the Association of American Geographers 88: 73–​87. Clark, G.L., Feldman, M.A., and Gertler, M.S. (2000). ‘Economic Geography: Transition and Growth’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 3–​17 (Oxford: Oxford University Press). Coe, N.M. (2012). ‘Geographies of production II: A global production network A-​Z’. Progress in Human Geography 36: 389–​402. Coe, N.M. and Hess, M. (2011). ‘Local and Regional Development:  A  Global Production Network Approach’ in A. Pike, A. Rodríguez-​Pose, and J. Tomaney (eds) Handbook of Local and Regional Development, pp. 128–​138 (London: Routledge). Coe, N.M. and Hess, M. (eds) (2013). ‘Theme issue on “Global production networks, labour and development” ’. Geoforum 44: 4–​92. Coe, N.M. and Jordhus-​Lier, D.C. (2011). ‘Constrained agency? Re-​evaluating the geographies of labour’. Progress in Human Geography 35: 211–​233.

The Logic of Production Networks    401 Coe, N.M. and Yeung, H. W.-​c. (2015). Global Production Networks:  Theorizing Economic Development in an Interconnected World (Oxford: Oxford University Press). Coe, N., Dicken, P., Hess, M., and Yeung, H. W.-​c. (2010). ‘Making connections: global production networks and world city networks’. Global Networks 10: 138–​149. Coe, N., Dicken, P., and Hess, M. (2008a). ‘Global production networks: realizing the potential’. Journal of Economic Geography 8: 271–​295. Coe, N., Hess, M., and Dicken, P. (eds) (2008b). ‘Theme issue on global production networks: debates and challenges’. Journal of Economic Geography 8: 267–​440. Coe, N., Hess, M., Yeung, H. W.-​c., Dicken, P., and Henderson, J. (2004). ‘“Globalizing” regional development: a global production networks perspective’. Transactions of the Institute of British Geographers, New Series 29: 468–​484. Coe, N.M., Lai, K., and Wójcik, D. (2014). ‘Integrating finance into global production networks’. Regional Studies 48: 761–​777. Cooke, P.N. and Morgan, K. (1993). ‘The network paradigm: new departures in corporate and regional development’. Environment and Planning D: Society and Space 11: 543–​564. Dicken, P. (1994). ‘Global-​local tensions:  firms and states in the global space-​economy’. Economic Geography 70: 101–​128. Dicken, P. (2000). ‘Places and Flows: Situating International Investment’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 275–​291 (Oxford: Oxford University Press). Dicken, P. (2015). Global Shift: Mapping the Changing Contours of the World Economy (7th edition) (London: SAGE). Dicken, P. and Thrift, N. (1992). ‘The organization of production and the production of organization:  why business enterprises matter in the study of geographical industrialization’. Transactions of the Institute of British Geographer 17: 279–​291. Dicken, P., Kelly, P., Olds, K., and Yeung, H. W.-​c. (2001). ‘Chains and networks, territories and scales: towards an analytical framework for the global economy’. Global Networks 1: 89–​112. Dixon, A.D. (2014). The New Geography of Capitalism: Firms, Finance, and Society (Oxford: Oxford University Press). Dörry, S. (2015). ‘Strategic nodes in investment fund global production networks: the example of the financial centre Luxembourg’. Journal of Economic Geography 15: 797–​814. Elms, D.K. and Low, P. (eds) (2013). Global Value Chains in a Changing World (Geneva: World Trade Organization). Ernst, D. and Kim, L. (2002). ‘Global production networks, knowledge diffusion and local capability formation’. Research Policy 31: 1417–​1429. Feenstra, R.C. (1998). ‘Integration of trade and disintegration of production in the global economy’. Journal of Economic Perspectives 12: 31–​50. Fernández, V.R. (2015). ‘Global value chains in global political networks: tool for development or neoliberal device?’ Review of Radical Political Economics 47: 209–​230. Gereffi, G. (1994). ‘The Organization of Buyer-​driven Global Commodity Chains: How U.S. Retailers Shape Overseas Production Networks’, in G. Gereffi and M. Korzeniewicz (eds) Commodity Chains and Global Capitalism, pp. 95–​122 (Westport, CT: Praeger). Gereffi, G. (1999). ‘International trade and industrial upgrading in the apparel commodity chain’. Journal of International Economics 48: 37–​70. Gereffi, G. (2005). ‘The Global EConomy: Organization, Governance, and Development’, in N.J. Smelser and R. Swedberg (eds) The Handbook of Economic Sociology (2nd edition), pp. 160–​182 (Princeton, NJ: Princeton University Press).

402   Yeung Gereffi, G. (2014). ‘Global value chains in a post-​Washington Consensus world’. Review of International Political Economy 21: 9–​37. Gereffi, G. and Korzeniewicz, M. (eds) (1994). Commodity Chains and Global Capitalism (Westport, CT: Praeger). Gereffi, G. and Luo, X. (2014). ‘Risk and opportunities of participation in global value chains’. Background paper to the 2014 World Development Report, Policy Research Working Paper 6847 (Washington, DC: The World Bank). Gereffi, G., Humphrey, J., and Sturgeon, T. (2005). ‘The governance of global value chains’. Review of International Political Economy 12: 78–​104. Gibbon, P. and Ponte, S. (2005). Trading Down: America, Value Chains and the Global Economy (Philadelphia, PA: Temple University Press). Gibbon, P., Bair, J., and Ponte, S. (eds) (2008). ‘Special issue on governing global value chains’. Economy and Society 37: 315–​459. Glassman, J. (2011). ‘The geo-​political economy of global production networks’. Geography Compass 5: 154–​164. Grabher, G. (ed.) (1993). The Embedded Firm:  The Socio-​Economics of Industrial Networks (London: Routledge). Gregson, N., Crang, M., Ahamed, F.U., Akter, N., Ferdous, R., Foisal, S., and Hudson, R. (2012). ‘Territorial agglomeration and industrial symbiosis: Sitakunda-​Bhatiary, Bangladesh, as a secondary processing complex’. Economic Geography 88: 37–​58. Grossman, G. and Rossi-​Hansberg, E. (2008). ‘Trading tasks: a simple theory of offshoring’. American Economic Review 98: 1978–​1997. Henderson, J., Dicken, P., Hess, M., Coe, N., and Yeung, H. W.-​c. (2002). ‘Global production networks and the analysis of economic development’. Review of International Political Economy 9: 436–​464. Herod, A., Pickren, G., Rainnie, A., and McGrath-​Champ, S. (2014). ‘Global destruction networks, labour and waste’. Journal of Economic Geography V14: 421–​441. Hess, M. (2004). ‘“Spatial” relationships? Towards a reconceptualization of embeddedness’. Progress in Human Geography 28: 165–​186. Hess, M. and Yeung, H.W.-​c (eds) (2006a). ‘Theme issue on global production networks’. Environment and Planning A 38: 1193–​1305. Hess, M. and Yeung, H.W.-​c. (2006b). ‘Whither global production networks in economic geography? Past, present and future’. Environment and Planning A 38: 1193–​1204. Horner, R. (2014). ‘Strategic decoupling, recoupling and global production networks: India’s pharmaceutical industry’. Journal of Economic Geography 14: 1117–​1140. Hotz-​Hart, B. (2000). ‘Innovation Networks, Regions, and Globalization’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 432–​454 (Oxford: Oxford University Press). Hudson, R. (2008). ‘Cultural political economy meets global production networks: a productive meeting?’ Journal of Economic Geography 8: 421–​440. Hughes, A. (2001). ‘Global commodity networks, ethical trade and governmentality: organizing business responsibility in the Kenyan cut flower industry’. Transactions of the Institute of British Geographers 26: 390–406. Hughes, A. and Reimer, S. (eds) (2003). Geographies of Commodity Chains (London: Pearson). Humphrey, J. (1995). ‘Industrial reorganization in developing countries: from models to trajectories’. World Development 23: 149–​162. Humphrey, J. and Schmitz, H. (2002). ‘How does insertion in global value chains affect upgrading in industrial clusters?’ Regional Studies 36: 1017–​1027.

The Logic of Production Networks    403 Kelly, P.F. (2009). ‘From global production networks to global reproduction networks: households, migration and regional development in Cavite, Philippines’. Regional Studies 43: 449–​462. Kelly, P.F. (2013). ‘Production networks, place and development: thinking through global production networks in Cavite, Philippines’. Geoforum 44: 82–​92. Koopman, R., Wang, Z., and Wei, S.-​J. (2014). ‘Tracing value-​added and double counting in gross exports’. American Economic Review 104: 459–​494. Lane, C. and Probert, J. (2009). National Capitalisms, Global Production Networks: Fashioning the Value Chain in the UK, US, and Germany (Oxford: Oxford University Press). Lee, R. and Wills, J. (eds) (1997) Geographies of Economies (London: Arnold). Lepawsky, J. (2015). ‘The changing geography of global trade in electronic discards: time to rethink the e-​waste problem’. The Geographical Journal 191: 147–​159. Leslie, D.A. and Reimer, S. (1999). ‘Spatializing commodity chains’. Progress in Human Geography 23: 401–​420. MacKinnon, D. (2012). ‘Beyond strategic coupling: reassessing the firm-​region nexus in global production networks’. Journal of Economic Geography 12: 227–​245. Milberg, W. and Winkler, D. (2013). Outsourcing Economics: Global Value Chains in Capitalist Development (Cambridge: Cambridge University Press). Murphy, J.T. (2012). ‘Global production networks, relational proximity, and the sociospatial dynamics of market internationalization in Bolivia’s wood products sector’. Annals of the Association of American Geographers 102: 208–​233. Murphy, J.T. and Carmody, P. (2015). Africa’s Information Revolution: Technical Regimes and Production Networks in South Africa and Tanzania (Chichester: Wiley). Murphy, J.T. and Schindler, S. (2011). ‘Globalizing development in Bolivia? Alternative networks and value-​capture challenges in the wood products industry’. Journal of Economic Geography 11: 61–​85. Neilson, J. (2014). ‘Value chains, neoliberalism and development practice: the Indonesian experience’. Review of International Political Economy 21: 38–​69. Neilson, J. and Pritchard, B. (2009). Value Chain Struggles: Institutions and Governance in the Plantation Districts of South India (Oxford: Wiley-​Blackwell). Neilson, J., Pritchard, B., and Yeung, H.W.-​c. (eds) (2014). ‘Special issue on global value chains and global production networks in the changing international political economy’. Review of International Political Economy 21: 1–​274. Neilson, J., Pritchard, B., and Yeung, H.W.-​c. (eds) (2015). Global Value Chains and Global Production Networks: Changes in the International Political Economy (London: Routledge). OECD (2011). ‘Global value chains: preliminary evidence and policy issues’ http://​www.oecd. org/​dataoecd/​18/​43/​47945400.pdf (last accessed 12 July 2014). OECD-​WTO-​UNCTAD (2013). ‘Implications of global value chains for trade, investment, development and jobs’. Report prepared for the G-​20 Leaders Summit, September 2013 http://​unctad.org/​en/​PublicationsLibrary/​unctad_​oecd_​wto_​2013d1_​en.pdf (last accessed 12 July 2014). Ouma, S. (2015). Assembling Export Markets. The Making and Unmaking of Global Market Connections in West Africa (Chicester: Wiley). Parrilli, M.D., Nadvi, K., and Yeung, H.W.-​c. (2013). ‘Local and regional development in global value chains, production networks and innovation networks: a comparative review and the challenges for future research’. European Planning Studies 21: 967–​988. Peck, J.A. (2000). ‘Doing Regulation’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 61–​80 (Oxford: Oxford University Press).

404   Yeung Peck, J.A. and Tickell, A.T. (1994). ‘Searching for a New Institutional Fix: The After-​Fordist Crisis and the Global–​Local Disorder’ in A. Amin (ed.) Post-​Fordism: A Reader, pp. 280–​315 (Oxford: Basil Blackwell). Pickles, J., Smith, A., and Begg, R. (2015). Articulations of Capital: Global Production Networks and Regional Transformations (Oxford: Wiley-​Blackwell). Pike, A., Rodríguez-​Pose, A., and Tomaney, J. (eds) (2011). Handbook of Local and Regional Development (London: Routledge). Ponte, S. and Sturgeon, T. (2014). ‘Explaining governance in global value chains: a modular theory-​building effort’. Review of International Political Economy 21: 195–​223. Porter, M.E. (2000). ‘Locations, Clusters, and Company Strategy’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 253–​274 (Oxford: Oxford University Press). Posthuma, A. and Nathan, D. (2011). Labour in Global Production Networks in India (Oxford: Oxford University Press). Rainnie, A., Herod, A., and McGrath-​Champ, S. (2011). ‘Global production networks and labour’. Competition and Change 15: 155–​169. Schoenberger, E. (1991). ‘The corporate interviews as a research method in economic geography’. Professional Geographer 43: 180–​189. Scott, A.J. (2000). ‘Economic Geography: The Great Half-​century’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 18–​44 (Oxford: Oxford University Press). Scott, A.J. (2012). A World In Emergence:  Cities and Regions in the 21st Century (Cheltenham: Edward Elgar). Selwyn, B. (2012). ‘Beyond firm-​centrism:  re-​integrating labour and capitalism into global commodity chain analysis’. Journal of Economic Geography 12: 205–​226. Selwyn, B. (2015). ‘Commodity chains, creative destruction and global inequality: a class analysis’. Journal of Economic Geography 15: 253–​274. Smith, A. (2015). ‘The state, institutional frameworks and the dynamics of capital in global production networks’. Progress in Human Geography 39: 290–​315. Smith, A., Rainnie, A., Dunford, M., Hardy, J., Hudson, R., and Sadler, D. (2002). ‘Networks of value, commodities and regions: reworking divisions of labour in macro-​regional economies’. Progress in Human Geography 26: 41–​63. Starosta, G. (2010). ‘Global commodity chains and the Marxian law of value’. Antipode 42: 433–465. Storper, M. (1997). The Regional World:  Territorial Development in a Global Economy (New York: Guilford Press). Storper, M. (2000). ‘Globalization, Localization and Trade’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 146–​ 165 (Oxford: Oxford University Press). Storper, M. (2013). Keys to the City: How Economics, Institutions, Social Interaction, and Politics Shape Development (Princeton, NJ: Princeton University Press). Sturgeon, T.J. (2002). ‘Modular production networks:  a new American model of industrial organization’. Industrial and Corporate Change 11: 451–​496. Sunley, P. (2008). ‘Relational economic geography: a partial understanding or a new paradigm?’ Economic Geography 84: 1–​26. Swyngedouw, E.A. (1992). ‘The Mammon Quest. “Glocalisation”, Interspatial Competition and the Monetary Order: The Construction of New Scales’ in M. Dunford and G. Kafkalas (eds)

The Logic of Production Networks    405 Cities and Regions in the New Europe: The Global-​Local Interplay and Spatial Development Strategies, pp. 39–​67 (London: Belhaven). Swyngedouw, E.A. (1997). ‘Neither Global nor Local: “Glocalization” and the Politics of Scale’ in K.R. Cox (ed.) Spaces of Globalization: Reasserting the Power of the Local, pp. 137–​166 (New York: Guilford). Taylor, M. and Asheim, B.T. (2001). ‘The concept of the firm in economic geography’. Economic Geography 77: 315–​328. Thrift, N. (2000). ‘Pandora’s Box? Cultural Geographies of Economies’ in G.L. Clark, M.A. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 689–​704 (Oxford: Oxford University Press). Thrift, N. and Olds, K. (1996). ‘Refiguring the economic in economic geography’. Progress in Human Geography 20: 311–​337. Tickell, A.T. and Peck, J.A. (1995). ‘Social regulation after-​Fordism: regulation theory, neoliberalism and the global-​local nexus’. Economy and Society 24: 357–​386. Tokatli, N. (2015). ‘Single-​firm case studies in economic geography:  some methodological reflections on the case of Zara’. Journal of Economic Geography 15: 631–​647. UNCTAD (2013). World Investment Report 2013: Global Value Chains: Investment and Trade for Development (New York: United Nations). Whitley, R. (1999). Divergent Capitalisms:  The Social Structuring and Change of Business Systems (New York: Oxford University Press). World Bank (2013). World Development Report 2014: Risk and Opportunity: Managing Risk for Development (Washington, DC: World Bank). WTO and IDE-​JETRO (2011). Trade Patterns and Global Value Chains in East Asia: From Trade in Goods to Trade in Tasks (Geneva and Tokyo: World Trade Organization and Institute of Developing Economies). Yeung, H.W.-​c. (1994). ‘Critical reviews of geographical perspectives on business organisations and the organisation of production: towards a network approach’. Progress in Human Geography 18: 460–​490. Yeung, H.W.-​c. (1998). ‘Capital, state and space: contesting the borderless world’. Transactions of the Institute of British Geographers 23: 291–​309. Yeung, H.W.-​c. (2000). ‘Organising “the firm” in industrial geography I: networks, institutions and regional development’. Progress in Human Geography 24: 301–​315. Yeung, H.W.-​c. (2002). ‘The limits to globalization theory: a geographic perspective on global economic change’. Economic Geography 78: 285–​305. Yeung, H.W.-​c. (2003). ‘Practicing new economic geographies: a methodological examination’. Annals of the Association of American Geographers 93: 442–462. Yeung, H.W.-​c. (2005a). ‘Rethinking relational economic geography’. Transactions of the Institute of British Geographers 30: 37–​51. Yeung, H.W.-​c. (2005b). ‘The firm as social networks: an organizational perspective’. Growth and Change 36: 307–​328. Yeung, H.W.-​ c. (2008). ‘Perspectives on Inter-​ organizational Relations in Economic Geography’ in S. Cropper, M. Ebers, C. Huxham, and P. Smith Ring (eds) The Oxford Handbook of Inter-​Organizational Relations, pp. 473–​501 (Oxford: Oxford University Press). Yeung, H. W.-​c. (2009). ‘Regional development and the competitive dynamics of global production networks: an East Asian perspective’. Regional Studies 43: 325–​351. Yeung, H. W.-​c. (ed.) (2010). Globalizing Regional Development in East Asia: Production Networks, Clusters, and Entrepreneurship (London: Routledge).

406   Yeung Yeung, H.W.-​c. (2014). ‘Governing the market in a globalizing era: developmental states, global production networks, and inter-​firm dynamics in East Asia’. Review of International Political Economy 21: 70–​101. Yeung, H.W.-​c. (2015). ‘Regional development in the global economy: a dynamic perspective of strategic coupling in global production networks’. Regional Science Policy & Practice 7: 1–​23. Yeung, H.W.-​c. (2016). ‘Strategic coupling: East Asian industrial transformation in the new global economy’, Cornell Studies in Political Economy Series (Ithaca, NY: Cornell University Press). Yeung, H.W.-​c. and Coe, N.M. (2015). ‘Toward a dynamic theory of global production networks’. Economic Geography 91: 29–​58.

Chapter 21

Gl obal Sou rc i ng of Bu si ne s s Pro cesses:  H i story, Effects , a nd Fu tu re T re nd s Stephan Manning, Marcus M. Larsen, and Chacko G. Kannothra Introduction The disintermediation and global sourcing of both administrative and more knowledge-​ intensive business processes—​a trend also referred to as ‘offshoring’ (Manning et  al., 2008)—​has been one of the most significant trends across industries and countries in the last few decades (UNCTAD, 2005; Kenney et al., 2009). It has not only changed the boundaries of the firm and the way firms perform their corporate functions (Sako, 2006) and organize innovation (Massini and Miozzo, 2012), but also how productive capabilities are distributed across geographies (Manning, 2013). It has brought about a new industry—​global business services—​and various new business models (Manning et al., 2015). It has led to fears of massive job losses in advanced economies (Blinder, 2006) and hopes for boosts of employment and development in emerging economies (Dossani and Kenney, 2007). Finally, it has inspired a rich stream of research across disciplines (Kenney et al., 2009). In the following, we provide a selective overview of the current debates and trends in global sourcing of business processes. We draw on research from multiple domains—​ international business, management, information systems, and economic geography—​to match the complexity and cross-​disciplinary importance of the phenomenon. While we apply a number of relevant theoretical angles, such as interdependency theory, transaction cost economics, and co-​evolutionary and institutional views, we stay close to the empirical phenomenon itself. Eventually, we invite readers to further explore the various debates and perspectives introduced here. The final section elaborates on future research questions which we regard important in moving research on global sourcing forward.

408   Manning et al.

Empirical Scope Many agree that global sourcing of business processes has become a mainstream practice. Business processes are typically defined as ‘structured, measured set[s]‌of activities designed to produce a specific output for a particular customer or market’ (Davenport, 1993, p. 5). While some processes may be specific to particular products, many are part of corporate functions supporting organizations across product lines (Sako, 2006). In particular large firms, but increasingly also small and mid-​size firms, mostly from advanced economies, engage in sourcing business processes from abroad in support of domestic or global operations (Manning et al., 2008). Such processes include information technology (IT) processes, such as server farm management and IT infrastructure, administrative human resource and legal processes, call centres, finance and accounting, product development functions, such as engineering support and software development, and analytical services (Lewin and Peeters, 2006; Manning et al., 2008). The global sourcing trend has grown rapidly in recent years. Whereas in 2000, only 10 per cent of US firms engaged in global sourcing, by 2007 the number had risen to 50 per cent (Lewin and Couto, 2007). Western European firms followed more recently, and today firms from Australia, Asia, Latin America, and even Africa engage in sourcing business processes globally (Manning et al., 2017). Firms source business processes mostly from developing countries. US firms have offshored mainly to India (50% of projects), Latin America (11%), China (9%), and other Asian countries (11%), notably Philippines (Manning, 2013). In India, around three million people work in the IT and business process outsourcing (BPO) sector (Sharma, 2015). The Philippines counts around 1.2 million BPO workers (Magkilat, 2015). Many firms from continental Europe initially focused on Eastern European locations, but, facing saturated labour markets, they have also gradually started sourcing from Asian locations. Whereas in the past most client firms would set up wholly owned captive centres for sourcing projects, today a large share is taken up by specialized external service providers, many of whom are based in the US or India, such as Accenture, Wipro, Infosys, and IBM Global Services (Couto et al., 2008). According to recent estimates by the Indian outsourcing association NASSCOM (2015), the total market size for outsourcing IT and business processes has grown rapidly to US$150 billion. Providers have not only learned to offer a variety of services (Athreye, 2005; Ethiraj et al., 2005), but also to set up delivery centres and tap into talent pools in locations across the world (Manning et al., 2015). Thus, firms’ choice of sourcing location is increasingly affected by the availability of providers in these locations. Therefore, clients not only outsource processes, but also delegate location decisions and associated risks to providers.

Drivers and Historical Development The trend of global sourcing of business processes has been driven by several technological, economic, and organizational factors. One key technological driver has been the

Global Sourcing of Business Processes    409 advancement of information and communication technology (ICT) (Mithas & Whitaker, 2007). On the one hand, the installment of high-​speed transatlantic fibre-​optic cables (Metters and Verma, 2008)  along with improved telecommunication infrastructure in developing countries, such as India (Dossani and Kenney, 2007), has rapidly reduced the cost and increased the capacity of long-​distance communication. On the other hand, advanced ICT has facilitated the digitization of processes and tasks, especially those whose ‘information intensity’ is high (Apte and Mason, 1995). This has led many scholars to argue that those tasks whose information intensity is high and whose performance needs neither physical presence nor personal face-​to-​face interaction with clients can, in principle, be disaggregated and sourced from separate locations (Apte and Mason, 1995; Blinder, 2006). Interestingly, this is particularly true for ‘knowledge work’, which is traditionally performed by skilled professionals (Sinha and Van de Ven, 2005). Not surprisingly, early offshoring experiments focused on IT, engineering, and software services, later followed by analytical work and product design. Beside these technological factors, one key economic driver has been the perceived labour cost advantages of sourcing from developing countries (Lewin and Peeters, 2006). This includes the increasing availability of qualified and potentially lower-​cost science and engineering graduates (Manning et al., 2008; Lewin et al., 2009). As the relative number of available young professionals in developing versus developed countries has been increasing rapidly in recent years (Freeman, 2006), sourcing high-​skilled work from India, China, and other emerging economies has become increasingly attractive—​even beyond initial cost advantages. Also, specialized service providers have become increasingly available and are able to take on numerous tasks and generate not only cost advantages, but also speed up processes for clients (Couto et al., 2008; Manning et al., 2015). One key facilitating factor has been the ability of service providers to ‘commoditize’ processes, that is, to make processes less firm-​, product-​, and industry-​specific (Davenport, 2005) and thus generate specialization advantages vis-​à-​vis client firms (Athreye, 2005; Ethiraj et al., 2005). Finally, aside from exploiting cost and specialization advantages, many client firms generate co-​location advantages, for example by bundling processes from across divisions in single locations, and creating synergies with expanding into new markets, for example by setting up regional headquarters. Finally, there have been important organizational factors partially explaining the trend of global sourcing in recent years. In particular, growing firm-​level experiences with global sourcing have accelerated recent sourcing decisions (Jensen, 2009; Gospel and Sako, 2010). For example, the 2006 decision of Cisco Systems to establish their second headquarters and global innovation centre in Bangalore, India, arguably built on a history of offshoring experiences in India prior to that investment (Cisco, 2016). Likewise, research suggests that, quite independent of the growing service provider industry, many firms have accumulated global sourcing experience and capabilities through experimenting with setting up global captive centres (Manning et al., 2008). These experiments have triggered investments into organizational capabilities, such as communicating between remote locations, that have facilitated future sourcing decisions and led to the development of global operational capacities (Jensen, 2009; Manning, 2014). Relatedly, prior studies suggest that early experiences of lead firms, such as Microsoft and Motorola, in India and other offshoring destinations have generated trust among followers in setting up their own operations in the same locations (Reddy, 1997; Patibandla and

410   Manning et al. Petersen, 2002). As a result, enclaves of foreign firms have established in particular offshore locations, such as German engineering firms that followed industry leaders in offshoring engineering work to Eastern Europe to tap into the growing pool of engineering graduates in these countries (Manning et al., 2012). Subsequent waves of foreign direct investment in particular locations have co-​evolved with the emergence of talent pools and provider capabilities which, in conjunction, have led to the emergence of ‘knowledge services clusters’ (Manning, 2013), which continue to attract global sourcing projects today. These drivers have taken effect over the course of around three decades. Arguably, the foundation for today’s global sourcing trend was laid in the early 1980s when a number of (mostly) US client firms began experimenting with offshore sourcing. One such experiment was Caribbean Data Services—​a captive offshoring unit created by American Airlines in Barbados in 1983 in order to collect revenue data from used airline tickets and, later, to handle insurance paperwork (Metters and Verma, 2008). Both used to be flown to Barbados, and data entry results were sent back electronically via satellite. The main driver for this experiment was labour cost savings of 50 per cent compared with doing the same work in the US (Tulsa, OK). Another parallel experiment was the setting up of captive centres by US insurance companies in Ireland to process health insurance claims. A number of US lead firms, such as Microsoft, General Electric (GE), Hewlett Packard, and Texas Instruments, further explored opportunities of setting up shared services centres and joint-​venture contracts for software development and other services in India in the mid-​1980s, thereby pushing the development of standards, graduate training programmes, and telecommunication infrastructure in that country (Patibandla and Petersen, 2002). By the mid-​1990s, US firms had created offshore jobs for up to 10,000 workers in the Caribbean, 3000 workers in Ireland, and 20,000 workers in Asia, mostly India (Wilson, 1995). From the mid-​1990s to the late 2000s, global sourcing experienced rapid growth stimulated by successful experiments of lead firms, supportive export promotion, and infrastructure policies of host governments, in particular India, and the emergence of specialized service providers, especially in India, such as Wipro, TCS, and Infosys. The latter would gain experience through contracts with lead firms coming to India, such as GE (Metters and Verma, 2008), as well as on-​site low-​level software service work, including coding, testing, and support, at US client firms in the early 1990s. This also allowed providers to gradually develop transferable client-​serving capabilities (Ethiraj et al., 2005). With an improving telecommunication infrastructure in India, growing trust of client firms in offshore resources, and an increasing capacity of service providers, more and more projects would be located and performed offshore. The growth trend was further promoted by two events: firstly, many client firms needed a massive number of IT professionals in 1999 to fix the Y2K bug, benefitting mainly Indian firms (Metters and Verma, 2008). Secondly, the stagnating number of Science and Engineering graduates in the US combined with an unexpected cut in H1B visas in 2003 created a temporary shortage of science and engineering professionals, which arguably led many US firms to search for personnel offshore (Lewin et al., 2009). As a result, client firms gained offshore experience and developed capabilities allowing them to expand offshore operations. Since the late 2000s, we have been experiencing a consolidation and transformation of the global sourcing trend. The increasing commoditization of processes has, on the one hand, led to a rapid expansion of the outsourcing business, but, on the other hand, increased

Global Sourcing of Business Processes    411 pressure on margins for service providers (Couto et al., 2008; Manning et al., 2015). Clients have gained experience with global sourcing, and expectations of service quality have increased. This has led to increasing investment of providers into new business models and service innovation. Firstly, providers have started to increasingly expand service operations and set up service delivery units across the world to better meet client demands. Secondly, providers are experimenting with using applied artificial intelligence to process client data more efficiently and make their services more attractive. Thirdly, new market niches have emerged, most notably so-​called ‘impact sourcing’, which serves the growing interest of clients in combining outsourcing decisions with corporate social responsibility considerations. We discuss these trends further below.

Management of Global Sourcing Relationships While many firms took initial global sourcing decisions for opportunistic reasons, such as saving costs, many have expanded both internal and external sourcing relationships, following the well-​known mantra ‘went for price, stayed for quality’ (Dossani and Kenney, 2007). Yet, managing global sourcing relations is not a straightforward task and may turn out more difficult than originally anticipated. For example, not only may global sourcing provoke internal resistance in the domestic organization, but it may also impede productivity due to lack of trust, status differences between domestic and foreign units, and poor communication and interaction in business process delivery (e.g. Levina and Vaast, 2008; Vlaar et al., 2008). Employees with cultural and language differences at geographically dispersed locations, whether offshored or outsourced, are refrained from informal face-​to-​face coordination, and are required to rely on inferior technology-​based coordination mechanisms (Storper and Venables, 2004; Manning et  al., 2013). Also, Larsen et al. (2013) find that growing complexity of global sourcing—​both with respect to the organizational configuration and the tasks being sourced—​produces ‘hidden costs’ as it undermines decision makers’ ability to estimate accurately the costs of sourcing activities abroad. As such, the complexities and uncertainties resulting from the relocation of processes may affect the ability of firms to reintegrate and perform processes effectively across locations, thus affecting their ability to achieve anticipated performance outcomes (Larsen et al., 2013). Managing the increased complexity of operations across locations may require larger investments into coordination, and firms must thus engage in the coordination of international operational networks across geographies, cultures, and different institutional systems (Niederman et al. 2006; Kumar et al., 2009; Srikanth and Puranam, 2011). As business processes, to various extents, interdepend with other processes and activities, research has stressed that firms need to devise appropriate mechanisms of communication and knowledge transfer—​ranging from often cost-​intensive personnel rotation and other informal practices, to implementing enhanced videoconferencing and other technologies. In a similar vein, Srikanth and Puranam (2011) argue that firms need to make additional investments in

412   Manning et al. new communication channels, shared training, coaching, and other ‘tacit forms of coordination’, to manage the interdependencies across locations. In managing global sourcing relationships, two decisions are particularly important: choice of governance mode and sourcing location. On the one hand, prior research indicates that choice of governance mode and supplier can be critical for clients in managing global sourcing complexity. Using transaction cost-​economics logic, Griffith et al. (2009) find that the asset specificity and uncertainty of the transaction has a direct impact on whether the business process is implemented internally or through an outsourcing arrangement. However, aside from transaction uncertainty, strategic and operational drivers play a similarly important role, including the ability of providers to drive down costs, provide access to talent and expertise, and speed up service delivery (see e.g. Manning et al., 2015). Also, research suggests that client–​provider relationships tend to sustain over time (Manning et al., 2011), even though switches in suppliers and governance modes may also happen (Petersen et al., 2010). On the other hand, selecting the right sourcing location is a key concern of firms as they build up globally dispersed operations. For example, firms have been found to choose locations with favourable wage differentials, knowledge infrastructure, availability of qualified personnel, and preferable country risks relative to the home country (Bunyaratavej et al., 2008; Doh et al., 2009). Also, firms are more likely to choose locations where they have previous experience (Demirbag and Glaister, 2010) or ethnic ties (Zaheer et al., 2009). Increasingly, however, client firms delegate location choices to international providers who operate global networks of delivery centres on behalf of clients (Manning et al., 2015). Another important concern is the ability of firms to build up and exchange, but also protect knowledge in global sourcing relations. For example, what kind of knowledge is necessary to ‘transfer’ to an outsourcing partner to facilitate an efficient process delivery? Can the communication channels be standardized and formalized without jeopardizing knowledge content (see e.g. Manning et al., 2013)? In this respect, while global sourcing is often portrayed as a learning-​by-​doing and opportunistic process (Maskell et al., 2007; Jensen, 2009; Asmussen et al., 2016), research shows that firms with previous sourcing experience generally display better performance in new sourcing ventures than firms with no or little experience. For example, Hutzschenreuter et al. (2007) argue that firms’ past sourcing experience may influence the range of issues and possibilities that managers consider when making global sourcing decisions. Equally, in a recent simulation study, Asmussen et al. (2016) find that when firms aim to source functions from geographically distant locations, pursuing a strategy based on prior experience is more effective, as it reduces the risk of being overwhelmed by coordination costs after the implementation. Finally, much literature has focused on performance implications of different global sourcing decisions and designs. For example, Larsen (2016) finds that a modular task design reduces hidden costs, whereas ongoing communication has a negative impact. Beside hidden costs (Larsen et al., 2013), research has looked into numerous other performance implications, including corporate financial performance (Mol et al., 2005), cost savings (Lewin and Peeters, 2006), export performance (Bertrand, 2011), and sales growth (Murray et al., 1995)—​and non-​financial performance measures, such as learning and organizational transformation (Maskell et al., 2007; Jensen, 2009; Asmussen et al., 2016) and innovation performance (Nieto and Rodríguez, 2011).

Global Sourcing of Business Processes    413

Emergence of Knowledge Services Clusters The recent global sourcing trend has had profound implications for the geographical distribution of work. Many scholars have wondered whether global sourcing, along with the digitization and commoditization of work, has made the world more ‘flat’, where location advantages become less important (Friedman, 2005; Mithas and Whitaker, 2007), or whether the world remains ‘spiky’ (Florida, 2005; Ghemawat, 2011). This question seems particularly relevant for the global distribution of so-​called knowledge work, including engineering, software development, product design, R & D, and analytical services. We argue that while a larger number of cities and regions participate in providing such work for global clients, the geography of knowledge production is rather ‘spiky’ and dominated by so-​called ‘knowledge services clusters’ (KSCs). Specifically, KSCs can be defined as geographical concentrations of lower-​cost technical and analytical skills serving a rising global demand for commoditized knowledge services (Manning, 2013). Examples for KSCs include Bangalore, Chennai, and Pune for software services (Zaheer et al., 2009; Sonderegger and Taeube, 2010), and Beijing, Sao Paolo, Moscow, and Bucharest for R & D services (e.g. GlobalServices, 2008). Similar to other clusters, KSCs feature geographical agglomerations of firms, labour pools, and institutions that are more or less specialized and interconnected, and that belong to a particular domain (e.g. Giuliani, 2005; Iammarino and McCann, 2006). KSCs also have two specific features (Manning, 2013). Firstly, they develop around business services, such as software development, testing, and computer-​aided design, rather than technologies or products. Secondly, they serve global clients, who are spread across rather than within particular industries. This is because knowledge services are increasingly decoupled from end products and market or industry specifics, that is, they are increasingly commoditized, which generates productivity gains for specialized service providers (Sako, 2006). To some extent, KSCs combine features of both high-​tech clusters (e.g. Silicon Valley, or Route 128 in the US), and low-​cost manufacturing clusters in emerging economies. Like high-​tech clusters, they rely on specialized providers and high-​skilled workers, as well as university programmes that produce such skills. Yet, like low-​cost manufacturing clusters, their existence also relies on significant labour cost advantages, which is why KSCs are mainly found in emerging, rather than advanced economies. Because of this dual nature, KSCs are subject to the ‘ambivalent effect’ of service commoditization (Manning, 2013, see Figure 21.1). On the one hand, increasing commoditization, for example of software and engineering support services, may increase client demand for such services across industries, which, in turn, helps expand markets and generate scale and scope economies for providers in KSCs. This may also explain how a growing number of locations have been able to provide business services to global clients. However, with increasing commoditization, location switching costs decrease for clients as well, as other KSCs may provide similar services and skills, which, in turn, increases competitive pressure on any particular location. In trying to reduce competitive pressure, KSCs may benefit only to some degree from building specificity. Unlike in the case of high-​tech clusters, whose skill sets serve highly specific client demands which allows them to develop a distinct competitive advantage owing to high imitation barriers, in the case of KSCs, high specificity of knowledge services

414   Manning et al. - High volume of transactions - Sufficient distinctiveness

Cluster attractiveness

High

Low

- Lack of global demand - Lack of scale economies

- High cost pressure - Global competition

Low

High Degree of service commoditization

Figure  21.1  Ambivalent Effect of Service Commoditization on Geographical Cluster Growth. Source: Manning (2013).

involves considerable disadvantages. Most importantly, high product or client specificity may lower the applicability of local service capabilities. Unlike high-​tech clusters whose success depends on highly specific expertise in technologies for end users in particular industries, KSCs are selected by clients because they provide more generic, often low-​value adding knowledge services, for example engineering tests, which feed into globally dispersed R & D client operations. Therefore, KSCs are more likely to grow and continuously attract client projects within a global competitive space if the level of service commoditization is ‘medium’. This allows for a sufficiently high volume of transactions and projects, while also generating some distinctiveness to lower the threat of imitation and to increase relocation costs for businesses operating in these clusters. One example of ‘medium’ commoditization is the provision of tech support to clients in the same time zone. While tech support can be highly commoditized, time zone proximity allows more immediate service and narrows down location options for clients demanding such service. Another example is high levels of service capability within a recognized standard system, such as the capability maturity model for software development, which meets standards requirements of clients, yet helps differentiate from locations with lower standards levels (e.g. Arora et al., 2001; Ethiraj et al., 2005). In the emergence of KSCs, linkages to advanced economies have been an important driver (Lorenzen and Mudambi, 2013). Many KSCs initially benefited from foreign direct investment of Western multinational enterprises (Patibandla and Petersen, 2002; Manning et al., 2010). Lead foreign firms often ‘customize’ local business conditions, for example by promoting process standards, building infrastructure, and sponsoring university programmes to produce the talent they need (see e.g. Manning et al., 2012). This has enabled KSCs to develop a strong global orientation, but has also limited their aspirations of becoming a new ‘Silicon Valley’. For example, Manning et al. (2012) describe how a German engineering firm

Global Sourcing of Business Processes    415 has transformed a local university in Romania into a provider of qualified engineering graduates, which has attracted numerous client firms interested in offshoring engineering work since then. At the same time, this firm has prevented the local university from launching more sophisticated R & D projects, which might compete with university alliances back in Germany. In other words, the engagement of multinationals often helps build and embed KSCs in global production networks (see also Humphrey and Schmitz, 2002), but it may also limit or slow down further upgrading aspirations. In addition, diaspora communities and returnee entrepreneurs have played a significant role in building KSCs (Kenney et al., 2013). Oftentimes, diaspora effects kick in after an emerging cluster already provides favourable conditions for employment and entrepreneurship, for example through supportive governmental policies and the arrival of lead multinationals (Kenney et al., 2013). The case of Bangalore is a good example (Lorenzen and Mudambi, 2013). Up to the early 1990s, many Indian science and engineering students and young professionals moved to Silicon Valley; when US visa policies got more restrictive, and conditions in cities like Bangalore became more attractive, many returned home, often to start their own business to serve US clients with whom they had already established relationships. These diaspora entrepreneurs have helped further ‘embed’ KSCs into global production networks (Saxenian, 2005)—​by transferring business models and practices from environments that are familiar to global clients and by adapting business models to specific local context conditions, such as lower cost labour. Recent studies suggest that new diaspora waves, for example of Indians into Africa, and the internationalization of global service providers, have promoted the emergence of new service hubs (PwC, 2011; Manning et al., 2017). Manning (2013) suggests that KSCs are more likely to attract client projects continuously if both globally operating MNCs (clients and/​or providers) and local entrepreneurial providers are located in that cluster. Dominance of either global or local players will lower the attractiveness of a KSC. However, the properties and importance of KSCs will also depend on at least two trends we discuss further in the next section: strategies of internationalization of service providers, and cloud and other technologies affecting the dependence of clients and providers on any one location.

New Trends in Global Sourcing Internationalization of Service Providers and Global Delivery Models One of the most important recent trends in the global outsourcing industry is the internationalization of service providers. For a long time, service providers mainly operated out of one location and occasionally sent on-​site teams to client locations. As services have become more commoditized and competition for global client projects has increased, especially larger providers have begun to establish more permanent delivery centres all over the world (Manning et al., 2015). Accenture, Infosys, and other major providers today have numerous delivery centres globally. In fact, Offshoring Research Network data suggests that over

416   Manning et al. 50 per cent of US providers have built up delivery centres in India, and over 50 per cent of Indian providers have established delivery centres in the US (see e.g. PwC, 2011). The way in which service providers have set up delivery centres not only resembles historical trends in manufacturing, but also shows some unique features. In manufacturing, such as automotive production, location choices of suppliers have to a large extent been explained by so-​called ‘follow-​the-​client’ strategies, in which suppliers typically follow their major clients in their international expansion in order to meet the expectations of clients to develop and maintain highly integrated relationships with their main suppliers (Erramilli and Rao, 1990). Co-​location can lower coordination and transportation costs and also enable better control of supplier performance (Yeung et al., 2006). Also it helps suppliers better match co-​ location advantages of their foreign rivals (Martin et al., 1998). This may partly explain the rationale of many Indian service providers, such as Infosys, to set up consulting units in the US that allow them to better initiate and manage deals with US clients. However, the recent study by Manning et al. (2015) suggests that another major driver for setting up global delivery hubs is the ability to better manage time zone differences and set up what many have called ‘global delivery models’ (GDMs) (Carmel, 2006; Ang and Inkpen, 2008). These enable a ‘service provider to deliver seamless services from an optimized delivery structure that involves resourcing skills and resources’ (Ang and Inkpen, 2008, p. 339). Unlike sales offices, GDMs constitute a globally integrated service delivery system which typically involves multiple centres at globally dispersed locations that contribute to the delivery of particular client services, for example IT system maintenance, call-​centre operations, or software development. GDMs thereby encapsulate two locational components (see Figure 21.2). Firstly, in order to establish GDMs, service providers set up international units that establish time zone proximity to core clients so that timely and efficient coordination and negotiation of orders and tasks can be carried out. While this does not exclude physical proximity with clients, it is not a necessity. Anecdotal evidence suggests that the reason why US or Indian providers have expanded into Central and South America (e.g. Costa Rica in the case of Infosys) or South Africa (in the case of Accenture) is a combination of resource access, language abilities, and time zone proximity to major US or European clients. At the same time, providers set up or maintain units that allow for time zone spread of operations to access various resources and to operate 24/​7. When asked about their new delivery centre in Brno, Czech Republic, Infosys CFO Mohandas Pai describes the approach of his company in the following way: ‘The Brno centre is part of our strategy to build nearshore centres in various parts of the globe. This, along with our large offshore

Clients

Service requests and coordination

Teams that deliver and coordinate tasks with core clients close to client sites

Service (change) orders

Teams that execute tasks with little direct client contact from various time zones

24/7 service delivery

Figure 21.2   Global Delivery Model. Source: Manning et al. (2015).

Global Sourcing of Business Processes    417 centres in India and the centre in China, gives us an expanded global network, allowing proximity to our clients and seamless flow of work on a 24 × 7 basis’ (Infosys, 2007 cited in Manning et al., 2015). Advanced ICT has thus enabled a new form of international expansion and coordination of business service delivery, in which location decisions need to be seen as part of configurations of interrelated client-​serving and back-​office units across time zones. This type of business model innovation may be the onset of a new rationale for internationalizing operations in other sectors as well. For example, whereas production facilities may continue to benefit from co-​location with client sites (Majkgard and Sharma, 1998; Yeung et al., 2006), supporting digitized service operations may follow different global distribution patterns where positioning in particular time zones may become a more important driver of resource allocation. At the same time, the increasing ability of global service providers to obtain relational quasi-​rents by bundling services and building hub-​and-​spoke operations targeting various clients (see also Sako, 2006) may help them take ‘service intermediary’ functions in other business-​to-​business industries as well.

New Technologies and Service Automation As ICT keeps advancing, global sourcing practices keep changing as well. A recent study by the World Economic Forum suggests that around five million administrative and office jobs across major economies will be made redundant by 2020 through advanced technology (World Economic Forum, 2016a). The study emphasizes, in particular, the driving force of mobile Internet applications and cloud technology, big data processing applications, and the ‘Internet of Things’—​the increasing remote accessibility and interconnectedness of physical objects and infrastructure, including transportation, energy supply, buildings, and mobile technology (World Economic Forum 2016b). In addition, artificial intelligence will be increasingly employed to process large amounts of data to operate such systems. This ongoing process is often associated with the ‘Fourth Industrial Revolution’, which marks the ‘fourth’ major technological transition—​from the introduction of water and steam power (first), electric power (second), digitization and automation (third)—​to combining and automating the use of artificial intelligence with the Internet of things and services. Davenport and Iyer (2015) suggest that the trend of service automation will radically impact global sourcing practices. They suggest that ‘automation, which uses algorithms and artificial intelligence to do tasks now done by humans, could reshape the entire IT services and business process outsourcing (BPO) landscape’ because, once set up, automated services may drastically reduce labour costs. This can already be observed in many ITO and BPO domains. For example, whereas twenty years ago many firms ran their own call centres in house, they then started gradually to outsource call c​ entre operations to providers in developing countries thereby benefiting from labour cost arbitrage. Today, however, many firms are transitioning to automated response services based on pre-​recorded scripts for incoming calls (Tufekci, 2015). Similarly, researching for court cases has transitioned from law firms internally processing masses of legal documents by hand to an industry where documents are increasingly analysed by data processing software semi-​automatically (Markoff, 2011).

418   Manning et al. These trends are predicted to have a significant impact on global employment dynamics. On the one hand, as predicted by the World Economic Forum (2016a) report, new service automation technology is likely, at least temporarily, to make human service jobs redundant. Observers predict, in particular, a reduction of jobs in developing countries (Treanor, 2016), specifically those that were previously created to cut costs—​at a time when replacing them with automated services was still too costly. By comparison, in the near future automation may undercut human labour cost. One obvious example is the processing of inbound calls using improved speech recognition and basic algorithms to direct callers to pre-​scripted standard answers. On the other hand, service automation may create new, semi-​skilled jobs to assist the ‘functioning’ of technology. This process has been historically observed whenever new digital and automation technology was introduced into the workplace, such as computerized numerical control machines in the 1970s and 1980s. Critical scholars, in the Marxist tradition, referred to this event as a threat to the profession of machinists. Francis (1986) describes how human labour was ‘reduced’ to monitoring automatic control systems and to trouble-​shooting in case computerized numerical control machines malfunction—​an effect sometimes referred to as ‘deskilling’, as it either replaces or reduces professions to technology ‘assistance’. Similarly, it can be predicted that semi-​automated service technology may continue to create more or less skilled human work to ‘assist’ new systems. For example, new software systems need to be installed, upgraded, and tested, and staff needs to be trained on new systems; new systems will continue to produce errors that need to be fixed and monitored; data processing software will always have restrictions in terms of what data can be processed, which requires ‘pre-​cleaning’ and manual input of ‘dirty data’; automated systems are unlikely to cover the entire workflow, which requires human labour to ‘connect workflows’; and client demands will continue to be negotiated and clarified—​a task left to ‘human labour’.

Inclusive Sourcing Practices: Rural Sourcing and Impact Sourcing The third trend we emphasize here relates to the increasing concern about employment and development effects in global sourcing. In particular, countries around the world have tried to develop an outsourcing industry as a way to promote economic development (Manning, 2013). However, these efforts have typically focused on a certain segment of urban, highly trained professionals, while neglecting less privileged—​for example, rural, unskilled, disadvantaged—​parts of the population. Recently, two related trends have emerged that may promote more inclusive employment and development through global outsourcing jobs. One trend that is mainly driven by the potential to further cut labour costs is so-​called ‘rural sourcing’—​the creation of outsourcing jobs in suburban and rural areas (Lacity et al., 2012). As service providers in urban centres (e.g. Bangalore) experienced rising infrastructure costs and wages, combined with client pressure to further reduce cost, they started exploring the option of moving to smaller cities and rural areas. Moving to rural locations helps lower local competition for talent and reduces operating costs. This has created job opportunities for college graduates and young people outside the main IT clusters like Bangalore (e.g. Kannothra and Manning, 2016).

Global Sourcing of Business Processes    419 In parallel, another trend has emerged many refer to as ‘impact sourcing’ (IS), which partly includes, but also extends beyond, ‘rural sourcing’. Unlike the latter, IS has been mainly driven by development initiatives and concerns for more inclusive employment. In particular, the Rockefeller Foundation has been instrumental in promoting IS—​a new model of global service outsourcing that focuses on providing employment opportunities to disadvantaged groups in society. This includes people in slums and minorities, whose access to education and income is limited, which prevents them from pursuing decent livelihoods and employment opportunities. It also includes people with physical disabilities (e.g. impaired hearing) whose access to regular jobs and careers is severely constrained (Hockerts, 2015). The Rockefeller Foundation (2013) first experimented with IS by sponsoring pilot programmes under the label ‘Digital Jobs Africa’ in Kenya, Ghana, South Africa, Nigeria, Egypt, and Morocco in 2013. The idea was to promote and fund so-​called ‘impact sourcing service providers’ (ISSP) that are profitable while achieving community impact by hiring and training staff from disadvantaged groups. IS service providers thus represent a new form of hybrid business model—​or ‘social entrepreneurship’—​in combining business and social objectives (Battilana and Dorado, 2011; Haigh and Hoffman, 2012). At the same time, it was anticipated that major clients would take an interest in IS as it helps better link outsourcing to corporate social responsibility initiatives (International Association of Outsourcing Professionals, 2012). However, it was equally expected that clients will continue to care mostly about cost and quality, which would pose a challenge to developing IS into a niche market (Bulloch and Long, 2012). Yet, IS has already developed into a successful new outsourcing business model, in particular in sub-​Saharan Africa, but also in India and to some extent in the US (Kannothra et al., 2017; Manning et al., 2017). According to the Avasant Group (2012), the market for IS is expected to grow rapidly and account for around 17 per cent of the global outsourcing industry thereby employing around three million people worldwide by 2020.

Directions for Future Research Following our review, we make some selective recommendations for future research. We focus, in particular, on questions of governance of global sourcing relationships, and geographical location and distribution of sourcing activities. Recommendations are equally driven by changing research agendas and the dynamics of the global sourcing trend itself. Firstly, we encourage future research to pay attention to the emergence of new intermediaries in global outsourcing. In particular, the advancement of ICT and the emergence of new sourcing models may generate incentives for new businesses to develop and offer new, specialized capabilities to both global clients and conventional providers. As business model innovation has become a growing domain of management research in recent years (Chesbrough, 2010), we argue that the global outsourcing industry may be an excellent example of an industry with frequent cycles of business model innovation, driven by the rapid advancement of ICT, global competitive pressure, and growing process commoditization. For example, the rise of Internet-​based sourcing platforms, such as e-​lance and Innocentive, suggest that the global sourcing space is gradually merging with crowdsourcing, open innovation and other Internet-​based sourcing practices (Baldwin and von Hippel, 2011; Fjeldstad et al., 2012; Bayus, 2013). Internet market platforms take intermediary roles

420   Manning et al. in managing and translating client requests for particular services or solutions into marketable transactions. More than conventional service providers, Internet market platforms are able to access globally dispersed pools of providers and problem solvers beyond established geographical clusters. However, this emerging space is also populated by innovation agents, such as Gen3, that specialize in building networks of freelancers they mobilize for particular client projects. In this regard, it will be also interesting to research to what extent regular service providers make extensive use of crowdsourcing and open innovation on behalf of their own global clients. Secondly, as the global outsourcing industry is further professionalizing, another important, yet under-​researched topic is the penetration and effects of standards on business models, governance, and location choices. Standards have become a pervasive part of organizational life. They are typically defines as ‘rules(s) for common and voluntary use’ (Brunsson et al., 2012, p. 616). As mentioned earlier, one growing concern in the global sourcing literature is the role of process standards, such as Capability Maturity Model Integration (CMMI), for attracting global clients (Athreye, 2005; Niosi and Tschang, 2009). A recent study by Larsen and Manning (2015) suggests that the level of adoption of CMMI in a global sourcing destination may lower the otherwise negative effect of institutional distance between home and host country in affecting location choices. Yet, it can be expected that other types of standards are on the rise. For example, with the advent of impact sourcing, it can be predicted that labour and social standards may become an increasingly important consideration for both clients and suppliers, similar to other sectors, such as textiles manufacturing (Reinecke and Donaghey, 2015) and coffee production (Reinecke et al., 2012). A better understanding of social standards in the context of global sourcing of processes may refine our understanding of labour governance in global supply chains (Donaghey et al., 2013). Thirdly, another interesting future research field is the integration of globally dispersed processes. Whereas prior studies have focused a lot on the rationale for process disintermediation and relocation (Mithas and Whitaker, 2007), companies increasingly face the challenge of re-​integrating globally distributed tasks (see e.g. Luo et al., 2012). Again, the service provider industry has been at the forefront of this process, by moving from the provision of independent services to integrated solution. Integration capabilities become important not least because service providers increasingly subcontract various services to specialized providers themselves (see also PwC, 2011). Learning more about process integration across geographical distances may advance long-​lasting research on systems integration and systems integrators, in terms of agents that ‘lead and coordinate from a technological and organizational viewpoint the work of suppliers involved in the network’ (Brusoni et al., 2001, p. 613; see also Hobday et al., 2005). In an organizational system consisting of a number of distributed components and entities, systems integrators thus become the architects that integrate and coordinate the different capabilities and resources of the different actors into a final output. Finally, we suggest future research to pay attention to global sourcing flexibility in terms of the ability of firms to adapt governance and location choices to changing environmental conditions. For example, recent studies suggest that firms increasingly move operations within their global networks in response to changing economic and political conditions in any one location (Manning, 2014; Jensen et al., 2015). This has important implications for both firms and regions. On the one hand, firms develop the capacity to flexibly shift operations

Global Sourcing of Business Processes    421 from one location to another. On the other hand, regions adapt to a reality where, owing to increasing commoditization of processes and standardization of skill sets needed to perform those processes (Davenport, 2005; Manning, 2013), firms adjust local investments and capacities to changing demands in their global network.

References Ang, S. and Inkpen, A. (2008). ‘Cultural intelligence and offshore outsourcing success: a framework of firm-​level intercultural capability.’ Decision Sciences 39: 337–​358. Apte, U. and Mason, R. (1995). ‘Global disaggregation of information-​intensive services.’ Management Science 41: 1250–​1262. Arora, A., Arunachalam, V., Asundi, J., and Fernandes, R. (2001). ‘The Indian software services industry.’ Research Policy 30: 1267–​1287. Asmussen, C.G., Larsen, M.M., and Pedersen, T. (2016). ‘Organizational adaptation in offshoring:  the relative performance of home-​and host-​based learning strategies.’ Organization Science 27: 911–​928. Athreye, S. (2005). ‘The Indian software industry and its evolving service capability’. Industrial and Corporate Change 14: 393–​418. Avasant Group (2012). Incentives and Opportunities for Scaling the ‘Impact Sourcing’ Sector (New York: Rockefeller Foundation). Baldwin, C. and von Hippel, E. (2011). ‘Modeling a paradigm shift: from producer innovation to user and open collaborative innovation’. Organization Science 22: 1399–​1417. Battilana, J. and Dorado, S. (2010). ‘Building sustainable hybrid organizations: the case of commercial microfinance organizations’. Academy of Management Journal 53: 1419–​1440. Bayus, B.L. (2013). ‘Crowdsourcing new product ideas over time: an analysis of the Dell Idea Storm community’. Management Science 59: 226–​244. Bertrand, O. (2011). ‘What goes around, comes around: effects of offshore outsourcing on the export performance of firms’. Journal of International Business Studies 42: 334–​344. Blinder, A. (2006). ‘Offshoring: the next Industrial Revolution? Foreign Affairs 85: 113. Brunsson, N., Rasche, A., and Seidl, D. (2012). ‘The dynamics of standardization: three perspectives on standards in organization studies. Organization Studies 33: 613–​632. Brusoni, S., Prencipe A., and Pavitt K. (2001). ‘Knowledge specialization, organizational coupling, and the boundaries of the firm:  why do firms know more than they make?’ Administrative Science Quarterly 46: 597–​621. Bulloch, G., and Long, J. (2012). ‘Exploring the value proposition of impact sourcing’ https://​ www.rockefellerfoundation.org/​ app/​ uploads/​ E xploring-​ t he-​ Value-​ Proposition-​ for-​ Impact-​for-​Impact-​Sourcing.pdf (last accessed 3 February 2016). Bunyaratavej, K., Hahn, E., and Doh, J. (2008). ‘Multinational investment and host country development:  Location efficiencies for services offshoring’. Journal of World Business 43: 227–​242. Carmel, E. (2006). ‘Building your information systems from the other side of the world: how Infosys manages time differences’. MIS Quarterly Executive 5: 43–​53. Chesbrough, H. (2010). ‘Business model innovation: opportunities and barriers.’ Long Range Planning 43: 354–​363. Cisco (2016). ‘Cisco India overview’ http://​www.cisco.com/​web/​IN/​about/​company_​overview.html (last accessed 3 February 2016).

422   Manning et al. Couto, V., Mani, M., Sehgal, V., Lewin, A., Manning, S., and Russell, J. (2008). Offshoring 2.0:  Contracting Knowledge and Innovation to Expand Global Capabilities (Durham, NC: Booz & Co. and Duke University). Davenport, T. (1993). Process Innovation: Reengineering Work Through Information Technology (Boston, MA: Harvard Business School Press). Davenport, T. (2005). ‘The coming commoditization of processes’. Harvard Business Review 83: 100–​108. Davenport, T. and Iyer, B. (2015). ‘Bringing outsourcing back—​to machines’ The Wall Street Journal http://​blogs.wsj.com/​cio/​2015/​07/​01/​bringing-​outsourcing-​back-​to-​machines/​ (last accessed 3 February 2016). Demirbag, M. and Glaister, K. (2010). ‘Factors determining offshore location choice for R&D projects: a comparative study of developed and emerging regions’. Journal of Management Studies 47: 1534–​1560. Doh, J., Bunyaratavej, K., and Hahn, E. (2009). ‘Separable but not equal: the location determinants of discrete services offshoring activities.’ Journal of International Business Studies 40: 926–​943. Donaghey, Jimmy, Reinecke, Juliane, Niforou, C., and Lawson, B. (2013). ‘From employment relations to consumption relations:  balancing labor governance in global supply chains’. Human Resource Management 53: 229–​252. Dossani, R. and Kenney, M. (2007). ‘The next wave of globalization: relocating service provision to India’. World Development 35: 772–​791. Erramilli, M. and Rao, C. (1990). ‘Choice of foreign market entry modes by service firms: role of market knowledge’. Management International Review 30: 135–​150. Ethiraj, S., Kale, P., Krishnan, M., and Singh, J. (2005). ‘Where do capabilities come from and how do they matter? A study in the software services industry’. Strategic Management Journal 26: 25–​45. Fjeldstad, Ø.D., Snow, C.C., Miles, R.E., and Lettl, C. (2012). ‘The architecture of collaboration’. Strategic Management Journal 33: 734–​750. Florida, R. (2005). ‘The world is spiky: globalization has changed the economic playing field, but hasn’t leveled it’. Atlantic Monthly 296: 48. Francis, A. (1986). New Technology at Work (Oxford: Clarendon Press). Freeman, R. (2006). ‘Does globalization of the scientific/​engineering workforce threaten U.S. economic leadership?’ Innovation Policy and the Economy 6: 123–​157. Friedman, T. (2005). The World is Flat: A Brief History of the 21st Century (London: Penguin). Ghemawat, P. (2011). World 3.0: Global Prosperity and How to Achieve it (Cambridge, MA: Harvard Business Press). Giuliani, E. (2005). ‘Cluster absorptive capacity: why do some clusters forge ahead and others lag behind?’ European Urban and Regional Studies 12: 269–​288. GlobalServices (2008). ‘The top 50 emerging global outsourcing cities’ Global Services Report. Gospel, H. and Sako, M. (2010). ‘The unbundling of corporate functions:  the evolution of shared services and outsourcing in human resource management’. Industrial and Corporate Change 19: 1367–​1396. Griffith, D.A., Harmancioglu, N., and Droge, C. (2009). ‘Governance decisions for the offshore outsourcing of new product development in technology intensive markets’. Journal of World Business 44: 217–​224. Haigh, N. and Hoffman, A. (2012). ‘Hybrid organizations: the next chapter of sustainable business’. Organizational Dynamics 41: 126–​134. Hobday, M., Davies, A., and Prencipe, A. (2005). ‘Systems integration: a core capability of the modern corporation’. Industrial and Corporate Change 14: 1109–​1143.

Global Sourcing of Business Processes    423 Hockerts, K. (2015). ‘A cognitive perspective on the business case for corporate sustainability’. Business Strategy and the Environment 24: 102–​122. Humphrey, J. and Schmitz, H. (2002). ‘How does insertion in global value chains affect upgrading in industrial clusters?’ Regional Studies 36: 1017–​1027. Hutzschenreuter, T., Pedersen, T., and Volberda, H. (2007). ‘The role of path dependency and managerial intentionality: a perspective on international business research’. Journal of International Business Studies 38: 1055–​1068. Iammarino, S. and McCann, P. (2006). ‘The structure and evolution of industrial clusters: transactions, technology and knowledge spillovers’. Research Policy 35: 1018–​1036. International Association of Outsourcing Professionals (2012). ‘Bi-​annual survey of corporate social responsibility in outsourcing’. Corporate Social Responsibility (CSR) Subcommittee of the Advocacy and Outreach Committee https://​www.iaop.org/​download/​download. aspx?ID=1875 (last accessed 4 May 2017). Jensen, P. (2009). ‘A learning perspective on the offshoring of advanced services’. Journal of International Management 15: 181–​193. Jensen, P., Manning, S., and Petersen, B. (2015). ‘Towards a theory of location flexibility in global sourcing’. Copenhagen Business School Working Paper. Kannothra, C.G. and Manning, S. (2016). ‘Impact Sourcing at ServImpact. Managing People, Clients, and Growth’ in Sydow, J., Schuessler, E., and Mueller-​Seitz, G. (eds) Managing Interorganizational Relationships—​Debates and Cases, pp. 139–​146 (Basingstoke: Palgrave Macmillan). Kannothra, C.G., Manning, S., Haigh, N. (2017). ‘How Hybrids Manage Growth and SocialBusiness Tensions in Global Supply Chains: The Case of Impact Sourcing’. Journal of Business Ethics. Forthcoming. Kenney, M., Breznitz, D., and Murphree, M. (2013). ‘Coming back home after the sun rises:  returnee entrepreneurs and growth of high tech industries’. Research Policy 42: 391–​407. Kenney, M., Massini, S., and Murtha, T. (2009). ‘Offshoring administrative and technical work: New fields for understanding the global enterprise’. Journal of International Business Studies 40: 887–​900. Kumar, K., van Fenema, P., and von Glinow, M. (2009). ‘Offshoring and the global distribution of work: implications for task interdependence theory and practice’. Journal of International Business Studies 40: 642–​667. Lacity, M., Rottman, J., and Carmel, E. (2012). Emerging ITO and BPO Markets: Rural Sourcing and Impact Sourcing (Washington, DC: IEEE Computer Society). Larsen, M. (2016). ‘Failing to estimate the costs of offshoring: a study on process performance’. International Business Review 25: 307–​318. Larsen, M. and Manning, S. (2015). ‘Does institutional distance still matter? Industry standards in global sourcing location decisions’. Working paper. Larsen, M., Manning, S., and Pedersen, T. (2013). ‘Uncovering the hidden costs of offshoring: the interplay of complexity, organizational design, and experience’. Strategic Management Journal 34: 533–​552. Levina, N. and Vaast, E. (2008). ‘Innovating or doing as told? Status differences and overlapping boundaries in offshore collaboration’. MIS Quarterly 32: 307–​332. Lewin, A. and Couto, V. (2007). Next Generation Offshoring: The Globalization of Innovation (Durham, NC: Duke University CIBER/​Booz Allen Hamilton). Lewin, A. and Peeters, C. (2006). ‘Offshoring work: business hype or the onset of fundamental transformation?’ Long Range Planning 39: 221–​239.

424   Manning et al. Lewin, A., Massini, S., and Peeters, C. (2009). ‘Why are companies offshoring innovation? The emerging global race for talent’. Journal of International Business Studies 40: 901–​925. Lorenzen, M. and Mudambi, R. (2013). ‘Clusters, connectivity and catch-​up: Bollywood and Bangalore in the global economy’. Journal of Economic Geography 13: 501–​534. Luo, Y., Wang, S., Zheng, Q. and Jayaraman, V. (2012). ‘Task attributes and process integration in business process offshoring: a perspective of service providers from India and China’. Journal of International Business Studies 43: 498–​524. Magkilat, B. (2015). ‘1.2  million BPO workers seen by December 2015’. Manila Bulletin http://​www.mb.com.ph/​1-​2-​million-​bpo-​workers-​seen-​by-​december-​2015/​ (last accessed 2 February 2016). Majkgård, A. and Sharma, D. (1998). ‘Client-​following and market-​seeking strategies in the internationalization of service firms’. Journal of Business-​to-​Business Marketing 4: 1–​41. Manning, S. (2013). ‘New Silicon Valleys or a new species? Commoditization of knowledge work and the rise of knowledge services clusters’. Research Policy 42: 379–​390. Manning, S. (2014). ‘Mitigate, tolerate or relocate? Offshoring challenges, strategic imperatives and resource constraints’. Journal of World Business 49: 522–​535. Manning, S., Hutzschenreuter, T., and Strathmann, A. (2013). ‘Emerging capability or continuous challenge? relocating knowledge work and managing process interfaces’. Industrial and Corporate Change 22: 1159–​1193. Manning, S., Kannothra, C.G., and Wissman-​Weber, N. (2017). ‘The strategic potential of community-based hybrid models: The case of global business services in Africa’. Global Strategy Journal, 7: 125–149. Manning, S., Larsen, M., and Bharati, P. (2015). ‘Global delivery models: the role of talent, speed and time zones in the global outsourcing industry’. Journal of International Business Studies 46: 850–​877. Manning, S., Lewin, A.Y., and Schuerch, M. (2011). ‘The stability of offshore outsourcing relationships: the role of relation specificity and client control’. Management International Review 51: 381–​406. Manning, S., Massini, S., and Lewin, A.Y. (2008). ‘A dynamic perspective on next-​generation offshoring: the global sourcing of science and engineering talent’. Academy of Management Perspectives 22: 35–​54. Manning, S., Ricart, J., Rosatti Rique, M., and Lewin, A. (2010). ‘From blind spots to hotspots: how knowledge services clusters develop and attract foreign investment’. Journal of International Management 16: 369–​382. Manning, S., Sydow, J., and Windeler, A. (2012). ‘Securing access to lower-​cost talent globally:  the dynamics of active embedding and field structuration’. Regional Studies 46: 1201–​1218. Markoff, J. (2011). ‘Armies of expensive lawyers, replaced by cheaper software’. The New York Times, 4 March http://​www.nytimes.com/​2011/​03/​05/​science/​05legal.html (last accessed 3 February 2016). Martin, X., Swaminathan, A., and Mitchell, W. (1998). ‘Organizational evolution in the interorganizational environment: Incentives and constraints on international expansion strategy’. Administrative Science Quarterly 43: 566–​601. Maskell, P., Dick-​Nielsen, J., Pedersen, T., and Petersen, B. (2007). ‘Learning paths to global offshore outsourcing—​from cost reduction to knowledge seeking’. Industry and Innovation 14: 239–​257. Massini, S. and Miozzo, M. (2012). ‘Outsourcing and offshoring of business services: challenges to theory, management and geography of innovation’. Regional Studies 46: 1219–​1242.

Global Sourcing of Business Processes    425 Metters, R. and Verma, R. (2008). ‘History of offshoring knowledge services’. Journal of Operations Management 26: 141–​147. Mithas, S. and Whitaker, J. (2007). ‘Is the world flat or spiky? Information intensity, skills, and global service disaggregation’. Information Systems Research 18: 237–​259. Mol, M., van Tulder, R., and Beije, P. (2005). ‘Antecedents and performance consequences of international outsourcing’. International Business Review 14: 599–​617. Murray, J.Y., Wildt, A.R., and Kotabe, M. (1995). ‘Global sourcing strategies of U.S. subsidiaries of foreign multinationals’. Management International Review 35: 307–​324. NASSCOM (2015). NASSCOM Annual Report 2014-​15 (New Delhi:  National Association of Software and Service Companies). Niederman, F., Kundu, S., and Salas, S. (2006). ‘IT software development offshoring’. Journal of Global Information Management 14: 52–​74. Nieto, M. and Rodríguez, A. (2011). ‘Offshoring of R&D: looking abroad to improve innovation performance’. Journal of International Business Studies 42: 345–​361. Niosi, J. and Tschang, F. (2009). ‘The strategies of Chinese and Indian software multinationals:  implications for internationalization theory’. Industrial and Corporate Change 18: 269–​294. Patibandla, M. and Petersen, B. (2002). ‘Role of transnational corporations in the evolution of a high-​tech industry: the case of India’s software industry’. World Development 30: 1561–​1577. Petersen, B., Welch, L.S., and Benito, G.R. (2010). ‘Managing the internalisation process’. Management International Review 50: 137–​154. PwC (2011). ‘The ever-​changing global service-​provider industry. Key findings for 2010’ https://​www.pwc.com/​us/​en/​increasing-​it-​effectiveness/​assets/​the-​ever-​changing-​global-​ service-​provider-​industry.pdf (last accessed 3 May 2017). Reddy, P. (1997). ‘New trends in globalization of corporate R&D and implications for innovation capability in host countries: a survey from India’. World Development 25: 1821–​1837. Reinecke, J. and Donaghey, J. (2015). ‘After Rana Plaza:  building coalitional power for labour rights between unions and (consumption-​based) social movement organisations’. Organization 22: 720–​740. Reinecke, J., Manning, S., and von Hagen, O. (2012). ‘The emergence of a standards market:  multiplicity of sustainability standards in the global coffee industry’. Organization Studies 33: 791–​814. Rockefeller Foundation (2013). ‘Digital jobs in Africa:  catalyzing inclusive opportunities for youth’ https://​assets.rockefellerfoundation.org/​app/​uploads/​20131217164951/​Catalyzing-​ Inclusive-​Opportunities-​For-​Youth.pdf (last accessed 3 February 2016). Sako, M. (2006). ‘Outsourcing and offshoring: implications for productivity of business services’. Oxford Review of Economic Policy 22: 499–​512. Saxenian, A. (2005). ‘From brain drain to brain circulation: transnational communities and regional upgrading in India and China’. Studies in Comparative International Development 40: 35–​61. Sharma, R. (2015). ‘IT and ITeS (BPO) industry overview in India 2015’ https://​www.rohitsharmalive.com/​it-​and-​ites-​bpo-​industry-​overview-​in-​india-​2015/​ (last accessed 3 May 2017). Sinha, K. and Van de Ven, A. (2005). ‘Designing work within and between organizations’. Organization Science 16: 389–​408. Sonderegger, P. and Taeube, F. (2010). ‘Cluster lifecycle and diaspora effects: evidence from the Indian IT cluster in Bangalore’. Journal of International Management 16: 383–​397. Srikanth, K. and Puranam, P. (2011). ‘Integrating distributed work: comparing task design, communication, and tacit coordination mechanisms’. Strategic Management Journal 32: 849–​875.

426   Manning et al. Storper, M. and Venables, A. (2004). ‘Buzz:  face-​to-​face contact and the urban economy’. Journal of Economic Geography 4: 351–​370. Treanor, J. (2016). ‘Fourth industrial revolution set to benefit richest’ UBS report says’ Guardian, 19 January http://​www.theguardian.com/​business/​2016/​jan/​19/​fourth-​industrial​revolution-​set-​to-​benefit-​richest-​ubs-​report-​says (last accessed 3 February 2016). Tufekci, Z. (2015). ‘The machines are coming’. New York Times, 18 April http://​www.nytimes. com/​2015/​04/​19/​opinion/​sunday/​the-​machines-​are-​coming.html?_​r=0 (last accessed 3 February 2016). UNCTAD (2005). Prospects for Foreign Direct Investment and the Strategies of Transnational Corporations 2005–​2008 (United Nations Report) (Geneva: UNCTAD). Vlaar, P., van Fenema, P., and Tiwari, V. (2008). ‘Co-​creating understanding and value in distributed work: How members of onsite and offshore vendor teams give, make, demand, and break sense’. MIS Quarterly 32: 227–​255. Wilson, M. (1995). ‘The Office Farther Back: Business Services, Productivity, and the Offshore Back Office’ in P. Harker (ed.) The Service Productivity and Quality Challenge, pp. 203–​224 (Boston, MA: Kluwer Academic Publishers). World Economic Forum (2016a). The Future of Jobs Employment, Skills and Workforce Strategy for the Fourth Industrial Revolution (Davos: World Economic Forum). World Economic Forum (2016b). ‘The fourth industrial revolution:  what it means, how to respond’ http://​www.weforum.org/​agenda/​2016/​01/​the-​fourth-​industrial-​revolution-​what-​ it-​means-​and-​how-​to-​respond (last accessed 3 February 2016). Yeung, H., Liu, W., and Dicken, P. (2006). ‘Transnational corporations and network effects of a local manufacturing cluster in mobile telecommunications equipment in China’. World Development 34: 520–​540. Zaheer, S., Lamin, A., and Subramani, M. (2009). ‘Cluster capabilities or ethnic ties? Location choice by foreign and domestic entrants in the services offshoring industry in India’. Journal of International Business Studies 40: 944–​968.

Chapter 22

Towards New E c onomi c Ge o graphies of Reta i l Gl obaliz at i on Neil M. Coe and Neil Wrigley Introduction By the end of the twentieth century, large retailers had replaced large manufacturers as the key organizers of the world economy … the retail revolution should be understood as a … fundamental transformation in the organization of the overall global economy, the transformation that continues to change not only the world of retailing, or even the relative power of retailers and their suppliers, but also the shape of international trade, economic development, product worlds, and consumption practices (Hamilton and Petrovic, 2011, p. 3).

Writing in the first edition of this Handbook in 2000, one of us highlighted the ‘myopic neglect of distribution systems and industries’ in the burgeoning literature on economic globalization (Wrigley, 2000, p. 294). In the subsequent fifteen years, understanding of the significance of retailers within the global economy has been transformed. As Hamilton et al.’s The Market Makers makes clear, there is now widespread recognition across the social sciences of the transformative power of retail capital. This ‘retail revolution’ has two interrelated aspects. On the one hand, the globalization of sourcing by large and powerful retailers has transformed a wide range of agricultural and manufacturing sectors, especially in the domains of agro-​food and consumer goods. In the process, this has facilitated not only new forms of south–​north trade flows but also, and very significantly, south–​south trade. On the other hand, the international expansion of the store networks of leading transnational retailers has played a key role in transforming retail structures, local supply networks, regulatory frameworks, and consumption patterns across a broad range of host economies. These two dimensions have important functional interconnections. Transnational retailers may seek to develop store operations in countries where

428   Coe and Wrigley they have established sourcing hubs. Conversely, the development of local supply capacity driven by investment in store networks may, in turn, generate regional or global sourcing relationships with other parts of the retailer’s operations—​in particular with the retailer’s home market (the Organisation for Economic Co-​operation and Development (OECD) study by Nordås (2008) demonstrating that the stimulation of imports into those markets was likely to be substantial). Taken together, therefore, the intertwined globalization of stores and sourcing underpin the pivotal role played by large retailers in the contemporary global economy. Given that these globalization processes are inherently geographical in nature, what has been the role of economic geographers in studying these dynamics? At the time of writing in the first edition, having already by that time conducted a series of studies on the initial efforts of European food retailers to penetrate the US market (e.g. Wrigley, 1997a, 1997b), Wrigley was essentially a lone geographical voice in a literature on international retailing that was otherwise dominated by scholars from business and management studies. That chapter, however, successfully diagnosed a new phase of retail globalization from the late-​1990s onwards—​incorporating the emerging markets of Latin America, Central and Eastern Europe, and South East and East Asia—​and the research imperatives therein (Wrigley, 2000). This call to arms attracted other economic geographers, and was integral to the subsequent development of an increasingly rich corpus of work encompassing management/​business scholars, economic geographers, agricultural economists and sociologists, among others (see, for overviews, Coe and Wrigley, 2007, 2009). Today’s multidisciplinary literature on retail globalization extends far beyond the traditional focus of international retailing research on firm structures and strategies, and considers the wide range of contested processes of societal and political change that the globalization process entails. Economic geographers have made significant theoretical contributions to this work (Alexander and Doherty, 2010), largely from a relational/​production network perspective, that have helped to frame both the distinctive characteristics of the retail transnational corporation (TNC) and the multifaceted and embedded nature of retail globalization dynamics (e.g. Wrigley et al., 2005). This chapter seeks to do five things against the backdrop of these intertwined real-​world and intellectual trends. We open by providing a quick snapshot of current levels of retail globalization (reflecting store as opposed to sourcing networks, as there is far more publicly available information on the former), highlighting the retailers that have been at the heart of these processes over the last two decades, before briefly reprising the received wisdom concerning the drivers and dynamics of retail globalization from the late 1990s onwards. We then seek to advance current understandings in two ways. Firstly, we review the available evidence on the impacts of recent globalization to highlight that, far from being an inevitable process of rapid retail-​TNC-​led modernization, the outcomes have, in fact, been highly variable and uneven, with profound variations across different national contexts. Secondly, and focusing, in particular, on the period since the global economic crisis of 2007–​08, we characterize a new era of globalized distribution in which economic crisis and profound Internet-​ induced structural shifts in retail industries have changed the dynamics of the process. To conclude, we briefly map out an economic geographical research agenda that reflects these two dimensions.

Towards New Economic Geographies of Retail Globalization    429

The State of Retail Globalization: A Contemporary Snapshot In aggregate sales terms, retail globalization has been dominated by a handful of grocery and general merchandisers that sell a broad assortment of goods; food and drink, non-​food groceries, such as health and beauty products, and non-​groceries such as electrical goods and houseware. With some notable exceptions (e.g. IKEA), the many internationalized specialty and clothing retailers (e.g. H&M) have overall revenues that are dwarfed by the grocery and general merchandise retailers, despite producing those revenues from what are frequently much larger store networks. And it was not until 2015 that the world’s first global pharmacy chain (Walgreens Boots Alliance) emerged, combining an extensive store network—​13,200 stores in eleven countries, with very significant revenues of over US$20 billion per year. Table 22.1 illustrates these points by capturing the top-​twenty transnational retailers in 2013. It should be noted that this is the best-​available listing of this class of retailers. That is to say, a ranking of the largest twenty in 2013 by strictly international revenues, not a ranking by overall revenue, a measure that may simply reflect the scale of the retailer’s domestic market (e.g. the USA). It can be seen that twelve of the top twenty in that year were food and general merchandise retailers—​including the top seven—​and a further three were involved in convenience stores/​general merchandise retailing. Of the fifteen food/​general merchandise/​convenience store retailers in the top twenty, only three, including the exceptional case of Walmart, were from the USA, demonstrating that contrary to some perceptions, it has been Western European retailers that have largely driven processes of retail globalization. While the individual firms in the table had shifted rankings during the preceding decade—​ for example Ahold had slipped down, while Amazon had progressively risen—​in large part the constitution of the top twenty remained fairly stable, with the small cohort of Walmart, Carrefour, Metro, Schwarz, Aldi, Tesco, and Auchan being permanent fixtures at the top of the listing, until a sudden retrenchment of Tesco began in 2013. After some fifteen years of ongoing globalization activity, the scale and scope of transnational retailing had become substantial (see Table 22.1). With only three exceptions, the leading transnational retailers had experienced a rise in their international sales as a percentage of total sales since the late 1990s (one exception being Ahold following its financial scandal of 2003). Many had seen substantial rises in the proportion of international sales of over thirty percentage points, with the net result being that twelve of the top twenty accrued over half their total sales from international markets, compared with just four in 1999. In terms of scale, all of the top twenty were deriving annual revenues of over US$10 billion from international markets, with the top fourteen notching over US$20 billion. In terms of scope, the average number of countries of operation of the top twenty was twenty-​seven, continuing a steady increase since 2000 (in 2000 the top twenty average was 15.4 countries, in 2005 it was 17.4, and in 2010 it was 23.6). The 2013 average, however, is lifted somewhat by the extensive networks of clothing, luxury good, and furniture retailers such as H&M and IKEA; the average number of countries of operation of the fifteen food, general merchandise, and convenience store retailers was 18.5. While these figures are short of the degree of internationalization which characterizes certain manufacturing sectors, the activities of these retailers

430   Coe and Wrigley Table 22.1 Leading Transnational Retailers, Ranked by International Revenue in 2013 Rank

Name of Country company of origin

Type of retailer

International International International revenue (USD, revenue, revenue, millions) 1999 (% 2013 (% of total) of total)

Change in % 1999–​ 2013

No. of countries of operation

1

Walmart

Food and general merchandise

135,201

14

29

+15

28

2

Carrefour France

Food and general merchandise

53,329

38

54

+16

31

3

Metro

Germany

Food and general merchandise

52,787

40

62

+22

32

4

Schwarz

Germany

Food and general merchandise

48,401

20

56

+36

26

5

Aldi

Germany

Food and general merchandise

38,312

33

54

+21

17

6

Tesco

UK

Food and general merchandise

34,431

10

34

+24

13

7

Auchan

France

Food and general merchandise

33,653

19

57

+38

13

8

IKEA

Sweden

Furniture

33,490

92

95

+3

41

USA

9

LVMH

France

Luxury goods

32,455

48

89

+41

76

10

Casino

France

Food and general merchandise

32,025

21

60

+39

26

11

Ahold

NetherlandsFood and general merchandise

28,002

76

66

–​10

12

12

Costco

USA

Food and general merchandise

27,758

19

28

+9

9

13

Amazon

USA

General merchandise

25,185

22

43

+21

11

14

Delhaize

Belgium

Food and general merchandise

22,926

83

78

–​5

11

15

H&M

Sweden

Clothing

16,999

84

96

+8

49

16

Inditex

Spain

Clothing

16,242

48

79

+31

88

17

Seven & I Japan

Convenience stores

15,982

30

27

–​3

18

Towards New Economic Geographies of Retail Globalization    431 Table 22.1 Continued Rank

Name of Country company of origin

Type of retailer

18

Couche-​ Canada Tard

Convenience stores

14,889

0

19

Rewe Combine

Food and general merchandise

13,813

20

The Home USA Depot

Home improvement

11,961

Germany

International International International revenue (USD, revenue, revenue, millions) 1999 (% 2013 (% of total) of total)

Change in % 1999–​ 2013

No. of countries of operation

45

+45

19

20

28

+8

11

4

16

+12

5

Sources: Deloitte (2015), the companies’ annual reports, and Euromonitor (GMID Passport Database).

have long since put an end to the assertion that retailing is essentially a domestic activity, inherently resistant to transnational expansion. Using Deloitte’s annual ‘Global Powers of Retailing’ surveys, we can extend our view beyond the top-​twenty cadre to the world’s largest 250 retailers, with the important caveat that their top 250 is defined in terms of overall, as opposed to international, sales. Again, there are strong globalization trends. Whereas in 2005, international sales accounted for only 14.4 per cent of the total sales of Deloitte’s top 250 retailers, it increased to 22.9 per cent by 2008, and in 2013 it reached 24.2 per cent. Moreover, the average number of countries of operation of these firms increased from 5.9 in 2005 to 10.2 in 2013. Just as ‘grocery’ retailers dominate the listing of leading retail TNCs in Table 22.1, so likewise they dominate Deloitte’s top 250 retailers, accounting for 132 of the top 250 in 2013. What Deloitte call ‘hardline and leisure’ retailers (IKEA, Home Depot, Amazon, Kingfisher, etc.) constituted the next largest group with fifty-​two of the top 250. On average, retailers in this sector operated in more countries than the ‘grocery’ retailers—​an average of 8.7 versus 4.9 in 2013. However, the difference in the percentage of the companies’ total sales, which are derived from international operations, was much narrower at 25.2 per cent versus 23.2 per cent. ‘Fashion goods’ retailers formed the third largest grouping in Deloitte’s top 250, contributing forty-​four of the top 250. Despite being the group with the highest number of both international market operations and international sales per company—​an average of 27.3 countries of operation and 31.6 per cent of sales—​only three had accrued sufficiently large annual international sales to appear in Table 22.1 (LVMH, Inditex, and H&M). Supporting evidence of these subsectoral variations in globalization levels drawn from a study of 323 leading retailers conducted by property company CB Richard Ellis is shown in Table 22.2. Geographical interrogation of these data (see Table 22.3) reiterates the argument that European retailers are the most globalized in terms of both revenues and number of countries: European retailers are shown to gain an average of 38.6 per cent of their revenue from 16.2 non-​domestic markets, while US retailers derive an average of 15.4 per cent of revenues

432   Coe and Wrigley Table 22.2 Differences in Level of Globalization by Retail Sector, 2010–​13 Retail sector (number of sample firms, 2013)

Average number Average number % change in % of firms with of countries in of countries in number of online offering, 2010a 2013 countries, 2012 2010–​13

Value and denim (24)

25.8

25.1

–​2.7

45

Luxury and business fashion (54)

24.2

24.4

0.8

36

Coffee and restaurants (19) 21.8

22.3

2.2

11

Specialist clothing (48)

19.4

19.3

–​0.5

29

Mid-​range fashion (70)

18.9

18.5

–​2.1

30

Other (books, music, pets, etc.) (34)

16.9

17.5

3.5

31

Consumer electronics (16)

16.6

14.6

–​12.0

35

Supermarkets (21)

10.8

10.5

–​2.8

7

8.0

8.2

2.5

20

Home and department stores (48)

aAs the sample shifted from 323 retailers across seventy-​three countries (2010) to 334 retailers

across sixty-​one countries (2013), the data from the two years are not, strictly speaking, directly comparable. Source: CBRE (2011, 2014).

from 8.5 markets. Japanese retailers are the least globalized, operating in an average of four countries and generating just 9.4 per cent of retail revenue in foreign markets.

Intense Globalization, mid-​1990s–​mid-​2 000s The extensive geographies of today’s leading retail TNCs have largely been constructed in the period since the mid-​1990s. While retail globalization has a long history dating back to the late nineteenth century (see Alexander, 1997, for a periodization), the late 1990s undoubtedly saw the initiation of a period of international expansion that was previously unparalleled in terms of scale, scope, and speed (Coe, 2004). The preconditions for the expansion were established in the home markets of these retailers during the 1980s and 1990s with the growing concentration—​albeit at variable rates across different advanced economies—​of retail capital, which saw market structures shift from those dominated by small businesses to ones which accounted for some of the largest firms in the national economy. This growing concentration of capital underpinned a broader shift from ‘supply push’ to ‘demand pull’ distribution systems in which retailers emerged as lead firms across a wide range of supply chains. They were able to use their increasing scale and market power to generate competition among suppliers and thereby counteract the power of large manufacturers, and also to develop their own private-​label brands in direct competition with supplier brands (Wrigley

Towards New Economic Geographies of Retail Globalization    433 Table 22.3 Level of Globalization of Top 250 Global Retailers by Region/​ Country, 2013a % retail Average % single-​ revenue number of country from foreign countries operators operations Top 250

24.4

10.2

Africa/​Middle East

25.1

12.1

0.0

Asia/​Pacific

14.0

5.4

43.6

9.4

3.9

45.2

Other Asia/​Pacific 18.9

7.4

41.7

Japan Europe

34.8

38.6

16.2

22.2

France

43.6

28.6

7.1

Germany

45.4

15.4

5.9

UK

21.5

16.1

21.4

Latin America

22.9

2.3

40.0

North America

14.7

7.8

44.3

15.4

8.5

40.5

USA

aResults reflect top 250 retailers headquartered in each

region/​country. Source: Deloitte (2015).

and Lowe, 2002). Growing concentration and retailer power intersected with the rise of so-​ called ‘lean retailing’ practices (Abernathy et al., 2000) in which integrated logistics and supply-​chain management methods were mobilized to reduce retailer inventories through just-​in-​time delivery systems that harnessed information and communications technologies to connect electronic point-​of-​scale information with the tracking of orders. By the mid-​1990s, these home-​market factors saw leading retailers accruing significant profits, in turn leading to pressure from financial markets to sustain the impressive earnings growth (and therefore equity valuations) of the preceding decade. Many had already established global sourcing strategies at this stage, but international store expansion remained a relatively untapped route. During the late 1990s, however, investment opportunities arose across a range of so-​called ‘emerging markets’—​providing those retailers with the promise of being able to maintain revenue growth in the medium-​to long-​term through expanded store networks. The emerging markets offered several important opportunities in this respect: potentially rapid economic development and rising levels of affluence, consumer spending, and retail sales, in combination with low levels of penetration of Western forms of large store retailing and associated distribution systems. Prior to investment, the majority of retail sales in these markets were usually in the hands of small independent retailers or informal retail channels. Leading international retailers were able to use their

434   Coe and Wrigley scale, lower costs of capital, and advanced distribution and logistics systems to obtain rapid revenue growth and high returns on their investment. Strong organic growth was possible in contexts where licences to develop were relatively easy to obtain, the costs of site acquisition and store construction were low, and existing local retailers were often relatively inefficient and uncompetitive. Initial entry was facilitated by the opening of emerging markets to retail foreign direct investment (FDI) through full or partial market access liberalization. Sometimes the opening of those markets was gradual and incremental, reflecting the inclusion of retail FDI liberalization in bilateral/​multilateral trade agreements and structural adjustment programmes throughout the 1990s. Sometimes, however, in regions where indigenous retail coalitions had been able to slow liberalization moves, a sudden exogenous shock, such as the Asian economic crisis of 1997–​98, was necessary to spur governments to undertake rapid liberalization of their retail sectors. In Thailand, for instance, the International Monetary Fund-​ inspired Foreign Business Act of 1999 increased the access of foreign TNCs to the retail and distribution sector. The net result of these intersecting home and host market trajectories was that by the early-​to-​mid 2000s international retail investment flows encompassed, to a significant level, a wide variety of emerging markets in Latin and Central America (especially Argentina, Brazil, and Chile), East Asia (especially China, Malaysia, South Korea, Taiwan, and Thailand), and Eastern Europe (most notably the Czech Republic, Hungary, Poland, and Slovakia). A strong strand of the literature that formed in parallel with these developments has provided corporate case studies of firms as they have engaged in this international expansion process. Collectively, these studies have allowed some broad conclusions about what retailers did well and what they did poorly during this wave of expansion (Wrigley and Lowe, 2014). In order to succeed, retailers had to ‘learn to be local’: responding to local tastes and consumer dynamics in the host markets they entered, using local managerial talent, and successfully integrating with local supply chains and planning and property systems (Coe and Lee, 2013). However, at the same time, many underestimated the local competition and the abilities of domestic retailers to imitate and respond to new retail practices within their territory (da Rocha and Dib, 2002). In financial terms, retailers understood that they had to invest consistently and substantially to achieve the scale required to succeed in the markets entered, but that in making those levels of investment, they must, above all, maintain the confidence of the capital markets (Wood et al., 2017). The flipside was that sometimes they allowed international investment to drain capital expenditure in core domestic markets, and, more frequently than might be imagined, they lost managerial control of the large dispersed organizations they had become (Wrigley and Currah, 2003). The most successful of these firms were often those that developed hybrid organizational cultures that were able to bridge home and host contexts (Coe and Lee, 2006), although they did not always comprehend the time and resource commitments for the organizational change that accompanied effective globalization. Generally, they were able to exploit their superior sourcing, distribution, and logistics systems (Reardon et al., 2007), but frequently were not able to demonstrate the wider potential and value of these systems in driving up local standards (e.g. food safety and quality). That is to say, while retail TNCs were able to use their scale to drive a wide range of changes in host economies, they were not always able to convince local consumers, suppliers, and regulators of the broader (‘scale for good’) benefits of their presence—​for example, acting as ‘export gateways’ for suppliers to access international markets. Finally, although

Towards New Economic Geographies of Retail Globalization    435 some learned the art of disinvestment and judiciously exiting markets in which they were not making headway (Alexander et al., 2005), for others this proved to be an extremely painful lesson. The case study research over the last fifteen years has also assisted development of theorizations of the retail TNC that are both grounded in wider literatures in economic geography and alive to the distinctive nature of retailing as an economic activity. There are three important dimensions to this conceptual work. Firstly, building upon global production network thinking (Henderson et al., 2002), there has been recognition of how retail TNCs are necessarily embedded and essentially networked (Wrigley et al., 2005). This focuses attention on the necessarily high levels of territorial embeddedness—​in local cultures of consumption, real-​estate and land-​use planning systems, and supply networks—​which retail TNCs must achieve in order to achieve organizational legitimacy within host markets. In turn, it leads to an appreciation of the mutual transformation of host economies by the retail TNCs and, reciprocally, of the retail TNCs themselves as they become more complex organizations through operating large dispersed store and sourcing networks across multiple national contexts. Secondly, research has enabled theorization of how retail TNCs act as learning organizations and generate/​transfer forms of knowledge through their management structures (Currah and Wrigley, 2004). Transnational retailers have established sophisticated ‘top-​ down’ knowledge management structures—​consisting of both IT networks and schemes for social interaction—​in an attempt to capture ‘bottom-​up’ forms of learning in an industry where each store is a potential source of innovation. Thirdly, the importance of the corporate culture of individual retail TNCs has been highlighted (Shackleton, 1998). The globalization strategies of retailers are highly variable across space and time, and, in important ways, are company-​specific. Corporate culture should thus be viewed as heavily influencing the nature of strategic localization and the success (or not) of the globalization process at the firm level.

Uneven Outcomes: The Mosaic of National Retail Markets A significant body of knowledge has thus emerged on the globalization strategies of retail TNCs and their varying levels of success. There may be a tendency, however, in such studies—​and indeed in interpreting the aggregate data presented earlier in this chapter—​to over-​generalize across host economies regarding the level and nature of the impacts of retail globalization. Moving from case studies of individual firms to market-​wide studies of processes of retail change reveals a much more complex picture. Rather than homogenous processes of change, the reality is a highly variegated map of different national retail dynamics. What is perhaps more appropriate to say is that the retail structures in developing/​emerging economies across the world have been transformed by the twin processes of formalization and consolidation. The former relates to the growth of Western/​modern retail formats such as supermarkets, hypermarkets, and convenience stores alongside ‘traditional’ formats such as fresh/​wet markets and small family-​run businesses. The latter describes the accumulation of retail capital among owners of multi-​site modern formats with the outcome that

436   Coe and Wrigley retail structures become less fragmented over time. Reardon et al. (2007) have described in detail the ‘waves’ of retail transformation across emerging markets, tracing four phases from the early 1990s to the late 2000s. The second (mid-​to-​late 1990s) and third (early 2000s) phases were, in particular, driven by the uplift in retail FDI described earlier, while the most recent phase (late 2000s onwards) has involved the initial transformation of retail structures in poorer countries in South Asia, South East Asia and sub-​Saharan Africa. Within the countries affected by such waves, the diffusion of modern retail formats has been well charted, spreading progressively from their original niches in major cities serving the rich and the middle classes, to smaller cities and rural towns where they serve the lower middle classes and working poor. At the same time, these spatial and socio-​economic diffusions are often accompanied by an expansion in the products provided by modern retail, from processed food and non-​food to semi-​processed products to fresh produce. This general account has been invaluable in highlighting the macro-​trends and generalizing across the myriad different markets in which such processes are occurring. There is a danger, however, that it obscures what are actually very uneven processes in terms of intensity and impacts. As Humphrey (2007, p. 434) cautions, ‘the depth and implications of retail transformation in developing countries is still unclear. A transformation is certainly taking place. The literature on the supermarket revolution captures this and highlights its potential implications. However, when a significant new trend is first isolated, it is quite common for the pioneering analyses to over-​generalise both its reach and its impact’. While the formalization and consolidation of retail have, indeed, become ubiquitous processes across most markets, the processes vary significantly along at least three axes. Firstly, speeds of transition have varied significantly across different markets. Secondly, the nature of the transformation process has often been markedly different. And, finally, the role played by foreign capital in the transformation process has shown important variation. Two brief examples of the unevenness of these processes are sufficient to illustrate. Firstly, while formalization may be occurring simultaneously across different markets, alternative explanations are required to understand why, among the formal formats, hypermarkets do best in some markets (e.g. Thailand), supermarkets in others (e.g. the Philippines), and convenience stores in yet others (e.g. Indonesia). Secondly, while in many markets retail TNCs have undeniably driven wider processes of change (e.g. Poland), in some they have essentially contributed in combination with significant local retailers (e.g. South Korea), and in others they have made little or no headway in markets either dominated by local capital (e.g. the Philippines) or in which there are still significant regulatory barriers to entry and operation (e.g. India). As Endo (2013, p. 2) suggests, such uneven outcomes draw attention to the continued importance of local contextual factors: ‘that the results of the same TNC differ depending on the host country it enters clearly suggests that success comes from not only the company’s management strategies but also from the host country’s particular circumstances; in other words the local context surrounding the retail industry in the host country’. While there are well-​established frameworks in the literature for understanding the range of different impacts of retail FDI (e.g. Dawson, 2003; Coe, 2004; Coe and Wrigley 2007), by definition they focus on the central role of the TNC. While a useful starting point, this may lead to overestimation of the importance of foreign capital in certain contexts. The ongoing transitions involve not just retailers, both foreign and domestic, but also huge numbers of consumers and farmers/​suppliers, as well as a wide variety of government institutions. The relationships

Towards New Economic Geographies of Retail Globalization    437 between retailer strategies, political conditions, regulatory frameworks, consumer cultures, and supply network structures serve to construct differentiated national retail markets that, while exhibiting a general tendency towards formalization and consolidation, also show highly significant differences in terms of the pace and nature of change. Hence, it is crucial to understand ‘the complex intersections of processes of retail globalization with evolving national retail and supply network structures, and institutional, regulatory and cultural formations’ (Coe and Bok, 2014, p. 493). In this context, it is now instructive to consider the four main interconnected domains of ‘resistance’ to retail transformation in order to understand more how these differentiated national markets are produced. We draw, in particular, on the various markets of South East Asia for empirical illustration but note related research in Latin America, in particular on Chile (e.g. Bianchi and Ostale, 2006; Bianchi, 2009). The first highly significant domain relates to processes of deregulation and re-​regulation. As noted earlier, retail globalization processes were, in part, driven in the late 1990s by deregulation of retail FDI across a range of emerging markets, allowing for significantly increased investment and ownership by foreign capital in the retail sector. In the period since, however, it has become clear that the initial removal of trade barriers has been superseded by new sets of regulatory barriers specifically designed to manage changing retail environments and protect domestic retailers. Nguyen et al. (2014, p. 378) use the term ‘re-​ regulation’ to denote these subsequent developments, and highlight the ways in which they ‘differentially impact on the operational costs of the multinational retailers and therefore become restrictive in terms of trade and investment’. Re-​regulation is usually driven by controversy over the desirability of retail TNC-​driven change, the perceived impacts on local small retailers, and retailer–​supplier tensions as supply systems are radically and quickly transformed by inward investors. Regulations affecting retailers can be divided into those governing initial entry versus those affecting subsequent operation and those that affect all inward FDI versus those that specifically target transnational retailers. In Vietnam, Nguyen et  al. (2014) show how, despite World Trade Organization accession and related deregulation moves, market entry remains very difficult for retail TNCs owing to the opaque Economic Needs Test with which they must comply. This serves to protect local retailers and pushes retail TNCs down the local partner route. More generally across South East Asia, there has been a shift in host economies towards forms of re-​regulation that have sought to target specifically transnational retailers in the post-​entry phase. Such measures can include equity thresholds, capital requirements, environmental and community/​business impact study requirements, zoning restrictions, building and outlet-​size codes, and store opening restrictions. However, within such trends, considerable country-​to-​country variation can be observed. Mutebi (2007) for example, profiled the different combinations of such measures that were used across Indonesia, Malaysia, and Thailand, with Malaysia being seen to be the most successful at restricting retail TNC growth, in particular through its use of zoning rules and outlet size codes. It is clear from these studies that the precise nature of re-​regulation is shaped by a range of contextual factors, including national politics (e.g. see Brillo, 2014, on the Philippines) and the level and effectiveness of lobbying from domestic retailer and supplier groups. Many studies have now shown how processes of re-​regulation intersect with a second dimension—​namely the variable resistance of two distinctive parts of the existing retail

438   Coe and Wrigley structure. In the modern formal sector, indigenous retailers have rapidly and successfully been able to imitate the organizational innovations and best practices of retail TNCs, while at the same time mobilizing local institutional knowledge and social/​political networks to erode the competitive advantage of retail TNCs. In many cases, this market strength was established before the entry of retail TNCs, such as in Chile, where large indigenous chains were able to fend off the competition of global players, including Ahold and Carrefour (Bianchi and Ostale, 2006). Some of these indigenous retailers have been strong enough to establish themselves as regional players through international expansion, as seen, for example, in the case of Chile’s Falabella (Bianchi, 2009)  or South Africa’s Shoprite (Dakora et  al., 2010). Even in 2012, Deloitte was still reporting that ‘one obstacle that foreign retailers often underestimate when assessing the opportunity in emerging markets is the growing power and sophistication of the local competition, which is often stronger than it may appear’ (Deloitte, 2012, p. 2). Similarly, in the pre-​existing informal retail channels, the persistent demand for food sourced from fresh/​wet markets in cultural contexts where ‘freshness’ is paramount for consumers, combined with the accessibility, low cost, and personalized shopping experience that such markets provide for low-​income consumers (Humphrey, 2007) provides a strong basis for diluting the supposedly irresistible competitive advantages of the retail TNCs. Resistance of indigenous retailers is thus a hugely important dynamic but is again a highly variable one from country to country, depending on a range of factors such as the strength of local retailers at the point of TNC entry, the effectiveness of re-​regulation in protecting local retailers (or not) in the post-​entry context, and how income levels and structures influence the demand for traditional retail channels. The third dimension similarly relates to the resistance of local supply systems to changes initiated by retail TNCs and/​or modernization of the retail system more generally. The highly influential oeuvre of work by Reardon and colleagues (e.g. Reardon et al., 2007) has served to highlight the generic processes of change associated with retail modernization, relating, most centrally, to the establishment of distribution centres and centralized procurement, use of advanced logistics techniques, shortening of supply networks and use of new forms of intermediaries, and the enforcement of private quality and safety standards. And for certain suppliers, successful enrolment in the procurement systems of retail TNCs can, in turn, lead to export opportunities to other parts of the retailer’s network. However, there is considerable unevenness in how these practices play out in reality. In part, this relates to the specifics of the individual commodities concerned (see e.g. Natwidjaja et al., 2014, on mangoes in Indonesia). More widely, the persistence of traditional retail channels such as wet/​fresh markets may also serve to sustain the ‘traditional’ supply networks on which they depend. At the same time, pre-​existing supply structures may be more durable than expected, even when supplying to the modern retail sector. In a comprehensive study of Thailand’s entire distribution system, for instance, Endo (2013) argues that the influence of retail TNCs on supply network restructuring has been exaggerated, and that traditional intermediaries such as wholesalers have proven to be remarkably resilient and adaptive in the face of wider changes. Again, then, supply network dynamics are shaped by contextual factors in particular countries; in the case of Thailand, focusing on the whole ‘mosaic structure’ of income groups and consumers provides a different picture than just considering the urban middle classes. These three dimensions in turn interact with the fourth—​consumer dynamics. In relation to the issues already introduced, consumers in particular contexts may show loyalty

Towards New Economic Geographies of Retail Globalization    439 to local stores/​brands, or may prefer to continue shopping in traditional markets and small stores. More broadly, however, it is well established that consumer preferences and practices vary considerably both across and within different societal contexts. Two brief examples will suffice here. Firstly, the expansion of retailer private-​label or own-​brand products—​an established way for retailers to exert supply network control and build relationships with consumers—​has been slower in some contexts than others. In general, in the markets of South East Asia, where a high precedence is placed on brands and brand loyalty, the expansion of private labels has been slow (see Shannon, 2014, for a discussion of how this relates to the cultural traits of Thai consumers). Secondly, as mentioned earlier, different formal retail formats have had varying levels of success across different markets. The preference for convenience stores in Indonesia, for instance, can partly be explained by the nature of regulatory controls, but also strongly reflects the cultural preferences for convenience and local shopping of large swathes of Indonesian consumers. Overall, it is clear that the four dimensions discussed here—​regulation, local retailers, supply networks, and consumers—​all vary significantly between different country contexts. Most importantly, the intersections of the domains serve to create highly differentiated national retail systems. The forces of retail globalization, although strong, are driving change across host economies in highly uneven and contingent ways rather than in the inexorable way that some narratives suggest.

The Next Phase? Emerging Tendencies in Retail Globalization While it is easy, then, to infer from aggregate data of the type presented earlier in the chapter that retail globalization over the last fifteen years or more has unfolded as a homogenous and consistent process, the previous section has demonstrated that has been far from the case. What we now consider is how the nature of the globalization process has changed in response to developments in the wider competitive landscape. We highlight three phases and dimensions of that changing nature of the process of retail globalization. Firstly, a shift from expansionary dynamics to a period of retrenchment. In this context, by the mid-​2000s, it had become apparent that the intense phase of rapid expansion into new markets was tapering off, shifting instead to a period of competitive ‘shake out’ as retail TNCs struggled with the task of achieving the market scale necessary to justify their investments in emerging economies. As Burt et al. (2008, p. 91) note in their study of Ahold, Carrefour, and Delhaize: ‘since 2002 there are signs of a consolidation process underway, with focus upon “strong” markets and a rigorous review of activities in “weaker” markets leading to divestment in several instances’. As initial retail FDI into those markets started to mature, achieving market scale and sustainable advantage became increasingly vital issues. Leading retail TNCs became less able to justify what has been termed ‘flag-​planting’ investment and focused on achieving and securing market leadership, if necessary within a smaller number and narrower range of markets—​a rationale driven, in part, by the demands of home-​ country financial markets (Wood et al., 2017). This led to increased incidence of strategic market divestment, as seen for instance in the market exits of Walmart and Carrefour from

440   Coe and Wrigley South Korea in 2006, leaving Tesco as the only significant retail TNC in the market until it, in turn, sold its operations in late 2015. Asset swaps provided another strategic means of firms re-​rationalizing their store portfolios—​one well known example in 2005 saw Carrefour exchange fifteen stores in the Czech Republic and Slovakia for six Tesco stores and two development sites in Taiwan. As Table 22.1 demonstrates, while overall proportions of foreign sales continued to grow among leading retail TNCs, albeit at a slower pace from the mid-​2000s onwards than in the late 1990s and early 2000s, concealed within the aggregate trends was an important reshuffling of assets and priorities as firms sought to concentrate on key markets. Secondly, a fundamental reassessment of the logic and practice of retail globalization consequent on global financial and economic crisis. With the onset of the global financial crisis in 2007–​08, however, the retail industries of most countries entered a period of profound and disruptive change that had important implications for globalization dynamics. Leading retail TNCs were faced with collapsing consumer confidence and increasingly difficult competitive conditions in their home markets. In this context, for many of those firms international expansion and development became a lower priority as, under the watchful gaze of the financial markets, they wrestled with these domestic challenges. Thirdly, a reassessment of the strategic possibility of e-​commerce-​based retail globalization. Back in 2000, one of us wrote that the ‘potential threat of the growth of e-​commerce and the emerging non-​store electronic retailers to the elite group of retail TNCs is likely, therefore, to be rather muted’ (Wrigley, 2000, p. 311). For a remarkably long period that prediction held true. More recently, however, after a decade and a half of progressively increasing e-​retail sales in domestic markets, there are signs that Internet retailing is finally emerging as a viable internationalization strategy. In many domestic markets, there is good evidence of the way online shopping has gradually become ‘normalized’. In the UK, for example, Internet sales as a percentage of total retail sales increased significantly from 3.0 per cent in 2007 to 11.0 per cent in 2014. That growth has been unevenly distributed across different retail sectors, with online penetration varying from 80.1 per cent in music and video and 52.9 per cent in books to 5.5 per cent in food and groceries, and the same level in health and beauty. Importantly, however, the Internet is also changing how people interact with ‘bricks and mortar’ stores, with customers increasingly doing extensive online research and price comparisons before making purchases. Overall, it appears that incumbent market leaders (with significant legacy costs) are being increasingly challenged by firms whose market presence was established with a much lower cost base as a result of Internet and mobile technologies. Two recent pieces of industry research allow us to gain more insight into this developing phenomenon. Firstly, Deloitte’s ‘Global Powers of Retailing’ reports have, since 2014, offered an assessment of the ‘e-​50’, the top fifty global retailers according to online business-​ to-​consumer sales—​the top fifteen of which, as measured by 2013 revenues, are shown in Table 22.4. While the list is undoubtedly tentative and there remain real problems in terms of defining and isolating online sales data, it is nonetheless revealing. The first key point to emerge is that three-​quarters of e-​50 (thirty-​seven firms) are also members of the overall top 250 global retailers, including all of the top fifteen in Table 22.4. Five firms—​Amazon, Walmart, Tesco, Casino, and Home Depot—​appear both in Tables 22.1 and 22.4. The second point that emerges is that the majority of the e-​50 retailers (thirty-​nine firms) are multichannel retailers, that is, they have both physical stores and online operations—​with only eleven being online-​only retailers (one of these, Amazon, is a huge and exceptional case with online

Towards New Economic Geographies of Retail Globalization    441 Table 22.4 Top Fifteen e-​Retailers, Ranked by Online Sales, 2013 Rank

Top 250 Name of company sales rank FY13

Country of origin

FY13 e-​commerce sales (USD, millions)

B2C e-​commerce % of total retail revenue

1

15

Amazon.com Inc.

USA

60,903

100.0

2

92

JD.com Inc.

China

10,827

100.0

3

1

Wal-​Mart Stores, Inc.

USA

10,000

2.1

4

46

Apple Inc.

USA

9000

30.8

5

70

Otto (GmbH & Co KG)

Germany

8189

61.3

6

5

Tesco plc

UK

5251

5.3

7

99

Liberty Interactive Corporation

USA

4884

47.4

8

13

Casino Guichard Perrachon S.A.

France

3953

6.2

9

59

Suning Commerce Group Co. Ltd.

China

3536

21.3

10

34

Macy’s, Inc.

USA

3100

11.1

11

2

Costco Wholesale Corporation

USA

3086

2.9

12

25

Best Buy Co., Inc.

USA

3044

7.2

13

117

Home Retail Group plc UK

2907

32.6

14

150

Lojas Americanas S.A.

Brazil

2838

45.4

15

9

The Home Depot, Inc.

USA

2750

3.5

FY, financial year; B2C, business to consumer. Source: Deloitte (2015).

sales of over six times that of the second-​place firm). These observations also confirm that the penetration of non-​store retailers has been relatively limited in terms of aggregate sales at the global level. In terms of domestic market origins of the leading online retailers, half (twenty-​five firms) are US based, a further nineteen are based in Europe, particularly in the UK and France, and only five are from ‘emerging’ markets (four from China, one from Brazil). However, the latter conclusion is clouded by definitional issues. According to Deloitte (2014, p. 27), while ‘Asia has a number of large and rapidly growing e-​commerce companies, online marketplaces tend to serve as the primary e-​commerce model in this region. Companies like Alibaba Group, operator of Taobao (China’s most popular consumer-​to-​consumer marketplace) and Tmall (an open business-​to-​consumer platform), as well as Japan’s Rakuten derive the majority of their revenue as facilitators, not as retailers’ and are therefore excluded from the e-​50 listing. The online sales of the e-​50 expanded rapidly in 2013, averaging growth of 26.6 per cent, and the overall contribution to total sales varied hugely from a level of 2 to 4

442   Coe and Wrigley per cent for food and general merchandise retailers and 8 to 16 per cent for department stores and specialty apparel retailers, through to more than 20 per cent in the case of consumer electronics retailers. The 323-​firm CBRE sample profiled in Table 22.2 also corroborates this variability, with only 7 per cent of supermarkets having an online operation compared with 45 per cent in the area of value and discount clothing. The second piece of evidence we can consider derives from research agency Conlumino. Based on interviews with sixty leading multichannel retailers in the UK drawn from a sample that covered all retail sectors and included retailers accounting for 71 per cent of UK retail sales, the key findings were as follows. Firstly, the global economic crisis appeared to not have dampened the desire to internationalize as much as many commentators had suggested. No less than 74 per cent of the sample were already trading internationally, with only 10 per cent determined to remain resolutely focused on just their domestic market. Secondly, a large majority (79%) believed that online retail had become far more important to international growth than having stores on the ground. Thirdly, online-​based expansion was seen as being much lower risk and more controllable than physical store expansion. Projecting forwards five years, online expansion was seen by members of the sample as likely to be three times as important as any other mode of international growth (stores, franchising, wholesaling, joint ventures, and licensing), with e-​retail being viewed as a low-​cost and low-​risk method of ‘trying out’ international markets and circumventing host economy regulatory requirements relating to the establishment of physical store networks. Significantly, the surveyed firms expressed a clear preference for operating overseas online businesses themselves rather than through a partnership with a local operator and/​or a franchise arrangement—​a sentiment that contrasts with previous research on store based expansion in which a capable local partner was often regarded as key to successful market entry and expansion in emerging economies (e.g. on Tesco in Asia, see Wood et al., 2016). However, there remain important caveats to be considered before any over-​hasty judgement is advanced regarding e-​commerce-​based international expansion. OECD analysis, for example, has consistently shown that despite the ‘normalization’ of online shopping over the last decade, consumers remain notably reluctant to make ‘cross-​border’ online purchases. In the case of UK and Dutch consumers, for example, Figure 22.1 shows that despite 93 and 95 per cent, respectively, of them claiming that they ordered online from national sellers, % 100 33 57 82 80 83 79 82 81 78 85 65 83 87 83 81 72 79 93 88 95 92 93 90 94 98 96 80

Percentage of individuals who ordered online from national sellers

60 40 20

LU X AU T CA N BE L ES T IR L FIN LV A PR T SV N IS L DN K SV K NO R ES P GR C ITA SW E FR A NL D HU N GB R CZ E DE U PO L TU R

0

From partner countries

From the rest of the world

Figure 22.1   Levels of Cross-​border Online Purchases. Source: Adapted from Polder (2015).

Towards New Economic Geographies of Retail Globalization    443 and despite the prevailing ‘trading nation’ traditions of their economies, simultaneously less than 20 per cent of those consumers claimed to have ordered online from cross-​border providers—​that is to say from sellers outside the domestic market. The lack of confidence and trust this implies is clearly a significant but little discussed constraint on the strategic possibility of an emerging era of e-​commerce based retail globalization. Of course, Amazon has demonstrated that it is possible, in part, to ease this constraint by establishing and investing in strong separate ‘country’ identities and ‘fulfilment’ regimes. But the Amazon model is not necessarily the low-​cost method of ‘trying out’ international markets and circumventing host economy regulatory requirements, which advocates of e-​commerce-​based international expansion have in mind.

Conclusion: A Developing Research Agenda Although great strides have been taken over the last fifteen years, research on retailing in economic geography still only accounts for a very small proportion of what we know is a sprawling and pluralistic sub-​discipline. Work on economic globalization still tends to prioritize manufacturing and advanced business services sectors. This is unfortunate, for retail globalization offers a powerful window into several essential characteristics of the evolving global economy: the continuing rise of ‘big buyers’ and demand pull supply networks; shifting patterns of global demand reflecting the rapid expansion of middle-​class consumers outside the Global North; complex mosaics of national-​level regulation, deregulation, and re-​regulation, and their effects on corporate strategy; and highly uneven patterns of international corporate expansion and their equally variable developmental impacts. Analytically, as key intermediaries between suppliers/​manufacturers and final consumers, retail globalization offers a bridge between research that prioritizes global production networks, and studies—​which tend to be more local or national in nature—​that explore shifting consumption patterns. Transnational retailers are key sourcing and selling agents connecting what can at times seem like separate domains—​that is, globalization trends in both production and consumption—​and their continued growth and influence serves to reinforce this point. There are myriad ways in which economic geographers can engage with these dynamics. Building upon the shifting nature of retail globalization profiled in the preceding two sections, however, we close by highlighting four specific high priority research areas. Firstly, while research over the last fifteen years has provided a reasonable window into the activities of the food and general merchandise retailers towards the top of Table 22.1, and one or two other distinctive cases such as IKEA and Seven & I, the retailers towards the bottom end of the table and below are less well understood. There are two angles here. On the one hand, the dynamics of other types of transnational retailing, notably fashion/​clothing, but also furniture, home improvement, consumer electronics, and pharmacies, undoubtedly merit further study. It is very likely, for example, that sourcing dynamics and geographies will be very different in such non-​food segments. On the other hand, in the context of a literature still dominated by studies of US and European retailers, it is important to discover more about the strategies of second-​tier and regional retailers, such as Aeon (Japan), Lotte (South Korea), Shoprite (South Africa), and Dairy Farm (Hong Kong). Broadening the

444   Coe and Wrigley range of sectors and origins of retailers will allow the development of more nuanced interpretations of the influence of retail type and home market on shaping globalization strategies and their consequent impacts. Secondly, more work is needed on how distinctive national retail markets are being produced on an ongoing basis. That is to say, not just through the globalization activities of different kinds of retailers from different home-​market contexts, but through the interaction of transnational retailers, where present, with local retailers, consumers, regulators, and suppliers. As Endo (2013) has argued so persuasively in the case of Thailand, the central focus in much of the literature on transnational retailers and their strategies may serve to exaggerate their impacts on host economy retail systems and may underestimate the resistance of pre-​existing structures to change. ‘Studies on the retail internationalization process … do not necessarily pay sufficient attention to specific particulars, such as the structure of consumer markets on the host countries; their traditional and changing distribution systems; the commercial practices and actions of local companies and businesses; governmental laws and regulations; and so on’ (Endo, 2013, p. 4). Further studies that focus on particular national markets, and the variable influence of transnational retail therein, will help reveal the context specific interactions of the multiple actors that shape retail development in particular territories. Thirdly, and as introduced above, there is an urgent need for research that considers how the increasingly rapid development of online retailing might be changing the dynamics of the retail globalization process. Up until relatively recently, online services were generally thought to follow the establishment of store operations in host economies. This followed the logic from home markets in that the most successful online operators were those that successfully added online operations to their existing distribution and retail infrastructure. However, recent evidence suggests that in the challenging home-​market context of the post-​ global economic crisis period, retailers may be turning to online expansion as the most cost-​ efficient and risk-​free way of entering foreign markets. This raises questions about the extent to which such strategies vary across retailers of different size and different market segments, and also whether retailers from particular markets are more successful, or not, in operationalizing this strategy. Importantly, it also alters the regulatory landscape in relation to retail globalization. That is to say, it raises issues relating to the extent to which different kinds of regulations come into play with respect to online retailing, as opposed to the relatively well studied regulatory processes surrounding international store expansion? Fourthly, within the wider literature on retail globalization, it is perhaps the consumer dimension that is still most poorly understood. More specifically, the recursive relationships in host economies between retail change and shifting consumer preferences require much further study and elucidation. To what extent, for example, do localized consumer cultures present resistance to certain forms of retailing, and, conversely, in what ways does the availability of new forms of retailing drive processes of consumption change and feed into wider dynamics, such as new middle-​class formation? And what is the relative role of transnational retail capital, vis-​à-​vis domestic retailers, in driving such dynamics? Two related questions concern the time frames over which changes are occurring—​initial evidence seems to suggest much faster shifts than took place historically in the home markets of leading transnational retailers—​and urban/​rural variations in shifting consumption practices. Are, as the supermarketization model suggests, formal retail practices ‘rolling out’ from cities into rural areas, or, in reality, are differences actually growing between the consumption practices

Towards New Economic Geographies of Retail Globalization    445 of the burgeoning urban middle classes and rural residents? Local and national consumption dynamics are clearly being reshaped by retail globalization processes—​both directly through foreign investment, and indirectly as domestic retailers imitate the business practices of leading global players. But the ways in which that is occurring are only starting to be understood. Ultimately, however, it is the position of retailers at the interface of production and consumption dynamics that makes the study of their globalization such a compelling topic for economic geographers.

References Abernathy, F.H., Dunlop, J.T., Hammond, J.H., and Weil, D.W. (2000). ‘Retailing and supply chains in the information age’. Technology in Society 22: 5–​31. Alexander, N. (1997). International Retailing (Oxford: Blackwell). Alexander, N. and Doherty A.M. (2010). ‘International retail research: focus, methodology and conceptual development’. International Journal of Retail & Distribution Management 38: 928–​942. Alexander, N., Quinn, B., and Cairns, P. (2005). ‘International retail divestment activity’. International Journal of Retail and Distribution Management 33: 5–​22. Bianchi, C. (2009). ‘Retail internationalisation from emerging markets: case study evidence from Chile’. International Marketing Review 26: 221–​243. Bianchi, C. and Ostale, E. (2006). ‘Lessons learned from unsuccessful internationalization attempts:  examples of multinational retailers in Chile’. Journal of Business Research 59: 140–​147. Brillo, B.B. (2014). ‘Shifting economic regimes for retail in the Philippines: external impetus amidst the workings of domestic politics’. International Review of Retail, Distribution, and Consumer Research 24: 516–​530. Burt, S., Davies, K., Dawson, J., and Sparks, L. (2008). ‘Categorizing patterns and processes in retail grocery internationalisation’. Journal of Retailing and Consumer Services 15: 78–​92. CBRE (2011). ‘How global is the business of retail?’ Executive summary http://​wall-​street.ro/​ files/​102460-​83.pdf (last accessed 10 April 2017). CBRE (2014). ‘How global is the business of retail?’ 2011 edition special report http://​www.cbre. com/​research-​and-​reports/​how-​global-​is-​the-​business-​of-​retail-​2014-​edition (last accessed 10 April 2017). Coe, N.M. (2004). ‘The internationalisation/​globalisation of retailing, towards an economic-​ geographical research agenda’. Environment and Planning A 36: 1571–​1594. Coe, N.M. and Bok, R. (2014). ‘Retail transitions in Southeast Asia’. International Review of Retail, Distribution and Consumer Research 24: 479–​499. Coe, N.M. and Lee, Y.S. (2006). ‘The strategic localization of transnational retailers: the case of Samsung-​Tesco in South Korea’. Economic Geography 82: 61–​88. Coe, N.M. and Lee, Y.S. (2013). ‘ “We’ve learnt how to be local”: the deepening territorial embeddedness of Samsung-​Tesco in South Korea’. Journal of Economic Geography 13: 327–​356. Coe, N.M. and Wrigley, N. (2007). ‘Host economy impacts of retail TNCs: the research agenda’. Journal of Economic Geography 7: 341–​371. Coe, N.M. and Wrigley, N. (2009). (eds). The Globalization of Retailing (2 vols) (Cheltenham: Edward Elgar).

446   Coe and Wrigley Currah, A.D. and Wrigley, N. (2004). ‘Networks of organizational learning and adaptation in retail TNCs’. Global Networks 4: 1–​23. Dakora, E.A.N., Bytheway, A.J., and Slabbert, A. (2010). ‘The Africanisation of South African retailing: a review’. African Journal of Business Management 4: 748–​754. da Rocha, A. and Dib, L.A. (2002). ‘The entry of Wal-​Mart in Brazil and the competitive responses of multinational and domestic firms’. International Journal of Retail and Distribution Management 30: 61–​73. Dawson, J. (2003). ‘Towards a Model of the Impacts of Retail Internationalisation’ in J. Dawson, M. Mukoyama, C.C. Sang, and R. Larke (eds) The Internationalisation of Retailing in Asia, pp. 189–​209 (London: Routledge Curzon). Deloitte (2012). ‘Retail globalization: navigating the maze’ http://​www2.deloitte.com/​global/​ en/​pages/​consumer-​business/​articles/​retail-​globalization.html (last accessed 10 April 2017). Deloitte (2014). ‘2014 Global powers of retailing’ https://​www2.deloitte.com/​global/​en/​pages/​ consumer-​business/​articles/​global-​powers-​of-​retailing.html (last accessed 10 April 2017). Deloitte (2015). ‘2015 Global powers of retailing’ https://​www2.deloitte.com/​global/​en/​pages/​ consumer-​business/​articles/​global-​powers-​of-​retailing.html (last accessed 10 April 2017). Endo, G. (2013). Diversifying Retail and Distribution in Thailand (Chiang Mai:  Silkworm Books). Hamilton, G.G. and Petrovic, M. (2011). ‘Introduction’ in G.G. Hamilton, M. Petrovic, and B. Senauer (eds) The Market Makers: How Retailers are Reshaping the Global Economy, pp. 1–​28 (Oxford: Oxford University Press). Henderson, J., Dicken, P., Hess, M., Coe, N.M., and Yeung, H.W-​C. (2002). ‘Global production networks and the analysis of economic development’. Review of International Political Economy 9: 436–​464. Humphrey, J. (2007). ‘The supermarket revolution in developing countries: tidal wave or tough competitive struggle?’ Journal of Economic Geography 7: 433–​450. Mutebi, A.M. (2007). ‘Regulatory responses to large-​format transnational retail in South-​east Asian cities’. Urban Studies 44: 357–​379. Natawidjaja, R., Rum, I., Sulistyowati, L., and Saidah, Z. (2014). ‘Improving the participation of smallholder mango farmers in modern retail channels in Indonesia’. International Review of Retail, Distribution, and Consumer Research 24: 564–​580. Nguyen, H.T.H., Deverteuil, G., Wrigley, N., and Ruwanpura, K.N. (2014). ‘Re-​regulation in the post-​WTO period? A case study of Vietnam’s food retailing sector’. Growth and Change 45: 377–​396. Nordås, H.K. (2008). ‘Gatekeepers to consumer markets: the role of retailers in international trade’. International Review of Retail, Distribution and Consumer Research 18: 449–​472. Polder, M. (2015). ‘Use of e-​ commerce data:  international comparisons and a micro-​ perspective’. Presentation given at ONS/​BIS Business Statistics User Event ‘How e-​commerce is changing the shape of business’, 8 October 2015, Department for Business, Innovation & Skills, London. Reardon, T. Henson, S., and Berdegué, J.A. (2007). ‘Proactive fast-​tracking diffusion of supermarkets in developing countries: implications for market institutions and trade’. Journal of Economic Geography 7: 399–​431. Shackleton, R. (1998). ‘Exploring corporate culture and strategy: Sainsbury at home and abroad during the early to mid 1990s’. Environment and Planning A 30: 921–​940.

Towards New Economic Geographies of Retail Globalization    447 Shannon, R. (2014). ‘The expansion of modern trade food retailing in Thailand’. International Review of Retail, Distribution, and Consumer Research 24: 531–​543. Wood, S., Coe, N.M., and Wrigley, N. (2016). ‘Multi-​scalar localization and capability transference: exploring embeddedness in the Asian retail expansion of Tesco’. Regional Studies 50: 475–​495. Wood, S., Wrigley, N., and Coe, N.M. (2017). ‘Capital discipline and financial market relations in retail globalization: insights from the case of Tesco plc’. Journal of Economic Geography 17: 31–​57. Wrigley, N. (1997a). ‘British food retail capital in the USA. Part 1: Sainsbury and the Shaw’s experience’. International Journal of Retail and Distribution Management 25: 7–​21. Wrigley, N. (1997b). ‘British food retail capital in the USA. Part  2:  Giant prospects?’ International Journal of Retail and Distribution Management 25: 48–​58. Wrigley, N. (2000). ‘The Globalization of Retail Capital: Themes for Economic Geography’, in G.L. Clark, M.P. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 292–​313 (Oxford: Oxford University Press). Wrigley, N. and Currah, A. (2003). ‘The stresses of retail internationalization: lessons from Royal Ahold’s experience in Latin America’. International Review of Retail, Distribution and Consumer Research 13: 221–​243. Wrigley, N. and Lowe, M. (2002). Reading Retail: A Geographical Perspective on Retailing and Consumption Spaces (Arnold: London). Wrigley, N. and Lowe, M. (2014). ‘Revisiting the globalization of trade in retail services’. Presentation to the OECD Global Forum on Trade, Paris, 4 November. Wrigley, N., Coe, N.M., and Currah, A. (2005). ‘Globalizing retail:  conceptualizing the distribution-​based TNC’. Progress in Human Geography 29: 437–​457.

Chapter 23

C orp orat e S o c ia l Resp onsibi l i t y a nd Standa rd s Alex Hughes Introduction Over the last two decades, there has been growing acknowledgement that corporations have responsibilities extending beyond the core capitalist objectives associated with the financial bottom line and epitomized by the arguments of the free-​market economist, Milton Friedman (1962). It is now a mainstream view, both within and outside the corporation, that business is also responsible for aspects of society and the environment affected by its operations. Terms such as corporate citizenship, corporate social responsibility, and sustainable business are now commonplace in the lexicon of what can, overall, be recognized as a corporate responsibility movement. And while there are many competing definitions of corporate responsibility, the groups and spaces most prominently regarded as being responsibilities of corporations are customers, employees, communities (broadly conceived), and natural environments, as well as shareholders (Blowfield, 2005; Blowfield and Frynas, 2005). The economics, politics, and development implications of this corporate responsibility movement have been of significant interest to social scientists, including economic geographers, often in connection with broader projects of understanding and theorizing neo-​liberalization, of which current articulations of corporate responsibility are viewed to be an integral part (Guthman, 2007; Sadler and Lloyd, 2009). This chapter sets out to capture why the field of corporate responsibility and standards has become important, what kinds of empirical and conceptual contributions have been made by economic geography, and how new research directions reflect both political–​ economic transformations and evolving intellectual projects. To do this, the chapter first presents a synopsis of the historical and political–​economic context of corporate responsibility and the rise of codes and standards as tools in the private regulation of the economy. While the corporate responsibility movement clearly spans both social and environmental fields, the emphasis of this chapter is on the social dimensions associated with corporate

Corporate Social Responsibility and Standards    449 social responsibility (CSR). The chapter then proceeds to place the critical spotlight on a particular field of CSR—​ethical and labour standards in global supply chains—​receiving significant attention in the social sciences, including within economic geography. Within this area, the different critical insights into CSR and standards offered by the global value chain (GVC) and global production network (GPN) frameworks, as well as postcolonial critique, theories of governmentality, and sociologies of standards and marketization, are summarized and debated. Finally, the chapter discusses some of the recent economic, geographical, and regulatory challenges to the ways in which CSR and standards are operating and transforming in practice, from the global economic downturn to the influence of ‘Rising Powers’ and emerging economies.

Corporate Social Responsibility and Standards: Political–​E conomic Context and Organizational Geographies Blowfield and Frynas (2005) explain that the ethics of business have been a subject addressed for centuries, dating as far back as the first century bc. However, current developments in corporate responsibility are argued by these authors to have their roots in late-​nineteenth-​ century regulation to check the development of the corporation, as well as food boycotts during the same time period. The influence of late-​nineteenth-​and twentieth-​century corporate philanthropy associated with family-​run businesses, such as Cadbury and Rowntree in the UK, is also acknowledged. However, corporate responsibility in its current form appeared to develop from the early 1990s, with a distinct incorporation of issues associated with development, human rights, and environmental protection on a more global scale (Blowfield and Dolan, 2014). Corporate responsibilities until this time tended to refer either to legal obligations to society or to philanthropic acts, including charitable giving and strategies of caring for employees and local communities. It now involves a greater integration of responsibility towards workers, customers, communities, and natural environments into core business strategies involving, for example, the forms of corporate compliance with labour and environmental codes of practice that have been so widely researched in the social sciences, including in economic geography. A range of processes has been argued to fuel the emergence of current developments in voluntary forms of corporate responsibility, including: (i) the globalization of corporations and their supply networks, and the associated political–​economic context of trade liberalization and deregulation since the 1980s (Blowfield, 2005); (ii) anti-​corporate activism (Bair and Palpacuer, 2012); (iii) the rise in the power of corporate brands and reputation, which makes large companies vulnerable to negative publicity (Hughes et al., 2008); (iv) an increase in public awareness of corporate activity and its impacts via improvements in global communications (Micheletti, 2006; Lyon and Montgomery, 2013); (v) the growing importance to the investment community of ethical performance on the part of public companies (Clark and Hebb, 2005); and (vi) the emergence of a strong business case for corporate responsibility in which economic performance in terms of profitability and social/​environmental responsibility are intertwined (Goger, 2013).

450   Hughes The context of globalization and the liberalization of trade and investment from the 1980s set the scene for voluntary corporate initiatives in the field of social (and environmental) responsibility as it engendered a decline in the regulation of corporate activity by traditional governmental institutions (Jenkins et al., 2002). The power of corporate capital encouraged by this climate of deregulation was increasingly called into question by campaigners, the critical media, and consumer groups during the 1990s. Campaigns concerning the contribution of corporate activity to environmental degradation and poor working conditions, for example, have been conducted by a host of campaigning groups, including non-​governmental organizations (NGOs) and trade unions, as well as college students and critical journalists (Johns and Vural, 2000; Littler, 2005; Sadler and Lloyd, 2009; Bair and Palpacuer, 2012). Many campaigns have involved alliances between these different groups, frequently originating in the Global North, but also, and increasingly, extending to the Global South. Most commonly, the object of anti-​corporate campaigning has been high-​ profile, often transnational, brand-​name companies such as Shell (in the extractive industries), McDonalds (in the fast-​food retail sector), and Nike and Gap Inc. (in the apparel sector). Such corporate names have become synonymous with global capitalism, and the adverse impacts they have sometimes had on the environment and worker welfare have been widely criticized. However, such prominent corporations, in part as a result of campaigning, have developed extensive programmes of work around corporate responsibility (Hughes et al., 2008). Added to the influence of anti-​corporate campaigning and critical journalism, institutional investors such as pension funds are ever more concerned with corporate social and environmental performance, with indicators of ethical responsibility increasingly influencing their investment decisions. Such indicators include the FTSE4Good Index Series. The corporate response to these political–​economic forces, in particular from transnational corporations and those at the leading edges of their industries, has been to introduce corporate responsibility programmes, which address the impacts of their business on social groups (often referred to as stakeholders) and the environment. The values of corporate responsibility are commonly crystallized into specific standards or codes set by institutions working at and across various local, national, regional, or global scales. Standards are broadly defined by Nadvi (2008, p. 325) as: commonly accepted benchmarks that transmit information to customers and end-​users about a product’s technical specifications, its compliance with health and safety criteria or the processes by which it has been produced and sourced … In addition, standards can extend to customers and end users a basis for attaching credence, or value, to particular claims made about a product’s characteristics and specification or the ways in which it has been produced.

Examples of standards specifically for CSR set by international organizations are the SA8000 workplace standard set by the US multi-​stakeholder organization Social Accountability International, the global International Organization for Standardization ISO 26000 standard for social responsibility, and international standards for the production of Fairtrade goods set by Fairtrade International (formerly Fairtrade Labelling Organizations International). The participation of companies is almost always voluntary. Some businesses choose to work with such formalized standards and labelling initiatives, while others work with their own or other organizational codes of conduct. Voluntary codes are guidelines drawn up by companies, sometimes aided by stakeholder groups, and serve to specify issues of business responsibility by constructing a set of minimum standards and procedures for companies to follow

Corporate Social Responsibility and Standards    451 in their routine business. Some codes incorporate conventions of international organizations such as the United Nations (UN) and the International Labour Organization (ILO). Such codes can apply to procedures covering all kinds of aspects of corporate activity, for example customer care, human resources, waste management, and the management of supply chains. Standards and codes have therefore become an accepted tool used in the management of corporate responsibility. Procedures for monitoring, reporting, and verifying the implementation of these standards and codes are equally important in order to ensure transparency and accountability. Tools and techniques of social and environmental auditing have therefore developed, influenced by traditional procedures of financial auditing. What auditing provides is a process of checking, monitoring and reporting evidence of corporate performance against the standards encapsulated in codes. Such voluntary codes and auditing procedures represent private-​interest forms of social regulation that appear to replace traditional, state-​led forms of regulation through political–​economic processes of neo-​liberalization (Guthman, 2007; Blowfield and Dolan, 2008). Against this backdrop, research has attended to various aspects of CSR’s institutional geographies, drivers, and outcomes in particular places. While case studies include the healthcare sector (Brando et al., 2013) and tourism (Sandve et al., 2014; Sin, 2014), the fields receiving the most significant attention by economic geographers are the extractive sector (Haalboom, 2012; Mayes et al., 2014; Billo, 2015) and global supply chains. The remaining sections of the chapter focus on the key dimensions of debate in the evolving field of research into CSR and standards in global supply chains, offering an assessment of the value of economic–​geographical perspectives in particular.

Corporate Social Responsibility and Standards in Global Supply Chains: Geographical Perspectives Perspectives of economic geography have been an important part of what has now become an established, extensive, and highly influential literature on global supply chains and standards. This interdisciplinary literature has sought in various ways to capture and theorize the processes through which CSR standards are set in international, national, and localized contexts, the part they play in supply-​chain governance (how power and authority is exercised through rules, norms, and values) and their localized outcomes for producers and communities. A geographical voice has been so crucial to this academic project because, as Riisgaard (2009, p. 328) argues, ‘standards are … not simply neutral market-​based instruments. When elaborated and implemented, they unavoidably play into complex international and local power politics’. The GVC framework established by Gereffi et al. (2005) and associated with economic sociology and development economics has arguably been the most dominant social science approach explaining the governance of global supply chains, and it has done so through a typology of five forms of governance—​hierarchy, captive, relational, modular, and market. The first of these types, ‘hierarchy’, describes a form of control through

452   Hughes the vertical integration of activities on the part of a firm. The last involves governance by ‘the invisible hand’ of the market. The three categories in between represent different network-​based forms of coordination lying along the continuum from hierarchies to markets. Research using this framework to understand CSR and standards has highlighted how ethical and labour standards tend to be implemented through relational (inter-​ firm coordination) and modular (more arms-​length and codified) modes of governance (Dolan and Humphrey, 2004; Nadvi, 2008). Gibbon and Ponte (2005) have also combined GVC analysis with French conventions theory to explain the role played by particular civic, as well as industrial, conventions and values in shaping the governance of socially responsible GVCs. With GVC perspectives critiqued for their narrowly firm-​centred explanations (driven, in part, by transactions cost economics) and their tendency to marginalize geographical and institutional context (Bair, 2005); however, the GPN framework associated with economic geographers has gained significance in the literature. Acknowledging not only how cross-​ border trade is driven by lead firms, but also recognizing the embeddedness of supply-​chain dynamics in different places and the importance of trans-​scalar spaces of governance, the GPN approach attends to the interaction of power, value, and embeddedness in the governance of global networks of supply (Coe et al., 2004). The approach is therefore more attentive to questions of spatiality and territory. GPN perspectives have shaped understandings of how CSR standards are embedded in various institutional contexts of retail and consumption, policymaking, and production. With respect to the embeddedness of CSR and labour standards in spaces of retail and consumption, the influences of corporate retail strategy, ethical consumption, and campaigning have been highlighted. Christopherson and Lillie (2005), for example, demonstrate the embeddedness of retail-​led global labour standards strategies in the particular national–​institutional contexts of retailers’ home economies, using the contrasting cases of Swedish-​based IKEA and US-​based Walmart. Similarly Hughes et  al. (2008) illuminate how UK–​US contrasts in retailers’ ethical codes and standards are shaped by different national–​institutional contexts of campaigning, as well as corporate retail structures. This resonates with the arguments of Bair and Palpacuer (2012) that national institutions and political cultures shape the particular articulations of anti-​sweatshop campaigns and corporate responses to them. Social science research, including economic geography influenced by GPN approaches, has also focused on the development of the Fairtrade standard represented by the global certification scheme of Fairtrade International (formerly known as Fairtrade Labelling Organizations International). In similar ways to the case of ethical trade and minimum labour standards referred to above, Fairtrade standard-​setting has been acknowledged to be driven by institutions in the Global North. The civic principles of partnership, dialogue, transparency, and empowerment are noted continually to drive Fairtrade, incorporating objectives of eliminating exploitative supply-​chain intermediaries, guaranteeing stable prices and fostering community development (the latter through a social premium paid to producer groups on top of the price paid for goods) (Raynolds et al., 2007). Although Fairtrade has been more developmental than minimum corporate labour standards and is rooted in the social values of civil society organizations, the progressive neo-​liberalization and mainstreaming of Fairtrade through northern spaces of retail and consumption are now widely critiqued (Blowfield and Dolan 2008; Neilson and Pritchard, 2009; Besky, 2010).

Corporate Social Responsibility and Standards    453 Both GVC and GPN perspectives have also been taken to evaluate the impacts and outcomes of ethical labour codes and Fairtrade standards on communities and spaces of production. Barrientos and Smith’s (2007) impact assessment of the UK’s Ethical Trading Initiative’s base code of conduct (built around core ILO conventions) used a GVC approach to trace GVCs in textiles, agriculture, and horticulture to sites of production in India, Vietnam, South Africa, Costa Rica, China, and the UK, and found that while workers were benefitting from labour codes in areas such as health and safety, less progress was being made in freedom of association and practices to tackle discrimination issues. GVC and GPN perspectives have similarly been used to understand the interconnections between global CSR standards and local governance in the football manufacturing industries of South Asia (Lund-​Thomsen and Nadvi, 2010; Lund-​Thomsen, 2013). Lund-​ Thomsen (2013), for example, points to the constraints on labour agency shaped by a combination of global supply-​chain dynamics, national–​institutional environment and local socio-​economic context in Sialkot, Pakistan (see also Lund-​Thomsen and Coe, 2015). And GVC/​GPN frameworks have been coupled with an institutional perspective in the work of Neilson and Pritchard (2009), who advance a geographical approach to understanding the outcomes of CSR and standards in South Indian tea and coffee plantation districts. Similarly, Franz (2012) uses the GPN framework to understand the role of civil-​society organizations and trade unions in resisting and thereby influencing the operations, including supply chains, of Metro Cash & Carry in Bangalore. GVC/​GPN approaches also have been taken to assess Fairtrade’s outcomes at sites of production. For example, GPN sensitivity to the territorial embeddedness of standards in particular localized institutional contexts has been used to appreciate some of the limitations, as well as gains, of Fairtrade for raisin producers in the Northern Cape of South Africa (Hughes et al., 2014) and tea-​producing communities in South India (Neilson and Pritchard, 2009). And Mutersbaugh (2005) draws on GVC perspectives, in particular value-​chain rent theory, to show how the costs of standards compliance (including Fairtrade and organics standards) can serve to disadvantage producers. To further understand the politics of governance executed through social standards and their links to neo-​liberalization, scholars have also looked beyond the more economic and firm-​centred frameworks of GVC/​GPN and applied the concept of governmentality in order to ‘recognize the diversity of authorities that have the capacity to govern different spaces according to different objectives’ (Hughes, 2001; Mutersbaugh, 2005; Gibbon and Ponte, 2008; Loconto, 2015, p. 3). The Foucauldian notion of governmentality refers to ‘how we think about governing others and ourselves in a wide variety of contexts’ (Dean, 1999, p. 209). While studies using the concept have concentrated on the changing ways in which the modern liberal state governs the economy and society (see e.g. MacKinnon, 2000), it can be applied to the field of CSR and standards in order to capture the practices of standard-​setting and implementation as part of private regulation ‘undertaken by a multiplicity of authorities and agencies’ (Dean, 1999, p. 11; Hughes, 2001). A focus on the politics and practices of CSR standards is also a key dimension of cultural and anthropological studies, contrasting the abstract standard of Fairtrade with the ‘lived experiences’ of Fairtrade producers (Getz and Shreck, 2006, p. 490). Studies of producer groups reveal cases where the benefits of Fairtrade make only a modest material difference, or quite commonly do not reach all members of the community in equal measure (Dolan, 2010). Such uneven participation in Fairtrade cooperatives can result in widening, rather

454   Hughes than narrowing, cleavages between different socio-​economic and cultural groups in producer contexts (Arce, 2009), which points to the challenges posed by complex and divided, or ‘fractured’, producer communities. Work by economic anthropologists is particularly illuminating in this regard, but the need to engage with lived experiences also has been argued by geographers (see e.g. Herman, 2010). Through ethnographic research with Costa Rican coffee growers, Luetchford (2008, p. 144) helpfully problematizes the monolithic model of the smallholder community, which effectively forms a ‘culturally appealing morality tale’ sold to consumers of Fairtrade in the North. By revealing the differences and tensions between landed farmers and landless migrant labourers at the site of production, he shows that, in practice, ‘synthetic categories break down and imagined communities of independent family producers melt into air’ (Luetchford, 2008, p. 165). A postcolonial lens has also been used by some scholars to capture the politics of CSR and standards. Both ethical labour codes and Fairtrade are based on standards frequently devised in the global North, which are often assumed to be universal and exportable to the global South. Ethical interventions are often seen to be driven by Northern ethical concerns with minimal consideration of socio-​economic realities in the South (Dolan, 2010), although more proactive roles in ethical trade are increasingly being played by Southern actors (Hughes et al., 2013). Studies have addressed how the discourses and legacies of colonialism influence export production. This is particularly evident in culturally inflected analyses of ethical standards for global supply chains (Freidberg, 2003; Dolan, 2005; Neilson and Pritchard, 2009). Colonial imaginaries and moralities, apparent both in the sphere of consumption and through corporate ethical codes applied to suppliers in the Global South, are significant themes through which ‘the colonial’ is acknowledged in research concerning global supply chains (Cook and Harrison, 2003). Alongside this, some research into global commodity chains takes a more historical perspective sensitive to the role of colonial pasts and cultural politics (Besky, 2010; Bair and Werner, 2011; Hough, 2011; Hughes et al., 2015). Another recent advancement of the theorization of CSR and standards in global supply chains comes via perspectives associated with the sociology of standards (Busch, 2011) and performative theories of markets, which have their roots in actor network theory (Callon, 2007; Berndt and Boeckler, 2009). Arising from a dissatisfaction with the top-​down understanding of governance prevailing in GVC/​GPN accounts that ascribe agency predominantly to ‘lead firms’, Loconto (2015) in sociology and Ouma (2015) in geography, for example, prefer to capture the networked and performative dynamics of standards in the Tanzanian Fairtrade tea sector and the Ghanaian export horticulture industry, respectively. For Loconto (2015, p. 3): [V]‌oluntary standards are mechanisms that facilitate the emergence of agencements, which are the collectives of human beings, technical devices, scientific tests, and written standards that are embedded in institutions, conventions, personal relationships and groups.

Ouma (2015) prefers to use the term ‘assemblage’ more commonly adopted in geography (Allen, 2011) to understand the (re)making of export markets in frontier regions of northern and southern Ghana in a similar way (see also Wanvik, 2014). What we gain from such perspectives is the prioritization of standard-​setting and governing practices through hybridized networks of ‘multiple material and discursive relationships’ (Loconto, 2015, p. 30) in an

Corporate Social Responsibility and Standards    455 analysis of how global and often abstract standards work through and influence particular localized spaces.

The Changing Landscape for Corporate Social Responsibility and Standards: Economic, Regulatory, and Geographical Challenges Until very recently, most of the literature on CSR and standards, including within economic geography, has focused on the politics and outcomes of various voluntary initiatives and standards. These standards have largely been implemented in the Global South but constructed by institutions in the Global North and often in the context of relative economic prosperity from the early 1990s onwards. Over the last few years, the CSR movement has faced challenges and has transformed in the context of a range of political–​economic pressures and transitions, two of which are discussed here—​economic recession in the Global North; and the influence of rising powers and emerging economies. Research in economic geography has contributed to understandings of how the CSR movement has responded at times of financial crisis, economic recession, and subsequent austerity, in particular since the global financial crisis of 2008, addressing the question of whether and how corporate responsibility is resilient in times of financial hardship. In line with the wider corporate responsibility movement, UK retailers’ ethical trading programmes for their global supply chains were shown to continue as integral dimensions of corporate reputational risk management. However, the prominence of ‘value’, low-​cost retailing, a potentially weakening voice of NGOs within multi-​stakeholder ethical trading initiatives, declining workplace conditions for many migrant labourers, and ethical trading programmes more tightly linked to strategies of corporate efficiency were found to present a changing set of challenges for actors confronting the economics and politics of the post-​ recessionary ethical trading landscape (Hughes, 2012). A sharper focus of retailers on price during the global economic downturn of 2008–​10 affected export producers by squeezing their orders and profit margins. Ruwanpura and Wrigley’s (2011) study of Sri Lankan garment producers showed retailers had a tendency to source more of their textiles from lower-​ cost countries such as Bangladesh during that period, as pressures were placed on producers to cut working hours and wages in order for their businesses to survive. Against the backdrop of a CSR field that has been seen both in theory and practice to emerge and develop largely from institutional and corporate contexts of the Global North, the last few years have witnessed far more active roles played by firms and organizations in the Global South. Whereas ethical and fair-​trade standards were seen to be actively developed in the Global North and applied to sites of production in the Global South, a recent body of literature now demonstrates ways in which national policy frameworks, civil-​society organizations, domestic businesses, and a globalizing middle class are more actively influencing CSR standards and their implementation, in particular in the ‘rising power’ economies of Brazil, India, China, and South Africa. Knorringa and Nadvi (2016)

456   Hughes point to the growing role played by the national policy frameworks of Brazil, India, and China to influence social compliance and regulatory labour standards in localized clusters of small-​and medium-​sized firms. They note a particularly strong regulation and enforcement of labour standards in Brazil, influenced, in part, by a tight relationship between Brazilian business and state institutions and an active engagement with international programmes such as the UN Global Compact, the Global Reporting Initiative, and the International Organization for Standardization. A more complex picture emerges in China, but one that has been influenced, nonetheless, by an evolving legal framework for labour standards. In addition to local-​government initiatives to raise minimum wages, the central government has passed a set of laws, including three in 2007—​the Employment Promotion Law, the Labour Dispute Mediation and Arbitration Law, and the Labour Contract Law—​and the Social Insurance Law in 2010, which Chan and Nadvi (2014, p. 522) argue: could suggest an end to the era of cheap labour in China and a relatively more optimistic future for Chinese workers, marked by increasing real wages, better social protection and pension benefits, and improved labour rights.

This growing influence of the national-​policy context in emerging economies and ‘rising powers’ is seen by commentators to contrast with the ‘soft regulation’ of voluntary corporate codes and to promise a kind of ‘ “regulatory renaissance”, highlighting the potential role that the State can play in the formulation and enforcement of labour standards and labour rights’ (Piore and Schrank, 2008; Pires, 2008; Chan and Nadvi, 2014, p. 517; see also Alford, 2015). A parallel body of work observes the effects of the globalizing middle classes on the growth of domestic markets for consumer goods in emerging economies and ‘rising powers’, upon which sustainability and ethical issues have a small but nonetheless growing influence (Belk et al., 2005; Guarin and Knorringa, 2014). Civil-​society organizations and business, rather than the state, are the key actors in the shaping of social compliance through new domestic ethical markets and are visible in the rapid growth of Fairtrade markets in emerging economies and ‘rising powers’ (Dombos, 2008; Doherty et al.,2015). In the wider context of increasing South–​South trade, emerging markets and growing middle classes in the Global South, Fairtrade International (2013) reports rising Fairtrade sales in emerging market economies in addition to growth in established markets of Western Europe, North America, Australia, and New Zealand. For example, South Africa’s Fairtrade sales increased by 220 per cent in 2012 alone. This was partially due to institutional support developing these markets, such as the work of organizations like Fairtrade South Africa. In attempts to match the success in South Africa, Fairtrade Eastern Africa was launched in 2013 to develop markets in Kenya focusing, in particular, on forging value chains between elite Kenyan consumers in cities like Nairobi and the country’s high-​profile Fairtrade producers in rural areas. A new Fairtrade marketing organization has also been established in Brazil and similar initiatives are being discussed in Argentina, India, Lebanon, and the Philippines (Fairtrade International, 2013). In all of these countries, there is an already strong and well-​ known Fairtrade supply base, implying that a key part of this new trend is a connection made between ethical consumers and producers within local and national economies of the Global South.

Corporate Social Responsibility and Standards    457

Conclusions CSR and accompanying codes and standards have developed over the past two to three decades as part of the private regulation of the global economy associated, at least in part, with neo-​liberalization. Critical studies in the social sciences have engaged with their political–​ economic contexts, the politics of their organizational energies, and their outcomes for firms, workers, and communities. As the chapter has shown, a wide range of conceptual tools have been used in this field, from GVC and GPN frameworks to postcolonial perspectives. The analytical contribution from economic geography has been wide ranging and often linked to interdisciplinary projects and academic endeavours. Arguably the most notable contribution of economic geography has been the application of a GPN approach to understanding the embeddedness of CSR and standards in particular institutional contexts of consumption and production. Most recently, Ouma’s (2015) networked and performative perspective on the construction of the Ghanaian export horticulture industry promises to advance economic–​geographical approaches by de-​centring lead firms per se in terms of locating agency in geographies of CSR and instead viewing standards as hybridized and networked assemblages. With the landscape of CSR and standards continually transforming, not least in terms of the recent acceleration in the influence of ‘rising powers’ and emerging economies through both a ‘regulatory renaissance’ and Southern-​based voluntary initiatives, there is scope to employ and develop such perspectives to grasp and challenge some of these new geographical articulations of responsibility and regulation.

References Alford, M. (2015). ‘Public governance and multi-​scalar tensions in global production networks:  crisis in South African fruit’ (unpublished Doctoral Dissertation, University of Manchester). Allen, J. (2011). ‘Powerful assemblages?’ Area 43: 154–​157. Arce, A. (2009). ‘Living in times of solidarity:  fair trade and the fractured life worlds of Guatemalan Coffee farmers’. Journal of International Development 21: 1031–​1041. Bair, J. (2005). ‘Global capitalism and commodity chains:  looking back, going forward’ Competition and Change 9: 153–​180. Bair, J. and Palpacuer, F. (2012). ‘From varieties of capitalism to varieties of activism:  the antisweatshop movement in comparative perspective’. Social Problems 59: 522–​543. Bair, J. and Werner, M. (2011). ‘The place of disarticulations: global commodity production in La Laguna, Mexico’. Environment and Planning A 43: 998–​1015. Barrientos, S. and Smith, S. (2007). ‘Do workers benefit from ethical trade? Assessing codes of labour practice in global production systems’. Third World Quarterly 28: 713–​729. Belk, R.W., Devinney T., and Eckhardt, G. (2005). ‘Consumer ethics across cultures’. Consumption Markets and Culture 8: 275–​289. Berndt, C. and Boeckler, M. (2009). ‘Geographies of circulation and exchange: constructions of markets’. Progress in Human Geography 33: 535–​551.

458   Hughes Besky, S. (2010). ‘Colonial Pasts and Fair Trade Futures: Agricultural Change and Regulatory “Friction” in Darjeeling Tea Production’ in S. Lyon and M. Moberg (eds) Fair Trade and Social Justice: Global Ethnographies, pp. 97–​122 (New York: New York University Press). Billo, E. (2015). ‘Sovereignty and subterranean resources: an institutional ethnography of Repsol’s corporate social responsibility programs in Ecuador’. Geoforum 59: 268–​277. Blowfield, M. (2005). ‘Corporate social responsibility: reinventing the meaning of development’. International Affairs 81: 515–​534. Blowfield, M.E. and Dolan, C.S. (2008). ‘Stewards of virtue? The ethical dilemma of CSR in African agriculture’. Development and Change 39: 1–​23. Blowfield, M. and Dolan, C. (2014). ‘Business as a development agent: evidence of possibility and improbability’. Third World Quarterly 3591: 22–​42. Blowfield, M. and Frynas, J.G. (2005). ‘Setting new agendas: critical perspectives on corporate social responsibility in the developing world’. International Affairs 81: 499–​513. Brando, C., Rego, G., and Duarte, I. (2013). ‘Social responsibility: a new paradigm of hospital governance?’ Health Care Analysis 21: 390–​402. Busch, L. (2011). Standards: Recipes for Reality (Cambridge, MA: MIT Press). Callon, M. (2007). ‘What Does it Mean to Say Economics is Performative?’ in D. MacKenzie, F. Muniesa, and L. Siu. (eds) Do Economists Make Markets? On the Performativity of Economics, pp. 311–​358 (Princeton, NJ: Princeton University Press). Chan, C.K-​C. and Nadvi, K. (2014). ‘Changing labour regulations and labour standards in China: retrospect and challenges’. International Labour Review 153: 513–​534. Christopherson, S. and Lillie, N. (2005). ‘Neither global nor standard: corporate strategies in the new era of labour standards’. Environment and Planning A 37: 1919–​1938. Clark., G.L. and Hebb, T. (2005). ‘Why should they care? The role of institutional investors in the market for corporate global responsibility’. Environment and Planning A 37: 2015–​2031. Coe, N.M., Hess, M., Yeung, H.W-​C., Dicken P., and Henderson, J. (2004). ‘ “Globalizing” regional development:  a global production networks perspective’. Transactions of the Institute of British Geographers 29: 468–​484. Cook, I. and Harrison, M. (2003). ‘Cross-​over food: re-​materializing postcolonial geographies’. Transactions of the Institute of British Geographers 28: 296–​317. Dean, M. (1999). Governmentality:  Power and Rule in Modern Society (Thousand Oaks, CA: SAGE). Doherty, B., Smith, A., and Parker, S. (2015). ‘Fair trade market creation and marketing in the Global South’. Geoforum 67: 158–​171. Dolan, C. (2005). ‘Fields of obligation: rooting ethical sourcing in Kenyan horticulture’ Journal of Consumer Culture 5: 365–​389. Dolan, C. (2010). ‘Virtual moralities:  the mainstreaming of Fairtrade in Kenyan tea fields’. Geoforum 41: 33–​43. Dolan, C. and Humphrey, J. (2004). ‘Changing governance patterns in the trade in fresh vegetables between Africa and the United Kingdom’. Environment and Planning A 36: 491–​509. Dombos, T. (2008). ‘ “Longing for the West”: The Geo-​Symbolics of the Ethical Consumption Discourse in Hungary’ in G.D. Neve, P. Luetchford, and J. Pratt (eds) Hidden Hands in the Market:  Ethnographies of Fair Trade, Ethical Consumption, and Corporate Social Responsibility, pp. 123–​141 (Bingley: Emerald Group Publishing). Fairtrade International (2013). ‘Unlocking the power: annual report 2012–​13’ http://​www.fairtrade.net/​annual-​reports.html (last accessed 7 May 2014).

Corporate Social Responsibility and Standards    459 Franz, M. (2012). ‘Resistance and strategic responses in food supply networks: metro cash & carry in Bangalore’. Geografiska Annaler Series B—​Human Geography 94B: 161–​176. Freidberg, S.E. (2003). ‘Culture, conventions and colonial constructs of rurality in south–​north horticultural trades.’ Journal of Rural Studies 19: 97–​109. Friedman, M. (1962). Capitalism and Freedom (Chicago, IL: University of Chicago Press). Gereffi, G., Humphrey, J., and Sturgeon, T. (2005). ‘The governance of global value chains’. Review of International Political Economy 12: 78–​104. Getz, C. and Shreck, A. (2006). ‘What organic and fair trade labels do not tell us: towards a place-​based understanding of certification’. International Journal of Consumer Studies 30: 490–​501. Gibbon, P. and Ponte, S. (2005). Trading Down: Africa, Value Chains, and the Global Economy (Philadelphia, PA: Temple University Press). Gibbon, P. and Ponte, S. (2008). ‘Global value chains: from governance to governmentality’. Economy and Society 37: 365–​392. Goger, A. (2013). ‘The making of a “business case” for environmental upgrading: Sri Lanka’s eco-​factories’. Geoforum 47: 73–​83. Guarin, A. and Knorringa, P. (2014). ‘New middle-​class consumers in rising powers: responsible consumption and private standards’. Oxford Development Studies 42: 151–​171. Guthman, J. (2007). ‘The Polanyian way? Voluntary food labels as neoliberal governance’. Antipode 39: 456–​478. Haalboom, B. (2012). ‘The intersection of corporate social responsibility guidelines and indigenous rights: examining neoliberal governance of a proposed mining project in Suriname’. Geoforum 43: 969–​979. Herman, A. (2010). ‘Connecting the complex lived worlds of Fairtrade’. Journal of Environmental Policy and Planning 12: 405–​422. Hough, P.A. (2011). ‘Disarticulations and commodity chains: cattle, coca, and capital accumulation along Colombia’s agricultural frontier’. Environment and Planning A 43: 1016–​1034. Hughes, A. (2001). ‘Global commodity networks, ethical trade and governmentality: organizing business responsibility in the Kenyan cut flower industry’. Transactions of the Institute of British Geographers 26: 390–​406. Hughes, A. (2012). ‘Corporate ethical trading in an economic downturn: recessionary pressures and refracted responsibilities’. Journal of Economic Geography 12: 33–​45. Hughes, A., McEwan, C., and Bek, D. (2013). ‘Retailers, supply networks and changing articulations of ethicality:  lessons from Flower Valley in South Africa’. Journal of Economic Geography 13: 211–​230. Hughes, A., McEwan, C., Bek, D., and Rosenberg, Z. (2014). ‘Embedding Fairtrade in South Africa:  global production networks, national initiatives and localized challenges in the Northern Cape’. Competition & Change 18: 291–​308. Hughes, A., McEwan, C., and Bek, D. (2015). ‘Postcolonial perspectives on global production networks: insights from Flower Valley in South Africa.’ Environment and Planning A 47: 249–​266. Hughes, A., Wrigley, N., and Buttle, M. (2008). ‘Global production networks, ethical campaigning, and the embeddedness of responsible governance’. Journal of Economic Geography 8: 345–​367. Jenkins, R., Pearson, R., and Seyfang, G. (2002). ‘Introduction’ in R. Jenkins, R. Pearson, and G. Seyfang (eds) Corporate Responsibility and Labour Rights: Codes of Conduct in the Global Economy, pp. 1–​10 (London and Stirling: Earthscan).

460   Hughes Johns, R. and Vural, L. (2000). ‘Class, geography, and the consumerist turn: UNITE and the Stop Sweatshops Campaign’. Environment and Planning A 32: 1193–​1213. Knorringa, P. and Nadvi, K. (2016). ‘Rising power clusters and the challenges of local and global standards’. Journal of Business Ethics 133: 55–​72. Littler, J. (2005). ‘Beyond the boycott:  anti-​consumerism, cultural change and the limits of reflexivity’. Cultural Studies 19: 227–​252. Loconto, A. (2015). ‘Assembling governance: the role of standards in the Tanzanian tea industry.’ Journal of Cleaner Production 107: 64e73. Luetchford, P. (2008). ‘The hands that pick fair trade coffee: beyond the charms of the family farm’. Research in Economic Anthropology 28: 143–​169. Lund-​Thomsen, P. (2013). ‘Labor agency in the football manufacturing industry of Sialkot, Pakistan’. Geoforum 44: 71–​81. Lund-​ Thomsen, P. and Coe, N.M. (2015). ‘Corporate social responsibility and labour agency: the case of Nike in Pakistan’. Journal of Economic Geography 15: 275–​296. Lund-​Thomsen, P. and Nadvi, K. (2010). ‘Clusters, chains and compliance: corporate social responsibility and governance in football manufacturing in South Asia’. Journal of Business Ethics 93: 201–​222. Lyon, T.P. and Montgomery, A.W. (2013). ‘Tweetjacketed: the impact of social media on corporate greenwash’. Journal of Business Ethics 118: 747–​757. MacKinnon, D. (2000). ‘Managerialism, governmentality and the state:  a neo-​Foucauldian approach to local economic governance’. Political Geography 19: 293–​314. Mayes, R., McDonals, P., and Pini, B. (2014). ‘ “Our” community: corporate social responsibility, neoliberlisation, and mining industry community engagement in rural Australia’. Environment and Planning A 46: 398–​413. Micheletti, M. (2006). ‘Political Consumerism:  Why the Market is an Arena for Politics’ in M. Kaiser and M.E. Lien (eds) Ethics and the Politics of Food, pp. 23–​27 (Wageningen: Wageningen Academic Publishers). Mutersbaugh, T. (2005). ‘Just-​in-​space: certified rural products, labour of quality, and regulatory spaces’. Journal of Rural Studies 21: 389–​402. Nadvi, K. (2008). ‘Global standards, global governance and the organisation of global value chains’. Journal of Economic Geography 8: 323–​343. Neilson, J. and Pritchard, B. (2009). Value Chain Struggles: Institutions and Governance in the Plantation Districts of South India (Chichester: Wiley-​Blackwell). Ouma, S. (2015). Assembling Export Markets:  The Making and Unmaking of Global Food Connections in West Africa (Chichester: Wiley Blackwell). Piore, M.J. and Schrank, A. (2008). ‘Toward managed flexibility: the revival of labour inspection in the Latin world’. International Labour Review 147: 1–​23. Pires, R. (2008). ‘Promoting sustainable compliance: styles of labour inspection and compliance outcomes in Brazil’. International Labour Review 147: 199–​229. Raynolds, L.T., Murray, D., and Wilkinson, J. (eds) (2007). Fair Trade: The Challenges of Transforming Globalization (London: Routledge). Riisgaard, L. (2009). ‘Global value chains, labor organization and private social standards: lessons from East African cut flower industries’. World Development 37: 326–​340. Ruwanpura, K. and Wrigley, N. (2011). ‘The costs of compliance? Views of Sri Lankan apparel manufacturers in times of global economic crisis’. Journal of Economic Geography 11: 1031–​1049. Sadler, D. and Lloyd, S. (2009). ‘Neo-​liberalising corporate social responsibility:  a political economy of corporate citizenship’. Geoforum 40: 613–​622.

Corporate Social Responsibility and Standards    461 Sandve, A., Mamburg, E., and Ogaard, T. (2014). ‘The ethical dimensions of tourism certification programs’. International Journal of Hospitality Management 36: 73–​80. Sin, H.L. (2014). ‘Realities of doing responsibilities: performances and practices in tourism’. Geografiska Annaler Series B—​Human Geography 96: 141–​157. Wanvik, T.I. (2014). ‘Encountering a multidimensional assemblage: the case of Norwegian corporate social responsibility activities in Indonesia’. Norsk geografisk Tidsskrift—​Norwegian Journal of Geography 68: 282–​290.

Pa rt  V

WOR K

Chapter 24

Plu ralizing L a b ou r Geo gra ph y Jamie Peck Introduction: Constructions of Labour Geography There are few issues that illustrate the differences between the generally heterodox field of economic geography and the more orthodox world of geographical economics than the treatment of labour. In orthodox terms, labour principally appears as an abstraction, traded on labour markets at a price (the wage), which in turn, is taken as a measure of human-​ capital endowments and the balance of supply and demand. In contrast, after economic geographers had opened up the black box of the production function, starting in the late 1970s—​venturing into workplaces as fieldwork sites and engaging directly with workers as research subjects and sometimes as collaborators—​labour would no longer be seen as a disembodied ‘factor’ of production. Having begun to pose the ‘labour question’, economic geographers soon found themselves on the terrain of class and gender relations; they were dealing with politics and institutions, with industrial disputes and union campaigns, and more. These were some of the origins of ‘labour geography’, the non-​linear trajectories and shifting concerns of which are the focus of this chapter. Necessarily selective, the chapter cuts a path through a variegated and evolving body of work that, among other things, has mapped the shifting politics of production, together with old and new forms of labour organization; that has problematized the workplace, as a site of struggle and as a place for the performance and (re)production of social identities; that has tracked the restructuring of labour markets, as spaces of socio-​institutional stress and regulatory transformation; and that has utilized labour (and labour relations) as a diagnostic, as the key to understanding different (local) varieties of capitalism, economies of care and reproduction, and alternative modes of socio-​economic organization. In as far as it is possible to generalize, these inherently more social, more political, and more cultural treatments of employment, labour, and work do not equate labour power (the socially shaped and institutionally regulated capacity to work) with a commodity, but instead recognize and grapple with labour’s pseudo-​commodity forum, as Polanyi called it (Peck, 1996). They

466   Peck understand labour to be different because production and reproduction entail social relations, and because labour itself is a social process, performed in many ways and in many sites, some waged and some not. But if this has been the ‘take’, broadly speaking, of labour geography, the field has always been tracking moving objects … and subjects; its projects are always in motion. While an Anglo-​American labour geography got its start in the late 1970s, not until almost two decades later did ‘Labour Geography’ take shape as a distinctive (and named) project, one that embraced as its central problematic the active roles of workers and workers’ organizations in remaking the landscapes of capitalism (see Herod, 1997). This line of work has been especially productive in opening up issues around the politics of (re)organizing labour movements and organizations, and around the causes and consequences of workplace-​cum-​ community struggles. But what is sometimes styled as Labour Geography ‘proper’ has also been but a particular moment in a broad and multifaceted engagement with issues of work, employment, and labour in the subdiscipline of economic geography. Foundational for these efforts was work on the political economy of restructuring, which spawned concepts like the spatial divisions of labour (see Bluestone and Harrison, 1982; Storper and Walker, 1983; Massey, 1984; Clark et al., 1986; Scott and Storper, 1986). Subsequent lines of enquiry would delve into the gendered character of industrial restructuring and service-​sector growth, ‘locality effects’, and local cultures of employment, moving on to explore the uneven advance of ‘post-​Fordist’ economies, and the formation of new industrial districts. Some of these concerns would later coalesce around the problematic of labour regulation, including the production of new workplace and employment norms, class and gender orders at work, and the social and institutional (re)organization of labour markets (see Storper and Scott, 1992; Hanson and Pratt, 1995; Peck, 1996; Storper, 1997). If the (industrial) restructuring rubric of the 1980s had morphed into concern with questions of social regulation by the 1990s, the arrival of Labour Geography proper at the end of that decade was associated with a sharper focus on activism and resistance, responding to what were seen as the limitations of ‘capital logic’ (and, by extension, ‘regulatory logic’) accounts by sampling purposefully in favour of labour’s agency (see Herod, 1998, 2001; Wills, 1998, 2001; Rutherford and Gertler, 2002). A crucial contribution of this work has been to analyse, in real time, the historic reorganization of the labour movement, rarely with any ambiguity as to which side it is on. After a decade of intensive case-​study work, the project has entered a phase of reflection and re-​evaluation (see Castree, 2007; Tufts and Savage, 2009; Rutherford, 2010; Coe and Jordhus-​Lier, 2011). Meanwhile, new rounds of small-l labour geography have been appearing, variously complementing and supplementing these long-​standing concerns with restructuring, regulation, and reorganization. Especially notable among these has been the growing concern with the reproduction of work (and the work of reproduction), including care economies, contingent employment, and the dynamics of contracted-​out capitalism (see Wills et al., 2010; Green and Lawson, 2011; Doussard, 2013; McDowell, 2014; Winders and Smith, 2015). Framed in these terms, this chapter offers a sympathetic rereading of what might be called pluralist labour geography in its various and evolving forms. Echoing one of economic geography’s most cherished metaphors, that of sedimented layers (or ‘rounds’) of investment, each marked by its own priorities and practices but at the same time interacting with what went before (see Massey, 1984; Scott, 2000), the roughly successive rounds of research on restructuring, regulation, reorganization, and reproduction are presented here not as

Pluralizing Labour Geography    467 competing efforts, contrary projects, or independent initiatives, but as moments in a cumulatively assembled palimpsest of analytical sensibilities and practices. Contemporary labour geographies are not (and cannot be) all of these things simultaneously. But they invariably do bear the imprint of what has been a productive series of past endeavours, shaping how the heterodox field of labour geography engages with the ‘restructuring now’. There is also both a need and an opportunity, as the chapter’s conclusion emphasizes, for consciously and reflexively ‘recombinant’ labour geographies to address, in new and creative ways, some of the most vexing contemporary challenges relating to work, employment, and livelihoods—​ not least those concerned with the entrenched processes of inequality, insecurity, and informalization.

Restructuring The emergence of labour geography—​in substance and in spirit if not in name—​can be traced to the crisis-​driven restructuring of the manufacturing economies of North America and Western Europe during the 1970s, the advent of de-​industrialization in what had complacently been labelled the advanced capitalist countries, and the (regionalized) waves of job losses and plant closures that followed in its wake. ‘Restructuring’, in this context, initially represented a concrete object of inquiry, although it would later spawn a distinctive analytical method. In the USA, Bluestone and Harrison (1982) had shown that blue-​collar job losses across what would subsequently be known as the ‘rustbelt’ could not be attributed to some sudden collapse of labour productivity or enterprise viability, but instead were the result of deliberate and planned corporate strategies for profit restoration by way of relocation (or its effective threat), fuelled by a determination to rewrite, or withdraw unilaterally from, the postwar social contract with organized labour. These were uniquely testing times for Bluestone and Harrison’s allies in the labour movement, as ‘runaway shops’ were seen to be heading for anti-​union enclaves, notably the ‘right to work’ states of the American south, and to yet-​lower-​cost production sites in Latin America and Asia. And while Bluestone and Harrison remained focused on opportunities for progressive reform (involving industrial retention efforts, workforce investment measures, fortified employment rights, and strategies for re-​industrialization), there were clear indications, in these early Reagan years, that some kind of ruptural transformation was under way—​if not the beginning of the end for the New Deal labour settlement (Peck, 2002). Not surprisingly, this was a time when ‘[e]‌conomic geography, and the questions of industrial location and labour markets in particular, took on a new political salience’ (Lovering, 1989, p. 216). Across the Atlantic, Massey and Meegan (1982) had been likewise convinced of the need to dig beneath the statistics on redundancies and factory closures to identify the drivers of employment change in their work on the industrial transformation of the UK. The framing of their analysis was clearly more Marxist in inspiration, but hardly of the conventional sort. Abstract tendencies like the deskilling of labour through workplace automation, or downward pressure on the rate of profit, were not finding singular or consistent expression in concrete outcomes, but were shown to be sectorally and geographically specific. These general tendencies were mediated and even transformed by an array of conjunctural, contextual, and contingent circumstances—​hence the need to bring to light what Massey and

468   Peck Meegan called the ‘anatomy of job loss’. Tracing the (differentiated) dynamics of investment, productivity, and employment across dozens of industrial sectors, they concluded that the ostensibly amorphous category of ‘job loss’ was, in fact, an outcome of the operation of three analytically distinct mechanisms, understood as a ‘repertoire’ of restructuring strategies: the productivity-​enhancing intensification of existing production systems; various forms of investment-​driven technical change, involving the reorganization of these systems; and the rationalization of employment and productive capacity. These were the intermediate steps, reformulated as mid-​level concepts or ‘mechanisms’, through which the abstract categories of Marxian political economy were rendered tractable and meaningful at the level of workplace restructuring. To borrow the language later used in Massey and Meegan’s (1985) influential compendium of research practices in industrial geography (as the field was still known), there would be implications for both politics and method. On the politics side, the Greater London Council and a network of other municipal–​socialist councils across the UK were actively engaged at the time in developing local economic strategies under the slogan ‘restructuring for labour’. On a collision course with the neoliberal programme of the Thatcher government, which would ultimately prove fatal for this suggestive cluster of left-​progressive innovations and demonstration projects, restructuring for labour involved the development of detailed strategies for the full span of (local) economic activities, from domestic labour and cultural work to more conventional industrial sectors like electronics, textiles, and defence (see Boddy and Fudge, 1984; Cochrane, 1986; Lovering, 1988). Understood as a method, however, ‘restructuring’ implied a non-​reductionist mode of analysis oriented to the creative and somewhat open-​ended exploration of the local political economies of employment change, corporate transformation, and strategic intervention. Informed by critical realism in the UK and by the tradition of progressive pragmatism in the USA, the restructuring approach was simultaneously skeptical of the positivist proclivity to infer causality from empirically observed patterns and of those styles of Marxian abstraction reliant on reductionist readings of cap­ ital logics and transformational tendencies. Instead, after having broken ‘into the chain of causation at the level of the [capitalist] system as a whole’ (Massey and Meegan, 1985, p. 6), the favoured approach was to identify and then intervene in local arenas of restructuring—​ understood in grounded and granular terms, as a socially shaped and politically contestable spaces. Counter to the view that these manifestly turbulent and disruptive economic geographies (understood as shifts in the ‘spatial structures’ of production) were trending towards long-​run equilibrium or that they were the result of ‘natural’ evolution (as orthodox accounts had it), or, for that matter, that they were governed by relatively immutable and mechanical forces like workplace deskilling and the falling rate of profit (as more conventional Marxisms would have it), in restructuring studies they were seen to be both inherently plastic and unpredictably political. These geographies were ‘established, reinforced, combatted and changed through political and economic strategies and battles on the part of managers, workers and political representatives’ (Massey, 1984, p. 85). Understandably, this did not lend itself to parsimonious or singular accounts of economic–​ geographical transformation, but rather to up-​close analyses of the construction, contingency, and contestability of restructuring pathways and regional-​adjustment models, together with explorations of (potential and actual) strategies for labour and for the local state (Storper and Walker, 1983; Clark et al., 1986; Lovering, 1989). As such, the restructuring approach was not associated with direct or deterministic theory claims, while its method was more about

Pluralizing Labour Geography    469 exploratory practice than formal procedure, its proponents insisting that it was no less potent or productive as a result. Likewise, the master concept of the spatial division of labour was often evoked in an almost metaphorical way (see Sayer, 1985; Warde, 1985), in the form of a methodological heuristic or political–​economic imaginary. Its central motif was the deepening separation between conception and execution, always expressed in social and spatial terms—​between disaggregated and detached elements of the labour process, like assembly work and R & D; between the branches and divisions of ‘stretched out’ corporate hierarchies and supply chains; and between peripheralized regions and centres of concentrated control. Crucially, the resulting economic geographies and reorganized power relations were to be understood together, in relational terms. Operationally, the spatial division of labour approach facilitated explorations of restructuring dynamics at the level of specific industries, or around new ‘rounds of (dis)investment’; and it enabled investigations of the diverse outcomes, mediations, and responses to these restructuring processes at the local scale, within ‘localities’. As a sensibility, this approach directed attention, then, towards the particularities of restructuring forms and dynamics; to non-​repeating patterns across usually local or regional cases; to the always-​distinctive positionality of these cases within wider divisions of labour, as uniquely nodal sites of restructuring; and to the fundamental character of restructuring as a contestable social and political process. The heyday of restructuring studies, as John Lovering (1989) later reflected, was a relatively brief one. The original remit may have been focused on a series of tightly bounded questions concerning the nexus of accelerating industrial transformation and workplace changes in technology and labour utilization, but this tight focus on the capitalist labour process had soon expanded to encompass shifts in household organization, gender relations, local politics and culture, and a range of other ‘locality effects’. Even if these wider concerns were licenced (and in some respects necessitated) by the foundational concept of restructuring—​which problematized qualitative transformations in the organization of the capitalist economy, including its (social) reproduction—​they implied a research agenda that was by its nature dissipative. A less than sympathetic charge was that the restructuring approach, and the locality studies that it would later inspire, represented an ‘empirical turn’, away from abstraction and towards local detail (Harvey, 1987; Smith, 1987). This critique may have been overdrawn, but it coincided with an apparent loss of theoretical edge and programmatic momentum. The project that in so many ways had been a product of the turbulent politics of the 1980s—​the ascendancy of Thatcherism and Reaganomics, and singular events like the year-​long coal miners’ strike and the abolition of the Labour Party-​controlled metropolitan councils in the UK—​did not survive that decade intact.

Regulation Axiomatic for the restructuring approach was the claim that employment relations, workplace transformations, and the ‘labour factor’ were diagnostically significant, both for shifts in the socio-​spatial organization of capitalism and for the moving terrain of politics. ‘Restructuring’, after all, was never simply about cyclical change or incremental adjustment. And although the language at the time had been more about Ford Motor Company than the Fordist regime of accumulation, as the coming wave of regulation–​theoretic treatments

470   Peck would have it, the earlier intuition that some kind of macro-​scale rupture was occurring seemed to have been confirmed. In the USA, de-​industrialization threatened the long-​run viability of once-​prosperous working-​class communities, to be superseded by a neoliberalized mode of growth rooted in and realized through widening inequalities (Bluestone and Harrison, 1982; Harrison and Bluestone, 1988; Peck, 2002). In the UK, there was a parallel sense of historical dislocation, accentuated by the polarizing stridency of the Thatcher government, its break with the postwar consensus on corporatist industrial relations, welfare-​ statist interventions, and Keynesian macroeconomic policy. Out went the commitments to full employment, regional policies, and public-​sector investment, which had reached their apogee with the social-​democratic modernization projects of the 1960s. In their place, after the turbulent decade of the 1970s, came the neoliberal doctrine of monetarism, along with the prioritization of entrepreneurship, finance, and the market. [T]‌he terms of the debate of the ‘modernisation’ period of the sixties were now reversed. This was not just in the most obvious senses, from Welfare State and public sector to cutback and privatisation. The attack on State expenditure and on public intervention was partly enabled by reworking the distinction between ‘productive’ and ‘non-​productive’ parts of the economy. In the sixties the productive sector was pre-​eminently manufacturing; services were relegated to second place … A decade later the distinction had come to mean something completely different; it was the market (private) sector which was productive, the public which was the parasitic dead-​weight … The periods of modernisation and monetarism were not just different because the wider economic situation had changed so dramatically, because ‘the requirements of accumulation’ were different; they were also dramatically contrasting in the dominant political interpretations of what those requirements were (Massey, 1984, pp. 265–​266).

By the early 1990s, a decade after these words were written, the political trajectory described by Massey had been consolidated, and the dominant concerns of transatlantic economic geography had shifted, too. The restructuring moment of the 1980s had transformed the field in at least one fundamental way: the concerns of economic geographers would from now on move, in more or less real time, with the constantly shifting economy itself. This would carry with it not only the virtues of social relevance and grounded engagement, but also certain risks associated with what is often a ‘presentist’ disposition, including a preoccupation with (presumed) historical novelty over evolutionary continuity, and ‘turns’ to new frameworks rather than the adaption or elaboration of existing tools (see Scott, 2000; Barnes et al., 2007; Sheppard et al., 2012). Allen Scott (2000) portrayed economic geography in terms of an uneven process of sedimentation, in which traces of previous concerns and practices invariably linger, in non-​systematic patterns of accumulation and habituated modes of analysis, while others are overwritten or forgotten. This culture of selective retention and open innovation would be revealed as the locus of concern began to shift from restructuring to regulation. The restructuring approach had assigned analytical priority to the (contested) workplace, positioned within transforming industries, framed in the first instance in national terms (as in Bluestone and Harrison’s concern with the US steel industry, or Massey’s work on British coal mining) in relation to ‘foreign’ competition, ‘overseas’ investment, or the ‘internationalization’ of production systems. The regulation optic, while in some senses internalizing many of these same concerns, tended to frame them in macroinstitutional and more explicitly historical terms. The analytical routines of regulation theory, especially in the Parisian

Pluralizing Labour Geography    471 variant that was most influential in economic geography, had been forged through studies of the eclipse of the Fordist–​Keynesian mode of growth, which in North America and Western Europe had entered its protracted crisis in the 1970s (Aglietta, 1979; Lipietz, 1986). This body of work had been concerned with the structural coupling of ‘regimes of accumulation’, patterned on the mass-​production methods of the Fordist factory and the norms of mass consumption and gendered reproduction on which they were predicated, and ‘modes of social regulation’, a co-​evolving complex of social, cultural, and political institutions indexed on varieties of the Keynesian welfare state (Tickell and Peck, 1992; Peck, 2000). While there was some engagement with the attendant ‘big geographies’ of macroeconomic and macroinstitutional transformation, the most sustained and consequential of these efforts took the transition from Fordism as a contextual frame, moving on to describe and debate a range of distinctively post-​Fordist dynamics at the regional and local scales (see Scott, 1988; Storper and Scott, 1992; cf. Gertler, 1988, 1992). The new motif was that of flexibility—​and flexibly specialized production systems, flexible firms, flexible labour markets, and flexible accumulation. If economic geography had experienced a moment of political relevance and interdisciplinary salience during the earlier moment of restructuring studies (Lovering, 1989), in some ways it would repeat the feat, albeit under very different circumstances, a decade later. In the ensuing post-​Fordism debate, the focus of attention moved from the bleeding to the leading edges of capitalist transformation, and to the upsides of restructuring through growth, in the high-​technology industries, in advanced business services and finance, and in resurgent forms of craft production. This was also the occasion of an ontological shift, from a firms-​ within-​industries approach to a nodes-​and-​networks orientation (see Amin and Thrift, 1992), in part as a reflection of the redundancy of the old, ‘standard’ industrial classifications, which looked increasingly archaic in a world of business-​to-​business subcontracting, vertical disintegration, and service-​sector growth, and in part as a recognition that a historically distinctive set of ordering principles seemed to be at work across the frontiers of a ‘new’ economy (Barnes et al., 2007; Sheppard et al., 2012), shaped by a quite different nexus of industries, institutions, and ideologies—​from financialized growth to flexible subcontracting, from cluster initiatives to collaborative production, from neoliberal deregulation to net-​ based commerce. The concern was less with industrial sectors, like engineering or car manufacturing, in the throes of restructuring, and more about new logics of growth and development, realized through multifunctional clusters and heterogeneous networks. About this time, there was something of a tonal transition, too. The target of restructuring studies had most often been big firms getting bigger and badder, and the pursuit of defensive or countervailing strategies on the part of organized labour and its allies. The fast-​growing industrial districts, however, were just as likely to be celebrated for their productive efficiency, adaptive capacity, and institutional superiority; stories from the flexible workplace were about responsibly autonomous work teams, cooperative upskilling, and the management of segmented workforces, rather than class conflict and industrial-​relations strife; and in place of the earlier emphasis on corporate hierarchies, running from the centres of coordination and control out to the peripheral branch plants and ‘runaway shops’, there came a rather more benign analysis of the potential of globalizing networks. Characteristically, there was continuing debate around these issues, although it is revealing that critical correctives called upon some of the old

472   Peck restructuring studies’ reflexes in emphasizing the ‘limits’ of the post-​Fordist vision and the ‘dark side’ of flexible production (see Gertler, 1988; Harrison, 1992, 1994). But the various sides of the post-​Fordism debate, even as they embraced questions of social and institutional transformation, still tended to be rather productivist at their explanatory core. Moving to enlarge the field of the analytically visible, Linda McDowell was among the first to call attention to the gender orders within which these ‘old’ and ‘new’ systems of production were embedded, raising questions of work, labour, and employment beyond the immediate confines of the restructured or flexibilized workplace, and indeed beyond the sphere of waged labour (McDowell, 1991, 2014). In a similar spirit, Hanson and Pratt (1995) made and demonstrated the case for exploring what they called ‘dynamic dependencies’ between households, (waged) workplaces, and local communities, reaching across the labour market as a space of analysis and intervention. As women had, in important respects, always been ‘flexible’ workers, and as they continued to shoulder most of the burden of unpaid domestic labour, transition models predicated on shifts in waged employment told only part of the story. Furthermore, the ongoing feminization of the labour market clearly spoke to more than the secular growth of the service sector (and in so-​called part-​time jobs), and also to mutually conditioning shifts in employment and gender norms. These processes, as Hanson and Pratt (1995, p. 1) revealed, are often intricately geographical in form and function: ‘social and economic geographies’, they argued, ‘are the media through which the segregation of large numbers of women into poorly paid jobs is produced and reproduced’. The supply of labour—​its quantity and quality—​is not mechanically determined by market demand; it is socially produced and regulated, as the complex outcome of cultural norms, state policies and programmes, household structures, gender relations, educational socialization, racial coding and discrimination, and much more. It follows that the labour market is a social institution, an irreducibly complex outcome of a host of relatively autonomous and often conflicting social forces, (but) one dimension of which is the production and reproduction of waged employment; beyond the factory gates and the office car park, the labour market, too, was recognized as a site of political struggle, social negotiation, and institutionalized compromise (Peck, 1996). If socio-​economic formations like labour markets are not unilaterally structured according to the needs of capital, and neither do they meaningfully resemble an idealized commodity market, then a wide array of demanding questions are raised concerning processes, practices, and patterns of regulation, governance, and institutional inclusion/​exclusion. These were questions that regulation theories had explored, in characteristically ‘macro’ terms, through a reading of Atlantic Fordist capitalism(s) anchored in the wage–​labour nexus, although it would later be suggested that the crux of this crisis-​prone system was migrating to the sphere of financialization, with significant consequences for social inequality, growth patterns, and employment security (Boyer, 2000). Labour geographers, for their part, have been rather less concerned with these big geographies, and instead have explored transformations in regulation and governance, often at the local and regional scale, in dialogue with mid-​level conceptions of contingent work, labour control, workfare, and forced labour (see e.g. Jonas, 1996; Peck, 2001; Kelly, 2002; Peck and Theodore, 2001, 2008; Strauss, 2012). Much of this work has been concerned with active moments of regulatory transformation, particularly at points of social stress in the job market, where the drivers of change are often traced to disciplinary state policies or exploitative business practices, and their neoliberalized combinations

Pluralizing Labour Geography    473 (Peck and Theodore, 2010). These incursions are often resisted—​sometimes defensively and sometimes more proactively—​but, in general, this line of work has been more inclined to ‘sample’ on moments of regulatory transformation than on moments of resistance per se. The latter has been the domain of Labour Geography ‘proper’.

Reorganization Labour geography found its name, and in a sense also its voice, in the late 1990s. In a decisive intervention, Andrew Herod (1997, p. 3) drew a distinction between ‘geographies of labour’, those neoclassical and Marxist analyses that in different ways conceived of the landscape of work, workers, and employment as secondary to enterprise decision-​making and the calculus of capital, and ‘labour geographies’, signifying a concern with the active roles, visions, and strategies of workers and workers’ organizations as the would-​be makers of worlds ‘in their own image’. What would become the self-​identifying project of Labour Geography was consequently defined in an orthogonal (and to some degree oppositional) relationship to those capital-​centric economic geographies that had gone before. While practically and normatively a project of the left, and therefore more sympathetic in its critiques of different currents in Marxian and neo-​Marxian economic geography than in what amounted to an affirmation of the earlier rejection of neoclassical market reductionism, it is nevertheless revealing that in the inverted optic of proper-​noun Labour Geography approaches as varied as those of August Lösch, Walter Isard, David Harvey, and Doreen Massey were marked by a shared failing, that of capital centricity, with its prioritization of corporate rationalities, dynamics, and decision-​making processes. Articulated against this composite foil, a self-​consciously labour-​centric Labour Geography would take as its licence and its problematic the question of how workers ‘actively produce economic spaces and scales in particular ways … as they implement in the landscape their own spatial fixes’, an endeavour understood as a ‘corrective to accounts that present workers as … inherently powerless and condemned only to follow the dictates of (global) capital’ (Herod, 2001, p. 46, original emphasis). This call would resonate strongly with a rising generation of activist scholars in and around the discipline of economic geography, and it would do so in the context of yet another moment of political salience. In the wake of the ‘Battle in Seattle’ and the millennial wave of protests against the institutions of the Washington Consensus, new energies were being channelled through various movements for global justice, many of which also gelled with a rising tide of locally based and/​or networked initiatives, including those for migrant-​worker rights and living-​wage ordinances, along with experiments in service-​sector organizing, community unionism, social-​justice campaigning, migrant-worker advocacy, and alternative economic development. On the ‘inside’, too, the project of Labour Geography would find itself embedded in a changed and still-​changing discipline. Economic geography was continuing to move with the churning political–​economic landscapes that for decades had been established its principal objects of study. Intellectually and sociologically, the field was also becoming more diverse and decentred than ever before. In contrast to the restructuring-​and-​regulation phase, which was marked by dominant and somewhat enduring centres of analytical gravity, engaged with a broadly shared methodological toolkit, the new orientation tended towards a diversity of parallel projects and

474   Peck proliferating sources of theoretical inspiration (see Chapter 8, and also Peck, 2005, 2012). From the mid-​1990s, the often taken-​for-​granted centrality of some of the old concerns—​its industrial–​productivist gaze, focused on leading sectors and related forms of waged work—​ was increasingly being questioned (and, indeed, transcended) from several quarters, including from feminist geographers, through post-​structuralist interventions, and from those engaged in various cultural, institutional, and relational turns (see Gibson-​Graham, 1996; Lee and Wills, 1997). By definition, the resulting intellectual milieu is not associated with a dominant analytical locus or singular direction of change. A ‘largely unplanned’ and somewhat spontaneous development, the emergence of Labour Geography reflected these environmental conditions (Castree, 2007, p. 854; Rutherford, 2010). As economic geography has cut its rather zigzagging course, so the palimpsest of ingrained habits and emergent practices has been repeatedly remade. It was in this context that, from the late 1990s, Labour Geography that exhibited on the character of a ‘project’, and with a unity of purpose that exhibited a simultaneously normative and analytical form. Normatively, this was reflected in an alignment with progressive currents within a (re)organizing labour movement, favouring (community) organizing and political mobilization models over more traditional (workplace) membership and representation-​based approaches, and reaching out to those groups that by virtue of race, ethnicity, gender, generation, or occupation had been marginalized (or excluded) from blue-​collar industrial unionism. Analytically, allegiance to the shared project of Labour Geography established lines of connection between what amounted to a series of quite disparate case studies of politically promising organizing campaigns and hard-​won battles. These combined acute analyses of (often local) disputes and campaigns with assessments of their potential as labour-​movement demonstration projects (see Walsh, 2000; Savage, 2006). The fact that these victories were being registered, as it were, against the grain of history—​on the heels of long-​established patterns of union-​membership decline across the most densely organized countries, an ideologically entrenched anti-​union (or ‘post-​union’) posture on the part of many functionally neoliberalized governments of the right and centre left, and some devastating strategic defeats for labour—​certainly added to their symbolic resonance if not (necessarily) their predictive significance. Nevertheless, a proclivity for sampling on positively realized expressions of workers’ agency, or what have been portrayed as ‘success stories’ (see Lopez, 2004; Coe et al., 2008; Tufts, 2009), tended to result in somewhat circular affirmations of Labour Geography’s founding charge. As the project of Labour Geography proper entered its second decade, it would become animated less by this original charter, or by propulsive, catalysing case studies, and more by reflective essays and stocktaking exercises. A recurrent theme here has been reflections on the changing position and prospects of a reorganized labour movement in the face, materially, of unrelenting systemic pressures on unions and, analytically, the challenge of situating labour’s agency (see Castree, 2007; Herod, 2010; Coe and Jordhus-​Lier, 2011). To be sure, the imperative of recovering labour’s agency was never conceived in naively unilateral terms, but its recovery as an antidote to capital centricity nevertheless produced a tendency to abstract from the wider political economy of corporate restructuring and regulatory transformation. Furthermore, while an atmosphere of (sometimes inescapably grim) realism pervaded the restructuring studies—​not only ontologically, but also politically—​there has been more than a whiff of idealism in the embrace of Labour Geography’s mission. ‘Many studies of labour union renewal’, as Tufts (2009, p. 981) pointed out, have been ‘largely prescriptive and

Pluralizing Labour Geography    475 often “idealize” labour transformation as an antithesis to the stagnant and defensive actions of retrenched business unionism’. The combination of an agency-​oriented ontology with a best-​practices selection strategy was never intended to yield a ‘representative’ reading of the actually existing state of the labour movement, but it has also produced a certain proclivity for affirmative sampling. This is more than a matter of pessimism of the intellect. The localized mobilization of labour’s agency, as Sweeney and Holmes (2013) have shown, may in some cases be detrimental to the wider strategic interests of the labour movement, or may undermine bargaining positions in other locations. As Labour Geography entered a more reflective and autocritical phase, recognition of the need to situate, contextualize, or ‘re-​embed’ labour’s agency has prompted calls for the simultaneous analysis of the strategies pursued by (fractions of) restructuring capital and (branches of) the (re)regulatory state, indeed for a recovered appreciation of structural constraints and contradictions of different kinds. To be sure, ‘integral’ modes of analysis conducted at the nexus productive and regulatory restructuring on the one hand, and the strategies of (re)organized labour on the other, have been a recurrent theme in the field (see Herod, 2000; Holmes, 2004; Peck, 2016), even if they have not defined its central tendency. Instead, this has more typically been focused on moments of labour activism, broadly defined. This galvanized the previously diffuse field of lower-​case labour geography, endowing the project with a strong sense of political and analytical purpose, while opening up new lines of dialogue with cognate fields like labour history, industrial relations, and working-​class studies. The capital-​L project can also be credited with progressively broadening, once again, the field of the analytically visible. Even if, ‘[i]‌n the early days, labour geographers had their eyes fixed firmly on paid employment and production issues’, particularly in the Global North (Castree, 2007, p. 855), more recent rounds of research have begun to reach beyond this, both socially and spatially (see Bergene et al., 2010; McGrath-​ Champ et al., 2010; Kelly, 2012; Coe, 2013). Like its predecessors, this too has become an expansive project.

Reproduction Issues of (social) reproduction have been a recurring presence across all stripes of labour geography, although often at the edges of the field of vision. Beyond the utilitarian construction of labour as ‘variable capital, an aspect of capital itself ’ (Harvey, 1982, p. 381, original emphasis), there has been a recognition that, unlike other factors of production, labour returns home at night (Harvey, 1989), although there has been rather less emphasis on the household, at least until recently, as a place of work, as a space of politics, or, indeed, as a theory-​making site (cf. McDowell and Massey, 1984; McDowell, 2014). Spatial divisions of labour was one of the first sustained efforts to explore the implications of the gendering of the ‘reserve army of labour’ for the geographies of industrial restructuring, albeit in rudimentary terms: in Cornwall, for example, ‘A married woman … was less likely to be “just” a wife and housekeeper to a “breadwinner” [than in the coalfields] and more likely to be involved some form of non-​domestic labour; keeping a bed-​and-​breakfast boarding house, maybe, or doing (paid or unpaid) work on the “family” farm’ (Massey, 1984, p. 225). The female workforces that were being targeted, at the time, by branch-​plant manufacturing firms seeking ‘green’

476   Peck (compliant and non-​unionized) labour exhibited pronounced geographies of their own, which, in turn, reflected spatially differentiated histories of patriarchal relations and gendered divisions of labour, waged and unwaged. These geographies of labour supply (again, which were not just matters of skills and wages, but involved structures of social reproduction and embedded patterns of socialization) have been shown to establish (pre)conditions for the creation and maintenance of different labour processes—​just one of the ways in which localized social geographies and racialized migration flows shape the form and spatiality of labour demand (Peck, 1996; McDowell, 2008; Samers, 2011). A principal concern for Hanson and Pratt (1995, pp. 122, 155) was with the ‘remarkable rootedness of many people’s (and especially women’s) lives’, along with its consequences for ‘limit[ing] women’s employment chances and [for the reproduction of] sex-​based occupational segregation’. This was an impetus for studies of suburban ‘pink-​collar ghettos’, for conceptions of the household as a ‘boundary institution’ of the labour market, and for explorations of the role of the sphere of social reproduction as a site of socialization and segmentation (England, 1993; Peck, 1996). As the commodification of labour is a necessarily incomplete process, labour’s ‘pseudo-​commodity’ form is inescapably social, and a joint product of the interaction and mutual conditioning of the waged economy and the domain of social reproduction. Social reproduction must therefore be understood as ‘a set of structured practices that unfold in dialectical relation with production, with which it is mutually constitutive and in tension’, Katz (2001, p. 711) has explained; by the same token, the abstraction that is the sphere of social reproduction also refers to the ‘fleshy, messy, and indeterminate stuff of everyday life’. An analytical vision that embraces what Mitchell et al. (2003, p. 429) have called ‘life’s work’ necessarily includes a host of activities (and social relations) that might be conventionally labelled as ‘non-​work’, from caring to learning, while also ‘blurring’ the boundaries between production and reproduction, paid and unpaid labour, and ‘work’ and its others. Against the accusation that this amounts to an all-​inclusive, catch-​all warrant for the study, as it were, of everything but the kitchen sink, there is the countercharge that to do otherwise is to ‘desocialize’ labour, to strip it of its inescapably human character, while also detaching the analysis of waged labour and (commodity) production from the circumstances of their existence and the circuits of their reproduction. The conditions under which labour is reproduced, on a daily and intergenerational basis, actually do include the kitchen sink, and properly so. At least since the ‘domestic labour’ debates of the 1970s, the household has been recognized not only as a hidden abode of reproductive work and unmeasured labour, but also as an alternate site of value creation (see Molyneux, 1979; Humphries and Rubery, 1984; Folbre, 2001). Beyond arguments around the ‘separate spheres’ of masculinized wage labour and feminized domestic work, and then their asymmetrical interaction, ‘the boundaries of and between the domestic and the economic, home and work [have since been recognized to be] more mobile, more porous, and contested, even if those boundaries did not melt away’ (Winders and Smith, 2015, pp. 11–​12). If the domains of domestic labour and household reproduction were once considered discrete, discreet, and ‘local’ matters, no longer is this the case: the work of social reproduction and the connections made by so-​called ‘care chains’ both exhibit an increasingly global reach. Feminist labour geographers, who led the drive to expand and rethink the definition of work, have since been at the forefront of a recent wave of work on (globalizing) care economies, many of which are simultaneously feminized,

Pluralizing Labour Geography    477 racialized, privatized, and (re)marginalized (see McDowell, 2004, 2008; Lawson, 2007; Massey, 2007; Wills et al., 2010; Pratt, 2012). Unhiding the economies (and geographies) of care reveals the social and institutional preconditions for the functioning of normalized economic subjectivities, like the rationally calculating, atomized, individual of neoclassical theory (‘economic man’). It demands an answer to the question, how is rational economic man reproduced? The answers invariably involve a great deal of female labour, and work that is (consequently) constructed as peripheral, secondary, and marginal, and as a result both undervalued and underpaid. The often-​ invisible labour of social reproduction, in turn, is a prerequisite for the high-​performance, high-​pressure, and individualized workplace that is so often celebrated in these after-​Fordist, neoliberalized times (see Henry and Massey, 1995; Ross, 2009; Brown, 2015). ‘Marginalizing care’, Victoria Lawson (2007, p. 5) writes, ‘furthers the myth that our successes are achieved as autonomous individuals and, as such, we have no responsibility to share the fruits of our success with others’. Questions of social reproduction therefore problematize political and normative constructions of (productive) work, as well as the gendered and institutionalized ‘boundaries’ that are conventionally established around workplaces, labour markets, households, and economies, with all their implications for socially differentiated processes of valuation, remuneration, and inclusion/​exclusion. In turn, these processes shape the measurement and meaning of ‘economy’ itself, from the calculation of national accounts (which exclude unpaid domestic work) to the production of un/​employment statistics (which exclude those not actively seeking waged employment), just as they frame and distort orthodox understandings of phenomena like labour-​market change or globalization. Social reproduction, it has been observed, is the ‘missing figure’ in globalization debates, as well as the marginalized Other of the ideologically and materially privileged market economy, while feminized care workers are among its missing subjects (Katz, 2001; Nagar et al., 2002; Winders and Smith, 2015). ‘Globalized capitalism has changed the face of social reproduction worldwide over the past three decades’, Cindi Katz has argued, ‘enabling intensification of capital accumulation and exacerbating differences in wealth and poverty’. The demise of the social contract as a result of neoliberalism, privatization, and the fraying of the welfare state is a crucial aspect of this shift … The flip side of the withdrawal of public and corporate support for the social wage is a reliance on private means of securing and sustaining social reproduction—​not just the uncompensated caring work of families, most commonly women, but also a shunting of responsibility, often geographically, that has clear class, race, and national components. For instance, the social reproduction of a migrant workforce is carried out in its members’ countries of origin. When they are employed elsewhere, this represents a direct transfer of wealth from generally poorer to richer countries. Variable capital produced in one site and tapped in another is no less a capital transfer than the extraction of raw materials, debt servicing, and the like. Yet this transfer seems to be of no moment to most theorists of globalization (Katz, 2001, p. 710).

The postwar social contract that prevailed across the Fordist–​Keynesian countries—​and which institutionalized a ‘social wage’ of non-​wage benefits, services, and supports, yielding mutual adjustments in the organization of households, enterprises, and the public sector—​has been variously deconstructed and disaggregated through several decades of neoliberal transformation. Tasks of social reproduction have been privatized, contracted out, and devolved (back) to households, to faith-​based, non-​governmental to ‘shadow-​state’

478   Peck institutions, and to an array of for-​profit organizations. This, in turn, has been associated with a significant transformation of the labour market, as women workers have been on the front lines not only of the rollback, restructuring, and privatization of public services, including health care, education, and childcare, but also the secular expansion of private-​ consumer services—​now itself one of the largest employers, economy-​wide (see Bowman et al., 2014; McDowell, 2014). All of this has pushed new costs, risks, and stresses onto households, and onto the shoulders of women in particular. In the context of real-​wage stagnation, growing precarity, and the normalization of ‘flexible’ work norms (including multiple job holding), women have often been confronted by double or triple burdens in ‘feminizing’ wage–​labour markets. ‘The need for multiple wage-​earners in a household’, Linda McDowell (2004, p. 150) has written, ‘means that all the work of social reproduction must be squeezed into a shorter and shorter time or redistributed among other networks’. In the process, under a particular configuration of historical and geographical circumstances, employment and gender norms are being transformed—​along with the boundaries (and patterned relationships) between enterprises and families, households and states, public and private, and production and reproduction.

Conclusion: For Recombinant Labour Geographies This chapter has highlighted the plurality of approaches to questions of labour, work, and employment in post-​1970s Anglo-​American economic geography—​before, during, and perhaps ‘after’ the moment of Labour Geography proper as an explicit and named project. The relationship between what have been rather distinctive currents in a pluralized and palimpsest-​ like labour geography—​ from restructuring through regulation and reorganization to reproduction—​has not been one of competition or simple succession, however. There has always been a degree of reciprocal learning and cross-​fertilization. Yet while these (and other) strands and projects of labour geography have existed for the most part in a state of comradely cohabitation and intermittent communication, they have also displayed some of the characteristics of self-​referential ‘cells’ of activity, latterly with closer relationships to extra-​disciplinary than to subdisciplinary communities (see Coe and Jordhus-​Lier, 2011; Peck, 2012). Furthermore, they have been arrayed more-​or-​less sequentially across the recent history of (critical) economic geography, and as such reflect a series of context-​specific (if also somewhat cumulative) encounters with various forms of Marxism, feminism, institutionalism, post-​structuralism, and so on. But for all of economic geography’s ‘turns’, and for all its theoretical promiscuity and magpie-​like eclecticism, traces of these earlier encounters invariably remain, in some form another, in the diversity of contemporary practice. In these circumstances, it may not be too fanciful to anticipate that some of the next moves for a pluralized labour geography might be integrative and intersectional, as opposed to additive or alternative ‘turns’ in different directions. There is still-​to-​be-​realized potential for dialogic deepening across the beyond these various modalities of small-l labour geography, excavating the combinational potential of which would amount to much more than turning inward or turning back. After all, even if they have not always been in deep dialogue,

Pluralizing Labour Geography    479 the various strands of actually existing labour geography have generally not been advanced through scorched-​earth critiques of their respective precursors and alternatives. None can claim a monopoly: each has yielded a distinctive analytical optic, along with characteristic principles of theoretical pertinence and methodological priority. The restructuring approach sought to connect the dynamics of capitalist restructuring with the particularities of workplace change and local social relations at a time of objective macroeconomic dislocation, with an emphasis on class and gender. Regulationist studies built from similar foundations, but were more explicitly oriented to the macro-​patterning of institutional transformations, extending (down) to considerations of localized governance and projects of reregulation. The advent of Labour Geography proper was marked by a defining concern with labour as a collective political actor, with its own rationales of action and organization, extending beyond the workplace and into local communities and transnational arenas. And explorations of the realm of social reproduction have sought to broaden both the field of vision and prevailing definitions of ‘work’, deconstructing the boundaries between the capitalist labour process and the household economy, between waged and unwaged work, and between production and reproduction. Arguably, the last of these moments has been the most encompassing, although none is fully a substitute for the others. Like the rudely heterodox field of economic geography more generally, labour geography seems in its very nature to be restlessly proliferative—​leading, predictably, to an undercurrent of concern about disciplinary fragmentation and amnesia. The promise of a reflexively ‘recombinant’ labour geography, in contrast, is not only to be cognizant of diverse contributions and perspectives over what has been a distinctively ‘layered’ intellectual history, but also to confront in bold and creative ways the structural challenges of the restructuring present. Foremost among these contemporary challenges is a knot of questions relating to the reproduction of decent work, fair employment, and sustainable livelihoods in global terms, not least those relating to inequality, insecurity, and informalization—​the scale and scope of which exceeds the capacity of siloed or sectional approaches, calling instead for more integral, intersectional, and integrative modes of analysis. In a post-​Kuznets curve world in which increased economic growth and accelerated development tend to produce more, rather than less, socio-​spatial inequality, endemic inequality has been established as something akin to a globalizing condition. Yet the diverse geographies of inequality in work and welfare, along with their causes and consequences, have yet to be interrogated in systematic or sustained ways by economic geographers—​one area where creative collaborations are surely called for, particularly between those trained in quantitative and qualitative methods. Furthermore, if insecurity was once regarded, at least in Fordist labour markets, as an affliction of marginal or ‘secondary’ workers, the widespread diffusion of contingent work and ‘precarity’, the fracturing and truncation of career paths, the privatization of education and training, and the erosion of employment protections threatens to confer normative status on employment and job insecurity, often under the euphemistic guise of ‘flexibility’. Similarly, the informalization and ‘deregulation’ of labour markets, long recognized as a structural feature of the economies of the Global South, have become entrenched phenomena, although hardly homogenously so. More expansively global labour geographies will need to trace these dynamics across cases and conjunctures, while continuing to unpack the meaning of concepts hitherto defined in little more than provisional and negative terms, according what they are lacking or not (as informal or deregulated labour markets).

480   Peck In the three decades that have passed since the publication of Doreen Massey’s Spatial divisions of labour, the terrain of labour geography has been mapped and remapped—​ perhaps not exhaustively but in a cumulative sense quite comprehensively. As economic geography has proved itself to be an almost spontaneously proliferative field, there is little doubt that productive boundary work will continue, not least through extensions of the social and spatial reach of various permutations of labour geography. The projects of labour geography began by reaching into the contested terrain of the industrial workplace before reaching out, more or less successively, into local communities and into post-​industrial workplaces, into the (de)regulatory folds of the state, into the collective institutions of the labour movement, and into the domains and circuits of reproduction. After three decades of development, there is reason for the diverse community of labour geographers to work across as well as out, and to enrich what ought to be mutually adaptive conversations. Pluralized labour geography has acquired a wide reach; maybe the next challenge should be to consolidate its intersectional grasp?

Acknowledgements The support of the Social Science and Humanities Research Council is gratefully acknowledged. I thank Kendra Strauss, Tod Rutherford, Nik Theodore, and the editors for comments on an earlier draft of this chapter, even as the responsibility for any remaining errors of omission and commission remains mine.

References Aglietta, M. (1979). A Theory of Capitalist Regulation: The US Experience (London: Verso). Amin, A. and Thrift, N. (1992). ‘Neo-​Marshallian nodes in global networks’. International Journal of Urban and Regional Research 16: 571–​587. Barnes, T.J., Peck, J., Sheppard, E., and Tickell, A. (2007). ‘Methods Matter: Transformations in Economic Geography’ in A. Tickell, E. Sheppard, J. Peck, and T.J. Barnes (eds) Politics and Practice in Economic Geography, pp. 1–​24 (London: SAGE). Bergene, A.C., Endresen, S.B. and Knutsen, H.M. (eds) (2010) Missing Links in Labour Geography (Aldershot: Ashgate). Bluestone, B. and Harrison, B. (1982). The Deindustrialization of America:  Plant Closings, Community Abandonment, and the Dismantling of Basic Industry (New York: Basic Books). Boddy, M. and Fudge, C. (1984). Local Socialism (London: Macmillan). Bowman, A., Irtürk, I., Froud, J., Law, J., Moran, M., and Williams, K. (2014). The End of the Experiment? (Manchester: Manchester University Press). Boyer, R. (2000). ‘Is a finance-​led growth regime a viable alternative to Fordism? A preliminary analysis’. Economy and Society 29: 111–​145. Brown, W. (2015). Undoing the Demos: Neoliberalism’s Stealth Revolution (New York: Zone). Castree, N. (2007). ‘Labour geography: a work in progress’. International Journal of Urban and Regional Research 31: 853–​862. Clark, G.L., Gerlter, M.S., and Whiteman, J. (1986). Regional Dynamics: Studies in Adjustment Theory (Boston, MA: Allen and Unwin).

Pluralizing Labour Geography    481 Cochrane, A. (1986). ‘What’s in a strategy? The London Industrial Strategy and municipal socialism’. Capital and Class 28: 187–​193. Coe, N.M. (2013). ‘Geographies of production III: making space for labour’. Progress in Human Geography 37: 271–​284. Coe, N.M. and Jordhus-​Lier, D.C. (2011). ‘Constrained agency? Re-​evaluating the geographies of labour’. Progress in Human Geography 35: 211–​233. Coe, N.M., Dicken, P., and Hess, M. (2008). ‘Global production networks: realizing the potential’. Journal of Economic Geography 8: 271–​295. Doussard, M. (2013). Degraded Work:  The Struggle at the Bottom of the Labor Market (Minneapolis, MN: University of Minnesota Press). England, K.V.L. (1993). ‘Suburban pink collar ghettos:  the spatial entrapment of women?’ Annals of the Association of American Geographers 83: 225–​242. Folbre, N. (2001). The Invisible Heart (New York: New Press). Gertler, M.S. (1988). ‘The limits of flexibility: comments on the post-​Fordist vision of production and its geography’. Transactions of the Institute of British Geographers 13: 419–​432. Gertler, M.S. (1992). ‘Flexibility revisited: districts, nation-​states, and the forces of production’. Transactions of the Institute of British Geographers 17: 259–​278. Gibson-​Graham, J-​K. (1996). The End of Capitalism (As We Knew It): A Feminist Critique of Political Economy (Minneapolis, MN: University of Minnesota Press). Green, M. and Lawson, V. (2011). ‘Recentring care: interrogating the commodification of care’. Social and Cultural Geography 12: 639–​654. Hanson, S. and Pratt, G. (1995). Gender, Work, and Space (New York: Routledge). Harrison, B. (1992). ‘Industrial districts: old wine in new bottles?’ Regional Studies 26: 107–​121. Harrison, B. (1994). ‘The dark side of flexible production’. National Productivity Review 13: 479–​501. Harrison, B. and Bluestone, B. (1988). The Great U-​Turn:  Corporate Restructuring and the Polarizing of America (New York: Basic Books). Harvey, D. (1982). The Limits to Capital (Oxford: Basil Blackwell). Harvey, D. (1987). ‘Three myths in search of a reality in urban studies’. Society and Space 5: 367–​376. Harvey, D. (1989). The Urban Experience (Oxford: Blackwell). Henry, N. and Massey, D. (1995). ‘Competitive time-​space in high technology’. Geoforum 26: 49–​64. Herod, A. (1997). ‘From a geography of labor to a labor geography’. Antipode 29: 1–​31. Herod, A. (ed.) (1998). Organizing the Landscape: Geographical Perspectives on Labor Unionism (Minneapolis, MN: University of Minnesota Press). Herod, A. (2000). ‘Implications of just-​in-​time production for union strategy: lessons from the 1998 General Motors-​United Auto Workers dispute’. Annals of the Association of American Geographers 90: 521–​547. Herod, A. (2001). Labor Geographies: Workers and the Landscapes of Capitalism (New York: Guilford). Herod, A. (2010). ‘Labour Geography:  Where Have We Been? Where Should We Go?’ in A.C. Bergene, S.B. Endresen, and H.M. Knutsen (eds) Missing Links in Labour Geography, pp. 15–​28 (Aldershot: Ashgate). Holmes, J. (2004). ‘Re-​scaling collective bargaining: union responses to restructuring in the North American auto industry’. Geoforum 35: 9–​21. Humphries, J. and Rubery, J. (1984). ‘The reconstitution of the supply side of the labour market: the relative autonomy of social reproduction’. Cambridge Journal of Economics 8: 331–​346.

482   Peck Jonas, A.E.G. (1996). ‘Local labour control regimes: uneven development and the social regulation of production’. Regional Studies 30: 323–​338. Katz, C. (2001). ‘Vagabond capitalism and the necessity of social reproduction’. Antipode 33: 709–​728. Kelly, P.F. (2002). ‘Spaces of labour control:  comparative perspectives from Southeast Asia’. Transactions of the Institute of British Geographers 27: 395–​411. Kelly, P.F. (2012). ‘Labor, Movement:  Migration, Mobility, and Geographies of Work’ in T.J. Barnes, J. Peck, and E. Sheppard (eds) The Wiley-​Blackwell Companion to Economic Geography, pp. 431–​443 (Oxford: Wiley). Lawson, V. (2007). ‘Geographies of care and responsibility’. Annals of the Association of American Geographers 97: 1–​11. Lee, R. and Wills, J. (eds) (1997). Geographies of Economies (London: Arnold). Lipietz, A. (1986). Mirages et miracles:  problèmes de l’industrialisation dans le tiers-​monde (Paris: La découverte). Lopez, S.H. (2004). Reorganizing the Rust Belt:  An Inside Study of the American Labor Movement (Berkeley, CA: University of California Press). Lovering, J. (1988). ‘The local economy and local economic strategies’. Policy and Politics 16: 145–​158. Lovering, J. (1989). ‘The Restructuring Debate’ in R. Peet and N.J. Thrift (eds) New Models in Geography, Vol. 1, pp. 213–​242 (London: Unwin Hyman). McDowell, L. (1991). ‘Life without father and Ford: the new gender order of post-​Fordism’. Transactions of the Institute of British Geographers 16: 400–​419. McDowell, L. (2004). ‘Work, workfare, work/​life balance and an ethic of care.’ Progress in Human Geography 28: 145–​163. McDowell, L. (2008). ‘Thinking through work: complex inequalities, construction of difference and trans-​national migrants’. Progress in Human Geography 32: 491–​507. McDowell, L. (2014). ‘Gender, work, employment and society: feminist reflections on continuity and change’. Work, Employment & Society 28: 825–​837. McDowell, L. and Massey, D. (1984). ‘A Woman’s Place?’ in D. Massey and J. Allen (eds) Geography Matters, pp. 128–​147 (Cambridge: Cambridge University Press). McGrath-​Champ, S., Herod, A., and Rainnie, A. (eds) (2010). Handbook of Employment and Society: Working Space (Cheltenham: Edward Elgar). Massey, D. (1984). Spatial Divisions of Labour: Social Structures and the Geography of Production (Basingstoke: Macmillan). Massey, D. (2007). World City (Cambridge: Polity). Massey, D. and Meegan, R. (1982). The Anatomy of Job Loss:  The How, Why and Where of Employment Decline (London: Methuen). Massey, D. and Meegan, R. (1985). ‘Introduction:  The Debate’ in Massey, D. and Meegan, R. (eds) Politics and Method:  Contrasting Studies in Industrial Geography, pp. 1–​ 12 (London: Taylor and Francis). Mitchell, K., Marston, S.A., and Katz, C. (2003). ‘Life’s Work: An Introduction, Review and Critique’ in Mitchell, K., Marston, S. A., and Katz, C. (eds) Life’s Work: Geographies of Social Reproduction, pp. 1–​26 (Oxford: Blackwell). Molyneux, M. (1979). ‘Beyond the domestic labour debate’. New Left Review 116: 3–​27. Nagar, R., Lawson, V., McDowell, L., and Hanson, S. (2002). ‘Locating globalization: feminist (re) readings of the subjects and spaces of globalization’. Economic Geography 78: 257–​284. Peck, J. (1996). Work-​place: The Social Regulation of Labor Markets (New York: Guilford).

Pluralizing Labour Geography    483 Peck, J. (2000). ‘Doing Regulation’ in G.L. Clark, M. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 61–​80 (Oxford: Oxford University Press). Peck, J. (2001). Workfare States (New York: Guilford). Peck, J. (2002). ‘Labor, zapped/​growth, restored? Three moments of neoliberal restructuring in the American labor market’. Journal of Economic Geography 2: 179–​220. Peck, J. (2005). ‘Economic sociologies in space’. Economic Geography 81: 129–​176. Peck, J. (2012). ‘Economic geography: island life’. Dialogues in Human Geography 2: 113–​133. Peck, J. (2016). ‘The right to work, and the right at work’. Economic Geography 92: 4–​30. Peck, J. and Theodore, N. (2001). ‘Contingent Chicago: restructuring the spaces of temporary labor’. International Journal of Urban and Regional Research 25: 471–​496. Peck, J. and Theodore, N. (2008). ‘Carceral Chicago: making the ex-​offender employability crisis’. International Journal of Urban and Regional Research 32: 251–​281. Peck, J. and Theodore, N. (2010). ‘Labor Markets From the Bottom Up’ in S. McGrath-​Champ, A. Herod, and A. Rainnie (eds) Handbook of Employment and Society: Working Space, pp. 87–​105 (Cheltenham: Edward Elgar). Pratt, G. (2012). Families Apart (Minneapolis, MN: University of Minnesota Press). Ross, A. (2009). Nice Work if You Can Get it: Life and Labor in Precarious Times (New York: NYU Press). Rutherford, T. (2010). ‘De/​re-​centring work and class? A review and critique of labour geography’. Geography Compass 4: 768–​777. Rutherford, T.D. and Gertler, M.S. (2002). ‘Labour in “lean” times: geography, scale and the national trajectories of workplace change’. Transactions of the Institute of British Geographers 27: 195–​212. Samers, M. (2011). ‘The Socioterritoriality of Cities: A Framework for Understanding the Incorporation of Migrants in Urban Labor Markets’ in N.G. Schiller and A. Caglar (eds) Locating Migration: Rescaling Cities and Migrants, pp. 42–​59 (Ithaca, NY: Cornell University Press). Savage, L. (2006). ‘Justice for janitors: scales of organizing and representing workers’. Antipode 38: 645–​666. Sayer, A. (1985). ‘Industry and space: a sympathetic critique of radical research’. Society and Space 3: 3–​29. Scott, A.J. (1988). New Industrial Spaces:  Flexible Production Organization and Regional Development in North America and Western Europe (London: Pion). Scott, A.J. (2000). ‘Economic Geography:  The Great Half-​Century’ in G.L. Clark, M. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 18–​44 (Oxford: Oxford University Press). Scott, A.J. and Storper, M. (1986). Production, Work, Territory: The Geographical Anatomy of Industrial Capitalism (New York: Allen & Unwin). Sheppard, E., Barnes, T.J., and Peck, J. (2012). ‘The Long Decade:  Economic Geography, Unbound’ in T.J. Barnes, J. Peck, and E. Sheppard (eds) The Wiley-​Blackwell Companion to Economic Geography, pp. 1–​24 (Oxford: Wiley). Smith, N. (1987). ‘Dangers of the empirical turn:  some comments on the CURS initiative’. Antipode 19: 59–​68. Storper, M. (1997). The Regional World:  Territorial Development in a Global Economy (New York: Guilford). Storper, M. and Scott, A.J. (eds) (1992). Pathways to Industrialization and Regional Development (London: Routledge).

484   Peck Storper, M. and Walker, R. (1983). ‘The theory of labour and the theory of location’. International Journal of Urban and Regional Research 7: 1–​43. Strauss, K. (2012). ‘Coerced, forced and unfree labour: geographies of exploitation in contemporary labour markets’. Geography Compass 6: 137–​148. Sweeney, B. and Holmes, J. (2013). ‘Problematizing labour’s agency: rescaling collective bargaining in British Columbia pulp and paper mills’. Antipode 45: 218–​237. Tickell, A. and Peck, J. (1992). ‘Accumulation, regulation and the geographies of post-​ Fordism: missing links in regulationist research’. Progress in Human Geography 16: 190–​218. Tufts, S. (2009). ‘Hospitality unionism and labour market adjustment: toward Schumpeterian unionism?’ Geoforum 40: 980–​990. Tufts, S. and Savage, L. (2009). ‘Labouring geography: negotiating scales, strategies and future directions’. Geoforum 40: 945–​948. Walsh, J. (2000). ‘Organizing the scale of labor regulation in the United States: service-​sector activism in the city’. Environment and Planning A 32: 1593–​1610. Warde, A. (1985). ‘Spatial Change, Politics and the Division of Labour’ in D. Gregory and J. Urry (eds) Social Relations and Spatial Structures, pp. 190–​212 (London: Macmillan). Wills, J. (1998). ‘Taking on the CosmoCorps? Experiments in transnational labor organization’. Economic Geography 74: 111–​130. Wills, J. (2001). ‘Community unionism and trade union renewal in the UK: moving beyond the fragments at last?’ Transactions of the Institute of British Geographers 26: 465–​483. Wills, J., Datta, K., Evans, J., and Herbert, J. (2010). Global Cities at Work: New Migrant Divisions of Labour (London: Pluto). Winders, J. and Smith, B E. (2015). ‘Social reproduction and capitalist production: a genealogy of spatial relations’. Mimeo, Department of Geography, Syracuse University.

Chapter 25

Precariou s Work a nd W inner-Tak e -​a l l Ec onom i e s Kendra Strauss Introduction It is 2 January 2015 and the first working day of the New Year dawns in the city of London. John Li, an Internet security consultant, has just stepped off of a red-​eye flight from Singapore. While waiting to clear customs, he calls home; Ana, his family’s Filipino nanny, answers and tells him sleepily that his wife and children are in bed. John has time to use a smartphone application to order a private car, and with no luggage to pick up, he is able to meet his driver, Ted, within ten minutes. On the way to Canary Wharf, Ted tells John that he has been driving for the ride-​share company for three months, ever since the security firm he used to work for decided he was ‘redundant’. John arrives at the office half an hour before his first meeting, so he asks the woman at the reception desk where he can get coffee. Alicia explains that she is a temp and does not know the area well, but thinks there is a café around the corner. John finds an outlet of a large sandwich chain, which, at 6:30 a.m., is still quiet; waiting for his coffee, he chats with the barista, who is from an area of Spain where John once went on holiday. The five workers in this scenario have at least one thing in common: they perform wage labour but are not in a standard employment relationship (SER). John is a contract worker from Hong Kong, living in Singapore for ten months while completing a project for HSBC Bank. Friends of John in Singapore originally hired Ana through an agency; he and his wife employed her informally when the other family left the city because Ana had reached the end of her two-​year foreign domestic worker permit. The company Ted drives for pays him for each client he successfully transports, but considers him an independent contractor rather than an employee of the company. Alicia, who was born in London to Nigerian parents, has been temping for Manpower for two years since finishing a degree in marketing. She has not been able to find a permanent job. The barista, who worked as a municipal planner in his

486   Strauss hometown in Southern Spain but was laid off following the 2008 financial crisis, is on a flexible contract that usually fails to deliver the full-​time hours he was promised. The notion that these are ‘non-​standard’ workers, whose situations deviate from a norm of full-​time, permanent, direct employment with a single employer or firm, has been both highlighted and challenged by the documented rise of precarious employment or ‘precariousness in work’ (Vosko, 2010). Highlighted by a large and evolving body of theoretical and empirical work on the changing political economy of labour markets and their regulation, which has drawn attention to the decline of the SER (McDowell and Christopherson, 2009; Kalleberg, 2011; Arthurs and Stone, 2013); challenged, in the sense of pointing out that the SER’s force, even in the Anglo-​European heartlands of Fordism, was for many women, immigrant, and racialized workers more normative than descriptive (Vosko, 2000; Strauss and Fudge, 2014). These developments not only signal the importance of changing employment norms, but also challenge a dichotomous conceptualization of standard and non-​ standard employment, suggesting instead the need for a more nuanced calibration of (in) security in work. The recalibration of the notion of ‘non-​standard’ work has happened for at least three reasons. Firstly, precarious work has increasingly become the standard for some groups in the labour markets of ‘advanced economies’, reinforcing and exacerbating segmentation and polarization between primary and secondary labour markets. Secondly, such polarization has not been the straightforward outcome of capital and the state producing precariousness through the deregulation of labour and employment regimes. Precarious employment is related to strategies of flexibilization which, as Peck (1996) has highlighted, involve the active re-​regulation of labour markets. Finally, polarization among workers in groups traditionally identified as disadvantaged—​including women, immigrant, and migrant workers—​signals the need to understand axes of differentiation as evolving and contested, rather than static and immutable categories. For Alicia, the temp receptionist in our scenario, being a young worker of colour increases the likelihood of being in temporary, precarious employment (Forde and Slater, 2005) more than being a woman, despite the association of the growth of precariousness with the feminization of work (Vosko, 2000). Precariousness as insecurity means that political economic accounts of precarious work overlap with ontological explorations of precarity as both universal and as socially and politically constructed (Butler, 2004; Ettlinger, 2007; see Strauss, 2017, for a discussion). There is a growing interest in the concept of precarity among political and development geographers and migration scholars in the field (see, e.g., Waite, 2009; Woon, 2011; Harker, 2012; Lewis et al., 2014; Waite et al., 2014). But excepting excellent work on London’s ‘migrant division of labor’ (Wills et al., 2009; Dyer et al., 2011), and on the creative industries (Donald et al., 2013; Watson, 2013; Warren, 2014), there has been relatively little engagement among economic and labour geographers with the work on precarious employment that has emerged in the last two decades in political economy, sociology, and labour law. My point in this chapter is not to castigate economic geographers for this lack of attention. Instead, I want to examine the relevance of the concepts of precarious employment and precariousness in work in light of the polarization that increasingly characterizes ‘winner-​take-​ all’ economies. Following Coe (2013, p. 276), who argues that ‘the role of agencies and other intermediaries . . . merits urgent further attention’, I examine the utility of these concepts to economic and labour geographers for theorizing and analysing recent trends in intermediated work. In particular, I am interested in the growing salience of intermediated self-​employment

Precarious Work and Winner-Take-all Economies    487 in the ‘sharing economy’, the evolving nature of intermediation itself, and the implications of changing patterns of employment for labour market regulation beyond the SER. The chapter is in three parts. I first provide a brief overview of how the concept of precarious employment has developed outside of the discipline of geography and reflect on its relationship with empirical trends in employment and income distribution. The second section looks at the relationship between intermediation and precariousness, arguing that geographical literatures on employment agencies and labour intermediaries need to be augmented with an expanded conceptualization of intermediation. The final part of the chapter, drawing on recent research on attitudes to labour market regulation in the UK, offers some thoughts on the implications of these developments for workers’ struggles against precarious employment.

Precarious Employment, Polarized Labour Markets, and Income Inequality As economic geographers have noted, precariousness is a complex concept that has emerged at the intersection of two political and intellectual traditions:  Marxist and neo-​Marxist theoretical work and activism, related to the Bourdieu’s concept of precarité and to the autonomist-​inspired San Precario movement in Italy; and more institutional approaches to the study of labour markets in labour studies, labour law, and sociology (McDowell and Christopherson 2009; Strauss, 2017). Both currents emphasize insecurity in employment, as well as the implications for more generalized forms of insecurity beyond the sphere of paid work (at the scale of the individual, the community, and the nation state). Research by the International Labour Organization in the 1980s was instrumental in defining precarious work in this broader sense: ‘precarious work goes beyond the form of employment to look at the range of factors that contribute to whether a particular form of employment exposes the worker to employment instability, a lack of legal and union protection, and social and economic vulnerability’ (Rodgers, 1989, p. 1, emphasis added). In practice, however, the Anglo-​North American literature on precarious work evolved more narrowly in relation to the aspects of precarious employment identified by Rodgers (1989). These included uncertainty around continuation of employment (temporary or contract work), the ability to exercise control in the labour process over such aspects as working conditions and the pace of work (linked to the presence or absence of collective representation), the presence and degree of regulatory protections, and sufficiency of income (Strauss and Fudge, 2014, p. 11). More recently, emerging concerns about increases in involuntary part time work and ‘zero-​hours contracts’ (the problem for our fictional barista in the UK) have illustrated the evolving nature of precarious employment relations.1 Feminist political economists and labour lawyers, building on critiques of the limitations models of labour market segmentation (Peck, 1996) and integrating concepts of the male breadwinner model of employment and gender regimes (Fudge, 2005), sought to re-​focus attention on the question of the relationship between the feminization of work, social location, and precarious employment (Vosko, 2000; Cranford and Vosko, 2005). This approach integrated categories of social difference relating to gender, race and racialization, and

488   Strauss occupational context, and in many cases included implicit and explicit critiques of the SER itself. In Western Europe and North America, the approach also translated into significant empirical projects for mapping increasing female labour market participation, the rise of precarious forms of employment, and precariousness in work. These overlapped with the increasing interest in geography in intermediated employment and the expansion of temporary employment agencies (Peck and Theodore, 2007; Coe et al., 2011). The uneven acceleration of economic migration, including within an expanded internal European Union (EU) labour market, and the imposition of regimes favouring temporary, circular forms of migration in a diverse range of countries, including Canada, China, and Israel, made legal status (related to citizenship, employment, and settlement rights) an increasingly salient and pressing consideration in frameworks for theorizing social location in relation to precarious employment (Goldring et al., 2009; Anderson, 2010). Like research on intermediated employment, these different literatures have had to grapple with questions of the role of choice and new cultures of work versus structural differences in worker power and voice. To what extent do workers in ‘the knowledge economy’—​like John Li, our fictional consultant—​actively embrace risk? Do they seek out variety and new opportunities, eschewing traditional forms of security and collective voice for more lucrative, and individualized, forms of opportunity? Critical approaches to the study of flexible employment point out that highly skilled mobile workers may benefit from such arrangements, but that such benefits accrue disproportionately to well-​paid workers in primary labour markets with high levels of human and social capital, and thus high levels of bargaining power (Kim, 2013; for a nuanced picture of flexibility and insecurity in the high-​ tech sector, see Benner 2002). These relations are thus gendered, classed, and racialized (Whitson, 2010). One of the key debates about precariousness versus flexibility addresses the degree to which forms of non-​standard work such as part-​time, contract, or temporary work are actively favoured by those seeking to achieve work–​life balance, in particular in relation to caring responsibilities—​and thus the gendered implications of flexible working (Brandth and Kvande, 2001; James, 2011; Smith et al., 2011). Euro-​centric approaches that are attentive to the different ways in which ‘choice’ is exercised by gendered, classed, and racialized workers have nevertheless also been critiqued for taking particular kinds of labour markets as given, for example in the lack of attention to the ‘informal’ sector (Williams and Round, 2007). Recent careful empirical work on the post-​recession UK labour market, characterized by relatively low levels of temporary work (6.2% in 2013 vs an EU 28 average of 13.8%) (Eurostat, 2014) and relatively high levels of part-​time work (24.5% in 2013 vs an EU 28 average of 17.8%) (OECD, 2015a), suggests that what Green and Livanos (2014, p. 8) call ‘involuntary non-​ standard work’ (INE) is a particular issue for ‘so-​called “vulnerable groups,” such as women, young workers and workers of non-​white ethnic background’. Their findings echo the warnings of scholars of precarious employment—​insecurity and a lack of power in the labour market (related to broader dimensions of social subordination) should not be confused with flexibility. But Green and Livanos also note that place matters: in their analysis, workers in the south-​west of England, and in Northern Ireland, Scotland, and Wales are most likely to be in INE. As these findings are not analysed in great detail, they provide a fruitful direction for future research by geographers, especially given that spatial difference in the literatures on precarious employment is generally approached by way of quantitative regional comparisons.

Precarious Work and Winner-Take-all Economies    489 Table 25.1 Full-​time, Part-​time, and Temporary Employment in the European Union, percentages, 2003, 2007, and 2014 2003

2007

2014

EMPa

PTb

TEMPc

EMP

PT

TEMP

EMP

PT

TEMP

EU 27

62.6

16.5

12.7

65.3

18.2

14.6

64.9

20.5

14.0

Denmark

75.1

21.3

9.3

77.0

23.7

9.1

72.8

25.5

8.5

Germany

65.0

21.7

12.2

69.0

26.1

14.6

73.8

27.6

13.0

France

64.0

16.8

13.4

64.3

17.3

15.1

64.3

18.9

15.8

Greece

58.5

4.3

11.2

60.9

5.7

11.0

49.4

9.5

11.7

Ireland

65.5

16.9

5.2

69.2

17.9

8.5

61.7

23.5

9.3

Italy

56.1

8.5

9.9

58.6

13.6

13.2

55.7

18.4

13.6

Netherlands

73.6

45.0

14.5

76.0

46.8

18.1

73.1

50.5

21.5

Poland

51.2

10.5

19.4

57.0

9.2

28.2

61.7

7.8

28.4

Slovenia

62.6

6.2

13.7

67.8

9.3

18.5

63.9

11.2

16.7

Spain

60.2

8.1

31.9

65.8

11.6

31.6

56.0

15.9

24.0

UK

71.5

25.6

6.1

71.5

25.1

5.8

71.9

26.8

6.4

Notes. aEMP indicates the employment rate (15–​64 years), annual averages. bPT indicates part-​ time workers in per cent of total employment. cTEMP indicates percentage of employees with temporary contracts. Source: Eurostat (2014), employment (main characteristics and rates).

If precarious work is defined against a normative model of standard employment that assumes full-​time, continuous, direct employment with access to both collective representation and occupational benefits, and paying an adequate wage, it begs the question of the utility of a dichotomous approach (in which precarious and secure employment are understood as distinct categories) relative to a more continuous and fluid understanding of precariousness in work. This is because of the decline of, or polarization in, levels of incomes and benefits within so-​called standard forms of employment in some places, and also because precariousness is not a static attribute of occupations, forms of employment, or labour markets: trends may move in different and contradictory directions, and may affect different dimensions of employment. Table 25.1 illustrates, for example, different levels of limited-​ duration contract employment (often considered a prototypical type of precarious work) relative to overall levels of employment and unemployment in selected European countries against the EU 27 average, in the years 2003, 2007, and 2013. Although there is a general trend towards increasing part-​time and temporary work, there is also considerable variation between countries and over time. While Ireland, Italy, and Greece saw small increases in temporary employment after the 2008 financial crisis, the proportion of temporary workers in Spain fell—​possibly because temporary workers have fulfilled the ‘shock absorbing and risk-​absorbing’ functions identified by Theodore and Peck (2014, p. 28) in their study of the temporary staffing industry in the USA, in the context of a Spanish economy still mired in recession.

490   Strauss At the same time other measures related to precariousness, like access to occupational pensions, prevalence of low pay, and union density, affect countries where the SER would otherwise seem to retain some of its dominance (in the UK, e.g., where part-​time employment—​mostly among women—​is high, but temporary employment is relatively low). In 2013 low pay (defined as the share of workers earning less than two-​thirds of median earnings) affected 20.5 per cent of British workers (OECD, 2016), whereas only thirty per cent were covered by an employer-​provided occupational pension (OECD, 2015b). Conversely, countries with growing levels of temporary work may nevertheless score relatively well in terms of relative adequacy of pay or pension coverage, whereas others with higher levels of permanent work may also have high levels of part-​time work or low levels of occupational benefits. One further, relatively consistent trend among richer nations is increasing income inequality—​ in only four Organisation for Economic Co-​ operation and Development (OECD) countries (Belgium, the Netherlands, France, and Greece) did the Gini measure of income inequality show no measurable increase between the mid-​1980s and 2013, and only in Turkey did it fall during that period (Keeley, 2015, p. 34). As levels of income and wealth polarization increase, the complex interrelationship of these processes with rising precarity in ‘winner-​take-​all’ economies—​where fractions of capital, especially financial capital, are seen as the unambiguous winners—​is becoming apparent.2 There has been considerable debate in geography about the relationship between neo-​ liberalization and labour market change, including in relation to the ‘austerity turn’ in politics (Featherstone et al., 2012; Torres et al., 2013a), whereas in the social sciences more broadly the research agenda on precarious employment has expanded and developed alongside political debates about flexicurity, immigration policy and migrant workers’ rights, and labour market re-​regulation favouring employers (Arthurs and Stone, 2013). Intermediated employment spans these agendas. What is evident, moreover, is that new and evolving forms of intermediation—​like the work situation of Ted, our fictional driver—​are developing concomitantly with the uneven and contested spread of ‘non-​standard’ work.

Self-​employment and the Technology Revolution: The New Choice Dilemma The link between precarious work and intermediation arises because, in legal and regulatory regimes that attach rights and benefits to the employment relationship, the existence and nature of such a relationship is difficult to establish where intermediation occurs. On the positive side, intermediation is understood as a functional role that emerges because of labour market inefficiencies and asymmetries: some agents (usually workers) lack perfect information to inform a job search, others (employers) lack perfect information about the size and composition of the pool of job candidates. Furthermore, employers can increase efficiency by outsourcing a function (recruitment) that they do not wish to perform in house. But the increase in intermediated employment is also associated with the transferal of risks from firms to workers, the shedding of ‘core’ workforces and the erosion of occupational benefits, and the dismantling of internal labour markets (Theodore and Peck, 2014, for a general discussion, see Strauss and Fudge, 2014, pp. 5–​7).

Precarious Work and Winner-Take-all Economies    491 Typically, research on intermediaries has focused on their roles in regional and sectoral space-​economies, and on temporary employment agencies and labour brokers. The assumption in much of this research is that the intermediary adds complexity to what is nevertheless a ‘triangular’ employment relationship (Vallee, 1999; Peck and Theodore, 2002): the intermediary connects the worker with a firm requiring labour power, and even if acting as the de facto employer, the intermediary is an agent that directly mediates between these labour market actors. Although challenged to an extent by the theorization of complexity in commodity chains, production networks, and their regulation (Yang and Coe, 2009; Jackson, 2002), this model also tends to inform the strategies of workers, unions, and labour lawyers seeking to assign responsibility for upholding labour rights to an identifiable employer. Such strategies often, although not always, focus on national legal regimes (the EU being an exception). Yet recent data suggest that increases in temp employment are slowing relative to a different kind of non-​standard arrangement. What has attracted less attention in debates about precarious work is the rise of self-​employment, including in national labour markets otherwise characterized by relatively low levels of temporary and contract work (in particular the UK, the USA, and Canada—​see, e.g., Hatfield, 2015, on rising self-​employment in the UK relative to the rest of Europe). In Canada, increases in self-​employment of those without paid help (own-​account self-​employed) followed recessions in the 1980s and 1990s, and occurred amid significant losses in paid employment of all types during the post-​2008 economic downturn (LaRochelle-​Côté, 2010). Moreover, LaRochelle-​Côté (2010) found that increases in self-​employment between 2008 and 2009 were within groups not normally associated with the highest levels of self-​employment, including women, those over the age of forty-​five, and those living alone or with a spouse not in work. The UK Institute for Public Policy Research noted an even more marked trend in the UK, where forty per cent of the rise in UK jobs since 2010 have been in self-​employment, and earnings among the self-​ employed are falling relative to employed workers. The UK had the third-​largest increase in self-​employed workers in Europe between 2010 and 2014 (after Luxembourg and Estonia), whereas Germany, Poland, Sweden, and Norway recorded reductions in self-​employment; the UK also recorded the lowest number of self-​employed workers with employees (Hatfield, 2015). In both Canada and the UK, there is an ongoing debate about the degree to which this increase in self-​employment is directly linked to the decline of the SER and the rise of precarious employment—​and the degree to which self-​employment is precarious employment for many. There is also a question to be asked about how much of this is ‘disguised’ or ‘false’ self-​ employment, in which firms (including temp agencies and labour brokers) require workers to set up as self-​employed even if they have only one, ongoing ‘client’. In 2013 the UK government was concerned enough to consider strengthening laws governing onshore employment intermediaries and false self-​employment, and there has been attention at the EU level to the connections between false self-​employment and precarious work in particular sectors like construction. The issue here is that employers are reacting to unions’ and workers’ attempts to embed employment protections within a triangular employment relationship in which an agency mediates between the worker and the firm requiring labour. Requiring workers to contract as self-​employed with an intermediary avoids those employment protections that do exist for agency workers, including the requirement to make social security, national insurance, or pension contributions on their behalf.

492   Strauss Although similar issues are also evident in the USA in industries like construction (Torres et al., 2013), attention there is being paid to new models of intermediated work beyond the employment agency. The knowledge economy broadly, and the heralded rise of the ‘sharing economy’ in particular, has produced business models and related applications that are grounded in concepts of intermediation: the firm owns rights over the ‘digital intermediary’ (usually an app or platform like Uber, Airbnb, TaskRabbit, or Amazon’s Mechanical Turk) that connects workers or owners with clients or customers, and takes some proportion of the fee paid for the service. In the case of the latter platforms, their services have also been called ‘micro-​outsourcing’. Much of the focus has been on services like Airbnb, Uber, and their competitors. These companies have been celebrated by commentators like Jeremy Rifkin (in part for their association with the shift to a ‘zero marginal cost society’) and Arun Sundararajan (for their role in ‘the emerging peer-​to-​peer, collaborative ‘sharing economy’ ”). The implications drawn from analyses of these shifts often highlight the benefits to consumers and ‘sharers’; much of the rhetoric explicitly avoids talking about workers. Yet the concerns of workers have been at the heart of some of the challenges to companies like Uber, which has found its model of intermediated self-​employment challenged in the US courts. As Malhotra and Van Alstyne (2014, p. 25) write, ‘Micro-​outsourcing that pays for only the task at hand can shed overhead but mortgage the future by covering only marginal costs and leaving nothing for new skills, health care, or retirement. If information goods are an indicator, marginal costs approach zero, so even covering them might not pay much’. Moreover, the parent firm generally owns the data associated with the myriad transactions its application facilitates. These data represent additional value, generated by workers and their clients/​customers, which is captured (and potentially monetized) by the owners of the digital intermediary. On the one hand, these changes could be seen to break down the normative and epistemological barriers between developed and developing economies vis-​à-​vis formal and informal labour markets—​with the decentring of the idea of ‘non-​standard’ employment, we are all portfolio workers with at least an element of informality to our ‘personal work relations’ (Freedland and Kountouris, 2011). Yet the firms themselves are increasingly at the centre, not the margins, of the economy, and are darlings of Wall Street and the City of London. Moreover, the shift towards new forms of digital intermediation co-​exist with diminished but still significant expectations centred on the welfare state, which developed in tandem with the SER. The arena of technologically mediated labour is thus a key one in which the tensions between flexibility, individual choice, and insecurity are playing out in relation to the challenges of regulating precarious work.

Conclusion: Regulating for Security in Precarious Labour Markets A key challenge for labour movements, therefore, as well as for broader political forces seeking to counter precarity, is to articulate what kinds of labour rights and what forms of regulation can be mobilized to this end. The barriers are formidable, and are spatial as well as

Precarious Work and Winner-Take-all Economies    493 political-​economic and ideological. If triangular employment and other forms of precarious work make traditional strategies for organization and collective action difficult, a workforce made up of self-​employed contractors, connected only by their relationship with a digital intermediary, presents an even tougher proposition. Indeed, and given the anti-​union animus of figures like Amazon’s Jeff Bezos, this is, in part, the point. Thus, despite the increased use of social media and other technologies by unions and social movements, the challenge to established models of organizing and collective bargaining is significant. This challenge is compounded by the apparent success of anti-​red tape rhetoric in relation to the de-​and re-​regulation of labour markets, including not only attacks on labour law relating to unions and collective bargaining, but also minimum employment standards that protect non-​unionized workers (on the UK, e.g., see Pollert, 2007; Jameson, 2012). Despite mobilizations against precarity in Europe and against assaults on workers’ rights in the USA, governments in Europe, North America, and, most recently, India have come to power on the back of promises to reduce the burdens on business and liberate entrepreneurial energies from constricting regulations.3 Where does this leave the struggles of workers and their representatives who are seeking more security and less precariousness? I want to conclude by reflecting on research on attitudes to workers’ rights, red tape, and employment regulation in the UK. An Internet survey carried out in 2013, on a representative sample of UK workers, included questions on attitudes to workers’ rights and regulation, as well as about their current working arrangements and conditions.4 Firstly, the survey found that seventy-​two per cent of respondents identified as permanent, direct employees, whereas nine per cent identified as self-​employed and seven per cent as directly employed on temporary contracts. Only one per cent stated that they were employed through a temp agency. Workers aged 18–​24 years had the lowest levels of permanent work (55%), and the highest levels of temp contract work (27%). Of all workers, half (52%) indicated that they were entirely happy with the organization of their employment, although the group of workers under twenty-​four had the highest proportion wanting permanent jobs (17%). More than thirty per cent of all workers stated that they would prefer to work the same number of hours but more flexibly (at home, or non-​standard hours), and this proportion was roughly the same for both men and women. Five per cent of respondents indicated that they did not have a permanent job but wanted one (the same proportion for both men and women). More than thirty per cent of those surveyed indicated that their workload had increased in the last few years (since 2011), that they and their colleagues were doing more work with fewer people, and that they were under pressure to do more work for the same salary.5 These responses seem to suggest that workers in the UK have experienced increased productivity pressures at work since the ‘recovery’ and that there is an unmet desire among a significant minority of workers for more flexible—​and for a much smaller group, more secure—​employment. How the employment relation should be regulated was addressed in four non-​sequential questions with different framing. Table  25.2, which summarizes responses, presents a mixed picture. Generally, most workers agreed that they have enough rights, and there was not support for more rules and regulations on businesses. Yet forty-​five per cent also indicated that they think the government should do more to protect employment rights, and this was fairly consistent across age groups and between men and women. Although this is only one piece of research, it is interesting to note that in what is the most ‘flexible’ labour market in Europe, there is both a desire for more flexible working arrangements (without a reduction or increase in hours) among men and women, and some

494   Strauss Table 25.2 Survey Results (Selected), UK Employees, 2013 Strongly agree

Agree

Disagree

Strongly disagree

11%

38%

19%

8%

Do more

Are doing right Do less amount

Don’t know

Do you think government should 45% do more or less to protect the employment rights of British workers (e.g. minimum wage legislation, maximum number of working hours in a day, etc.), or do you think they are already doing the right amount?

41%

6%

8%

When thinking about human 33% resources policies in businesses in Britain (e.g. hiring and firing), do you think the government should do more or less to make it easier for businesses to operate, or do you think they are already doing the right amount?

42%

11%

15%

Too many About right

Too few

Don’t know

34%

14%

14%

To what extent do you agree or disagree with the following statement? Overall, employed workers in Britain today have enough rights in terms of legal rights, employment rights and salaries.a

Do you think that there are too many or too few employment rules and regulations for businesses in Britain, or is it about right?

37%

Notes. aResults exclude ‘Neither agree nor disagree’ (19%) and ‘Don’t know’ (4%), which is why they don’t add up to 100%.

ambivalence about regulation. I want to set these findings against a backdrop of an also ambivalent, and geographically uneven, set of conditions faced by national and international labour movements. Trade union density, for example, has been stable or falling in most OECD countries in the last fifteen years; Italy and Spain, however, which have faced severe structural adjustment policies following the 2008 financial crisis, are interesting exceptions (OECD 2015c). Yet social movements, for example in the USA, relating to minimum and living wages and the right to organize among low-​wage workers (often in food services and hospitality), represent significant worker mobilizations against precarious employment. Anti-​austerity protests in Southern Europe, demonstrations in support of public higher

Precarious Work and Winner-Take-all Economies    495 education in the UK, Chile, and the Canadian province of Quebec, and large-​scale climate marches with a strong union presence all suggest a shifting but significant set of progressive alliances protesting multidimensional precariousness. There is, nevertheless, rather scant evidence of the emergence of a class consciousness among what Guy Standing (2011) calls the precariat, which unites diverse workers like John, Ana, Alicia, and Ted in a common struggle against employment-​related insecurity. Nevertheless, demands for a ‘floor’ of basic rights and entitlements for all workers (regardless of their employment status)—​including to a living wage—​contra the vagaries of insecure, low-​paid, work and increasing levels of income polarization, is echoed across diverse labour and social struggles. These demands could be reflected in the more positive responses of those surveyed in the UK to the question of whether the government should do more to protect workers’ rights. One could argue that precarious employment—​and economic inequality—​are defining, rather than anomalous, conditions of capitalism, and in this sense the post-​World War II heyday of the SER is the anomaly. This suggests that the transformation of our socio-​ economic system is the only way to address and stem the growth of precariousness in work, because there is no going back, even if we wanted to. Nevertheless, current socio-​economic, political, and legal trends, especially those that encompass and give primacy to developments beyond the Anglo-​American and European labour markets (not addressed in this chapter), suggest much more diverse and contingent responses. This evolving landscape presents a fertile area of future research for economic geographers.

Acknowledgements Funding for the research discussed in this chapter was made possible by a Regional Studies Association (RSA) Early Career Grant. The author thanks the RSA for their generous support.

Notes 1. Zero-​hours contracts are employment contracts that do not stipulate or guarantee a minimum number of working hours. This also means no minimum or guaranteed income; in the UK, zero-​hours contracts are thus associated with low pay and insecurity (Monaghan, 2014). 2. There is a vigorous debate about how to measure income and wealth polarization, and its significance (Palma, 2014; Piketty and Saez, 2014). The attention garnered by the work of Thomas Picketty and his collaborator, Emmanuel Saez, points to concern even among many orthodox economists about distributional imbalances. 3. There are parallels here with tax policy. Despite seeming widespread dissatisfaction with income and wealth inequality (Hayes, 2014), this does not necessarily translate into policy action (nor the election of parties that promise to raise taxes on the rich). 4. The survey was commissioned by the author and was carried out by YouGov on behalf of Progressive Partnership. Total sample size was 2031 adults. Fieldwork was undertaken between 3 and 7 May 2013. The figures have been weighted and are representative of all UK adults (aged 18+). 5. Respondents could tick all statements that applied in this question.

496   Strauss

References Anderson, B. (2010). ‘Migration, immigration controls and the fashioning of precarious workers’. Work Employment and Society 24: 300–​317. Arthurs, H.W. and Stone, K.V.W. (eds) (2013). Rethinking Workplace Regulation: Beyond the Standard Contract of Employment (New York: Russell Sage Foundation). Benner, C. (2002). Work in the New Economy: Flexible Labor Markets in Silicon Valley (Oxford and Malden, MA: Blackwell Publishers). Brandth, B. and Kvande, E. (2001). ‘Flexible work and flexible fathers’. Work Employment and Society 15: 251–​267. Butler, J. (2004). Precarious Life:  The Powers of Mourning and Violence (London and New York: Verso). Coe, N.M. (2013). ‘Geographies of production III: making space for labour’. Progress in Human Geography 37: 271–​284. Coe, N.M., Johns, J., and Ward, K. (2011). ‘Variegated global expansion: internationalization strategies in the temporary staffing industry’. Geoforum 42: 61–​70. Cranford, C.J. and Vosko, L.F. (2005). ‘Conceptualizing Precarious Employment:  Mapping Wage Work Across Social Location and Occupational Context’ in Vosko, L.F. (ed.) Precarious Employment:  Understanding Labour Market Insecurity in Canada, pp. 43–​66 (Montreal: McGill-​Queen’s University Press). Donald, B., Gertler, M.S., and Tyler, P. (2013). ‘Creatives after the crash’. Cambridge Journal of Regions Economy and Society 6: 3–​21. Dyer, S., McDowell, L., and Batnitzky, A. (2011). ‘Migrant work, precarious work-​life balance: what the experiences of migrant workers in the service sector in Greater London tell us about the adult worker model’. Gender Place and Culture 18: 685–​700. Ettlinger, N. (2007). ‘Precarity unbound’. Alternatives 32: 319–​340 ER. Featherstone, D., Ince, A., Mackinnon, D., Strauss, K., and Cumbers, A. (2012). ‘Progressive localism and the construction of political alternatives’. Transactions of the Institute of British Geographers 37: 177–​182. Eurostat (2014). ‘Employees with a contract of limited duration (annual average)’ http://​ ec.europa.eu/​eurostat/​tgm/​table.do?tab=table&init=1&language=en&pcode=tps00073&pl ugin=1 (last accessed 1 January 2015). Forde, C. and Slater, G. (2005). ‘Agency working in Britain: character, consequences and regulation’. British Journal of Industrial Relations 43: 249–​271. Freedland, M. and Kountouris, N. (2011). The Legal Construction of Personal Work Relations (Oxford: Oxford University Press). Fudge, J. (2005) ‘Beyond vulnerable workers: towards a new standard employment relationship’. Canadian Labour and Employment Law Journal 12: 151–​176. Goldring, L., Berinstein, C., and Bernhard, J.K. (2009) ‘Institutionalizing precarious migratory status in Canada’. Citizenship Studies 13: 239–​265. Green, A.E. and Livanos, I. (2014). ‘Involuntary non-​standard employment and the economic crisis: regional insights from the UK’. Regional Studies 49: 1223–​1235. Harker, C. (2012). ‘Precariousness, precarity, and family: notes from Palestine’. Environment and Planning A 44: 849–​865. Hayes, T.J. (2014). ‘Do citizens link attitudes with preferences? Economic inequality and government spending in the “new gilded age” ’. Social Science Quarterly 95: 468–​485. Hatfield, I. (2015). Self-​Employment in Europe (London: IPPR).

Precarious Work and Winner-Take-all Economies    497 Jackson, P. (2002). ‘Commercial cultures: transcending the cultural and the economic’. Progress in Human Geography 26: 3–​18. James, A. (2011). ‘Work-​life (im)“balance” and its consequences for everyday learning and innovation in the New Economy:  evidence from the Irish IT sector’. Gender Place and Culture 18: 655–​684. Jameson, H. (2012). ‘The Beecroft Report: pandering to popular perceptions of over-​regulation’. Political Quarterly 83: 838–​843. Kalleberg, A.L. (2011). Good Jobs, Bad Jobs: The Rise of Polarized and Precarious Employment Systems in the United States, 1970s to 2000s (New York: Russell Sage Foundation). Keeley, B. (2015). Income Inequality:  The Gap between Rich and Poor, OECD Insights (Paris: OECD Publishing). Kim, Y. (2013). ‘Diverging top and converging bottom: labour flexibilization and changes in career mobility in the USA’. Work Employment and Society 27: 860–​879. LaRochelle-​Côté, S. (2010). ‘Self-​employment in the downturn’. Perspectives on Labour and Income 22: 5. Lewis, H., Dwyer, P., Hodkinson, S., and Waite, L. (2014). Precarious Lives:  Forced Labour, Exploitation and Asylum (Bristol: Polity Press). McDowell, L. and Christopherson, S. (2009). ‘Transforming work: new forms of employment and their regulation’. Cambridge Journal of Regions, Economy and Society 2: 335–​342. Malhotra, A. and Van Alstyne, M. (2014). ‘The dark side of the sharing economy . . . and how to lighten it’. Communications of the ACM 57: 24–​27. Monaghan, A. (2014). ‘TUC warns of stark wage gap in two-​tier workforce’, Guardian http://​ www.theguardian.com/​uk-​news/​2014/​dec/​15/​tuc-​pay-​two-​tier-​workforce-​zero-​hours-​ contracts (last accessed 9 January 2015). OECD (2015a). ‘Part-​time employment rate (indicator)’ https://​data.oecd.org/​emp/​part-​time-​ employment-​rate.htm (last accessed 8 January 2015). OECD (2015b). ‘Coverage of private pensions’ in OECD (ed.) Pensions at a Glance 2015: OECD and G20 indicators (Paris: OECD Publishing). OECD (2015b). ‘Trade union density (indicator)’ http://​stats.oecd.org/​Index.aspx?DataSetCode=UN_​ DEN (last accessed 14 January 2015). OECD (2016). ‘Wage levels (indicator)’ http://​www.oecd-​ilibrary.org/​employment/​wage-​ levels/​indicator/​english_​0a1c27bc-​en (last accessed 9 March 2016). Palma, J.G. (2014). ‘Has the income share of the middle and upper-​middle been stable around the ‘50/​50 rule’, or has it converged towards that level? The ‘Palma ratio’ revisited’. Development and Change 45: 1416–​1448 Peck, J. (1996). Work Place: The Social Regulation of Labor Markets (London and New York: The Guilford Press). Peck, J. and Theodore, N. (2002). ‘Temped out? Industry rhetoric, labor regulation and economic restructuring in the temporary staffing business’. Economic and Industrial Democracy 23: 143–​175. Peck, J. and Theodore, N. (2007). ‘Flexible recession: the temporary staffing industry and mediated work in the United States’. Cambridge Journal of Economics 31: 171–​192. Pollert, A. (2007). ‘Britain and individual employment rights: “Paper tigers, fierce in appearance but missing in tooth and claw” ’. Economic and Industrial Democracy 28: 110–​139. Piketty, T. and Saez, E. (2014). ‘Inequality in the long run’. Science 344: 838–​843. Rodgers, G. (1989). ‘Precarious Work in Western Europe: The State of the Debate’ in G. Rodgers and J. Rodgers (eds) Precarious Jobs in Labour Market Regulation: The Growth of Atypical Employment in Western Europe, Ch. 1 (Geneva: ILO).

498   Strauss Smith, F., Wainwright, E., Buckingham, S., and Marandet, E. (2011) ‘Women, work-​life balance and quality of life: case studies from the United Kingdom and Republic of Ireland’. Gender Place and Culture 18: 603–​610. Standing, G. (2011). The Precariat: The New Dangerous Class (London: Bloomsbury Academic). Strauss, K. (2017). ‘Precarious Work’ in International Encyclopedia of Geography: People, the Earth, Environment, and Technology, pp. 1–​9 (Washington, DC: Wiley-​AAG). Strauss, K. and Fudge, J. (2014). ‘Temporary Work, Agencies, and Unfree Labor: Insecurity in the New World of Work’ in J. Fudge and K. Strauss (eds) Temporary Work, Agencies, and Unfree Labor: Insecurity in the New World of Work, pp. 1–​25 (New York: Routledge). Theodore, N. and Peck, J. (2014) ‘Selling Flexibility: Temporary Staffing in a Volatile Economy’ in J. Fudge and K. Strauss (eds) Temporary Work, Agencies and Unfree Labour: Insecurity in the New World of Work, pp. 26–​47 (New York and London: Routledge). Torres, R., Heyman, R., Munoz, S., Apgar, L., Timm, E., Tzintzun, C., et al. (2013) ‘Building Austin, building justice:  immigrant construction workers, precarious labor regimes and social citizenship’. Geoforum 45: 147–​157. Vallee, G. (1999). ‘The growth of non-​standard forms of employment and the protection of human rights:  What role for labour law?’ Relations Industrielles-​ Industrial Relations 54: 277–​312. Vosko, L.F. (2000). Temporary Work:  The Gendered Rise of the Precarious Employment Relationship (Toronto: University of Toronto Press). Vosko, L.F. (2010). Managing the Margins: Gender, Citizenship, and the International Regulation of Precarious Employment (Oxford: Oxford University Press). Waite, L. (2009). ‘A place and space for a critical geography of precarity?’ Geography Compass 3: 412–​433. Waite, L., Valentine, G., and Lewis, H. (2014). ‘Multiply vulnerable populations:  mobilising a politics of compassion from the “capacity to hurt” ’., Social and Cultural Geography 15: 313–​331. Warren, A. (2014). ‘Working culture: the agency and employment experiences of nonunionized workers in the surfboard industry’. Environment and Planning A 46: 2300–​2316. Watson, A. (2013). ‘ “Running a studio’s a silly business”: work and employment in the contemporary recording studio sector’. Area 45: 330–​336. Whitson, R. (2010). ‘The reality of today has required us to change:  negotiating gender through informal work in contemporary Argentina’. Annals of the Association of American Geographers 100: 159–​181. Williams, C.C. and Round, J. (2007). ‘Beyond negative depictions of informal employment: Some lessons from Moscow’. Urban Studies 44: 2321–​2338. Wills, J., Datta, K., Evans, Y., Herbert, J., May, J., and McIlwaine, C. (eds) (2009) Global Cities At Work: New Migrant Divisions of Labour (London: Pluto Press). Woon, C.Y. (2011). ‘Undoing violence, unbounding precarity: Beyond the frames of terror in the Philippines’. Geoforum 42: 285–​296. Yang, D.Y. and Coe, N.M. (2009). ‘The governance of global production networks and regional development:  a case study of Taiwanese PC production networks’. Growth and Change 40: 30–​53.

Chapter 26

Ta lent, Skil l s , a nd Urban Ec onomi e s Richard Florida and Charlotta Mellander Introduction A large and influential body of research on the role of talent and skills, or what economists refer to as human capital, has emerged in urban economics and economic geography over the last couple of decades, greatly expanding our understanding of the factors and forces that drive urban and regional development. This is a fairly dramatic shift, as the core unit in urban and regional research has long been the firm. Marshall (1890) long ago noted the tendency of firms to cluster or agglomerate around each other. Classical location theory (e.g. Weber, 1909; Christaller, 1933; von Thünen, 1966) emphasized transportation costs and the trade-​offs made by large industrial firms. This analysis focuses on the firm carried over into modern research on industrial clusters and districts. During the 1980s, economic geographers noted the rise of industrial districts (Porter, 1990) and flexibly specialized industrial networks (Piore and Sabel, 1984; Christopherson and Storper, 1986; Saxenian, 1994; Scott, 2000). The shift to human capital and talent coincides with the shift from industrial to knowledge-​based capitalism. And while it is true that talent has become a more important and obvious factor of production in knowledge-​based capitalism, interest in talent, skills, and human capital as economic factors dates a long way back. In his classic work The Wealth of Nations (1776), Adam Smith identified the ‘acquired and useful abilities of all the inhabitants or members of the society’ as essentially the ‘fourth factor of production’ (e.g. Samuelson and Nordhaus, 2005)  operating alongside land, labour, and production. ‘The greatest improvement in the productive powers of labour, and the greater part of the skill, dexterity, and judgment with which it is anywhere directed, or applied’, Smith wrote, ‘seem to have been the effects of the division of labour’(Smith, 1776, book 1, p. 7). Interest in human capital was rekindled by Gary Becker’s path-​breaking research (1962, 1964, 1993) and a large body of empirical research grew up examining the association between human capital and the economic growth of nations (Barro, 1991, 1997; Mankiw et al., 1992). Jane Jacobs (1961, 1969) later argued that what distinguished cities and propelled their economic growth and development was not firms, but the clustering of talented and creative

500   Florida and Mellander people. By the mid-​1980s, Wilbur Thompson (1986) noted that occupations and occupational analysis offered perhaps a more powerful way to understand cities and regional analysis than just looking at firms and industries. Building explicitly on Jacobs, Lucas (1988) suggested that knowledge is embodied in human beings rather than firms, and that the human capital externalities that stem from concentrations of highly educated individuals provide the key motivating force in economic growth and development. Over the past decade or so, a growing body of research has noted a growing divergence in the geographical location and concentration of talent and human capital and their increasing importance to regional innovation, productivity, and growth (Glaeser and Maré, 2001; Florida, 2002; Glaeser and Saiz, 2003). This has occurred alongside and has been spurred on by a broader shift from industrial to knowledge-​based capitalism. As a result, the locus of competitive advantage has shifted from firms to talent and cities and urban agglomerations have come to replace large firms and the nation state as the central social and economic organizing units of our time. This chapter reviews the rise of talent, human capital, and skills as key analytic categories and of the city as the central organizing unit for the knowledge-​and talent-​driven economy. It begins by reviewing the historic firm and industry focus in urban economics and economic geography. The following section outlines the rise of talent, human capital, and skills alongside the rise of the knowledge economy. After that we turn to the ongoing and productive debate over how best to operationalize talent or human capital. We first review the literature that measures human capital in terms of educational attainment. After that we turn to the growing body of studies that examine talent in terms of occupations, including research on the rise of the creative class and creative occupations. We then discuss more recent studies that examine human capital in terms of underlying occupational skills, such as physical, cognitive, and social skills. The second part of this chapter discusses the connection between talent and cities, outlining the increasingly important role played by cities as the key social and economic organizing unit of knowledge-​based capitalism. This section examines the role of cities in both the location and clustering of talent, as well as the factors that are argued to attract talent to cities, including the labour market, amenities, universities, and knowledge-​based institutions, diversity, tolerance, and quality of place. The penultimate section examines the growing literature on the downside and externalities produced by talent clustering, including increased inequality and socio-​economic segregation. The conclusion briefly summarizes the broad thrusts of this research on talent, skills, and urban economies.

The Firm and Industry Focus in Urban Economics and Economic Geography Historically, urban economists and economic geographers have focused on the firm as the principal unit of analysis (e.g. Weber, 1909; Christaller, 1933; Ohlin, 1933; Hoover, 1937, 1948; Lösch, 1954; von Thünen, 1966). This stems from foundational work by Marshall (1890), who argued that firms cluster to achieve the advantages of collocation, such as shared labour markets, shared inputs, risk minimization, and knowledge spillovers. Hotelling (1929) showed how firms producing similar types of products that do not compete based on price have incentives to collocate next to one another.

Talent, Skills, and Urban Economies    501 This focus on the firm developed as the industrial revolution was reaching full maturity and large vertically organized industrial firms were coming to the forefront. Transportation costs were high and location decisions were heavily influenced by proximity to natural resources. As production became less place dependent, Vernon (1963) advanced a simple model of industrial location based on the product cycle—​firms would decentralize production through branch plants once production processes became standardized. Others focused on the growing international spatial division of labour informed by the location decisions and global reach of multinational firms (Froebel et al., 1979; Massey, 1984). Labour, when it was considered at all, was mostly seen as a cost to minimize. But by the 1980s, as researchers began paying closer attention to the ways that clusters, industrial districts, and flexibly specialized networks were allowing smaller firms to achieve economies of scale (Piore and Sabel, 1984; Christopherson and Storper, 1986; Scott, 2000), labour began to be seen in a different light. One stream of research focused on a rising trend in Japanese factories, in which the knowledge and intelligence of factory workers was tapped to form new systems of industrial production (Florida and Kenney, 1991, 1993; Florida, 1995). These new, post-​Fordist production processes were much less standardized than they had been in the past. Workers were forced to adopt a wider range of expertise, which led to a continuous process of learning (Piore and Sabel, 1984; Storper and Christopherson, 1987). Social factors, including trust and social capital, became important elements of industrial clusters (Saxenian, 1994).

The Rise of Talent, Skills, and Human Capital The role of skills and talent had not been overlooked entirely; rather they were understood strictly in relation to the interests of firms. Writing in the Grundrisse, Karl Marx (Marx and Nicolaus, 1993), pointed to the rise of knowledge, human capital, and science as increasingly important forces of production. He also noted that workers’ individual investments in knowledge and skills overwhelmingly benefit capital in the form of firms, which sees gains in overall productivity and thus profit (Marx and Nicolaus, 1993). Pigou (1920) distinguished investments in human capital that combine consumption and investment from investments in physical or material capital. Becker (1964, 1993) developed the micro-​economic foundations for understanding human capital and human capital investment as key drivers of economic advancement. But skills, talent, and human capital took on a new dimension as the shift from the industrial to the post-​industrial economy began. Machlup (1962) and Drucker (1969; 1993) described the rise of ‘knowledge workers’ alongside a ‘knowledge economy’. Bell (1973) identified the shift to a ‘post-​industrial society’, with a new class structure based on scientists, managers, administrators, and engineers. Romer’s (1990) theory of endogenous growth formalized the role of human knowledge and talent in overall economic growth. Introducing the concept of creativity to regional analysis, Andersson (1985) found that major creative city regions have historically had six fundamental conditions in common:  flexible financial capital; deep knowledge and competence; an unbalance between experienced needs and actual resources; environmental diversity; well-​functioning transportation and communication; and structural instability or a genuine insecurity. Landry

502   Florida and Mellander 70

60

Millions

50

40

30

20

10

Service Class

09

00

20

90

20

80

19

70

19

60

19

50

19

40

19

30

19

20

19

10

19

00

Agriculture

19

90

19

80

18

70

18

60

Creative Class

18

50

18

40

18

30

18

20

18

10

18

18

18

00

0

Working Class

Figure 26.1   The Growth of the Creative Economy Share (Florida, 2012, p. 45). (2008) advanced the construct of the ‘creative city’ and Howkins (2001) introduced the notion of the ‘creative economy’, spanning the technology and culture industries. Florida (2002) tracked the rise of the creative class of knowledge workers alongside the decline of the traditional working class. Up until 1960, creative-​class workers (people who work with their minds) accounted for just 12 to 16 per cent of the US workforce. By 1970 their share was 19 per cent; it rose to 24 per cent by 1980; and it is about 33 per cent today. The industrial working class accounted for some 40 per cent of the workforce in the 1970s. Since then, its share has been halved (see Figure 26.1). With the rise of this post-​industrial knowledge economy, talent and skills became a more important factor of production, if not the predominant factor. A detailed study by Glaeser (1999) found the clustering of talent to be a more important driver of urban and regional performance than that of firms. While the increasingly important role of talent and skills has been recognized by a large and growing group of scholars in economics, urban economics, and economic geography, there is still considerable debate about how best to measure, operationalize, and define human capital.

Talent as Education The great bulk of the empirical work on human capital and economic growth explicitly defines human capital in terms of educational attainment.

Talent, Skills, and Urban Economies    503 A large literature documents the role of educational attainment in economic growth at the national level (Barro, 1991, 1997; Mankiw et al., 1992; de la Fuente and Domenech, 2006; Cohen and Soto, 2007). While these studies are important, they do not account for regional spillovers or interdependencies. The political boundaries between nations and other large jurisdictions can be arbitrarily drawn; metropolitan regions tend to make up more economically delimited units (Duranton, 2007). Other researchers have shown that the metropolitan is a more appropriate context within which to evaluate the effects of human capital than nations (Mathur, 1999; Glaeser, 1998, 1999; Glaeser and Saiz, 2003; Glaeser et al., 2004). Utilizing spatial econometrics, Fischer (2011) examined the connections between knowledge diffusion and growth across regions in a study involving twenty-​two European countries. The relationship between educational attainment and regional development has been acknowledged by economists since at least 1958, when Ullman first wrote about the significance of education to regional economic growth. A considerable body of research has built off Ullman’s initial work to uncover significant relationships between educational levels and wages within and across urban regions (Ullman, 1958). Simply put, this research has found that cities with higher levels of educational attainment are more productive. An increase by one year in education has been found to increase productivity in a region by 3 per cent (Rauch, 1993). Simon (1998) found a large positive relationship between educational attainment and employment growth in metropolitan regions between the years 1940 to 1986. He not only found spillover effects between cities within the same metropolitan region, but also noted that cities with higher levels of human capital grew faster than other cities within the same metropolitan region, which suggests that human capital effects are partly localized to the city itself. Simon and Nardinelli (1996) conducted a similar analysis but for a longer time period and reached similar conclusions. Places with larger shares of educationally measured human capital, they found, have also been more resistant to external shocks over time. More recent research suggests a growing divergence in talent and human capital over time (Berry and Glaeser, 2005). This is an endogenous process, in which places with initially high educational attainment values have increased their human capital levels more over time than places that started out with lower values. Stolarick et al. (2010) explored intra-​regional relations between education and economic performance to find out if it matters where within metropolitan regions educated people cluster. They found strong connections between high levels of human capital and incomes in the suburbs of small-​or medium-​sized metropolitan areas. However in metropolitan regions with populations of one million or more, they found that city-​centre human capital is relatively stronger in relation to economic performance.

Talent as Occupation Educational attainment, while a useful operational measure, captures only a part of what economists mean by skill or human capital. Mincer’s classic study (1974) of the returns on human capital used census data from the 1950s and 1960s to show that income increased by 5 to 10 per cent for every additional year of schooling. But he also found that skills and age had a significant impact on earnings. Thompson long ago suggested the need to utilize

504   Florida and Mellander occupational analysis in regional development research (Thompson, 1965; Thompson and Thompson, 1987). Florida (2002) used occupations to divide the workforce into three main occupational classes—​the creative class, the working class, and the service class. His approach focuses on creativity as opposed to education as a proxy for skill. Research in psychology has shown that creativity is a fundamental and intrinsic human capability. Where Marx and Engels (1848) and other classical economists looked at physical labour—​in other words, the ability of humans to transform nature, create farms, and build manufactured products as the definitive human trait—​in reality it is our underlying creativity that differentiates us from other species and it is what entrepreneurs, chief executive officers, artists, and technologists have in common. As Sternberg and Lubart note (1999, p. 3), ‘If one wanted to select the best novelist, artist, entrepreneur, or even chief executive officer, one would most likely want someone who is creative’ (Figure 26.2). Using data on occupations to define the creative class, which includes workers in science and technology, arts, design, media, and entertainment; business and management; law and health care. Florida identified three mutually dependent and self-​reinforcing types of creativity as key to economic activity: (1) technological creativity or innovation; (2) economic creativity or entrepreneurship; and (3) artistic or cultural creativity. While there is some overlap between educational attainment and creative occupations, they are not the same. By focusing on the actual skill content of specific occupations—​what people actually do—​it is possible to analyse talent clusters in much finer detail than if we

Creative Class Share 25

30 35

Figure 26.2   Creative Class Shares in US Metropolitan Regions.

40

Talent, Skills, and Urban Economies    505 simply use education as a broad proxy for talent. Bill Gates, for example, who dropped out of Harvard, would not be counted under the human-​capital approach. In the USA for example, nearly three-​quarters of adults with college degrees are members of the creative class. But less than 60 per cent of the people whose occupations qualify them as members of the creative class have college degrees (Stolarick and Currid-​Halkett, 2013). In other words, four in ten members of the creative class would not be counted as high human-​capital individuals under the educational attainment measure. In Sweden, 37 per cent of the population holds a creative-​class job, but only one-​quarter of those have a university degree. However, about 90 per cent of the highly educated hold a creative-​class job (Mellander, 2009). A significant body of research shows that the occupation-​based creative-​class measure operates in addition to and through other channels than the standard education-​based human capital variable. Independent research by Gabe (2009) shows that the creative class continues to have a substantial effect on regional economic growth, even when controlling for the effects of education and other factors. Having a creative-​class job brings economic benefits that extend beyond those of going to college. A college graduate working in the same occupation as a non-​college graduate earns approximately 50 per cent higher wages. But having a creative-​class job adds another 16 per cent, about the same as another 1.5 years of additional education, according to Gabe’s (2009) research. McGranahan and Wojan (2007) used sophisticated statistical techniques to gauge the effects of the creative class versus human capital on regional growth. To do so, they used systems of simultaneous equations rather than the conventional simple regression models to control for the endogeneity of population and employment growth, as well as influences from a range of other local conditions and attributes. Their key findings overwhelmingly confirm the ‘strong independent influence on employment growth from both the initial share employed in the recast creative class occupations and its growth over the decade. By contrast, the statistical association with human capital variables is quite weak’. And they add: ‘the econometric test of the creative class thesis provides strong support for the notion that creativity has an effect on growth independent of the endowment of human capital’ (McGranahan and Wojan, 2007, p. 213). A detailed study of regional growth in the Netherlands found that the creative-​class measure considerably outperformed the standard education-​based human capital measure in accounting for employment growth. This led its authors to conclude that the creative-​class measure sets a ‘new standard’ for measuring skill and talent, especially when considering regional labour productivity (Marlet and van Woerkens, 2004). ‘With our Dutch data set we do find evidence that Florida’s creative class is a better predictor of city growth than traditional education standards’, they wrote. ‘Therefore we conclude that Florida’s major contribution is his successful attempt to create a population category that is a better indicator for levels of human capital than average education levels or amounts of highly educated people. The point is, as Florida stated, not which or how much education people can boast of, but what they really do in working life’ (Marlet and van Woerkens, 2004, p. 2620). A 2012 study used advanced statistical models to compare the effects of the creative class and human capital across the 257 European Union regions. ‘Our results’, it concluded, ‘indicate that highly educated people working in creative occupations are the most relevant component in explaining production efficiency’ (Marrocu and Paci, 2012, p. 369).

506   Florida and Mellander Florida et al. (2008) used structural-​equation models and path analysis to examine the effects of educational versus occupational measures on regional income and wages, and also to isolate the effects of tolerance, consumer service amenities, and universities on their distribution. Educationally derived human capital and the creative class both effect regional development, they found, but through different channels. The creative class outperforms conventional educational attainment measures in accounting for regional labour productivity measured as wages, while conventional human capital better accounts for regional income or wealth. Educational levels may reflect richer places, but it seems that the creative class actually makes a place more productive.

Talent as Underlying Skill An important wave of recent studies has zeroed in on the specific skills that occupations require. This research uses new data from the U.S. Bureau of Labor Statistics O*NET database, which collects detailed information on the actual skill content of work for more than 800 individual occupations. Bacolod and Blum (2005) examined the association between four key skills—​physical, motor, cognitive, and social skills—​and found that cognitive and social skills have the highest returns. Another study by Glaeser and Ressenger (2009) observed a geographical connection between skill types, finding social and cognitive skills to be associated with larger cities and metros. ‘Urban density’, they noted, ‘is important because proximity spreads knowledge, which either makes workers more skilled or entrepreneurs more productive. Bigger cities certainly attract more skilled workers, and there is some evidence suggesting that human capital accumulates more quickly in urban areas’ (Glaeser and Ressenger, 2009, p. 1). Motor skills and physical strength were less rewarded in cities. In a follow-​up study, Bacolod et al. (2009, p. 1) write that, ‘Urbanisation thus enhances thinking and social interaction rather than physical abilities’. Feser (2003) identified the general knowledge requirements across occupations and the economic returns they generate. Gabe (2009) differentiated between skill requirements and the returns they offer in private and public sectors, showing how spillover effects enhance earnings in metropolitan regions with higher shares of high-​knowledge occupations. Scott (2009) examined the connection between skills and regional employment, finding the largest increases in regional employment to be associated with cognitive-​ intensive occupations, with substantial employment declines for occupations that depend on physical skills. Using data from 1999 and 2008, Florida et al. (2012) distilled three broad categories of skills—​analytical, social intelligence, and physical—​from 87 occupations (Figure  26.3). Utilizing cluster analysis, they found that analytical and social intelligence skills have a significant positive effect on regional wages, while physical skills have a negative effect. Analytical skills, they found, are also somewhat more closely associated with regional wages than social intelligence skills, after controlling for education, industry, immigration, and regional size. Furthermore, wage returns to analytical and social intelligence skills have increased over time, and the return to physical skills has declined significantly. They also

Talent, Skills, and Urban Economies    507 (a)

Social 35

40

45

(b)

Analytical 30

35

40

Figure 26.3   (a) Social, (b) Analytical; and (c) Physical Skills in US Metropolitan Regions.

508   Florida and Mellander (c)

Physical 35

40

45

50

Figure 26.3  (continued). showed that larger cities reward analytical and social intelligence skills to a higher degree, whereas smaller cities rely more on physical skills.

The Organizing Role of Cities and Urban Economies Because of their central role in organizing, harnessing, attracting, and retaining talent, cities have emerged as the central social and economic organizing unit of global, knowledge-​based capitalism. If industrial production was organized in and around firms, knowledge-​based or creative production is organized in and around cities. Globalization has been widely associated with the ‘the end of geography’. In Friedman’s (2005) famous catchphrase, ‘the world is flat’. But Friedman missed the key role played by the clustering of economic activity in cities, or what Porter (2006) called the ‘location paradox’. ‘Location still matters’, he said. ‘The more things are mobile, the more decisive location becomes. This point has tripped up a lot of really smart people’ (Porter, 2006,). Florida (2005) argues that instead of being flat, the world is spiky, organized around cities, metropolitan areas, and mega-​regions. The importance of talent to cities and vice versa was recognized early on by scholars who paid attention to the geographical concentration of human capital (Ullman, 1958). In a 1969

Talent, Skills, and Urban Economies    509 critique of Adam Smith’s paradigmatic pin factory, Jacobs argued that if the division of labour improves productivity and profitability for firms, cities enable the constant combining and recombining of the key inputs—​including talented people—​that are the real drivers of economic growth (Jacobs, 1969). If firms improve efficiency, cities give rise to new products, new enterprises, and whole new industries. Inspired by Jacobs, Lucas (1988) showed how collocations of skilled, talented, ambitious, and entrepreneurial people lead to the so-​called ‘human capital externalities’ that are the basic underlying mechanism of economic development. Urban density reduces the cost of knowledge transfer and speeds the transmission of ideas. Workers are at their most productive when they are located around other workers at similar levels of education or skills, who can challenge them and push their own work further (Eaton and Eckstein, 1997; Black et al., 2000). The clustering of talent is even more important in creative industries. Caves (2000) showed how creative industries are more likely to be organized as geographical clusters of creative individuals as opposed to vertically integrated firms. Creative industries have higher levels of uncertainty and production challenges due to multiplicative production functions, in which every input is non-​substitutable and therefore must be present in order to produce. Further, shorter life cycles and a constant need for reinvention demands a closer interaction between consumers and producers, as well as new skill combinations for the faster generation of ideas. Talented, highly skilled individuals exercise wide choices about where they can live (Mincer, 1974; Graves, 1979; Graves and Linneman, 1979; Becker, 1993; Pandit, 1997; Edlund, 2005). The role that amenities play in their location choices is the subject of a growing body of empirical research (see Rosen, 1979; Roback, 1982; Gottlieb, 1995; Glaeser, 2001; Florida, 2002). Glaeser and Maré (2001) also examined the factors that attract skilled labour to cities and found that higher-​amenity cities attract more skilled labour and consequently grow faster. Clark et  al. (2002) argued that successful talent-​attracting cities maximize individuals’ overall utilities, especially their quality-​of-​life preferences, not just their incomes. A German study of the economic effects of Baroque opera houses (Falck et al., 2011) found that ‘proximity to a baroque opera house significantly affects the spatial equilibrium share of high-​human-​capital employees’, even though those opera houses were built hundreds of years before the high skill jobs were ever thought of. Albouy (2009) showed that both natural and built amenities are strongly associated with economic productivity, in both US and Canadian cities. A detailed study by Shapiro (2006, p. 1) found that ‘roughly 60% of the employment growth effect of college graduates is due to enhanced productivity growth, the rest being caused by growth in quality of life. This finding contrasts with the common argument that human capital generates employment growth in urban areas solely through changes in productivity’. Tolerance and openness is another factor that has been shown to affect the distribution and location choices of talent. Simonton, a leading student of creativity, identified four key characteristics of places that generate substantial creative activity:  domain activity, receptiveness to different intellectual modes, ethnic or cultural diversity, and an openness to political views (Simonton, 2000). Florida captured these attributes in his theory of the 3Ts of economic development—​technology, talent, and tolerance, in which each alone is a

510   Florida and Mellander necessary but insufficient condition for talent attraction and creative economic development. Technology and talent are better understood as flows rather than stocks. The most competitive regions draw in the broadest range of talent by age, ethnicity, marital status, and so on. Tolerance encourages diversity, and diversity, particularly in the form of immigration, has been shown to increase regional productivity by introducing different but complementary skills to an area (Ottaviano and Peri, 2005). Highly innovative places like Silicon Valley have higher rates of immigration. Studies by Wadhwa et al. (2007, 2008) found that immigrants are among the founding teams of between a third and a half of high-​tech Silicon Valley start-​ups. Tolerant attitudes to the gay and lesbian community have also been found to be associated both with higher levels of innovation and high-​tech industry at the metropolitan level (Florida and Gates, 2002) and with better economic outcomes across nations (Noland, 2005). This is not because gays and lesbians are more likely to be directly involved in innovative activity, but because their presence reflects an open-​minded and tolerant social and economic ecosystem. Page (2008), for example, shows that cognitive diversity results in greater idea generation, creativity, and innovation, and notes that cognitive diversity is, in turn, associated with demographic diversity. Tolerance is not just an advantage for attracting talent; it increases the probability that new ideas can be turned into economic value. Becker (1957) shows how discrimination can lead to decrease in productivity when employers choose people who may not be the most appropriate for the job due to racial or sex discrimination. Research by Inglehart and his collaborators found associations between openness and economic growth in studies covering more than sixty countries over four decades (Inglehart, 1989, 1997; Inglehart and Baker, 2000; Inglehart, et al., 2000; Inglehart and Norris, 2003). Inglehart argues that the best indicator of national tolerance is openness to gay and lesbian people. Florida (2002) used gay friendliness as a proxy for regional openness in the USA (Florida et al., 2008) and found a strong correlation with the distribution of both the highly educated and the creative class across metropolitan areas. Talent and the people who hold it are more likely to flow to places that have lower barriers of entry (Florida and Gates, 2002). Florida (2002; Florida et al., 2012) advanced the construct of quality of place to identify the varied ways that places attract and harness talent. Ultimately, quality of place can be understood in terms of Maslow’s (1943) famous hierarchy of needs, which is often represented as a pyramid with physiological needs at its base, ascending through security, love, self-​esteem, and, ultimately, self-​actualization. Quality of place encompasses five major needs:  physical and economic security (public safety, jobs); basic services (schools, health care, housing, transportation); leadership (political and business); openness (tolerance for diversity); and aesthetics (physical beauty, culture, amenities). A large-​scale Gallup–​Knight survey found that aesthetic values, along with a welcoming attitude and the ability to meet other people, were key factors in attracting and retaining talent (Carlino and Saiz, 2008; Florida, 2008; Florida et al., 2011; Mellander et al., 2011). Mellander et al. (2011) illustrated how place-​specific characteristics, such as the ability to meet and make friends, high-​quality public schools, and ease of getting from one place to another without too much traffic, are significantly more related to community satisfaction and the likelihood of staying in a place, than such economic variables such as the likelihood of getting a job.

Talent, Skills, and Urban Economies    511

Talent Clustering and Spatial Inequality It has been widely argued that the knowledge economy rose at the expense of the industrial economy; that as manufacturing jobs have been eliminated, the labour market has cleaved into high-​paying knowledge and creative jobs and lower-​paying, more contingent service sector jobs. Florida et al. (2012) estimate that as much as two-​thirds of the potential workforce of the USA is either under-​or unemployed. Thomas Piketty (2014), among others, has argued that we are entering a new age of ‘patrimonial capitalism’, in which a handful of fortunate families control greater and greater shares of wealth. While a large body of research pins the cause of rising inequality on so-​called skill-​biased technical change (Autor et al., 1998, 2003, 2006), a main factor, if not the main factor itself, is clustering. Clustering not only powers innovation and economic growth, but it also leads to greater geographical sorting of people and places by skill, talent, and ability. The geographical clustering of talent gives rise to both increasing inequality and an even more vexing problem of geographical segregation. As the incomes and purchasing power of the middle class declines, so does the diversity that fuels creativity and innovation. The very clustering that makes growth possible may also bring it to an end (Florida, 2017). A growing number of studies document the increasing divergence of talent and human capital across cities and metropolitan areas (Berry and Glaeser, 2005; Florida, 2006). This affects not just economic growth, but also related economic factors like housing values (Gyourko et al., 2006; Shapiro, 2006). Inequality has been found to be closely associated with city or metropolitan size. Size accounted for roughly 25 to 35 per cent of the total increase in economic inequality over and above the effects of skills, human capital, industry composition, and other factors, an important 2011 study found. This effect is more pronounced among lower-​wage earners. City size explains 50 per cent more of the increase in inequality for the lower half of the wage distribution than for the upper half (Baum-​Snow and Pavan, 2012). Florida and Mellander (2014) found that wage inequality explains just 15 per cent of the variation of income inequality across regions. Furthermore, while wage inequality is associated with skill-​biased technical change, income inequality is more associated with the enduring legacy of race and poverty at the bottom of the socio-​economic order, as well as the unraveling of the postwar social compact between capital and labour. We have already touched on some of the underlying reasons for growing spatial inequalities and geographical segregation. Places with existing concentrations of talent tend to attract more talent, compounding their advantages. The cognitive and especially social skills that drive economic growth tend to be concentrated in larger cities and metros (Bacolod et al., 2009). The result is growing class divisions within, as well as between, cities and metropolitan areas (Florida et al., 2014). The location choices and clustering of the higher skilled and more affluent also affect and structure the location choices and geography of less advantaged groups. The more affluent and skilled members of the creative class are drawn to central neighbourhoods, around transit and knowledge institutions, and near natural amenities. Rising housing costs exacerbate the ensuing class segregation and division as lower-​paid service workers are forced to live further out. Diamond (2012) shows how this sorting process involves migrations away from,

512   Florida and Mellander as well as to, knowledge-​based metros. ‘[T]‌he combination of desirable wage and amenity growth for all workers causes large amounts of in-​migration, as college workers are particularly attracted by desirable amenities, while low skill workers are particularly attracted by desirable wages’, she notes. But this leads directly to higher housing costs, which ‘disproportionately discourage low skill workers from living in these high wage, high amenity cities’ (Diamond, 2012, p. 5). This creates an additional level of inequality—​inequality of well-​being—​in which more skilled workers not only take home more money, but also benefit from better neighbourhoods, superior amenities, and better schools. This well-​being inequality, Diamond explains, is an additional 20 per cent higher than can be explained by the simple wage gap between college and high-​school graduates. The effects of neighbourhood poverty and disadvantage are long lasting and persistent (Sampson, 1995; Sharkey, 2013). The result is a divided city and metropolis increasingly defined by areas of concentrated advantage juxtaposed against areas of concentrated disadvantage. For these reasons, strategies to cope with geographical inequality and spatial segregation have implications that go beyond social welfare; they are critical components of local competitiveness strategies. Some policies that are called for are coordinated efforts to upgrade service jobs by developing strategies to tap the cognitive and social skills of workers, increasing density to provide more housing, and expanding access to transit in outlying neighbourhoods.

Conclusion The long sweep of urban economics, regional science, and economic geography has focused on the firm and industry as the key units of analysis. But the last couple of decades have seen an increasing interest in talent, skills, and human capital. The rise of talent, human capital, and skills in urban and regional research is rooted in the larger shift from an old industrial economy to a newer one based upon knowledge, innovation, and skill. Broadly speaking, talent has become the key input in economic growth and development in global, knowledge-​ based capitalism. Cities have emerged as the central economic organizing unit of talent and economic development. Size and density matter, but so, too, does the ability to deliver quality of place broadly, made up of services and amenities, openness to diversity, and lower barriers to entry for talent. But if clustering powers economic development and competitiveness it also generates increasing divides. Inequality has grown; economic segmentation and segregation have increased within cities. This threatens to undermine the neighbourhood-​level diversity, which functions as a basic mechanism for urban and regional economic growth in the first place.

References Albouy, D. (2009). ‘What are cities worth? Land rents, local productivity, and the capitalization of amenity values’. Working Paper No. 14981 (Cambridge, MA:  National Bureau of Economic Research).

Talent, Skills, and Urban Economies    513 Andersson, Å.E. (1985). Creativity—​The Future of Metropolitan Regions (Stockholm: Prisma). Autor, D.H., Katz, L.F., and Kearney, M.S. (2006). ‘The Polarization of the U.S. Labor Market.’ Working Paper No. 11986 (Cambridge, MA: National Bureau of Economic Research). Autor, D.H., Katz, L.F., and Krueger, A.B. (1998). ‘Computing inequality:  have computers changed the labor market?’ Quarterly Journal of Economics 113: 1169–​1213. Autor, D.H., Levy, F., and Murnane, R.J. (2003). ‘The skill content of recent technological change: an empirical exploration’. Quarterly Journal of Economics 118: 1279–​1333. Bacolod, M. and Blum, B.S. (2005). ‘Two sides of the same coin: U.S. “residual” inequality and the gender gap’. Working Paper (Toronto: University of Toronto). Bacolod, M., Blum, B.S., and Strange, W.C. (2009). ‘Skills in the city’. Journal of Urban Economics 65: 136–​153. Barro, R.J. (1991). ‘Economic growth in a cross section of countries’. The Quarterly Journal of Economics 106: 407–​443. Barro, R.J. (1997). Determinants of Economic Growth:  A  Cross-​Country Empirical Study (Cambridge, MA: MIT Press). Baum-​Snow, N. and Pavan, R. (2012). ‘Understanding the city size wage gap’. Review of Economic Studies 79: 88–​127. Becker, G. (1957). The Economics of Discrimination (Chicago, IL: University of Chicago Press). Becker, G. (1962). ‘Investment in human capital: a theoretical analysis’. The Journal of Political Economy 70: 9–​49. Becker, G. (1964). Human Capital (New  York:  Columbia University Press for the National Bureau of Economic Research). Becker, G. (1993). Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education (Chicago, IL: The University of Chicago Press). Bell, D. (1973). The Coming of Post-​ Industrial Society:  A  Venture in Social Forecasting (New York: Basic Books). Berry, C.R. and Glaeser, E.L. (2005). ‘The divergence of human capital levels across cities’. Papers in Regional Science 84: 407–​444. Black, D., Gates, G., Sanders, S., and Taylor, L. (2000). ‘Demographics of the gay and lesbian population in the United States:  evidence from available systematic data sources’. Demography 37: 139–​154. Carlino, G. and Saiz, A. (2008). ‘City beautiful’. Working Paper No. 08-​22 (Philadelphia, PA: Research Department, Federal Reserve Bank of Philadelphia). Caves, R.E. (2000). Creative Industries:  Contracts Between Art and Commerce (Cambridge, MA: Harvard University Press). Christaller, W. (1933). Die Zentralen Orte in Süddeutschland (Jena:  Gustav Fisher Verlag) (English translation Baskin, CW (1967); Englewood Cliffs, NJ). Christopherson, S. and Storper, M. (1986). ‘The City as studio, the world as back lot: the impact of vertical disintegration on the location of the motion-​picture industry’. Environment and Planning D: Society and Space 4: 305–​320. Clark, T.N., Lloyd, R., Wong, K.K., and Jain, P. (2002). ‘Amenities drive urban growth’. Journal of Urban Affairs 24: 493–​515. Cohen, D. and Soto, M. (2007). ‘Growth and human capital: good data, good results’. Journal of Economic Growth 12: 51–​76. De la Fuente, A. and Domenech, R. (2006). ‘Human capital in growth regressions: how much difference does data quality make?’ Journal of the European Economic Association 4: 1–​36. Diamond, R. (2012). ‘The determinants and welfare implications of US workers’ diverging locations choices by skill: 1980–​2000’ https://​web.stanford.edu/​~diamondr/​jmp_​final_​all_​ files.pdf (last accessed 4 April 2017).

514   Florida and Mellander Drucker, P. (1969). The Age of Discontinuity: Guidelines to Our Changing Society (New York: Harper and Row). Drucker, P. (1993). Post-​Capitalist Society (New York: HarperCollins). Duranton, G. (2007). ‘Human Capital Externalities in Cities: Identification and Policy Issues’ in R.J. Arnott and D.P. McMillen (eds) A Companion to Urban Economics, pp. 24–​39 (Oxford: Blackwell Publishing). Eaton, J. and Eckstein, Z. (1997). ‘Cities and growth: theory and evidence from France and Japan’. Regional Science and Urban Economics 27: 443–​474. Edlund, L. (2005). ‘Sex and the city’. Scandinavian Journal of Economics 107: 25–​44. Falck, O., Fritsch, M., and Heblich, S. (2011). ‘The phantom of the opera: cultural amenities, human capital, and regional economic growth’. Labor Economics 18: 755–​766. Feser, E. (2003). ‘What regions do rather than make: a proposed set of knowledge-​based occupation clusters’. Urban Studies 40: 1937–​1958. Fischer, M.M. (2011). Spatial Data Analysis: Models, Methods and Techniques (Heidelberg and New York: Springer). Florida, R. (1995). ‘Toward the learning region’. Futures 27: 527–​536. Florida, R. (2002). The Rise of the Creative Class: And How it’s Transforming Work, Leisure, Community and Everyday Life (New York: Basic Books). Florida, R. (2005). ‘The world is spiky’. Atlantic Monthly October https://​www.theatlantic.com/​ past/​docs/​images/​issues/​200510/​world-​is-​spiky.pdf (last accessed 4 April 2017). Florida, R. (2006). ‘Where the brains are’. Atlantic Monthly, October http://​www.theatlantic. com/​magazine/​archive/​2006/​10/​where-​the-​brains-​are/​305202/​2/​ (last accessed 4 April 2017). Florida, R. (2008). Who’s Your City? How the Creative Economy is Making Where to Live the Most Important Decision of Your Life (Toronto: Random House of Canada). Florida, R. (2012). The Rise of the Creative Class, Revisited (New York: Basic Books). Florida, R. and Gates, G. (2002). ‘Technology and tolerance’. The Brookings Review 20: 32–​36. Florida, R. and Kenney, M. (1991). ‘Japanese foreign direct investment in the United States: the case of the automotive transplants’ in J. Morris (ed.) Japan and the Global Economy: Issues and Trends in the 1990s, pp. 91–​114 (London: Routledge). Florida, R. and Kenney, M. (1993). Beyond Mass Production:  The Japanese System and its Transfer to the U.S. (New York: Oxford University Press). Florida, R. and Mellander, C. (2014). ‘The geography of inequality: difference and determinants of wage and income inequality across US metros’. Regional Studies 50: 1–​14. Florida, R., Matheson, Z., Adler, P., and Brydges, T. (2014). ‘The divided city: and the shape of the new metropolis’ http://​community-​wealth.org/​sites/​clone.community-​wealth.org/​files/​ downloads/​report-​florida-​et-​al_​0.pdf (last accessed 4 April 2017). Florida, R., Mellander, C., and Stolarick, K. (2008). ‘Inside the black box of regional development—​ human capital, the creative class and tolerance’. Journal of Economic Geography 8: 615–​649. Florida, R., Mellander, C., and Stolarick, K. (2011). ‘Beautiful places, the role of perceived aesthetic beauty in community satisfaction’. Regional Studies 45: 33–​48. Florida, R., Mellander, C., Stolarick, K., and Ross, A. (2012). ‘Cities, skills and wages’. Journal of Economic Geography 12: 355–​377. Florida, R. (2017) The Next Urban Crisis: How Our Cities Are Increasing Inequality, Deepening Segregation, and Failing the Middle Class - And What We Can Do About It (New York: Basic Books).

Talent, Skills, and Urban Economies    515 Friedman, T.L. (2005). ‘It’s a flat world, after all’. The New  York Times, 3 April http://​www. nytimes.com/​2005/​04/​03/​magazine/​its-​a-​flat-​world-​after-​all.html (last accessed 4 April 2017). Froebel F., Heinrichs, J., and Krey, O. (1979). The New International Division of Labour (Cambridge: Cambridge University Press). Gabe, T.M. (2009). ‘Knowledge and earnings’. Journal of Regional Science 49: 439–​457. Glaeser, E.L. (1998). ‘Are cities dying?’ Journal of Economic Perspectives 12: 139–​160. Glaeser, E.L. (1999). ‘Learning in cities’. Journal of Urban Economics 46: 254–​277. Glaeser, E.L. (2001). ‘Consumer city’. Journal of Economic Geography 1: 27–​50. Glaeser, E.L. and Maré, D.C. (2001). ‘Cities and skills’. Journal of Labor Economics 19: 316–​342. Glaeser, E.L. and Resseger, M.G. (2009). ‘The complementarity between cities and skills’. Working Paper 15103 (Cambridge, MA: National Bureau of Economic Research). Glaeser, E.L. and Saiz, A. (2003). ‘The rise of the skilled city’. Working Paper No. 10191 (Cambridge, MA: National Bureau of Economic Research). Glaeser, E.L., La Porta, R., Lopez-​de-​Silanes, F., and Shleifer, A. (2004). ‘Do institutions cause growth’. Journal of Economic Growth 9: 271–​303. Gottlieb, P.D. (1995). ‘Residential amenities, firm location and economic development’. Urban Studies 32: 1413–​1436. Graves, P.E. (1979). ‘A life-​cycle empirical analysis of migration and climate, by race’. Journal of Urban Economics 6: 135–​147. Graves, P.E. and Linneman, P.D. (1979). ‘Household migration:  theoretical and empirical results’. Journal of Urban Economics 6: 383–​404. Gyourko, J., Mayer, C., and Sinai, T. (2006). ‘Superstar cities’. Working Paper No. 12355 (Cambridge, MA: National Bureau of Economic Research). Hoover, E.M. (1937). Location Theory and the Shoe and Leather Industries (Cambridge, MA: Harvard University Press). Hoover, E.M. (1948). The Location of Economic Activity (Cambridge, MA: Harvard University Press). Hotelling, H. (1929). ‘Stability in competition’. Economic Journal 39: 41–​57. Howkins, J. (2001). The Creative Economy:  How People Make Money from Ideas (London: Penguin). Inglehart, R. (1989). Culture Shifts in Advanced Industrial Society (Princeton, NJ: Princeton University Press). Inglehart, R. (1997). Modernization and Post-​ Modernization (Princeton, NJ:  Princeton University Press). Inglehart, R. and Baker, W.E. (2000). ‘Modernization, cultural change, and the persistence of traditional values’. American Sociological Review 65: 19–​51. Inglehart, R. and Norris, P. (2003). Rising Tide: Gender Equality and Cultural Change Around the World (Cambridge: Cambridge University Press). Inglehart, R., Aguir, C., Ahmad, A., Aliev, A., Alishauskiene, R., and Andreyenkov, V. (2000). World Values Surveys and European Values Surveys, 1981–​1984, 1990–​1993, and 1995–​1997 (Ann Arbor, MI: Institute for Social Research). Jacobs, J. (1961). The Death and Life of Great American Cities (New York: Random House). Jacobs, J. (1969). The Economy of Cities (New York: Vintage). Landry, C. (2008). The Creative City: A Toolkit for Urban Innovators (London: Earthscan). Lösch, A. (1954). Die Räumliche Ordnung Der Wirtschaft [The Economics of Location] (New Haven, CT: Yale University Press).

516   Florida and Mellander Lucas, R.E. (1988). ‘On the mechanics of economic development’. Journal of Monetary Economics 22: 3–​42. McGranahan, D. and Wojan, T. (2007). ‘Recasting the creative class to examine growth processes in rural and urban counties’. Regional Studies 41: 197–​216. Machlup, F. (1962). The Production and Distribution of Knowledge in the United States (Princeton, NJ: Princeton University Press). Mankiw, N.G., Romer, P.M., and Weil, D.N. (1992). ‘A contribution to the empirics of economic growth’. The Quarterly Journal of Economics 107: 407–​437. Marlet, G. and van Woerkens, C. (2004). ‘Skills and creativity in a cross-​section of dutch cities.’ Discussion Paper Series 04–​29 (Tjalling: C. Koopmans Research Institute). Marrocu, E. and Paci, R. (2012). ‘Education or creativity: what matters most for economic performance’. Economic Geography 88: 369–​401. Marshall, A. (1890). Principles of Economics (London: Macmillan). Marx, K. and Engels, F. (1848). Manifesto of the Communist Party (London:  Lawrence and Wishart). Marx, K. and Nicolaus, M. (1993). Grundrisse:  Foundations of the Critique of Political Economy (Rough Draft) (London; New York: Penguin Books in association with New Left Review). Maslow, A.H. (1943). ‘A theory of human motivation’. Psychological Review 50: 370–​396. Massey, D.B. (1984). Spatial Divisions of Labor:  Social Structures and the Geography of Production (New York: Methuen). Mathur, V.K. (1999). ‘Human capital-​based strategy for regional economic development’. Economic Development Quarterly 13: 203–​216. Mellander, C. (2009). ‘Creative and knowledge industries:  an occupational distribution approach’. Economic Development Quarterly 23: 294–​305. Mellander, C., Florida, R., and Stolarick, K. (2011). ‘Here to stay—​the effects of community satisfaction on the decision to stay’. Spatial Economic Analysis 6: 5–​24. Mincer, J. (1974). Schooling, Experience and Earnings (New York: Columbia University Press for the National Bureau of Economic Research). Noland, M. (2005). ‘Popular attitudes, globalization and risk’. International Finance 8: 199–​229. Ohlin, B. (1933). Interregional and International Trade (Cambridge, MA: Harvard University Press). Ottaviano, G.I.P. and Peri, G. (2005). ‘Cities and cultures’. Journal of Urban Economics 58: 304–​337. Page, S.E. (2008). The Difference:  How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies (New Edition) (Princeton, NJ: Princeton University Press). Pandit, K. (1997). ‘Cohort and period effects in U.S. migration: how demographic and economic cycles influence the migration schedule’. Annals of the Association of American Geographers 87: 439–450. Pigou, A.C. (1920). The Economics of Welfare (London: Macmillan). Piketty, T. (2014). Capital in the Twenty-​First Century (Cambridge, MA: Belknap Press). Piore, M.J. and Sabel, C.F. (1984). The Second Industrial Divide:  Possibilities for Prosperity (New York: Basic Books). Porter, M. (1990). The Competitive Advantage of Nations (New York: Free Press). Porter, M. (2006). ‘Q and A with Michael Porter.’ Business Week, 21 August https://​www. bloomberg.com/​news/​articles/​2006-​08-​20/​online-​extra-​q-​and-​a-​with-​michael-​porter (last accessed 4 April 2017).

Talent, Skills, and Urban Economies    517 Rauch, J.E. (1993). ‘Productivity gains from geographic concentration of human capital: evidence from cities’. Journal of Urban Economics 34: 380–​400. Roback, J. (1982). ‘Wages, rents, and the quality of life’. The Journal of Political Economy 90: 1257–​1278. Romer, P.M. (1990). ‘Endogenous technological change’. Journal of Political Economy 98: 71–​102. Rosen, S. (1979). ‘Wage-​based indexes of urban quality of life’ in P. Mieszkowski and M. Straszheim (eds) Current Issues in Urban Economics, pp. 74–​104 (Baltimore, MD: Johns Hopkins University Press). Sampson, R.J. (1995). ‘The community’ in J.Q. Wilson and J. Petersilia (eds) Crime and Public Policy, pp. 193–​216 (San Francisco, CA: Institute of Contemporary Studies Press). Samuelson, P.A. and Nordhaus, W.D. (2005). Macroeconomics (Boston, MA:  Irwin McGraw-​Hill). Saxenian, A. (1994). Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA: Harvard University Press). Scott, A.J. (2000). The Cultural Economy of Cities: Essays on the Geography of Image-​Producing Industries (London: SAGE). Scott, A.J. (2009). ‘Human capital resources and requirements across the metropolitan hierarchy of the USA’. Journal of Economic Geography 9: 207–​226. Shapiro, J.M. (2006). ‘Smart cities: quality of life, productivity, and the growth effects of human capital’. Review of Economics and Statistics 88: 324–​335. Sharkey, P. (2013). Stuck in Place: Urban Neighborhoods and the End of Progress toward Racial Equality (Chicago, IL: The University of Chicago Press). Simon, C.J. (1998). ‘Human capital and metropolitan employment growth’. Journal of Urban Economics 43: 223–​243. Simon, C.J. and Nardinelli, C. (1996). ‘The talk of the town: human capital, information and the growth of English cities, 1861–​1961’. Explorations in Economic History 33: 384–​413. Simonton, D.K. (2000). ‘Creativity:  cognitive, developmental, personal, and social aspects’. American Psychologist 55: 151–​158. Smith, A. (1776). The Wealth of Nations (London: W. Strahan and T. Cadell). Sternberg, R. and Lubart, T. (1999). ‘The concept of creativity: prospects and paradigms’ in R. Sternberg (ed.) Handbook of Creativity, pp. 3–​15 (New York: Cambridge University Press). Stolarick, K. and Currid-​Halkett, E. (2013). ‘Creativity and the crisis: the impact of creative workers on regional unemployment’. Cities 33: 5–​14. Stolarick, K., Mellander, C., and Florida, R. (2010). ‘Creative jobs, industries and places’. Industry and Innovation 17: 1–​4. Storper, M. and Christopherson, S. (1987). ‘Flexible specialization and regional industrial agglomerations: the case of the U.S. motion picture industry’. Annals of the Association of American Geographers 77: 104–​117. Thompson, W.R. (1965). A Preface to Urban Economics (Baltimore, MD:  Johns Hopkins University Press). Thompson, W.R. (1986). ‘Cities in transition’. The Annals of the American Academy of Political and Social Science 488: 18–​34. Thompson, W.R. and Thompson, P. (1987). ‘National industries and local occupational strengths: the cross-​hairs of targeting’. Urban Studies 24: 547–​560. Ullman, E.L. (1958). ‘Regional development and the geography of concentration’. Papers and Proceedings of the Regional Science Association 4: 179–​198.

518   Florida and Mellander Wadhwa, V., Saxenian, A., Rissing, B.A., and Gereffi, G. (2007). ‘America’s new immigrant entrepreneurs: part I’. Duke Science, Technology and Innovation Paper (Chapel Hill, NC: Duke University Press). Wadhwa, V., Saxenian, A., Rissing, B.A., and Gereffi, G. (2008). ‘Skilled immigration and economic growth’. Applied Research in Economic Development 5: 6–​14. Weber, A. (1909). Über den Standort der Industrien, Teil I:  Reine Theorie des Standorts (Tübingen: J.C.B. Mohr). Vernon, R. (1963). Metropolis 1985:  An Interpretation of the Findings of the New  York Metropolitan Region Study (Garden City, NY: Doubleday). Von Thünen, J.H. (1966). Isolated State; An English Edition of Der Isolierte Staat, translated by C. Wartenberg (Oxford: Pergamon Press).

Chapter 27

Imm igrati on a nd the P olitics of Sk i l l Natasha Iskander and Nichola Lowe Introduction Skill has played a central role in immigration scholarship, most notably in a protracted debate over whether ‘unskilled’ immigrants threaten job security for less or moderately educated native-​born workers. That is, are immigrants to blame for job losses experienced by their unskilled native counterparts? Economists especially have kept this discussion alive over several decades, introducing new and innovative methodological techniques to study the effects of immigration on native-​worker job displacement and wage compression in Europe and the USA. Through their careful analysis, they have tipped the debate in favour of immigrants, indicating they create minimal labour market disruption and even complement the domestic workforce. In recent years, a new and important twist has been added to this debate that questions the logic that immigrant workers, particularly those with limited formal education, are unskilled. Recent research, including our own, has helped to reveal the skills of less educated immigrant workers, including transferable knowledge and expertise developed initially through work experience in communities of origin (Iskander and Lowe, 2010, 2013; Lowe et al., 2010; Hagan et al., 2015). Revealing sources of ‘hidden’ skill allows immigration scholars and advocates to ask the ever-​important question of why immigrant skills go unrecognized and justifies claims that immigrant workers deserve greater recognition and reward for their skill contribution, which includes extending legal protections that are granted on the basis of that skill. Still, while reclassifying less educated immigrants as skilled is a useful starting place, in isolation it misses a related contribution that immigrant workers make to skill-​development processes. Immigrants are not simply individuals that possess, acquire, and apply their skill. They are also contributors to collective learning processes through which industry skills are developed, replenished, and recombined overtime. What is defined as individual skill is actually a time-​stamped artefact of ongoing, collective, often tacit learning processes—​a by-​product that might be easier to measure in time and place and by using

520   Iskander and Lowe individual-​level proxies, but ultimately needs to be contextualized in social relations of work in order to understand fully why and when skill changes, and how and under what conditions those changes benefit and gain recognition from employers. While methodologically more challenging, making this connection opens opportunities for immigration and labour scholarship. Less educated immigrants are not the only ones that suffer skills misclassification. Many workers that toil in low-​wage, high-​turnover labour markets, including native-​born workers, are undervalued and underpaid for skill development and training services they provide to the businesses and industries that employ them. Like immigrants, they, too, carry a disproportionate share of the cost and risk associated with industry training and upskilling, thus calling into question traditional economistic models of human capital, which presume low wages are paid to workers because they are low-​skilled, or as a means to offset investments in training by employers. Beyond an example of labour-​market failure, this disconnect presents an opportunity to explore the underlying political and institutional factors that obscure worker contributions to skill development and make it especially difficult for workers in low-​wage segments of the labour market to leverage their collective contribution for economic gain. If this skills oversight affects non-​immigrant workers as well, why should we focus much of this chapter on less-​educated immigrant workers as a special skills case? Immigrants are especially vulnerable to skill misclassification because they lack access to institutions that can protect and defend spaces for, and contributions to, collective learning. Focusing on immigrant contribution to industry skill reproduction in the absence of institutional protections allows us to reflect on the role those institutions play in shaping and reshaping the politics of skill. As we will illustrate, those institutions—​most notably labour unions—​are rarely the initiators or essential sources of industry knowledge and skill. This is not to say they play no role in training and skill development, but rather their primary role has been that of skill protector—​helping groups of workers reveal their collective contribution to industry skill-​building processes and using that to initiate collective bargaining for higher wages and improved working conditions. Starting from this institutional vantage point enables us not only to think critically about why protective institutions might matter for skill development and for whom, but also to interrogate which institutional forms best support worker efforts to reveal their ongoing contribution to industry upskilling. This query has implications for immigration advocacy, but equally suggests a new direction for the growing movement in support of low-​wage workers more generally—​that is to say, a redirection that would call for increased wages in light of evidence of worker-​initiated skill contributions, rather than presuming skill is initially absent without these wage gains. In this chapter, we expand this discussion and provide an illustrative example from our empirical study of Latino immigrant construction workers in two new destination cities. The construction industry has long provided an empirical arena for debate about the value of worker skill and the institutional safeguards necessary to defend it. Thanks in large part to mobilization by organized labour, construction work was once appraised as highly skilled. Now, however, as construction has experienced an influx of new immigrant workers, jobs in this industry have been recast as low-​skill and low-​wage. While industry experts themselves push back against this reclassification, the narrative linking skill decline to Latino immigrant incorporation in construction prevails in popular media accounts, immigration policy

Immigration and the Politics of Skill    521 discourse, and even academic scholarship (Borjas, 2003; Cortes, 2008). This revised view of construction work has taken hold, even though the industry has experienced no technological shifts and the skill base required to complete construction tasks has remained fundamentally unchanged. The construction industry therefore offers as a poignant illustration of skill misclassification, and we use it as a staging ground to reflect on what is at stake for both equity and innovation if contributions to industry skill by low-​waged workers remain undervalued.

Skill as Both Means and Ends For many economists, the relationship between skill development and wages is easily explained and predicted. To economists, wages are signifiers of individual skill and reflect skill specificity. The more specific a set of skills to a company or task, the more investment an employer is willing to make in training individual workers in that skill. But given their investment, the employer will also be less willing to pay that worker a higher wage. Equally, the specificity of that skill makes it more difficult for individual workers to move to another company on the basis of that skill and, as such, that skill is less valued on the open market and commands a significantly lower price. In contrast, general skills—​which is to say skills that are valued widely and not specific to any one company—​are more portable and often receive higher wages on the open market. They also tend to be developed through formal education programmes and through investments in higher education that are made by individual workers and job seekers, rather than the employer. By extension, a worker who is more highly educated can expect higher wages, both justifying and offsetting their educational investment. Employers, for their part, are often willing to pay a higher premium, or ‘efficiency’ wage, to keep more educated workers from leaving. In the eyes of economists, wage differences help to signal to a worker which general skill sets are more valued and thus guide them towards the right external educational and training supports. Of course, economists have extended their work beyond this stylized framework, at times considering how market frictions—​including those introduced by labour market institutions—​shift individual incentives (i.e. wage returns) to invest time and resources in education and training (Acemoglu and Pischke, 1998). Still, what most studies in the human capital tradition share is a view that skill is an easy-​to-​measure, individual attribute, whose responsibility it is to develop is determined primarily by its relative transferability. A rich body of scholarship has raised important challenges about economists’ oversimplified categorization of skill (see Paul Attewell’s thorough review (Attewell, 1990)). For our purposes, an especially problematic feature of the human-​capital approach is the limited attention given to how skill is developed and through what processes and social interactions. Viewed through a human-​capital lens, the skill development process remains an unexplored black box, thereby reinforcing a focus on the ends—​in this case an ‘individual’s fund of knowledge and skill’ (Attewell, 1990, p. 425) measured by years of education or work experience—​rather than the means to reaching these ends. By failing to unpack processes of skill development, we risk missing the way that skill is socially and institutionally produced, which, in turn, can affect the value that employers (and others) assign it.

522   Iskander and Lowe Perhaps the most stalwart critiques of human-​capital conceptions are grounded empirical observations of skill and its reproduction. Authored by sociologists primarily in the 1970s and 1980s, scholarship in this vein has looked at how work is organized and carried out on the factory floor or in specific industry settings, and has focused on the implications of the labour processes for the way that skill is understood and developed (Doeringer and Piore, 1971; Braverman, 1998; Burawoy, 1979; Juravich, 1985). These workplace ethnographies have paid attention to the way that the relations between workers and management inform what is valued as skill (Lee, 1981). They have revealed how worksite interactions not only shape worker access to training opportunities that are organizationally sanctioned, but also support worker-​initiated processes of learning that are informal, unplanned, and often transgressive (Juravich, 1985). These accounts have also called into question the easy binary between general and specific skills advanced by economists, often noting the fluid and contested quality of most skill. Furthermore, they have shown that skill may not, in fact, be the property of any individual worker (or employer for that matter) but rather is most accurately understood as a collective achievement, held in the workplace interactions through which it is enacted and cultivated (Van Maanen and Barley, 1982; Scribner, 1984; Lave, 1988; Cobble, 1991). The construction industry has long offered a rich terrain through which the collective development of skill has been documented, even prior to immigrant incorporation. As labour process ethnographers have noted, the skill of construction workers is critical to industry performance and, consequently, has been a central industry concern: the construction process is inherently unpredictable, and production depends on workers with the expertise to adapt to new conditions and to resolve the unique challenges that emerge on the construction site (Reimer, 1979). Each building structure reflects a particular design, and is fixed to an exact physical location with idiosyncratic characteristics that inform how the building process unfolds. Moreover, construction crews contend with factors such as the vagaries of weather, the site-​specific sequencing of construction tasks and coordination with other trades, and differences in the qualities of materials. Construction tasks also involve working conditions that can be hazardous, such as work that occurs at heights or working with hand-​held pneumatic tools, and where safety outcomes are dependent on the construction abilities (Iskander and Lowe, 2010). Given these conditions, the industry depends on skill that is deep enough so that workers can also resolve the specific problems that they may confront as they complete their work (Silver, 1986; Applebaum, 1999). Given the centrality of construction expertise to day-​to-​day production, how do industry actors perceive and identify skill? The workers, supervisors, and contractors queried by labour ethnographers about definitions of skill concur that a good craftsman is an experienced worker (Reimer, 1979; Steiger, 1993). The industry equation of skill with experience combines an attention to learning and skill; and because it blends a concern with the means through which skill was learned with the technical competence that is its end, it stands in sharp contrast to the human-​capital model of skill that focuses too narrowly on the outcome. Ethnographies of skill development in construction have considered both union and non-​ union settings, and find that despite the institutional difference, the social practices that foster learning are remarkably consistent. Across institutional context, novices learn through cumulative experience under the loose tutelage of more seasoned workers (Applebaum, 1981; Emmitt and Gorse, 2006). This informal teaching system, seemingly casual and

Immigration and the Politics of Skill    523 unstructured, is described as deceptively complex and sophisticated, built around an array of teaching modalities (Graves, 1958; Steiger, 1993; Worthen and Berchman, 2010). Guided demonstration emerges as a core pedagogical practice: in this teaching approach, an experienced worker brackets the flow of construction work and slows down the execution of a task to reveal to the less experienced worker the component movements and logic required to complete it, sometimes accompanying this demonstration with a verbal explanation. During slack times on a construction site, or even off-​site during non-​work times, novices engage in supervised experimentation of those same tasks, with experienced workers providing active coaching or critique of these assays. More experienced workers also provide mentorship in problem-​solving. In Socratic form, they lay out the building challenge to be addressed and solicit ideas from the novice for how to resolve it. Just as often, the experienced worker creates a problem to be solved by misdirecting a worker—​asking for the wrong tool, for example—or by giving incomplete instructions—ordering a worker to carry over lumber or piping but without specifying where the material should be brought. Storytelling is another pedagogical tactic detailed by labour ethnographers (Graves, 1958; Worthen and Berchman, 2010). Journeymen share ‘war stories’ to not only convey ways to cope with technical difficulties on the construction site, trouble with faulty equipment, for example, but also to model strategies to resist management attempts to undermine worker autonomy (Reimer, 1979; Applebaum, 1999). A motivating puzzle for construction studies—​and for other industry studies as well—​ has been how these intricate social systems for learning are reproduced when management plays only a minimal training role. Here industry ethnographies help draw attention to the larger social systems that tether construction-​worker identity to their participation in social processes of teaching and learning. This occupational identity is often overlaid onto broader ethnic or neighbourhoods communities; family ties and ethnic affiliation have historically offered privileged access to construction jobs and to the union (Silver, 1986; Finkel, 1997). A tradition of workers following fathers and grandfathers into the building trades has made norms of mutual obligation around training novices compelling: a journeyman who trained someone else’s kin likely did so with the understanding that others would train his own son or nephew (Applebaum, 1999). But while workers have tended to gain access to the industry through social networks, their ongoing participation depends on their adherence to the occupational culture in the industry, and especially on their deference to ‘craft’ mastery and experience on the job site. ‘Horseplay’ or ‘hazing’ of novices is used on site to enforce the status associated with experience and to enact an occupational hierarchy based on skill. In this context, learning is not merely or even primarily about the acquisition of technical expertise; it is an induction into a way of life (Applebaum, 1981; Lave and Wenger, 1991; Worthen and Berchman, 2010). For most of the twentieth century, building trade unions were the most vocal and powerful representatives of this occupational community, and they used skill to defend the position of their members (Palladino, 2005). Unions asserted that the craft expertise of their members justified the higher wages and preferential hiring they demanded (Finkel, 1997). They lobbied hard to maintain craft standards in the industry, despite contractor and, occasionally, government efforts to reduce industry skill requirements, either to reduce labour costs or increase employment. Unions pledged to guarantee skill, and by implication, the production quality it supported. To that end, they vouched for the competence of union journeymen and instituted an apprenticeship system to train new workers.

524   Iskander and Lowe However, ethnographies of skill development have revealed that unions were less involved in skill development than they claimed publicly. Instead, unions provided institutional cover for the informal social system through which skill was created and reproduced in the industry. They protected social learning processes to which they were largely superfluous. Even in unionized segments of the construction labour force, hiring and training decisions were made informally, often outside the purview of the building trades. As Applebaum, in his monograph on unionized construction, reports, workers were recruited for jobs through personal connections, not through union hiring halls: ‘It is not unusual for a super or foreman, address book in hand, to drive around at night or on the weekend, contacting workers in the homes, trying to round up crews for a job […] The hiring hall is mainly for men either not well-​known or not competent’ (1981, p. 26). Similarly, studies of the industry observe that only a minority of tradesmen entered the industry through formal, union-​managed apprenticeships. Most learned their trade on the periphery of this formal training system, many acquiring foundational knowledge from friends or relatives on smaller side projects (Silver, 1986, p. 111). They entered the union after showing proof of qualification by passing a trade test or clearing other forms of evaluation administered by their local. Through their push for higher wages and employer contribution training programmes (under the Taft–​Hartley act), unions pressed employers to compensate workers for the social system through which the skill for their industry was developed. In an industry where demand was variable and workers faced long and unpredictable stretches of unemployment, unions compelled employers to share in the costs of training industry newcomers and upskilling incumbents, such that when production resumed after a hiatus, employers had a replenished skill pool on which to draw. In the 1970s, building trade unions came under concerted attack from contractors and the Ford administration, and faced a decline in membership, which would drop from a little less than 50 per cent of the labour force in the 1970s to about 15 per cent when it finally plateaued in the 2000s (Palladino, 2005). As they confronted this crisis, which would only become more acute under the Reagan administration, they pushed to further formalize the apprenticeship system and make their contribution to skill development a more tangible source of leverage. Unions systematized their programme of instruction: they added new rubrics for assessment, bolstered classroom instruction, and, in many cases, expanded classroom hours and forged collaborations with community colleges for degree-​granting programs (Weil, 2005). But, in making the training provided under the union umbrella more visible, unions ultimately obscured and supplanted the casual social exchanges through which most industry learning occurred. Once the unions drew an institutional boundary that divided learning from work, separating out pedagogical transfers from informal social interactions, employers could point to the training offered by unions, now clearly defined, and claim that employers, too, could provide such upskilling (Erlich and Grabelsky, 2005). In the 1970s and 1980s, contractor associations expanded their training programmes, offering mostly short-​term, specialized, and modular courses (Northrup and Northrup, 1984). These did not replicate the robust social system of on-​the-​job learning, but because unions, in their push to ring fence and claim training, had removed the notion of occupational community from negotiations over skill, they were unable to invoke the value of informal learning to counter contractor encroachment. Paradoxically, by establishing an institutional proxy for social processes of learning while failing to provide an institutional shield for the processes themselves,

Immigration and the Politics of Skill    525 unions implicitly advanced a definition of skill as individual attribute, forsaking the rich social system of learning that had offered the empirical basis for rebuttals to human capital notions of skill to begin with.

Learning Processes Among Latino Workers in Construction The gap between the interactive social system for skill development and the institutionalized structures that represent training raises questions about learning processes in construction labour markets today. In the 2000s, construction labour markets experienced a sharp increase of Latino immigrant workers, with the proportion they represent more than doubling from 11 per cent in 2003 to 24 per cent in 2012 (Kochhar, 2008). In cities in the south and west of the USA, they make up more than two-​thirds of the labour pool. The majority of Latino immigrants in construction are employed in the residential segment of the industry, but many also work on commercial and infrastructure projects. Despite the proportion of the industry labour market they now represent, these immigrants have widely been portrayed as unskilled, and have been blamed for undercutting wages and turning construction work into a low-​wage occupation (Erlich and Grabelsky, 2005). In addition to the anti-​immigrant rhetoric underlying this claim, it suggests that an industry that has depended on the expertise of workers to function had, in the short span of a decade, somehow become devoid of skill. This unsettling assertion could only be accurate if the longstanding equivalence in the industry between skill and experience had suddenly been broken, and that learning supported through social processes on the job no longer produced competence. Only then would it make sense to assert that workers who had no access to training that was institutionally recognized also had no skill. To interrogate the basis for claims that this new Latino immigrant workforce in construction was unskilled, we conducted an empirical study in two new destination labour markets: Philadelphia, Pennsylvania, where industry training and credentialing processes are tightly controlled by labour unions but closed to immigrant workers; and Raleigh–​Durham, North Carolina, a region in a ‘right-​to-​work’ state where union density is extremely low and therefore there are fewer institutional obstacles to immigrant participation in mainstream construction markets. Still, formal construction training and apprenticeship programmes are sparse in North Carolina and, when they exist, provide limited access for immigrant workers. Drawing on over 200 interviews across both sites over five years (2007–​2011), we explored how skills that immigrants had developed, often before immigrating, informed their participation in construction labour markets (Iskander and Lowe, 2010). Contrary to the prevailing representation of Latino construction workers as unskilled, we found that over half of the immigrant workers we interviewed had acquired significant construction experience before emigrating to the USA. Moreover, they used their competence as a base from which to continue to develop their skill in US labour markets, adapting their knowledge to localized construction techniques and deepening their expertise in one or multiple building trades. They also developed collective learning processes, elaborating learning strategies

526   Iskander and Lowe that consolidated their foothold in the labour market and enabled them to create pathways for occupational advancement. The organization of construction on project sites that employed immigrant workers differed in our two research locations: in Philadelphia, immigrants worked in small teams that completed the range of tasks involved in residential construction and rehabilitation with minimal supervision, whereas in Raleigh–​Durham, immigrants worked on large, hierarchically organized construction sites where crews were assigned specialized construction tasks. Despite these differences, the collective learning processes we observed in each city were more alike than they were different, and they also displayed remarkable continuity with those reported in earlier labour ethnographies of the construction industry. Immigrants at both sites relied heavily on guided demonstration, and workers with more experience in the USA showed their co-​workers how to execute building techniques and use unfamiliar construction tools (Iskander and Lowe, 2013). This teaching method was supported with supervised experimentation. In Philadelphia, workers practised building skills on the job site, relying on team members for help and correction, and in Raleigh–​Durham, where workers experienced more intensive and rigid supervision, they experimented instead on small side projects with more experienced workers instructing novices. Workers also provided one another with mentorship in problem solving. Still, unlike most US workers who learned new skills through mentorship, the primary challenge for the immigrant workers in our study was how to translate the skill they brought with them to their new construction contexts, so that they might apply the full extent of their technical expertise to the construction problems they confronted on the job site. In both cities, workers used stories to share knowledge about how to deal with difficulties they encounter on the job: how to handle supervision, how to avoid injury, and how to resist exploitative practices. They also cultivated a set of norms—​or workplace customs—​that governed learning on the job site and established skill as shared resource. In a telling echo of the occupational culture described in earlier accounts, deference to craft mastery emerged as a constitutive value of the learning practices in both cities. ‘The most important thing for learning and teaching on the job is respect […] respect for each other and for your compadres who have worked in construction for a long time’, explained Rafael from Philadelphia. Their view of skill reasserted the equation of skill with experience that had historically been at the core of definition of competence in industry. The immigrants in our study also merged the means and ends of skill development: for them, the means of acquiring skill was to become part of an occupational community and to participate the social process of learning that characterized it. ‘It’s all relationships’, added Rafael, ‘You treat your co-​workers with respect, with solidarity. It is your friends who teach you …’. In the institutions that governed the construction industry in both cities, Latino immigrants found little protection for the learning processes that were at the heart of their occupational community. There were no formal structures that they could draw on to defend training system they had fostered. In response, immigrants in our study cobbled together a kind of institutional cover to safeguard and assign value to the social exchanges through which they developed their competence, rather than to the skill that was its outcome. In their approach, they focused on the means rather than the ends. In this respect, their strategies resembled those adopted by unions before they found themselves in crisis and responded by elevated the visibility of their apprenticeship programmes in order to make their training contributions explicit (Iskander and Lowe, 2010, 2013; Lowe et al., 2010).

Immigration and the Politics of Skill    527 In both cities, immigrants drew on the organization of production on job sites to defend and expand their social systems of learning. In Philadelphia, for example, where immigrants worked in loosely supervised teams, they pushed for collective pay increases and quality bonuses. In this way, they safeguarded the cohesion of their work group from employer attempts to create divisions by paying workers differently. They protected the least skilled among them, giving them time to learn as they pre-​empted any employer attempts to winnow out the novices or to pit workers on the team against one another as a strategy to drive down wages. In Raleigh–​Durham, where the organization of work was hierarchical and highly specialized, immigrant workers replicated certain dominant practices in order initially to shield more transgressive practices of skill development. In several instances, highly ranked immigrant workers used their labour market status to promote learning practices in the crews they were assigned to supervise. Aware of expertise within their crews, yet also mindful of the need to support learning among newcomers, these immigrant supervisors took steps to relax job categories and flatten job ladders, thereby encouraging members of their work crew to engage in cross-​task training and job rotation (Iskander and Lowe, 2013). Latino immigrants in both cities also used broader institutional structures as a resource to open up additional spaces for learning. Government regulations and safety rules comprised one institutional area that immigrants drew on. In Philadelphia, immigrant workers used government licence inspections of building permits opportunistically to query city officials about scaffolding techniques and learn about the structural organization of the historic row houses they were remodelling. In Raleigh–​Durham, immigrant workers used state-​ mandated on-​site safety training as an arena to explore how to repurpose previous skills for local construction and to acquire new ones: safety training videos were used as a pedagogical break in which workers could collectively reflect on building techniques, and safety monitoring was transformed into a vehicle for mentoring around skill and team-​building (Lowe and Iskander, 2015). Efforts like these demonstrate the value that workers place on collective learning processes as not only a means for occupational advancement, but also as a cultural norm of an occupational community. They also display the resourcefulness and ingenuity of workers as they draw on workplace practice and institutional structure to create proto-​institutions to protect social processes. A subset of the immigrant workers in our study were even able to achieve some improvement in job quality, including increased worker prestige, more secure and predictable hours, and even small increases in compensation. But, ultimately, with no institutional mechanism to compel employers to compensate them for their industry skill contribution, their jobs often remained low-​wage and their employment insecure. Immigrant workers in non-​construction industries also struggle to advance their labour-​ market position on the basis of their contribution to collective learning processes. Rich ethnographies by Roger Waldinger, Leslie Salzinger, Travis Du Bry, Manuel Adrián Hernández Romero, Ruth Gomberg-​Muñoz, and Miriam Wells, among others, provide hints of this struggle through their detailed accounts of immigrant workers in agriculture, housekeeping, apparel, janitorial services, and hospitality. Admittedly, learning and skill development were not their primary research concern, and so these descriptions of immigrant-​authored learning processes are less detailed and drawn out. However, similar to what we uncover for construction, their depictions offer evidence that group learning practices of immigrant workers contribute to a shared sense of pride in the work, and also make it difficult to assign attribution for quality performance to any one individual within the group (Zlolniski, 2006;

528   Iskander and Lowe Newman, 2009). We flag these implicit references to collective learning as an opportunity for coordinating worker advocacy in multiple industries where low-​wage work has become the new norm. Combining our research in construction with related ethnographies of work creates an opening for organizing efforts in the name of social and economic justice to be more inclusive of these learning processes, especially when making demands for higher wages and stronger institutional protection at the bottom of the labour market. It is this connection to which we next turn our attention.

The Low-​wage Turn: Bringing Collective Learning Back in Immigrant advocates increasingly situate demands for improved working conditions and wages within a larger critique of low-​wage work generally. The focus on the low-​wage economy has intensified in recent years and has gained considerable traction given national policy concern over income inequality in the wake of the Great Recession. Tethering the immigrant experience to that facing all low-​wage workers, regardless of national origin, is both necessary and strategic. It is impossible to study the low-​wage economy without recognizing the sizeable share of immigrant workers within many low-​wage industries and labour markets. Politically, too, the low-​wage umbrella provides a helpful buffer against anti-​immigrant rhetoric and backlash, ultimately shifting the focus to questions of moral economy, rather than citizenship. But while this close coupling has helped to bring much needed research attention to the plight of immigrant workers, this low-​wage turn—​at least in its current iteration—​has pushed important insights about worker contributions to skill and learning into the shadows. As research on low-​wage jobs progresses, so, too, must our understanding of workplace practices and learning processes that low-​wage workers contribute to and collectively inform. Existing studies of low-​wage work help catalogue many features of low-​wage jobs that make them demeaning and unrewarding for workers (Appelbaum et al., 2003; Bernhardt et al., 2008; Doussard, 2008). To be sure, the defining characteristic of these jobs is inadequate compensation, but scholars also point out that low wages are accompanied by a compendium of other traits that undercut job quality. These include managerial practices that are degrading and exploitative, restrictions on worker autonomy and decision-​ making, and stratagems to siphon off some portion of worker earnings, either through overt forms of wage theft or pernicious employment practices that drive down compensation (Milkman et al., 2010; Kalleberg, 2011). Moreover, low wages are often correlated with employer tactics to exert direct control over workers both on and off the job site: scheduling practices that prevent workers from holding additional jobs or managing non-​work obligations, employer codes specifying appropriate social behaviour, and other obstacles to career advancement (Peck and Theodore, 2001; Standing, 2011; Weil, 2014). Much of this scholarship has attributed the rise of low-​quality, low-​wage jobs to a wholesale gutting of protective employment institutions. The macro-​institutional shifts since the late 1970s include not only de-​industrialization and subsequent growth in low-​end

Immigration and the Politics of Skill    529 service jobs, but also de-​unionization and weakened labour laws that make it harder for workers to defend themselves against employer abuse (Appelbaum et al., 2003; Bernhardt et  al., 2008). Scholars have also considered micro-​institutional changes, most notably transformations in the organization of work and related normative shifts that affect how employers treat and value their workforce. They have documented the steady rise in contingent work arrangements, including labour subcontracting, that not only create considerable distance between workers and the final beneficiaries of their effort, but, more importantly, make it harder to identify which companies and employers should be implicated for unsafe, unjust, and at times deadly, working conditions (Applebaum et al., 2003; Kalleberg, 2011; Weil, 2014). A sub-​stream of scholarship unpacks industry-​specific reconfigurations in order to explore why some groups of workers within an industry or organizational setting might experience these changes differently—​for example, why flexible employment practices might lead to greater worker empowerment in some organizational environments or occupational categories but not in others (Bailey and Bernhardt, 1996; Batt et al., 2003). In highlighting ‘varieties’ of employment practices, this scholarship helps recast low-​wage work as the product of actions taken by employers when institutional protections are absent, and suggests the ways in which labour advocacy can change the calculus on which employers’ decisions are based. The research on advocacy efforts for low-​wage workers often starts with the insight that low wages are not inevitable or a predetermined outcome of economic change, and then looks for potential inroads for securing better jobs and better pay in an era where traditional labour institutions have collapsed or been greatly weakened. The advocacy efforts depicted in this scholarship have used low wages as a central organizing principle, but their strategies reflect an understanding that low wages are associated with job characteristics that make organizing strategies within a single firm or industry unsuccessful. When jobs are contingent and when the offending employer is hard to pinpoint within numerous contractual layers, organizers find it difficult to use the traditional labour union strategies to bargain collectively. Instead, organizers invoke community belonging or other forms of identity politics to mobilize groups of workers (Piore and Safford, 2006; Ghandnoosh, 2013). Studies of these new forms of advocacy have shown that unions representing low-​ wage workers, such the Service Employees International Union and its affiliates, have pioneered new forms of community organizing that link job quality concerns to identity politics (Milkman, 2006). Here, we also find examples of hybrid community-​labour approaches that target low-​income immigrant workers in particular (Milkman, 2000; Milkman and Wong, 2000; Osterman, 2002). Worker centres are a case in point, providing a laboratory for pioneering new approaches to organizing and outreach that are often based in immigrant neighbourhoods and that seek to promote job quality in immigrant-​heavy industries, such as construction, housekeeping, and hospitality (Fine, 2006; Valenzuela Jr, 2014). Still, while this scholarship provides us with a rich and textured empirical understanding of the world of low-​wage work and illuminates a pathway for labour advocacy, it shares a familiar blind spot when it comes to the question of skill development and, more specifically, the collective learning processes that workers contribute to, even within this low-​wage context. Admittedly, some labour scholars have chosen to back away from

530   Iskander and Lowe questions of worker skill altogether, fearing that this discussion will shift the focus away from job quality to worker characteristics. Arne Kalleberg summarized this concern best, arguing that a focus on skill can lead to the problematic reframing that there ‘might be good and bad workers, but not good and bad jobs’ (2011, p. 6). Moving the focus away from skill and towards institutional erosion has ultimately allowed labour scholars to mount a powerful challenge to long-​standing claims by neoclassical economists that rising income inequality is simply the outcome of declining or irrelevant skills (Osterman, 2000; Levy and Temin, 2007). But, paradoxically, the lack of attention given to learning processes by institutionally minded scholars only reinforces many of the same assumptions that traditional theories of human capital make about skill and how it gets produced and by whom. As in the standard human capital model, workers are presumed to be compensated less because they lack skill at that moment in time. Thus, many scholars advocating for improvements at the bottom of the labour market also seem comfortable using the terms low-​wage and low-​skilled interchangeably—​and when describing skill, present it as an individual attribute, fixed in time. Additionally, in the same way that educational programmes rank highly as essential sources of general skills in the traditional human capital narrative, so, too, do formal external institutions in the low-​wage camp, even if the mix of training institutions is more varied. Some scholars even make the case that workers remain trapped in low-​wage, dead-​end jobs because of their exclusion from formal educational and training institutions (Appelbaum et al., 2003; Grimshaw, 2011). By extension, this leads to calls for expanding access to formal training systems through which low-​wage workers can develop their skill and, in turn, gain access to better jobs. Admittedly, some studies of low-​wage work do make an important conceptual break from traditional human capital theory by challenging the very notion that skill is a prerequisite to higher wages and instead outlining a compelling argument that payments of higher wages will lead to industry upskilling. However, this revised twist is often predicated on the logic that employers will invest in worker training in order to offset those higher wage costs, thus reinforcing yet another tenet of traditional theory. What is missing from this research is an analytical space for also revealing how learning happens within low-​wage jobs, even if employer recognition and institutional support is initially absent. Paying attention to skill development not only turns our attention back to the job, but also brings squarely into the frame the social processes through which workers actively shape their jobs through their collective learning. In the example of low-​wage construction jobs we provide in this chapter—​which is implicitly echoed in other ethnographies of low-​wage work—​workers used these social learning processes to improve the quality of their jobs and press for higher wages, more discretion, and greater job security. In a virtuous cycle, they used social interactions on the job site to strengthen collective learning practices, thus deepening their skill contribution and increasing their leverage to push for better working conditions. And equally the immigrants we studied also attempted to create norms and other proto-​institutions to protect their skill-​building processes and contributions. However, these efforts at bottom-​up institution building have proved insufficient to compel the majority of employers to compensate and reward workers fully for the skills they collectively create and replenish. This suggests the need for further analysis that connects investigations of skill and collective learning in low-​wage jobs to an exploration of avenues for further advocacy.

Immigration and the Politics of Skill    531

A New Politics of Skill The current and ever-​growing spotlight on low-​wage work has been essential for inspiring policy change and is clearly behind recent advances in minimum-​and living-​wage legislation and efforts to shore up state and federal labour laws. Still, there is an opportunity for the next round of scholarship on low-​wage work to advance worker bargaining power by situating workers as active participants in collective processes of workplace learning and, thus, as contributors to industry performance and innovation. Retraining the research eye on collective learning processes can also help inspire low-​wage advocacy organizations to embed fully worker contributions to industry skill within a larger narrative around worker rights and demands for social and economic justice. Linking justice with skill thus necessitates a unifying conceptual platform for combining two currently divergent approaches to scholarship on labour advocacy (Adler, 2006; Burawoy, 2008)—​firstly, an older tradition that uses ‘thick description’ to reveal complex social dynamics within occupational communities and in ways that reveal existing power struggles around skill; secondly, a newer focus on innovative strategies for worker empowerment and mobilization, including new partnerships to reinvigorate the US labour movement. Connecting these two threads requires not only that we integrate labour processes within studies of the new labour movement, but also that we reposition worker advocacy efforts to engage explicitly with the politics of skill. A useful starting place for this combined research endeavour involves documenting the diverse ways in which labour and community advocates currently link collective skill development processes to gains in worker compensation and worker rights more generally. While a voluminous body of research is dedicated to institutional practices that support collective learning within highly paid professions with particular attention to managerial strategies to reward and strengthen knowledge sharing, we know very little about related experiments that cultivate and protect collective skill development within the low-​wage economy. What might existing models look like for low-​wage workers in particular? How can advocates help increase the visibility of worker-​initiated learning processes and use this to secure better wages and improved working conditions at the bottom of the labour market? To what extent are these experiments inclusive of employers, and how adaptive are they to contingent forms of ‘fissured’ work that might limit employer involvement and accountability? An example by Leslie Salzinger (1991), grounded in an earlier labour process research tradition, helps illustrate what can be gained from more documentation in this area. Through her comparative case study of immigrant housekeepers in the San Francisco Bay Area, Salzinger describes one successful cooperative, called ‘Choices’, as an ‘occupational community’ through which women immigrant refugees shared knowledge and insights, and, in turn, gained valuable lessons from older, more experienced housekeepers and nannies. As she put it, ‘members trade tips constantly, developing and sharing strategies to deal with dirty houses and impossible employers in the same breath’ (Salzinger, 1991, p. 149). A crucial focus of those discussions was communication with clients, and specifically how to use those conversations to draw attention to worker skills and a related commitment to quality service. This included empowering all members to make it clear at the time of service that quality required sufficient time be spent on a given task, but equally that members would

532   Iskander and Lowe ‘not work extra time for free’ (Salzinger, 1991, p. 148). Members of the cooperative ultimately encouraged each other to walk away from any job that undervalued their knowledge or undermined their sense of control over their work schedules and routine. The cooperative reinforced a space not only for collective skill-​building, but also for using that shared knowledge as the basis for worker power, which was especially important as housework was typically performed individually, not by multiple members of the group. The immigrant advocates at the non-​profit that created the Choices cooperative built on the skill exchanges initiated by cooperative members, and brought additional training opportunities into the social interactions of the group, enabling members, for example, to learn about new cleaning products and what might be required when encountering a new kind of appliance, surface, or material. What is key, therefore, was the mutually reinforcing quality of this exchange. Rather than positioning the non-​profit as the initiator of the learning process, staff saw its role as learning partners that contributed and helped facilitate a virtuous skill-​building cycle. Their role also ensured the collective skill of the cooperative remained visible and formed the basis for securing wages that were almost twice the legislatively mandated minimum wage. Fundamentally, the Choices experiment speaks to the need to document whether and how community and labour advocates today are transforming group learning into a collective and visible resource from which to garner better wages and improve working conditions. It is quite possible that local actors, like workers’ centres, are developing strategies to cultivate and defend skill-​building among low-​wage workers, but research attention to efforts in this direction may have been edged out by the focus on wage-​based mobilization. Even as we propose an exploration of the way that local labour advocates foster and protect collective skill-​building, however, we are also mindful of the resource constraints facing many non-​profit advocacy groups, especially workers centres. This suggests the need for additional research that helps link localized advocacy efforts to a larger macro-​institutional push to strengthen the conceptual link between collective learning contributions and efforts to secure broad-​based gains in wages and worker rights. The current movement for an overall increase to the national minimum wage and regional increases above that to compensate for variable cost of living is grounded in an argument for social justice and inclusion. This is a powerful framework, but without inclusive language around skill, it risks homogenizing jobs at the bottom of the labour market and, in the process, erasing industry specific skills contributions upon which further wage gains can be secured. Its emphasis on justice without equal attention to worker contribution through skill complicates larger conversations about how to create institutions that defend collective learning and ensure that workers are compensated for the full support they provide to industry. In this chapter, we call attention to the ways in which workers in low-​wage jobs develop skills—​skills that are specific to their firms and general to their industry, and skills that they hold individually and that they hold collectively in interactions at the job site and in their communities. We have highlighted the contribution their collective learning makes to production and productivity. Finally, we have argued that what is needed is a more visible and nuanced role of skill in the push for higher wages. Taken together, these observations suggest there is a case to be made that low-​wage work itself represents a form of wage theft. Workers disproportionately carry the burden for industry training and skill development; non-​existent or insufficient compensation for their investment in skill creation allows the

Immigration and the Politics of Skill    533 jobs they hold to remain low-​wage. As the movement for higher wages progresses, a stronger call should be made for employers to absorb more of the costs associated with this collective learning and not just reap its benefits. Even with mandated wage raises, jobs in which workers are paid anything short of an amount that reflects their full contribution still remain low-​ wage, and still undermine workers’ ability to build solid livelihoods.

References Acemoglu, D. and Pischke, J.-​S. (1998). ‘The structure of wages and investment in general training’. National Bureau of Economic Research. Adler, P.S. (2006). ‘From Labor Process to Activity Theory’ in P. Sawchuk, M. Elhammoumi, and N. Duarte (eds) Critical Perspectives on Activity Theory, Education and Work:  An International Collection, pp. 160–​192 (Cambridge: Cambridge University Press). Applebaum, H.A. (1981). Royal Blue: The Culture of Construction Workers (New York: Holt Rinehart & Winston). Applebaum, H. (1999). Construction Workers, U.S.A. (Westport, CT: Greenwood Press). Appelbaum, E., Bernhardt, A., and Murnane, R.J. (2003). ‘Low Wage America: An Overview’ in R. Murnane and E. Applebaum (eds) Low Wage America: How Employers Are Reshaping Opportunity in the Workplace, pp. 1–​29 (New York: Russell Sage Foundation). Attewell, P. (1990). ‘What is skill?’ Work and Occupations 17: 422–​448. Bailey, T. and Bernhardt, A. (1996). ‘In Search of the High Road in a Low-​Wage Industry’. IEE Working Paper No. 2. Batt, R., Hunter, L.W., and Wilk, S. (2003). ‘How and When Does Management Matter? Job Quality and Career Opportunities for Call Center Workers’ in R. Murnane and E. Applebaum (eds) Low-​wage America:  How Employers are Reshaping Opportunity in the Workplace, pp. 270–​315 (New York: Russell Sage). Bernhardt, A., Boushey, H., Dresser, L., and Tilly, C. (eds) (2008). The Gloves-​off Economy: Workplace Standards at the Bottom of America’s Labor Market (New York: Cornell University Press). Borjas, G. (2003). ‘The Labor Demand Curve is Downward Sloping: Reexamining the Impact of Immigration on the Labor Market’. No. w9755. National Bureau of Economic Research. Braverman, H. (1998). Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century (New York: NYU Press). Burawoy, M. (1979). Manufacturing Consent: Changes in the Labor Process Under Monopoly Capitalism (Chicago, IL: University of Chicago Press). Burawoy, M. (2008). ‘The public turn from labor process to labor movement’. Work and Occupations 35: 371–​387. Cobble, D.S. (1991). Dishing it Out: Waitresses and Their Unions in the Twentieth Century (Chicago, IL: University of Chicago Press). Cortes, P. (2008). ‘The effect of low‐skilled immigration on U.S. prices: evidence from CPI data’. Journal of Political Economy 116: 381–​422. Doeringer, P.B. and Piore, M.J. (1971). Internal Labor Markets and Manpower Analysis (New York: ME Sharpe). Doussard, M. (2008). Degraded Work: Industry Restructuring, Immigration and the New Low-​ wage Labor Market (Chicago, IL: University of Illinois at Chicago).

534   Iskander and Lowe Emmitt, S. and Gorse, C. (2006). Communication in Construction Teams (New York: Routledge). Erlich, M. and Grabelsky, J. (2005). ‘Standing at a crossroads: the building trades in the twenty-​ first century’. Labor History 46: 421–​445. Fine, J.R. (2006). Worker Centers: Organizing Communities at the Edge of the Dream (Ithaca, NY: ILR Press). Finkel, G. (1997). The Economics of the Construction Industry (Armonk, NY: ME Sharpe). Ghandnoosh, N. (2013). ‘Organizing Workers along Ethnic Lines’ in R. Milkman, J. Bloom, and V. Narro (eds) Working for Justice: The LA Model of Organizing and Advocacy, p. 49 (Ithaca, NY: Cornell University Press). Graves, B. (1958). ‘ “Breaking out”:  an apprenticeship system among pipeline construction workers’. Human Organization 17: 9–​13. Grimshaw, D. (2011). ‘What do we know about low wage work and low wage workers: analysing the definitions, patterns, causes and consequences in international perspective’. International Labour Organization http://​www.ilo.org/​travail/​whatwedo/​publications/​ WCMS_​157253/​lang-​-​en/​index.htm (last accessed 4 April 2017). Hagan, J., Hernandez-​Leon, R., and Demonsant, J.L. (2015). Skills of the ‘Unskilled’: Work and Mobility among Mexican Migrants (Berkeley, CA: University of California Press). Iskander, N. and Lowe, N. (2010). ‘Hidden talent: tacit skill formation and labor market incorporation of Latino immigrants in the United States’. Journal of Planning Education and Research 30: 132–​146. Iskander, N. and Lowe, N. (2013). ‘Building job quality from the inside-​out: Mexican immigrants, skills, and jobs in the construction industry’. Industrial & Labor Relations Review 66: 785–​807. Juravich, T. (1985). Chaos on the Shop Floor:  A  Worker’s View of Quality, Productivity, and Management (Philadelphia, PA: Temple University Press). Kalleberg, A.L. (2011). Good Jobs. Bad Jobs: The Rise of Polarized and Precarious Employment Systems in the United States, 1970s to 2000s (New York: Russell Sage Foundation). Kochhar, R. (2008). ‘Latino labor report 2008: construction reverses job growth for Latinos’. Pew Research Center http://​www.pewhispanic.org/​2008/​06/​04/​latino-​labor-​report-​2008-​ construction-​reverses-​job-​growth-​for-​latinos/​ (last accessed 17 April 2017). Lave, J. (1988). Cognition in Practice (New York: Cambridge University Press). Lave, J. and Wenger, E. (1991). Situated Learning:  Legitimate Peripheral Participation (Cambridge: Cambridge University Press). Lee, D.J. (1981). ‘Skill, craft and class:  a theoretical critique and a critical case’. Sociology 15: 56–​78. Levy, F. and Temin, P. (2007). ‘Inequality and institutions in 20th century America’. Working Paper 07-​17. Lowe, N. and Iskander, N. (2015). Power through Problem Solving: Latino Immigrants and the Inconsistencies of Economic Restructuring (Chapel Hill, NC: University of North Carolina Press). Lowe, N., Hagan, J., and Iskander, N. (2010). ‘Revealing talent: informal skills intermediation as an emergent pathway for immigrant labor market incorporation’. Environment and Planning A 42: 205–​222. Milkman, R. (2000). Organizing Immigrants:  The Challenge for Unions in Contemporary California (Ithaca, NY: Cornell University Press). Milkman, R. (2006). L.A. Story: Immigrant Workers and the Future of the U.S. Labor Movement (New York: Russell Sage Foundation).

Immigration and the Politics of Skill    535 Milkman, R. and Wong, K. (2000). ‘Organizing the Wicked City: The 1992 Southern California Drywall Strike’ in R. Milkman (ed.) Organizing Immigrants: The Challenge for Unions in Contemporary California, pp. 169–​198 (Ithaca, NY: Cornell University Press). Milkman, R., Bloom, J., and Narro, V. (eds). (2010). Working for Justice:  The LA Model of Organizing and Advocacy (New York: ILR Press/​Cornell University Press). Newman, K.S. (2009). No Shame in My Game:  The Working Poor in the Inner City (New York: Vintage Books). Northrup, H.R. and Northrup, D.O. (1984). Open Shop Construction Revisited (Philadelphia, PA: University of Pennsylvannia Center for Human Resources). Osterman, P. (2000). Securing Prosperity: The American Labor Market: How it has Changed and What to do About it (Princeton, NJ: Princeton University Press). Osterman, P. (2002). Gathering Power: The Future of Progressive Politics in America (Boston, MA: Beacon Press). Palladino, G. (2005). Skilled Hands, Strong Spirits: A Century of Building Trades History (Ithaca, NY: Cornell University Press). Peck, J. and Theodore, N. (2001). ‘Contingent Chicago: restructuring the spaces of temporary labor’. International Journal of Urban and Regional Research 25: 471–​496. Piore, M.J. and Safford, S. (2006). ‘Changing regimes of workplace governance, shifting axes of social mobilization, and the challenge to industrial relations theory’. Industrial Relations: A Journal of Economy and Society 45: 299–​325. Reimer, J. (1979). Hard Hats:  The Work World of Construction Workers (Beverly Hills, CA: SAGE). Salzinger, L. (1991). ‘A Maid by Any Other Name:  The Transformation of “Dirty Work” by Central American Immigrants’ in M. Burawoy, A. Burton, A. Arnett Ferguson, K.J. Fox, J. Gamson, et al. (eds) Ethnography Unbound: Power and Resistance in the Modern Metropolis, pp. 139–​160 (Oakland, CA: University of California Press). Scribner, S. (1984). ‘Studying Working Intelligence’ in B. Rogoff and J. Lave (eds) Everyday Cognition: Development in Social Context (Cambridge, MA: Harvard University Press). Silver, M.L. (1986). Under Construction:  Work and Alienation in the Building Trades (New York: SUNY Press). Standing, G. (2011). ‘Labour market policies, poverty and insecurity’. International Journal of Social Welfare 20: 260–​269. Steiger, T.L. (1993). ‘Construction skill and skill construction’. Work, Employment and Society 7: 535–​560. Valenzuela Jr, A. (2014). ‘Regulating Day Labor:  Worker Centers and Organizing in the Informal Economy’ in V. Mukhija and A. Loukaitou-​Sideris (eds) The Informal American City: From Taco Trucks to Day Labor, p. 261 (Cambridge, MA: MIT Press). Van Maanen, J. and Barley, S. R. (1982). ‘Occupational communities: culture and control in organizations’. DTIC document. Weil, D. (2005). ‘The contemporary industrial relations system in construction:  analysis, observation, and speculations’. Labor History 46: 447–​471. Weil, D. (2014). The Fissured Workplace (Cambridge, MA: Harvard University Press). Worthen, H. and Berchman, M. (2010). ‘Apprenticeships: what happens in on-​the-​job training (OJT)?’ Learning Through Practice 1: 222–​239. Zlolniski, C. (2006). Janitors, Street Vendors, and Activists: The Lives of Mexican Immigrants in Silicon Valley (Berkeley, CA: University of California Press).

Pa rt V I

F I NA N C E

Chapter 28

F inance and Fi na nc ia l Systems: Evolv i ng Ge o graphies of C ri si s and Insta bi l i t y Gary A. Dymski Introduction This chapter explores geographical approaches to financial systems, with special attention to their instability. This review focuses on research by geographers, while also taking into account the implications for spatial analysis of economists’ approaches to financial instability. Financial dynamics since the late 1970s have been defined by relentlessly expanding globalization and by instability of two different forms: ever-​deepening financialization, and increasingly frequent and virulent financial crises. Financialization, that is, ‘the increasing role of financial motives, financial markets, financial actors and financial institutions in the operation of the domestic and international economies’ (Epstein, 2006, p. 3), has gradually transformed microeconomic and social dynamics during this period. And at the macro-​ level, as Laeven and Valencia (2012, p. 10) document, crisis cycles have coincided with (or followed from) credit cycles during this period: financial crises have occurred from the late 1970s onward, finally reaching a crescendo with the double (subprime and Eurozone) meltdown of the late 2000s. These micro-​and macro-​phenomena are mutually reinforcing: growing household and firm debt, one of the most visible manifestations of financialization, has fed financial fragility and led to the destabilization of credit flows and economic growth the world over, slowing economic growth, and increasing households’ and firms’ dependence on debt and financial manipulation, leading to further financial instability, and so on. Despite the centrality of these developments in global economic outcomes, economists have reached no agreement on the sources of financial instability, the relationship between local (national) and global financial dynamics, or whether and how public-​policy responses

540   Dymski can mitigate the social costs of financial losses and crises at either the micro-​or macro-​levels. On the contrary, they have proposed very different understandings, which say as much about the differences in their theoretical entry points as about financial processes and instability. Economists who use market equilibria as their points of departure and of reference—​that is, those in the mainstream—​tend to trace the financial losses and crises of recent years to factors interfering with markets’ microeconomic workings: government policy failures or perverse incentive mechanisms in financial markets. Financial crises, when they arise, reflect in-​principle-​avoidable coordination failures; financial instability arises as because of disturbances to market logic, not as a consequence of market logic. The notion that financial losses and crises are inevitable components of economic processes is maintained by heterodox economists, whose entry points typically include uncertainty, power, and/​or class. Of special interest in this chapter is Minsky’s conception of financial instability as a core component of capitalist systems with advanced financial systems. Geographers’ investigations of how space matters in financial processes and systems, like those of economists, are shaped by their analytical entry points. And those pioneering the emerging field of the geography of finance have used very different entry points and theoretical frameworks—​Marxian crisis theory, critical social geography, and the institutional analysis of global hybridity, in particular—​to shed light on many different spatial aspects of financial structures, behaviours, and outcomes. This chapter reflects on some of the principal lines of development and findings of this literature, as well as exploring one area ripe for further examinations of how space matters in finance—​that is, the theory of financial instability itself. The next section examines some of the foundational contributions that launched the geography of finance. ‘The Spatial Logic of Globalizing Finance’ then describes spatial research on the global spread of innovative practices in finance—​an initial focal point of this young field. ‘Why Most Geographers and Economists Overlooked the 1980s–​1990s Financial Crisis Wave’ asks why so little attention was paid to macro aspects of financial crises prior to September 2008. ‘Geographers on Sub-​prime Lending and Crisis’ reviews geographers’ investigations of subprime lending and of the subprime crisis; missing from this work, as from the pre-​crisis geography of finance, is attention to the macro-​dimension of financial instability. ‘Minskyian Financial Instability in a Spatial Context’ describes the most prominent macro approach to financial instability and crisis, that of economist Hyman Minsky, and elaborates both on its Keynesian foundations and on how incorporating spatial dimensions into this approach can generate new insights into financial crisis and instability. The chapter ends with ‘Conclusion: Debating Post-​crisis Finance and Financial Instability’.

Foundational Explorations in the Geography of Finance Aalbers’ (2014) overview essay on the evolution of financial geography dates its beginnings to several mid-​1990s publications, including Andrew Leyshon’s three Progress in Human Geography essays (1995, 1997, 1998) on geographies of money and finance. Aalbers points out that geographers began examining spatial aspects of financial processes in the 1970s. Two such early contributions set the pattern for subsequent geographical work on financial

Finance and Financial Systems    541 systems and their instability. One direction, cited by Aalbers, is the work of Boddy (1976) and Williams (1978) on the UK mortgage market. Boddy discusses the links between housing finance and inequality, and Williams investigates race-​and income-​based redlining by UK lenders in inner-​city areas of UK cities.1 These two early studies parallel the earliest work on redlining in US mortgage markets (Alhbrandt 1977; Bradford and the Urban–​Suburban Investment Study Group, 1977) in recognizing the destabilizing impact of redlining on the economic trajectories of affected urban areas. The other notable 1970s geographical contribution on finance, David Harvey’s (1973) work on the political economy of the city, focused more on its systemic (macro) role in capitalist reproduction. This and subsequent works by this author inspired much of contemporary social geography, by way of imitation, modification, or critique. As Ira Katznelson points out in his foreword to the 1988 edition of Harvey’s 1973 volume Social Justice and the City, Harvey’s goal, to ‘embed . . . [geography] in a theoretical project’, found a locus in Marxism, ‘and an object of analysis for this theory within geography, the city’. The tension between a surplus-​generating mode of economic production and the spatial location of production and surplus absorption provides a robust anchor for Marxian work in geography. Marx’s characterization of the accumulation process as contradictory and crisis prone finds ready interpretation in the contradictory demands made on urban space: while productive assets and housing are spatially fixed and long lived, reorganizing surplus-​generating activities and renewing accumulation requires the destruction and renewal of places within the urban grid. Finance plays a central role, as Harvey recognized, in putting expensive, long-​lived assets in place; and these financial commitments can undermine capitalist growth or recovery when debt obligations remain after the real capital they have financed is devalued. So finance is seen both as functionalist and as potentially dysfunctional: both a means of resolving one set of contradictions (financing gaps) and the source of another (debt repayment gaps). Harvey was especially attracted to Hilferding’s notion of ‘finance capital’, wherein the allocation of credit and circulation of monies is driven by the needs of surplus-​ generating accumulation processes. This notion is embodied in the idea of a city as a growth machine (Molotch, 1976). Harvey’s subsequent work (especially Harvey, 1982) was based on a more comprehensive reading of Marx on capitalist accumulation and crisis. Regarding finance, he emphasized Marx’s analysis of how the pre-​conditions for crisis are created in cyclical upturns when promises to pay multiply without adequate attention to ability to pay. In these foundational texts, Harvey poses a question that he leaves open: is urban development explained by capitalism’s broader cyclical dynamics—​its boom–​bust cycle—​or is it independent of any ‘cycle’, with its own momentum? These early contributions, then, confront the question of whether financial relations are spatially embedded or, instead, are determined primarily by cyclical economic forces. Leyshon’s (1995) survey of the emerging political economy of finance identifies three approaches to this question. ‘Geopolitical economy’ examines hegemonic dominance and transitions amongst national currencies. The ‘geoeconomics of finance’ opposes the notion that financial globalization means ‘the end of geography’ (O’Brien, 1991) by investigating the distinctive national, and even regional and local, basis on which financial systems remain organized.2 The third approach identified by Leyshon, the geography of financial exclusion, encompasses both nations that have suffered debt crises and sub-​national spaces subject to ‘the closure of banking infrastructures, with catastrophic economic consequences for populations abandoned in this way’ (Leyshon, 1995, p. 538)—​a phenomenon that he viewed as being most acute in the USA but that he saw as spreading to the UK and elsewhere.

542   Dymski Leyshon’s essay links destabilizing dynamics at the macro-​level with the deepening of uneven development at the micro-​level: he cites exchange-​rate pressures due to imbalances among European economies, deregulation of financial services, and competition for privileged customers as factors that are driving processes of micro-​level exclusion. The possibility that micro-​level mechanisms could undermine macro-​level financial stability was not yet in view: subprime lending and private-​label securitization were in their infancy, and not until the early 2000s would housing price bubbles bring subprime mortgages to Main Street. The mid-​1990s stream of work on financial exclusion saw it as encompassing ‘processes that prevent poor and disadvantaged social groups from gaining access to the financial system. . . [with] implications for uneven development’ (Leyshon and Thrift, 1995, p. 312) Consequently, scholarly work on financial exclusion encompassed bank branch closures (Pollard, 1996), the role of finance in uneven spatial development (Dymski and Veitch, 1996), the shifting boundaries of the financial system, and the need for policy change instituting a concept of ‘financial citizenship’ (Leyshon and Thrift, 1996). In effect, financial exclusion was viewed as a generalization of the case of redlining.

The Spatial Logic of Globalizing Finance The years immediately after the 1973 demise of the Bretton Woods system saw some nations succumb to recession and inflationary pressure, while others continued to expand. And epochal institutional changes—​including financial deregulation, an end to pattern bargaining, and deindustrialization—​were underway in some but not all countries. Some scholars identified patterns in these contrasting trends: it seemed that institutional and regulatory differences might account for these varying macroeconomic growth patterns. Most notably, Zysman (1983) argued that the superior industrial competitiveness of Germany and Japan, as against France, the UK, and the USA, could be traced to the organization and governance of their financial systems. His suggestive contrast between Anglo-​American and German–​Japanese approaches to finance shifted attention from the nature of capitalist crisis per se to the question of how nations’ financial structures and regulations affect economic performance.3 This reframing of finance as a component of industrial strategy in a competitive global marketplace raised yet a new question: how would the global shift to financial deregulation and freer capital movement affect individual firms and nation states? Geographers began to undertake mappings that not only responded to O’Brien (1991), but also began to identify a distinctive ‘geography of finance’ approach. As Clark and Wójcik (2007) termed it: This conceptual approach to understanding institutional change differs from more functionalist approaches . . . The main premise of the approach is not only to take the convergence forces behind global finance seriously—​as more and more political economies engage with global finance—​but also to recognize the variety of geographically and historically contingent institutional responses and filters, which result in hybridity rather than homogeneity.

These mappings focused on the role of practices (Jones and Murphy, 2010) and institutions in reproducing or challenging the loci of control over financial decisions and allocations. In a study of pension fund decision making, Clark (2008) argued that good financial governance

Finance and Financial Systems    543 would emerge if financial decision making reflected a balance between expertise, on one hand, and community representation and political legitimacy, on the other. The globalization of share-​holding threatened this balance by creating pressures for homogenized decision making. The quantification of financial expertise (Hall, 2006) drove this homogenization, as did the spread of formalized business courses (Hall and Appleyard, 2009), which both legitimized expertise and distributed it along the hubs and spokes of financial supply chains. Given the turn away from functionalism in explanation, whether the financial system would continue to meet different spatial areas’ diverse needs then became an open question. Geographers’ institutional focus on hybridity was augmented by an analysis of global networks and of the presence or absence of global convergence in standards and practices. The evidence on global convergence was mixed. Mason and Harrison (2002) found that UK venture capital investment had become more equal across space, but investments in younger UK companies remains concentrated in the broader London area. Zook (2002) argued that the spatially uneven financing needs of the Internet industry were being adequately met precisely because venture capitalists were resisting centralization and homogenization. Wójcik (2006) found a trend towards global convergence in corporate governance, involving shareholder and board roles. Clark and Wójcik (2007) used German experience to demonstrate the dynamic erosion of static ‘varieties of capitalism’ distinctions among national economies. To explain this meta-​trend, Clark et al. (2006, p. 303) observed: ‘The continuity of different regimes of governance is subject to inter-​market arbitrage’. In turn, Dixon and Monk (2009) contrasted the harmonization of accounting standards and spread of globalized corporate governance practices with the uneven spatial distribution of defined-​benefits pensions in the UK and the Netherlands. Examining Swiss pension funds in this same (pre-​crisis) time period, Corpataux et al. (2009) found that both homogenizing and ‘territorial’ forces were at work: ‘the mobility/​liquidity of capital and the changing dimensions of new regions and countries are central to the finance industry’s functioning’. This emerging literature not only explored the scope and extent of financial globalization, but also had a critical edge. As Leyshon has written, a paper by Clark and Wójcik (2007) ‘seeks to account for the growing power of money and finance within contemporary economic life’ (Leyshon, 2008, p. 262). Clark and Knox-​Hayes (2007) found that ‘social status’ crucially determined pension holdings. Leyshon et al. (2008) found that building society closures were concentrated in poorer areas.

Why Most Geographers and Economists Overlooked the 1980s–​1990s Financial Crisis Wave Given the critical edge that characterized financial geography from its origins, Aalbers’ 2014 observation in his overview of financial geography contributions to the geography journal Transactions comes as a shock: ‘In the 1970s and early 1980s Transactions published no papers that discuss the economic crises of that decade in primarily monetary or financial terms’. While geographers were engaged in critical analysis of subprime lending well before

544   Dymski the subprime crisis (see the next section), Aalbers has a point; indeed, his comment on the absence of attention to financial crisis can be extended to other geography journals, and to the 1990s, as well as the late 1970s and 1980s. A keyword search of the Web of Science database for thirteen leading geography and regional-​studies journals found zero appearances of the term ‘financial crisis’ in the 1980s, and only one in the 1990s (in Transactions in 1995).4 This observation is surprising because these years were accompanied by successive financial crises: citing only the most severe episodes, the years prior to 2000 included the US savings and loan crisis, the Latin American debt crisis, the 1987 stock market crash, the junk bond bubble and crash, the Mexican crisis of 1994–​95, the East Asian financial crisis of 1997, and the Russian–​Brazilian–​Turkish foreign exchange crisis of 1998. Economists too paid little or no attention to financial crisis in these years. A parallel search for a sample of leading mainstream economics journals in the 1980s and 1990s also finds virtually no keyword references to ‘financial crisis’; a search of heterodox economics journals, only a handful. In any event, most academics working on finance were subsequently caught unawares by the subprime meltdown. And that event concentrated minds well outside of academia on the disruptive power of finance. Indeed, the enormity of the September 2008 crisis led Queen Elizabeth II to ask, in a November 2008 visit to the London School of Economics, why economists did not foresee it. A July 2009 response by representatives of the Royal Academy mentioned economists’ belief that ‘banks knew what they were doing’, and the fact that low inflation and modest economic growth had lulled economists into ignoring the growing imbalances; thus, there was a ‘failure of the collective imagination of many bright people, both in this country and internationally’ (Besley and Hennessey, 2009). A Financial Times columnist observed, in response, that ‘economists [had] shuffle[d]‌the deckchairs’ (Brittan, 2009). Why then did geographers and most economists largely ignore the global wave of financial crises until the catastrophic events of Fall 2008 made it impossible to look away? Geographers’ neglect of financial crises can be traced to three factors. Firstly, the geography of finance had not yet cohered as a subfield. Secondly, when it did, it was initially oriented towards the role of finance in industrial competitiveness. Thirdly, emerging work on the geography of economics and finance relied primarily on two very different entry points that were predisposed to overlooking the possibility of a cataclysmic financial crisis that could bring global capitalism to its knees: Marxian models of accumulation and crisis, and neoclassical equilibrium models. We consider these in order. Most geographers who implicitly or explicitly reject neoclassical economic theory, and instead view the capitalist economy as prone to booms and busts, tend to follow David Harvey’s conception of Marxian crisis theory. In his framework, finance has an assigned role to play in urban (or capitalist) reproduction. Geographical discourses that use Harveyesque lenses to elaborate Marxian ideas about capitalist crises arguably view them as crises of capitalism tout entier. To put it boldly, in the Marxian approaches followed within geography, economic crises might be accompanied by financial crises, but they are not caused by financial crisis. Neoclassical economic models, in turn, typically assign a passive role to finance. Efficient-​ market theorists such as Fama (1980) regard financial structure as a passive element in economic outcomes, on the basis that the self-​interest of participants in hyper-​competitive markets would lead them ruthlessly to eliminate any inefficiencies in resource allocation; the key implication is that market processes leave no potentially profitable holes in the financial services landscape. Neoclassical theorists prefer to use the general competitive equilibrium

Finance and Financial Systems    545 as their point of analytical reference, complexifying it as needed to explain why deviations from this equilibrium might occur. In the postwar period, this analytical approach characterized microeconomic thinking, not macroeconomics. But since the overthrow of Keynesian structural macro-​modeling at the end of the 1970s, the terrain of macro-​modelling has been left to equilibrium models—​most recently, the dynamic stochastic general equilibrium (DSGE) model. The defining characteristic of the DSGE model is its general equilibrium approach. In any general equilibrium model (including DSGE), analysis focuses on the determinants of the supply of goods and services, that is, on available technology, the prior distribution of wealth, and agents’ preferences over consumption and leisure (more consumption can be had if more labour is supplied, hence more supply created, ceteris paribus). Demand is passive; any supply generated will be sold as long as prices are right. The prices may not be right, owing to informational rigidities or transactions costs in particular. But what should be emphasized is that this theoretical approach is antithetical to Keynes (1936) and more broadly to Keynesian macroeconomics, which asserts that the level of effective demand determines income flows and, in turn, the level of output: if demand falls, so will employment and income levels. This emphasis on the centrality of demand contrasts markedly with the orthodox perspective in macroeconomics, wherein demand responds to supply and does not independently affect the level of economic activity. The standard DSGE model excludes the possibility that aggregate spending can fall short in the aggregate, because it anticipates that prices should always be able to fall so as to close the gap. The idea that we may live in a world in which prices are ‘downward sticky’ is ruled out. Further, the DSGE model, in its unadulterated form, abstracts from the financial sector altogether; so there is no space for even conceptualizing, much less explaining financial crisis, except by reference to deviations from the conditions required for market equilibrium. These deviations tend to be explored in microeconomic models of banking, given the absence of analytical oxygen in the DSGE model. And, indeed, coincident with the Latin American debt crisis, Diamond and Dybvig (1983) developed an asymmetric information banking model that generated a bank run as one possibility.5 This framework gave rise to a sizeable literature exploring the implications of asymmetric information and one of its consequences, moral hazard, for financial outcomes. Subsequently, then, problems in financial markets could be attributed by mainstream theorists to mechanism or design failures, moral hazard problems, ‘sunspots’, or ‘sudden stops’.6 These categories neatly account for the above list of crises: the Latin American crisis can be attributed to debtor nations’ moral hazard (Eaton et al., 1986), as can the East Asian crisis (Krugman, 1998); the savings and loan crisis can be explained by thrift managers’ and regulators’ moral hazard (Kane, 1990); and the 1997 and 1998 crises to sudden stops (Calvo, 1998).7 Evidence for the idea that financial crises could be blamed on flaws in market processes became available when, in the wake of the East Asian financial crisis, economists associated with the International Monetary Fund (IMF) and World Bank pioneered a new empirical approach to financial crisis. They created a database consisting of aggregate statistics drawn from the years immediately before and after financial-​crisis episodes in affected countries; the idea was to identify causal patterns in these episodes. Econometric tests using this database generate cross-​national answers that abstract from differences in time and space (Beck et al., 2010) This database has been expanded and used as the basis of numerous studies; almost invariably these identify lax financial oversight as a root cause of financial crises.

546   Dymski

Geographers on Subprime Lending and Crisis While geographers had largely ignored the 1980s Latin American and 1990s East Asian financial crises, they did extensive work on subprime lending prior to the outbreak of the crisis. While, as noted, this work focused on the microfoundational aspects of this lending and not on the build up of macrostructures of financial imbalances, geographical work on financial exclusion pre-​dated the subprime crisis by three decades. Further, some of those researching examining credit-​market discrimination and financial exclusion turned their attention to subprime lending soon after it emerged in the 1990s. The first rounds of work on this new form of predatory financial inclusion forced researchers to ask basic institutional questions: who was authorizing subprime loans, and who was funding them? How were bank and non-​bank markets connected, locally and nationally? Where was the regulatory oversight of these new loan types? Were race and gender differences targeted? Answers were proposed in a number of studies, including Listokin and Wyly (2000), Bradford (2002), Newman and Wyly (2004), Wyly et al. (2006), and Williams et al. (2006). UK-​based research on subprime lending posed a further question: would the spread of these instruments to the UK homogenize these markets across national borders? Wainwright (2009a, 2009b) considered this question, based on research undertaken before the 2008 crisis. He showed how the process of feeding locally generated mortgages into the globally linked securitization process reflected a mixture of influences: ‘Big Bang’ deregulation of the City of London, the broader financial deregulation process, and the specific legal, political, and social institutions that defined UK housing finance. In other words, UK housing finance did not involve cross-​border cloning, but instead retained key differences from US practices (see Ashton, 2009). Aalbers et al. (2011, p. 1779), in turn, showed that mortgage practices in the Netherlands ‘reflects Dutch corporatist institutional arrangements, implying that both geography and states do matter for the supposedly aspatial process of securitization’. What are the implications of the subprime crisis for the geography of finance? One was described picturesquely by Lee et al. (2009): ‘When all that is solid is seen to melt into air, [geographers] are forced into the role of the reporter who sketches the first draft of history’. Aalbers (2009a, 2009b) took up this challenge; his essays delineated the dynamics and institutional context of the near collapse of global finance in September 2008. Engelen and Faulconbridge (2009, p. 591) called for ‘financial geographies which are historically situated and focused not on the epochal but the conjectural and . . . relevant to academic and policy debates’. The four contributors to another reactive essay broadly agreed (Lee et al., 2009); for example, Clark emphasized the need to study ‘the interplay between financial markets, between market players and institutions and between markets and political institutions’ (Lee et al., 2009, p. 734). And, indeed, a continuing stream of research in economic geography has pursued this ‘first draft of history’ by unearthing ‘the long chains associated with the credit crunch’ (Wainwright, 2011, p. 1301). For example, Marshall et al. (2011) showed how the September 2007 run on Northern Rock resulted from the interplay between global financial centres and peripheral financial hubs such as Newcastle, and then deepened development in the urban landscape; Pani and Holman (2013) showed how even localities at a ‘fictitious distance’ from

Finance and Financial Systems    547 global booms and busts—​such as Norwegian municipalities—​were entangled in the crisis owing to intertwined cross-​border financial cash flows; and Hendrikse and Sidaway (2014) showed how the German city of Pforzheim experienced crisis-​linked losses due to derivative contracts it had signed with Deutsche Bank. Another reaction to the financial crisis was registered in Lee et al. (2009, pp. 740–​1) by Leyshon, who asserted that ‘we need to know more about the geographies of asset creation and destruction’, ‘the regulatory geographies of the global financial system’, and the ‘geopolitical consequences of finance’. Following this lead, French et al. (2009) set out a framework describing the compositional architecture of the crisis. These authors identified four geographical spaces that had combined to generate the extraordinary force of this sequence of events: (i) the international financial centres, especially London and New York; (ii) the growth of insular financial practices that disregard the risks arising from small-​margin bets on highly leveraged asset positions; (iii) structural imbalances in the global economy, especially the US–​China linkage; and (iv) the growing power of financial media in shaping ‘the behavior and culture of financial agents and institutions’ (French et al., 2009, p. 287). As these authors observed, while this crisis marked one logical end of the globalization of finance, it did not clearly indicate the end of global finance. They worried that ‘the financial system may already be reinventing itself in the midst of crisis. . . . [It] is highly likely that a new financial paradigm is already in the making’ (French et al., 2009, p. 299). But what paradigm? What if the restless dynamism of the financial system, its tem­ poral instability, is part of what defines it? How could the architecture of an unstable global financial system be described across space? Pollard signalled the need for understanding time/​space dynamics when she called (in the same Lee et al. (2009) essay) for analysis that went beyond ‘the scientism of technical, purportedly objective metrics of liquidity, rate of return, shareholder value and so forth’ (p. 738). But doing so would require moving from the description of spatially differentiated structures to an understanding of how these structures moved and combined in real time (Hall, 2013); it would also require moving further away from the efficient-​market approach to finance and even further in the direction of ideas about finance which Hyman Minsky had initially advanced in the 1970s, and which were ever more critical in comprehending the evolving shape of financial relations and financial crises.

Minskyian Financial Instability in a Spatial Context Shifting from a description of how spatial structures of finance could generate unstable dynamics to a spatial dynamics of unstable finance requires two steps. The first is to build a model that unfolds the motion of unstable financial processes across time. The second is to understand the spatial scope and logic of such a model. We consider these in turn. The first of these steps requires a conscious break from the view that financial crises arise from regulatory disturbances or informational barriers that prevent financial processes from achieving socially optimal equilibria. Understanding a financial crisis as a disturbance from an otherwise stable financial equilibrium rules out seeing it as one logical end point of core

548   Dymski financial processes. University of California, Los Angeles, economist Axel Leijonhufvud (2014, pp. 761–​2) put it this way: . . . the economists who in the last 20 or so years have based their macroeconomics on GE [general equilibrium] constructions have shown little interest in investigating their stability properties. Stability has been taken ‘on faith’ . . . It is my belief that this stability-​with-​impediments approach is quite wrong, that it does not explain recent events, and that it fails to suggest the right policies.

The diagnoses of our current problems that we get from DSGE practitioners all tend to run in terms of stable general equilibrium systems beset with ‘ “frictions”. . . .Walrasian constructions, even those of recent vintage . . . are hopelessly inadequate for dealing with financial crises and their aftermaths’ (Leijonhufvud, 2014, pp. 771–​2). Contributions to the geography of finance do not generally identify their degree of reliance on efficient-​market or equilibrium frameworks. Pollard (2003) had recognized some years before that implicit reliance on efficient-​market theory made it impossible to understand the credit starvation of small businesses as disequilibria, the result of imperfect real-​time decision-​making. This provides a start towards a larger-​scale breakdown of supply-​demand relations, triggered by malfunctioning financial contracts and spreading to the broader real-​sector effects, but it does not go all the way. Establishing that disequilibria in individual markets could generate broader-​based financial malfunctioning and macroeconomic disturbances requires an analytical framework in which the level of aggregate demand can differ systematically from aggregate supply—​that is, a Keynesian macro-​framework. At the behavioural base of this process is necessarily the assumption that economic agents are irrational, or that some agents are misinformed and taken advantage of by others, or that agents are operating with incomplete information sets. The latter two assumptions set up ‘real-​time’ decisional contexts, as opposed to the notional time context required to establish stable market equilibria. Fortunately, real-​time financial analysis rooted in a Keynesian macro-​framework does not have to be invented de novo. This approach constituted the analytical baseline of the work of the Keynesian economist Hyman Minsky (1975, 1982). Minsky and his followers were among the heterodox Keynesian economists who had been warning about financial instability and crises for nearly three decades (see Galbraith, 2009). Hyman Minsky based his theory of capitalist dynamics on the centrality of money and credit. He argued that a defining characteristic of capitalist economies was chronic financial fragility due to the tendency of debt commitments to outpace available cash flows over the business cycle. Built up financial fragility would generate financial instability once debt/​cash flow gaps put downward pressure on financial market prices. Following Keynes, Minsky argued that when the panic came, a run to liquidity would result. Whether a financial crisis then ensues depends on whether the central bank and fiscal authorities in the affected nation acts as a lender of last resort (satisfying liquidity demand) and undertakes counter-​cyclical spending. Minsky encapsulated this ‘financial instability hypothesis’ with his oft-​repeated phrase, ‘stability is destabilizing’. In this ‘Wall Street view’, financial crisis arises as part of the normal cyclical rhythm of capitalist economies. This brings us to the second step noted earlier: linking this logic to the space of interlinked global financial crises. Heterodox Keynesians have begun exploring how to adapt Minsky’s nation state-​and US-​centric ideas about the crisis-​prone trajectory of financialized

Finance and Financial Systems    549 capitalism to the case of globally interlinked crises in the neo-​liberal era. Consider Keen’s (2015, p. 298) summary of the heterodox Keynesian approach to crisis: Post Keynesian economics has two complementary theories of crisis that were used to predict the 2007 crisis and diagnose its causes: Minsky’s financial instability hypothesis and Godley’s stockflow-​consistent approach. Both theories take a monetary perspective on capitalism and argue that the dynamics of private debt caused the crisis. . . .both theories imply that the current recovery will be short-​lived because the underlying cause of the last crisis has not been addressed by subsequent economic policy.

Here, structural imbalances at the global level are combined with Minskyian dynamics to form the heart of the analysis. Crisis is an expression of imbalances at the level of the whole, driven by debt overloads and balance-​sheet inconsistency. The stock-​flow consistent approach makes it clear that imbalances across the globe—​trade balances, savings–​ investment balances, and government revenue–​ expenditure balances—​ inevitably arise, and must equal zero as a matter of logic. What makes it Keynesian is a dual assertion: firstly, each of the aforementioned couplets includes one of the elements of aggregate demand; secondly, when aggregate demand is not sufficiently high, across the globe, only government action is capable of assuring that the required balance will not be forced by global economic shrinkage—​that is, crisis.8 In the heterodox view, then, falling levels of investment or consumption due to financial market collapse will unbalance the set of interacting macro-​imbalances and force downward shifts in economic activity unless an external force (such as government stimulus) steps in. This is what Leijonhufvud (2014) meant in observing that mainstream models have failed to identify the balance-​sheet recession. In the heterodox view, an analysis of a global economic crisis requires a global structural perspective. Structural rigidities are not disturbances that must be overcome if a more desirable market equilibrium is to emerge; they are the model. Geographers have, until now, overlooked Minsky’s analysis of financial instability. There are several reasons. Firstly, Minsky produced his opus before the geography of finance came of age; and the papers he published prior to his death in 1996 were published almost exclusively in heterodox economics journals. So while these journals frequently reference Minsky’s touchstone concepts and often use ‘Minsky’ as a keyword, geography journals do not. Underlying this lacuna, in turn, is the fact that Minsky’s ideas are rooted in Keynes’ central ideas—​ that fundamental uncertainty, not probabilistic risk, underlies investment (Keynes, 1936, Ch. 12), and that inadequate aggregate demand will trigger stagnation (Keynes, 1936, Chs 2–​4). And geographers have, to this point, engaged only minimally with the core Keynesian concepts of fundamental uncertainty and independent aggregate demand.9 A third reason is that Minsky himself did not write about how space and place might affect the dynamics or outcomes of financial instability. He was always focused on a stylized depiction of financial instability and crisis, with the stages enumerated—​from a robust to a fragile to a Ponzi financial structure—​purposely left loosely defined. These stages, by design, could refer to an economic unit or to an economy as a whole; so the notions of variations across space, or how this framework might change in national settings other than the US, were not considered.

550   Dymski

Conclusion: Debating Post-​crisis Finance and Financial Instability While geographers have clearly demonstrated that taking space into account can more clearly identify the scale and scope of financial processes, the economic approaches to finance on which much inquiry in the geography of finance implicitly relies have largely ignored its spatial dimensions. Mainstream theoretical and empirical models of financial crisis abstract from space; the heterodox approach highlights imbalances across space but is indifferent to which areas are experiencing imbalance and why. The seriousness of this lacuna became evident when the subprime crisis exposed the fact that geographers had largely ignored financial crises until then - as one manifestation of an overall breakdown in the capitalist accumulation process (per Harvey). Closing this gap in the wake of the subprime crisis will require a consideration, within the geography of finance, of how financial instability plays out through time and in space. In effect, what is needed is a spatialization of Minsky’s financial instability framework. While some economists have explored this terrain (Dymski, 1999; Schroeder, 2002), a fuller mapping depends on expanding the dialogue between geographers and heterodox economists (Dymski, 2010), especially those who work with Minsky’s ideas. Meanwhile, the importance of analysing instability more directly is evident in two open debates in the geography of finance. The first of these debates concerns the role of finance in mediating relations between the global economy and national capitalist formations. If, as Sokol (2013) has suggested, geographers of money and finance see ‘financialisation as an inherently spatial process—​as part of the search for spatial–​temporal fix’ does global finance operate flexibly, so as to permit several semi-​autonomous ‘varieties’ of capitalist societies to coexist? Or does it instead operate as an empowered forcefield, breaking down whatever barriers any given nation state or region attempts to place on its freedom of movement? Geographers have expressed increasing doubts about the former view. For example, Dixon (2010), hoping for a ‘common agenda’ between the geography of finance and varieties of capitalism, argued that the institutional mechanisms of financialization could be usefully explored within the varieties-​of-​capitalism framework. But in a later paper, he criticizes the latter literature for presuming that ‘function follows from form’ (Dixon, 2012, p. 279). Engelen et al. (2010) also critique the varieties of capitalism framework for its ‘productivist’ approach to finance, and argue that geographies of financialization are ‘in disarray’. These doubts lead to an open question for further research: are the global forces creating this disarray homogenizing global space, or do they undermine some national differences while leaving variegated remnants of differentiation behind? French and Leyshon (2010, p. 2549) pointed out the irony that while the financial crisis required massive state intervention, ‘rather than developing a form of capitalism wherein the state exerted more control over the economy it seems conversely to have heralded an age of austerity and an emboldened form of hyper-​neoliberalisation’. But what then are the spatial and temporal characteristics of this hyper-​neoliberalism? The second ongoing debate turns to whether global finance organized through financial centres is a source of economic growth or instability and crisis. It was brought into focus

Finance and Financial Systems    551 by Martin (2011), who emphasized the uneven regional impacts of the crisis in the UK and the USA, and then argued that the costs of hosting a global financial centre may outweigh the benefits. Martin et al. make a further argument: ‘spatial economic imbalance in the UK has to do with the progressive concentration of economic, political and financial power in London and its environs’ (2015, p. 1). Wójcik (2012, 2013) argues that global financial centres have undercut effective regulation, both before and after the subprime crisis: ‘the global financial crisis 2007–​09 originated to a large extent in the [New York-​London] axis rather than in an abstract space of financial markets . . . contrary to expectations the axis is not in decline’ (Wójcik, 2013, p. 2736). This debate also involves the question of how to understand financial centres’ spatial footprints. On one hand, Cook et al. (2012), elaborating a theme introduced by Thrift (1994) and Leyshon and Thrift (1997), show how City of London insiders use cultural capital to reproduce its social exclusivity. On the other, Taylor et al. (2009) identify twenty command-​and-​ control financial centres throughout the global economy; and Wainwright argues that the subprime crisis hit as deeply as it did precisely because the growth of ‘communities of practice’, and expertise in peripheral regions (such as Leeds) linked by growing network relations to financial hubs (London) helped to ‘expose British mortgage lenders to the crash’ (2013, p. 1041). In conclusion, Queen Elizabeth II could not accuse geographers, as she did mainstream macroeconomic theorists, of being blind to the perverse financial dynamics that preceded the subprime crisis. On the contrary, geographers had undertaken in-​depth analyses of key elements undergirding these dynamics—​from the global spread of innovative financial instruments and practices to the growth of sub​prime lending. But geographers did have an analytical blind spot: their work paid almost no attention to the increasingly severe financial crises that dotted the globe prior to 2007. The few economists who did ‘see it coming’ largely built on the fragility–​instability framework of Hyman Minsky. We have argued here that the problem of how to incorporate financial instability into spatial analysis is among the logical next steps for research in the geography of finance. Spatializing the analysis of instability in finance, in effect, will shed further light on some open debates among geographers. In the two debates reviewed immediately above, the dynamic of financial instability is present but not explicit. This is not to say that there is one clear answer to the question of how space and financial instability are interrelated. Given the plethora of institutional structures and regulatory regimes now characterizing financial systems around the globe, it would be surprising if one analytical conclusion sufficed. But the ongoing economic and financial crises clearly demonstrate the urgency of rendering more visible the links between financial instability and the space of financial systems.

Notes 1. Aalbers (2005) has himself published work on mortgage redlining in Rotterdam. 2. Leyshon cites Lash and Urry (1987, 1994). Also see Corpataux et al. (2009). 3. Zysman’s work renewed the tradition pioneered by Gurley and Shaw (1955) and Gerschenkron (1962). 4. One important exception is Corbridge (1984). Details on the method used and on the journals included in this investigation are set out in Dymski and Shabani (2017).

552   Dymski 5. Their framework falls into the category of multiple-​ equilibrium ‘sunspot’ models discussed above. 6. A sunspot model can shift among multiple equilibria when market participants’ beliefs change. This same mechanism underlies the ‘sudden stop’ model, which has been used to explain sovereign debt crises (such as that in East Asia) that cannot be traced to borrowers’ moral hazard. 7. Dymski (2011) reviews economists’ writings on international debt crises. 8. Significantly, IMF economists have developed and begun to explore DSGE macro-​models that pay attention to stock-​flow-​consistent linkages in global dynamics; see Kumhof et al. (2010). 9. Storper (2011) and Bhattacharjea (2010) discuss the non-​Keynesian basis of the New Economic Geography.

References Aalbers, M.B. (2005). ‘Who’s afraid of red, yellow and green? Redlining in Rotterdam’. Geoforum: 562–​580. Aalbers, M.B. (2009a). ‘Geographies of the financial crisis’. Area 41: 34–​42. Aalbers, M.B. (2009b). ‘The sociology and geography of mortgage markets: reflections on the financial crisis’. International Journal of Urban and Regional Research 33: 281–​290. Aalbers, M.B. (2014). ‘Financial geography: introduction to the Virtual Issue’. Transactions of the Institute of British Geographers (New Series) 40: 300–​305. Aalbers, M.B., Engelen, E., and Glasmacher, A. (2011). ‘ “Cognitive closure” in the Netherlands: mortgage securitization in a hybrid European political economy’. Environment and Planning A 43: 1779–​1795. Ahlbrandt, R.S., Jr. (1977). ‘Exploratory research on the redlining phenomenon’. Journal of the American Real Estate and Urban Economics Association 5: 473–​481. Ashton, P. (2009). ‘An appetite for yield: the anatomy of the subprime crisis’. Environment and Planning A 41: 1420–​1441. Beck, T., Demirgüç-​Kunt, A., and Levine, R. (2010). ‘Financial institutions and markets across countries and over time: the updated financial development and structure database’. World Bank Economic Review 24: 77–​92. Besley, T. and Hennessey, P. (2009). Letter to Her Majesty Queen Elizabeth II, 22 July 2009 (London: British Academy). Bhattacharjea, A. (2010). ‘Did Kaldor anticipate the New Economic Geography? Yes, but . . . ’. Cambridge Journal of Economics 34: 1057–​1074. Boddy, M.J. (1976). ‘The structure of mortgage finance: building societies and the British social formation’. Transactions of the Institute of British Geographers New Series 1: 1976: 58–​7 1. Bradford, C. (2002). Risk or Race? Racial Disparities and the Subprime Refinance Market. A Report of the Center for Community Change (Washington, DC: Center for Community Change). Bradford, C. and the Urban-​Suburban Investment Study Group (1977). ‘Redlining and disinvestment as a discriminatory practice in residential mortgage loans’ Center for Urban Studies, University of Illinois (Washington, DC:  Department of Housing and Urban Development, Office of the Assistant Secretary for Fair Housing and Equal Opportunity). Brittan, S. (2009). ‘Economists shuffle the deckchairs’. Financial Times, 6 August 2009.

Finance and Financial Systems    553 Calvo, G.A. (1998). ‘Capital flows and capital-​market crises: the simple economics of sudden stops’. Journal of Applied Economics 1: 35–​54. Clark, G.L. (2008). ‘Governing finance: global imperatives and the challenge of reconciling community representation with expertise’. Economic Geography 84: 281–​302. Clark, G.L. and Knox-​Hayes, J. (2007). ‘Mapping UK pension benefits and the intended purchase of annuities in the aftermath of the 1990s stock market bubble’. Transactions of the Institute of British Geographers New Series 32: 539–​555. Clark, G.L. and Wójcik, D. (2007). The Geography of Finance: Corporate Governance in the Global Marketplace (Oxford: Oxford University Press). Clark, G.L., Wójcik, D., and Bauer, R. (2006). ‘Geographically dispersed ownership and inter-​ market stock price arbitrage—​Ahold’s crisis of corporate governance and its implications for global standards’. Journal of Economic Geography 6: 303–​322. Cook, A.C.G., Faulconbridge, J.R, and Muzio, D. (2012). ‘London’s legal elite:  recruitment through cultural capital and the reproduction of social exclusivity in City professional service fields’. Environment and Planning A 44: 1744–​1762. Corbridge, S. (1984). ‘Crisis, what crisis? Monetarism, Brandt II and the geopolitics of debt’. Political Geography Quarterly 3: 331–​345. Corpataux, J., Crevoisier, O., and Theurillat, T. (2009). ‘The expansion of the finance industry and its impact on the economy: a territorial approach based on Swiss pension funds’. Economic Geography 85: 313–​334. Diamond, D.W. and Dybvig, P.H. (1983). ‘Bank runs, deposit insurance, and liquidity’. Journal of Political Economy 91: 401–​419. Dixon, A.D. (2010). ‘Variegated capitalism and the geography of finance: towards a common agenda’. Progress in Human Geography 35: 193–​210. Dixon, A.D. (2012). ‘Function before form: macro-​institutional comparison and the geography of finance’. Journal of Economic Geography 12: 579–​600. Dixon, A.D. and Monk, A.H.B (2009). ‘The power of finance:  accounting harmonization’s effect on pension provision’. Journal of Economic Geography 9: 619–​639. Dymski, G. (1999). ‘Asset bubbles and Minsky crises in East Asia:  a spatialized Minsky approach’. Working Paper, Department of Economics, University of California, Riverside. Dymski, G. (2010). ‘Confronting the quadruple global crisis’. Geoforum 41: 837–​840. Dymski, G. (2011). ‘The International Debt Crisis’ in J. Michie (ed.) Handbook of Globalisation (2nd edition), pp. 117–​134 (Cheltenham: Edward Elgar). Dymski, G. and Shabani, M. (2017). ‘On the Geography of Financial Bubbles and Financial Crises’ in R. Martin and J. Pollard (eds) Handbook of the Geographies of Money and Finance, pp. 29–​51 (Cheltenham: Edward Elgar). Dymski, G. and Veitch, J. M (1996). ‘Financial transformation and the metropolis:  booms, busts, and banking in Los Angeles’. Environment and Planning A 28: 1233–​1260. Eaton, J., Gersovitz, M., and Stiglitz, J. (1986). ‘The pure theory of country risk’. European Economic Review 30: 481–​513. Engelen, E. and Faulconbridge, J. (2009). ‘Introduction:  financial geographies—​the credit crisis as an opportunity to catch economic geography’s next boat?’ Journal of Economic Geography 9: 587–​595. Engelen, E., Konings, M., and Fernandez, R. (2010). ‘Geographies of financialization in disarray: the Dutch case in comparative perspective’. Economic Geography 86: 53–​73. Epstein, G. (2006). ‘Introduction: Financialization and the World Economy’ in G. Epstein (ed.) Financialization and the World Economy, pp. 3–​29 (Cheltenham: Edward Elgar).

554   Dymski Fama, E. (1980). ‘Banking in the theory of finance’. Journal of Monetary Economics 6: 39–​57. French, S. and Leyshon, A. (2010). ‘ “These f@#king guys”: the terrible waste of a good crisis’. Environment and Planning A 42: 2549–​2559. French, S., Leyshon, A., and Thrift, N. (2009). ‘A very geographical crisis: the making and breaking of the 2007–​2008 financial crisis’. Cambridge Journal of Regions, Economy and Society 2: 287–​302. Galbraith, J.K. (2009). ‘Who are these economists, anyway?’ Thought and Action Fall: 85–​98. Gerschenkron, A. (1962). Economic Backwardness in Historical Perspective (Cambridge, MA: Belknap Press). Gurley, J.G. and Shaw, E.S. (1955). ‘Financial aspects of economic development’. American Economic Review 45: 515–​538. Hall, S. (2006). ‘What counts? Exploring the production of quantitative financial narratives in London’s corporate finance industry’, Journal of Economic Geography 6: 661–​678. Hall, S. (2013). ‘Geographies of money and finance III: financial circuits and the “real economy” ’. Progress in Human Geography 37: 285–​292. Hall, S. and Appleyard, L. (2009). ‘ “City of London, city of learning”? Placing business education within the geographies of finance’. Journal of Economic Geography 9: 597–​617. Harvey, D. (1973). Social Justice and the City (Baltimore, MD: Johns Hopkins University Press). Harvey, D. (1982). The Limits to Capital (Oxford: Basil Blackwell). Harvey, D. (1988). Social Justice and the City (Oxford: Basil Blackwell). Hendrikse, R.P. and Sidaway, J.D. (2014). ‘Financial wizardry and the Golden City: tracking the financial crisis through Pforzheim, Germany’. Transactions of the Institute of British Geographers 39: 195–​208. Jones, A. and Murphy, J.T. (2010). ‘Theorizing practice in economic geography: foundations, challenges, and possibilities’. Progress in Human Geography 35: 366–​392. Kane, E.J. (1990). The S&L Insurance Mess: How did it Happen? (Washington, DC: Urban Institute Press). Keen, S. (2015). ‘Post Keynesian theories of crisis’. American Journal of Economics and Sociology 74: 298–​324. Keynes, J.M. (1936). The General Theory of Employment, Interest, and Money (London: Macmillan). Krugman, P. (1998). ‘What happened to Asia?’ Working Paper, MIT Department of Economics. Kumhof, M., Laxton, D., Muir, D., and Mursula, S. (2010). ‘The global integrated monetary and fiscal model (GIMF)—​theoretical structure’. IMF Working Paper WP/​10/​34 (Washington, DC: International Monetary Fund). Laeven, L. and Valencia, F. (2012). ‘Systemic banking crises database: an update’. IMF Working Paper WP/​12/​163 (Washington, DC: International Monetary Fund). Lash, S. and Urry, J. (1987). The End of Organised Capitalism (Cambridge: Polity Press). Lash, S. and Urry, J. (1994). Economies of Signs and Space (New York: SAGE). Lee, R., Clark, G.L., Pollard, J., and Leyshon, A. (2009). ‘The remit of financial geography—​ before and after the crisis’. Journal of Economic Geography 9: 723–​747. Leijonhufvud, A. (2014). ‘Economics of the crisis and the crisis of economics’. European Journal of the History of Economic Thought 21: 760–​774. Leyshon, A. (1995). ‘Geographies of money and finance I’. Progress in Human Geography 19: 531–​543. Leyshon, A. (1997). ‘Geographies of money and finance II’. Progress in Human Geography 21: 381–​392.

Finance and Financial Systems    555 Leyshon, A. (1998). ‘Geographies of money and finance III’. Progress in Human Geography 22: 433–​446. Leyshon, A. (2008). ‘Book review of Gordon L. Clark and Dariusz Wójcik, “The geography of finance: corporate governance in the global marketplace” ’. Journal of Economic Geography 8: 262–​264. Leyshon, A. and Thrift, N. (1995). ‘Geographies of financial exclusion: financial abandonment in Britain and the United States’. Transactions of the Institute of British Geographers New Series, 20: 312–​341. Leyshon, A. and Thrift, N. (1996). ‘Financial exclusion and the shifting boundaries of the financial system’. Environment and Planning A 28: 1150–​1156. Leyshon, A. and Thrift, N. (1997). Money Space:  Geographies of Monetary Transformations (London: Routledge). Leyshon, A., French, S., and Signoretta, P. (2008). ‘Financial exclusion and the geography of bank and building society branch closure in Britain’. Transactions of the Institute of British Geographers New Series 33: 447–​465. Listokin, D. and Wyly, E.K. (2000). ‘Making new mortgage markets: case studies of institutions, home buyers, and communities’. Housing Policy Debate 11: 575–​643. Marshall, J., Pike, A., Pollard, J. Tomaney, J., Dawley, S., and Gray, J. (2011). ‘Placing the run on Northern Rock’. Journal of Economic Geography 11: 1–​25. Mason, C.M. and Harrison, R.T. (2002). ‘The geography of venture capital investments in the UK’. Transactions of the Institute of British Geographers New Series 27: 427–​451. Martin, R. (2011). ‘The local geographies of the financial crisis: from the housing bubble to economic recession and beyond’. Journal of Economic Geography 11: 587–​618. Martin, R., Pike, A., Tyler, P., and Gardiner, B. (2015). Spatially Rebalancing the UK Economy: The Need for a New Policy Model (London: Regional Studies Association). Minsky, H.P. (1975). John Maynard Keynes (New York: Columbia University Press). Minsky, H.P. (1982). Can ‘It’ Happen Again? Essays on Instability and Finance (Armonk, NY: M.E. Sharpe). Molotch, H.S. (1976). ‘The city as a growth machine:  toward a political economy of place’. American Journal of Sociology 82: 309–​332. Newman, K. and Wyly, E.K. (2004). ‘Geographies of mortgage market segmentation: the case of Essex County, New Jersey’. Housing Studies 19: 53–​83. O’Brien, R. (1991). Global Financial Integration: The End of Geography? (London: Pinter). Pani, E. and Holman, N. (2013). ‘A fetish and fiction of finance: unraveling the subprime crisis’. Economic Geography 90: 213–​235. Pollard, J. (1996). ‘Banking on the margins: geographies of financial exclusion in Los Angeles’. Environment and Planning A 28: 1209–​1232. Pollard, J. (2003). ‘Small firm finance and economic geography’. Journal of Economic Geography 3: 429–​452. Schroeder, S.K. (2002). ‘A Minskian analysis of financial crisis in developing countries’. CEPA Working Paper 2002-​09 (New York: Center for Economic Policy Analysis, New School for Social Research). Sokol, M. (2013). ‘Towards a “newer” economic geography? Injecting finance and financialisation into economic geographies’. Cambridge Journal of Regions, Economy and Society 6: 501–​515. Storper, M. (2011). ‘From retro to avant-​garde: a commentary on Paul Krugman’s “the new economic geography, now middle-​aged” ’. Regional Studies 45: 9–​15.

556   Dymski Taylor, P.J., Ni, P., Ben Derudder, B., Hoyler, M., Huang, J., Lu F., et al. (2009). ‘The way we were: command-​and-​control centres in the global space-​economy on the eve of the 2008 geo-​economic transition’. Environment and Planning A 41: 7–​12. Thrift, N.J. (1994). ‘On the Social and Cultural Determinants of International Financial Centres’ in S. Corbridge, R.L. Martin, and N. Thrift (eds) Money, Power and Space, pp. 327–​ 355 (London: Routledge). Wainwright, T. (2009a). ‘Laying the foundations for a crisis:  mapping the historico-​ geographical construction of residential mortgage backed securitization in the UK’. International Journal of Urban and Regional Research 33: 372–​388. Wainwright, T. (2009b). ‘Tax doesn’t have to be taxing: London’s “onshore” finance industry and the fiscal spaces of a global crisis’. Environment and Planning A 43: 1287–​1304. Wainwright, T. (2011). ‘Tax doesn’t have to be taxing: London’s “onshore” finance industry and the fiscal spaces of a global crisis’. Environment and Planning A 43: 1287–​1304. Wainwright, T. (2013). ‘Finance’s outsiders? Networks, knowledge and power beyond the City’. Journal of Economic Geography 13: 1041–​1058. Williams, P. (1978). ‘Building societies and the inner city’. Transactions of the Institute of British Geographers New Series 3: 23–​34. Williams, R., Nesiba, R., and Diaz McConnell, E. (2006). ‘The changing face of inequality in home mortgage lending’. Social Problems 52: 181–​208. Wójcik, D. (2006). ‘Convergence in corporate governance: evidence from Europe and the challenge for economic geography’. Journal of Economic Geography 6: 639–​660. Wójcik, D. (2012). ‘Where governance fails: Advanced business services and the offshore world’. Progress in Human Geography 37: 330–​347. Wójcik, D. (2013). ‘The dark side of NY–​LON: financial centres and the global financial crisis’. Urban Studies 50: 2736–​2752. Wyly, E.K., Atia, M., Foxcroft, H., Hammel, D.J., and Phillips-​Watts, K. (2006). ‘American home: predatory mortgage capital and neighbourhood spaces of race and class exploitation in the United States’. Geografiska Annaler 88B: 105–​132. Zook, M.A. (2002). ‘Grounded capital: venture financing and the geography of the Internet industry, 1994–​2000’. Journal of Economic Geography 2: 151–​177. Zysman, J. (1983). Governments, Markets and Growth:  Financial Systems and the Politics of Industrial Change (Oxford: Martin Robertson).

Chapter 29

The Gl obal Fi na nc ia l Net work s Dariusz Wójcik The remit of financial geography Financial geography can be defined as the study of the spatiality of finance, guided by three basic questions. First comes the what question. What is the spatiality of finance? What spatial forms, structures, and manifestations does finance take? This question is about mapping finance in a broad sense of the word ‘mapping’. Second comes the why question. Why does the map of financial phenomena look the way it does? How can we explain the spatiality of finance? This question requires the study of the forces that shape the spatiality of finance, including the behaviour of agents, as well as the investigation of economic structures, institutions, and ideas. This implies that while finance is a fundamentally economic phenomenon, of most importance to economic geography, social, political, and cultural geography are also indispensable for financial geography. Rational and irrational behaviour, property laws, financial regulation, trust, and confidence are just a few examples that make clear how central social, political, and cultural geographies are for understanding finance. It should be stressed that financial geography is more than a section of economic geography with elements of social, political, and cultural geography. It is about informing geography with understanding of finance. This ambition is best justified with the so what question of financial geography. What are the consequences of the spatiality of finance for the economy, society, and environment? As a type of capital and incentive, finance is fundamental to economic growth and innovation. Its distribution as capital and store of value is key to inequality, and its circulation is decisive for crises and stability. Finally, relationships between financial, natural, and social capital are crucial to environmental and social sustainability. Beyond the contours of geography as a discipline, financial geography is also about informing economics and other social sciences with understanding geography. If financial economics, as a field of economics most directly concerned with finance, dealt with the spatiality of finance effectively, we would not need financial geography. In reality, however, mainstream economics, to which financial economics very much belongs, fails to consider space in finance. According to mainstream economics, financial systems channel funds

558   Wójcik effectively and efficiently from utility-​maximizing savers to borrowers, using costless and symmetric information, and engaging in frictionless transactions in the conditions of perfect information. As a result, funds simply flow to best projects regardless of location, rendering the spatial structure of a financial system irrelevant (Mishkin, 2006). In a much more realistic, but difficult-​to-​model, world of economic and financial geography, savers and borrowers act in conditions of uncertainty and imperfect competition, using costly and asymmetric information (with lenders knowing less than borrowers about the projects for which money is borrowed), and incurring high transaction costs. What obtains is a financial system with an ‘over-​accumulation of credit (debt) and investment in one period, sector of the economy, or geographical location, and under-​accumulation in others’ (Klagge and Martin, 2005, p. 394). In its critique of mainstream economics and financial economics, financial geography can find allies in heterodox economics, including the behavioural, institutional, Marxian, and post-​Keynesian strands, political economy of finance, sociology and anthropology of finance, legal studies of finance, and—​not to be forgotten—​the history of finance. Within financial geography such defined, this chapter focuses on financial centres and networks as the key spatial manifestations of finance. Just as pyramids are the symbols of many ancient civilizations, and cathedrals are associated with medieval Europe, the skyscrapers decorating the skylines of largest cities are arguably the main symbol of the twentieth-​and early-​twenty-​first-​century capitalism. The main tenants and often owners of the skyscrapers are financial and related firms piled on the top of each other in a quintessentially geographical pursuit of maximizing their centrality, an economy-​wide outlook, and proximity to the commanding heights of the economy. Financial centres will hereby be defined simply as concentrations of financial organizations and their activities. Financial organizations consist of financial firms and public financial institutions, such as central banks and regulatory agencies. Our definition of financial firms will, however, be extended beyond financial services firms to include business services (services firms serving other firms) in accountancy, law, business consultancy, recruitment services, and information technology (IT) services related to financial services (Dicken, 2011). As such we will refer to them as financial and business services (FABS). Financial networks will be defined as networks of FABS activity linking financial centres, offshore jurisdictions, and the rest of the world. This chapter is about financial centres, relationships between them, their place in financial networks, and the position of financial networks in the global economy. It is about what financial centres and networks are, why they exist, and what implications they have for global finance and economy.

Placing Finance in Economic Geography Let us first outline the broad structure of finance, focused on different types of financial products or instruments, and markets (Figure 29.1). To start the list of financial products, there are basic assets, either tangible, such as cash, real estate, plant, and machinery, or intangible, such as the value of patents or brands. Loans represent more sophisticated assets, as they give the lender the right to a stream of payments from the borrower. Securities are yet more sophisticated as they give the owner the right to a stream of payments linked to an underlying asset or assets. In case of equity securities (shares), they offer a share of profit

The Global Financial Networks    559

Trading

Derivatives Securities Securities Loans Basic Assets

Financial Vehicles

Figure 29.1   The Structure of Finance. Source: author.

from a company. In case of debt securities such as bonds, they offer a predetermined stream of payments from the general revenues of the borrower (issuer), whether these are corporate revenues or tax revenues of a government. Securities can be created from basic assets, loans, or other securities. Wherever there is a more or less predictable stream of payments, it can be securitized (Leyshon and Thrift, 2007). It can be a bond based on royalties predicted from copyrights to music. It may be a set of mortgage loans, repackaged into a mortgage-​backed security. Or it could be a set of securities backed by mortgages and other types of loans repackaged into a collateralized debt obligation. As we move from basic assets, through loans to layers of securities, the level of financial abstraction and financial engineering grows. The highest level of innovation is achieved with derivatives, such as futures, options, and swaps, which entitle the owner to payments depending on the future value of underlying assets or securities. These can be used to hedge production or investment activities or speculate. Just as the ‘genius’ of securities is that one can share in the gains of the issuing company, without owning a share of each asset of the company, the ‘genius’ of derivatives is that one can benefit from the movement of prices in the underlying financial instruments without owning them. After they are first created (issued) in primary markets, financial instruments can be traded in secondary markets. The tradability differs from instrument to instrument. Gold and foreign exchange, for example, are easier to buy or sell than a house. The invention of securities and derivatives has a lot to do with the demand to make financial instruments easier to trade (make them more liquid). Equities mean that you can trade rights to corporate profits without trading actual assets of the company. Derivatives can also be traded without trading the underlying assets. After the values of financial instruments are first set in primary markets, they can change constantly in secondary markets. The third building block of the world of finance that complements primary and secondary markets are financial vehicles—​containers that hold assemblages of financial instruments. These include companies themselves, containing tangible and intangible assets to produce goods and services, holding companies that hold the controlling equity stakes in other companies, and investment funds and trusts, which hold and trade financial instruments to produce returns for their beneficiaries. Financial vehicles have to be registered somewhere and be subject to laws and regulations of a jurisdiction. In addition to assets, which they own

560   Wójcik and in which they invest, they have to have clearly identified sources of funding, referred to as liabilities. To be sure, a liability of one vehicle can be an asset of another. For example, stocks or bonds issued by one company, when sold, become assets of another company or an investment fund. As such, financial vehicles are financial, legal, and accounting abstractions superimposed on each other (Wójcik, 2013a). The three-​part structure of finance can be translated into a map (Figure 29.2). Although not distributed evenly, assets, including real estate, potential borrowers (individuals and companies), and potential issuers of securities (companies and governments), are scattered all over the world. In contrast to relatively dispersed primary markets, secondary markets concentrate in financial centres. Proximity to each other gives financial firms in these centres better access to non-​standardized, tacit knowledge relevant to trading. If buy-​and-​sell orders are matched or even generated by computers, professionals creating orders and programming computers to generate them still need access to such information and benefit from proximity to other professionals in secondary markets. Their concentration spreads the costs of trading infrastructure, such as high-​speed fibre-​optic communication networks, and contributes to a large and deep pool of labour, attractive to both finance professionals and their employers. Being centres of financial expertise, financial centres are also centres of financial innovation, creating new types of securities, derivatives, and other financial instruments. As information relevant to secondary markets is also highly relevant to primary markets, and large financial firms typically provide services in both types of markets, financial centres are also key centres for the management of primary market transactions. Financial vehicles have a distinctive geography. As legal and accounting abstractions, they can be registered in a place different than the place of the sponsor—​an individual or a company that creates a financial vehicle. We will refer to these places as offshore jurisdictions. Registration in an offshore jurisdiction takes place because the sponsor wants the financial vehicle, its activities, and proceeds to be subject to different laws, including tax laws, and regulations than those prevalent in the jurisdiction where the sponsor itself resides or is registered. As an example, consider a company operating in Central and Eastern Europe but owned by a holding company registered in the Netherlands or Switzerland; a hedge fund operating in London or New York, registered in the Cayman Islands; or a trust containing the wealth of a family living in London, registered in the Channel Islands. To be sure, not all financial vehicles are registered offshore. Millions of existing offshore companies and investment funds, however, warrant offshore jurisdictions a place on the map of finance. We use

Financial Centres Financial and Business Services The World

Offshore jurisdictions

Figure 29.2   The Map of Finance. Source: author.

The Global Financial Networks    561 the term offshore jurisdiction rather than tax haven or offshore financial centre. These places are not only about tax, but also about other laws and regulations that offer financial vehicles more flexibility in international operations or secrecy (Palan et al., 2010). In contrast to financial centres, their attraction lies in jurisdictional autonomy, rather than in being centres of finance professionals or expertise. While jurisdiction comes with state sovereignty, the latter is not necessary for an offshore jurisdiction. The state of Delaware, for example, can function as an offshore jurisdiction within the USA, attracting thousands of business incorporations, because US states have a high degree of autonomy in corporate law. Hong Kong can function as an offshore jurisdiction within China, owing to its separate legal system. Hong Kong also provides an example that a place can serve both as a financial centre and an offshore jurisdiction, as these categories are not mutually exclusive. What joins financial centres, offshore jurisdictions, and the rest of the world together are FABS. With financial services firms in the lead, this sector includes law and accounting firms. These are necessary to operate financial vehicles as legal and accounting constructs, as well as to design and manage contracts underpinning financial transactions. Relevant business services also include management consultancies, recruitment agencies, and IT firms. Financial transactions such as mergers and acquisitions involve corporate restructuring. Financial services, particularly at the high-​finance, wholesale end, have high employee turnover, and are some of the savviest users of information and communication technology. The complexity of the financial sector and its activities thus generate demand for related business services. FABS firms, headquartered mainly in financial centres, link financial centres with the rest of the world through their networks, reaching places that offer opportunities to manage financial assets, grant loans, and conduct securitization of existing assets and loans (Taylor, 2004). Crucially, FABS also link financial centres and the rest of the world with offshore jurisdictions. For example, a wealthy individual from Texas or Russia keen on minimizing their tax obligations is not going to choose a financial vehicle and offshore jurisdiction by visiting a number of Caribbean or other island states. Rather, they would approach one or more financial, legal or accounting firms and choose from a menu of options presented by them. To develop such options and service clients, large international FABS firms often maintain branches in offshore jurisdictions. The next step in our framework is to fit this conceptual map of finance into a map of the global economy (Figure 29.3). To conceptualize the map of the world economy, we will use the global production networks (GPN) approach, according to which regional development is the outcome of the strategic coupling process between transnational corporations (TNCs) and territories (Henderson et al., 2002). The approach relies on the traditional focus of economic geography on the firm–​territory nexus (Dicken and Malmberg, 2001), but it puts regional development in a truly global context, recognizing the power of TNCs and their supply chains and networks. To be sure, TNCs and territories interact through a multitude of formal and informal institutions operating at international (e.g. World Trade Organization), national (e.g. national laws and regulations), and local levels (including business customs). The firm–​territory nexus is key to establishing the correspondence between the map of world finance and the GPNs. FABS are obviously on the firm side (Coe et al., 2014). They are among the world’s most globalized and networked companies, and perform important functions intermediating between other TNCs and territories. In order to expand internationally and manage their supply chains, TNCs hire FABS firms to help them raise capital, understand foreign laws, regulations, and markets, as well as hire foreign labour. Governments

562   Wójcik

Transnational Companies

FIRM

Global Production Networks

Financial and Business Services Financial Centres

TERRITORY

Offshore Jurisdictions Territories

Institutionally Mediated Interface

Global Financial Networks

Figure 29.3   Finance in the Global Economy. Source: author, based on Coe, Lai, and Wójcik (2014) published by Taylor & Francis Ltd. www.tandfonline.com.

representing territories also hire FABS to help them attract TNCs directly or create conditions making their territories attractive to TNCs. FABS influence the institutional interface between TNCs and territories, as they provide experts to shape international accounting rules, trade and investment treaties, and so on. Offering knowledge, FABS affect the very way that TNCs and governments think about the world economy. Consider that the concepts of BRICS (Brazil, Russia, India, China, and South Africa), emerging economies, shareholder value, and value at risk were all invented and/​or popularized by FABS (Wójcik, 2012). Financial centres and offshore jurisdictions represent special types of places. What distinguishes them from the rest of the world is their centrality to the world of finance. By consequence, they may have more in common with each other than with other places with which they coexist as part of nation states. If we scan the map of the world, we find that the wealthiest economies in terms of gross domestic product per capita, other than resource-​rich countries, include Switzerland, Ireland, Luxembourg, Liechtenstein, Bermuda, Singapore, and Hong Kong, which clearly built their fortunes to a large extent as financial centres and/​or offshore jurisdictions. In analogy with GPN, we could refer to the conceptual map of finance, as global financial networks (GFN), defined as networks of FABS activity linking financial centres, offshore jurisdictions, and the rest of the world. Put together, GPNs and GFNs offer a schematic to understand the world transformed by arguably two of the most important economic phenomena of the last decades: globalization and financialization. Indeed, in the light of the framework the latter could be described as the growing significance of FABS, financial centres, and offshore jurisdictions in the world economy. While, the framework is simplified, two final examples should make clear how central GFNs are to GPNs and understanding the map of the world economy. Firstly, a study of over 43,000 companies operating in more than one country showed that nearly 40 per cent of the value of these companies is controlled by a super-​core of 147 companies, which have almost complete control over themselves, and three-​quarters of which are financial companies (Vitali et al., 2011). Another series of studies demonstrates that 30 to 50 per cent of the global stock of foreign direct investment (FDI)

The Global Financial Networks    563 is found in offshore jurisdictions. Moreover, offshore FDI is as important to emerging and developing economies as it is for advanced economies (Haberly and Wójcik, 2014).

The Global Network of Securities Centres FABS are all at the heart of the GFNs, but different types of FABS play slightly different roles. While financial services provide both expertise and capital (or intermediate in its provision), business services provide only expertise. Financial services firms are typically much larger than firms in any type of business services. The securities industry arguably represents the elite of financial services. The industry is involved in the production, distribution, and circulation of securities and derivatives, and consists of the sell side, serving the issuers of securities (companies and governments), and buy side, serving the buyers of securities (individual and institutional investors). The sell side is sometimes referred to as investment banking, and the buy side as asset management. Institutionally, the industry is made of securities departments or subsidiaries of universal banks (e.g. Citigroup or Deutsche Bank) or specialized securities firms (e.g. Goldman Sachs or Morgan Stanley, hedge funds, or investment consulting firms). Serving corporations, governments, and investors, securities firms are privy to the most valuable information in capital markets (Morrison and Wilhelm, 2007). Their elite character is reflected in salaries and bonuses much higher than in any other financial or business services, job titles (with vice presidents, senior vice presidents, directors, managing directors, etc.), and prestigious office locations. This section will focus on the securities industry, its geography (the what question), as well as the mechanisms explaining this geography (why) and its implications (so what), in a series of five propositions (Wójcik, 2011). The first proposition is that the process of securitization, understood as the growth in the volume and value of securities, involves an institutional and geographical dispersion of asset (including corporate) ownership. The objective of a company issuing stocks or bonds, or a bank converting loans into securities, is to raise as much capital from investors as possible, which requires selling securities to a large number of investors. As there are limits to how many investors can be found in any one place, institutional dispersion has to be complemented with geographical dispersion, with securities sold in as many places as possible. This spatial dispersion or expansion is thus in the ‘DNA’ of securities markets. Take Dutch East India Company, the world’s first company that issued shares, as an example. The Company was chartered in 1602, and its start-​up capital was collected in Amsterdam and five other port cities of today’s Netherlands. The Company listed on the Amsterdam Stock Exchange, raised more capital, and extended its ownership basis to over 1000 shareholder over the next twenty years, including investors from today’s Belgium and Luxembourg, as well as many immigrants from other countries (de Vries and van der Woude, 1997). Securitization implies not only the growth of the securities markets, but also the securities industry that serves them. Firstly, there is a growing demand for ‘match-​making’, with intermediaries connecting an increasing number of investors from an increasing number of places with issuers (in primary markets), and with each other (in secondary markets). Secondly, there is growing demand for intermediaries addressing information asymmetry and agency problems amplified by institutional and geographical dispersion of asset ownership. Recall that information asymmetry means that borrowers or users of capital know

564   Wójcik better how they are going to use the capital than lenders or investors. Agency problem involves the risk that the user of capital (e.g. the manager of a company) will use funding against the interests of investors, for example by hiring inefficient subcontractors related to the manager. Information asymmetry and agency problems apply to small and remote investors in the most acute form as such investors have limited means and face high costs of gathering information and undertaking action to mitigate these risks. This implies that as securitization unfolds and with it the institutional and geographical dispersion of ownership, information and agency problems grow as well, and so does demand for an industry to deal with them, including equity analysts and investment advisors. In a typical fashion for financial services, the securities industry also generates work for business services, including securities lawyers, audit and due diligence services, and corporate governance rating agencies and consultancies. In line with galloping securitization, the securities industry has experienced a huge boom in the last few decades. Its share (excluding related business services) in total US employment grew from 0.3 per cent in 1978 to 0.8 per cent in 2008, and its slice of total payroll in the same period increased from 0.5 per cent to 3.6 per cent. In France, Germany, and the UK, employment in the securities industry grew by over 50 per cent between 1998 and 2008, incomparably more than in other parts of the financial sector (Wójcik, 2012). The third and fourth propositions can be considered in tandem. The third is that securities firms benefit from co-​location, forming securities centres. The fourth is that these securities centres are also important concentrations of corporate headquarters and investors. Proximity among securities firms, issuers, and investors matters both in primary and secondary markets. Take an initial public offering as an example of a primary market transaction. An issuer needs a trustworthy investment bank to price and underwrite its securities. The bank needs to engage closely with both the issuer to be able to judge its business prospects and key potential investors to judge the demand for the issue. Investment banks often form syndicates to promote an initial public offering. Proximity helps all these interactions. For secondary markets, consider the case of HSBC, with its shares listed and traded on exchanges in Hong Kong, London, and New York. Despite the fact that HSBC operates in all time zones, and has over 200 thousand shareholder in more than 100 countries, price discovery for the HSBC shares is concentrated in London, the seat of the bank’s headquarters, home to most of its decision makers and key business relationships. This is where the most strategic and sensitive information about HSBC is available from, with markets in Hong Kong and New York mostly just following the price movements on the London market (Wójcik, 2011). Co-​location of securities firms, and corporate and investors’ headquarters also contributes to and is affected by access to specialized labour market for finance and business professionals, and access to specialized infrastructure. In its pursuit of proximity, and agglomeration benefits, securities industry is much more concentrated spatially than other parts of the financial sector. New York MSA, for example, accounted for over 25 per cent of US employment in the securities industry in 2008, compared with approximately 7 per cent in credit and insurance. In the same year, the securities industry accounted for over 34 per cent of total payroll in Manhattan, compared with a 12 per cent share of credit, insurance, and real estate. In 1978 the proportions were 6 to 16 per cent. We might even say that Manhattan has evolved from a general financial centre to a securities centre (Wójcik, 2012).

The Global Financial Networks    565 The fifth and final proposition is that while issuer-​and investor-​specific knowledge is important to securities firms, so is global knowledge on exchange and interest rates, and macroeconomic and sector-​wide trends. The necessity of combining local knowledge from different places with global knowledge pushes securities firms to create extensive international branch networks. As for any other FABS, the creation of these networks is also related to the fact that their main customers operate globally, and need global service providers (Sassen, 1991). Morgan Stanley, a leading securities firm, with focus on investment banking, is a good example of such a global network (Table 29.1). Morgan Stanley has over fifty offices, with global headquarters in New York, and regional ones in London and Hong Kong, the main financial trading hubs of the three major time zones. Its presence in Menlo Park (Silicon Valley) and Tel Aviv is likely to be driven by access to customers in primary markets, as these places are concentrations of high-​technology firms and venture capital. Its presence in Baltimore and Boston may be related to the presence of large institutional investors (such as T. Rowe Price and State Street, respectively). Its presence in Amsterdam, Dublin, Hong Kong, and Singapore is also influenced by the feature of these places as offshore jurisdictions,

Table 29.1 The Global Network of Morgan Stanley Americas

Europe, Middle East, Africa

Asia-​Pacific

New York, NY Atlanta, GA Baltimore, MD Boston, MA Boca Raton, FL Chicago, IL Houston, TX Los Angeles, CA Menlo Park, CA Miami, FL Purchase, NY San Francisco, CA Washington, DC West Conshohocken, PA Buenos Aires (Argentina) Calgary (Canada) Lima (Peru) Mexico City (Mexico) Montreal (Canada) São Paulo (Brazil) Vancouver (Canada) Toronto (Canada)

London (UK) Amsterdam (the Netherlands) Budapest (Hungary) Dubai (United Arab Emirates) Dublin (Ireland) Frankfurt (Germany) Glasgow (UK) Istanbul (Turkey) Johannesburg (South Africa) Madrid (Spain) Milan (Italy) Moscow (Russia) Paris (France) Riyadh (Saudi Arabia) Stockholm (Sweden) Tel Aviv (Israel) Warsaw (Poland) Zürich (Switzerland)

Hong Kong (China) Beijing (China) Bangalore (India) Bangkok (Thailand) Hanoi (Vietnam) Jakarta (Indonesia) Melbourne (Australia) Mumbai (India) Labuan (Malaysia) Manila (Philippines) Seoul (South Korea) Shanghai (China) Singapore (Singapore) Sydney (Australia) Taipei (Taiwan) Tokyo (Japan)

Source: Author, based on information from morganstanley.com as of 3 August 2016.

566   Wójcik attracting the establishment of financial vehicles, which need the services of securities firms and related business services. Put together, firms like Morgan Stanley, through their networks of offices, and both intra-​and inter-​firm transactions link securities centres, giving rise to a global network of securities centres. This network can be seen as a type or a part of the GFNs, one that is at the centre of the GFNs owing to its privileged access to the most sensitive and strategic financial information. Similar networks could be drawn for credit banking, insurance, accounting, and other types of FABS. The borders between these networks are fuzzy owing to the very nature of FABS. Many of the largest banks perform both credit and securities industry functions. Big Four companies engage in accounting and legal services (including tax advisory), as well as securities transactions. Initial public offerings and mergers and acquisitions and many other transactions involve securities firms, accountancy, and law and management consultancy firms, cooperating with each other on the same deals. This complex, interrelated nature of FABS justifies their treatment as a group underpinning financial centres.

The Global Financial Networks in History While we have outlined the development of the global network of securities centres, we need to elaborate on the evolution of the GFNs as a whole. One way to analyse the dynamics of the GFN is to chart the history of financial instruments and markets, from which the GFN framework is derived (Figure 29.1). While coins and trade credit have existed since ancient times, bank loans and the first securities in the form of government bonds, first emerged in Italy in the thirteenth/​fourteenth centuries (Kindleberger, 1984). These financial innovations coincided in time and place with accounting and legal inventions of double-​entry bookkeeping and limited liability (Braithwaite and Drahos, 2000), which, in turn, laid the foundations for the development of joint stock companies as financial vehicles. The Dutch East India Company, as the archetypal early joint stock company, was instrumental in developing the Amsterdam Stock Exchange. While simple derivatives, like forward and futures contracts on physical commodities, were traded already in seventeenth-​century Amsterdam, the first derivatives exchanges were created in the US in the nineteenth century, and the first derivatives on financial instruments such as interest rates were invented in the 1970s (Bernstein, 2007). Further advancements in the use of financial vehicles in the form of trusts and holdings companies, which enabled new forms of investment management and industrial consolidation, came in the nineteenth century (Geisst, 1997). The key message from charting the evolution of financial instruments is that it involves the history of financial, as well as legal and accounting, innovation, justifying the focus of the GFN framework on financial, legal, and accounting services. Another way to approach the history of the GFN is by focusing on its spatial elements (Figure 29.2). With regard to financial centres we can think about the well-​documented succession from Amsterdam, as the world’s leading centre in the seventeenth and eighteenth centuries, to London, taking over in the nineteenth century, and challenged by New York in the twentieth century (Arrighi, 1994; Cassis, 2006). Moving to FABS, we can consider the succession of securities firms as leading financial firms of their times, from Hope & Co. merchant bank serving the Dutch East India Company, to Barings and Rothschild, as the

The Global Financial Networks    567 most powerful merchant banks of nineteenth-​century London, through JP Morgan dominant in the late nineteenth and early twentieth centuries in New York, all the way to Morgan Stanley and Goldman Sachs, leading investment banks of the mid-​to late twentieth and early twenty-​first centuries. To be sure, the fortunes of financial centres and leading firms are closely interrelated. Hope & Co.’s success was built partly on its expansion in London, JP Morgan had its roots in London’s JS Morgan, while Rothchild’s inability to develop US operations brought an end to its dominance in high finance (Ferguson, 1998). Offshore jurisdictions have a shorter history, as they developed mainly in response to stricter financial regulation introduced in the twentieth century, creating incentives for sponsors to place financial vehicles beyond the reach of their home country’s laws and regulations. The absolute majority of offshore jurisdictions emerged in former British colonies, with Bermuda, the Cayman Islands, and the British Virgin Islands in the lead, offering British and UK-​based sponsors a common language and law combined with low levels of taxation and regulation, and high levels of secrecy, giving rise to what some referred to as ‘the second British empire’ (Palan et al., 2010). London itself became the coordinating centre of offshore finance, complemented by its functions as the centre of Eurodollar markets (Strange, 1986; Shaxson, 2011). The US-​sponsored offshore network imitated the British one, focusing on the Caribbean islands, and former US colonies like Panama. A new offshore sub-​ network has developed since the 1990s, serving Eastern European sponsors, with Cyprus as the primary offshore jurisdiction (Haberly and Wójcik, 2014). Finally, the youngest sub-​network, serving mainly Chinese sponsors, emerged in the 2000s, with Hong Kong, Singapore, the Cayman Islands, and the British Virgin Islands as the leading offshore jurisdictions. Notwithstanding the role of former British colonies, London, common law, and the English language, it should be noted that leading offshore jurisdictions connecting the British, US, Eastern European, and Chinese networks include Luxembourg, Switzerland, and first of all the Netherlands (alongside the UK and the British Virgin Islands). In a perverse way, the action of Organisation for Economic Co-​operation and Development (OECD) and other organizations against tax havens and money laundering since the late 1990s has reinforced the position of offshore jurisdictions that are OECD members, while undermining the pos­ ition of smaller and weaker jurisdictions (Sharman, 2006). What the histories of financial centres, FABS, and offshore jurisdictions have in common is growing interconnectivity. Arguably, today’s landscape of financial centres can be characterized not by the dominance of a single financial centre, but rather the centrality of the New York–​London axis, itself part of a dense network of financial centres (Wójcik, 2013b). The corporate supercore made of financial firms controlling a big chunk of all TNCs in the world demonstrates the complex network of FABS and GPNs. Recent research on offshore FDI also presents a very dense network of connections (Haberly and Wójcik, 2014, 2015). Overall, the history of the GFN is the history of leading financial firms, centres, and offshore jurisdictions co-​evolving in a changing context of the global economy. To place the evolution of the GFNs in a broader political and economic context, let us focus on the shift from Fordism to flexible accumulation of the last half century, when contemporary GFNs have become fully formed. Fordism describes a regime of accumulation in the period from the end of World War II to the mid-​1970s, characterized by mass production of standardized goods dominated by large, vertically integrated companies, exploiting internal economies of scale. The key to the regime’s maintenance was a compromise between workers and corporate owners, manifested in collective wage bargaining, and underpinned

568   Wójcik by welfare state, following the Keynesian economic principles of full employment and control over cross-​border financial flows (Glyn et al., 1991; Webber and Rigby, 1996). Put simply, Fordism can be characterized as a system of social relations featuring big government (with large public sector and significant state ownership of industry), big labour (with strong unions), and big business, sheltered to a large extent from disruptive external influence by a supranational framework aimed at economic stability through regulation of exchange rates, international trade, financial markets, and development assistance (Figure 29.4a). Elements of the new regime of ‘flexible accumulation’ emerged in the 1970s in response to the crisis of Fordism and its rigidities, including the power of trade unions, the oligopolistic corporate sector, and large public sector, which together exacerbated the problems of the high inflation (a wage price spiral) of the 1970s, and contributed to relatively low levels of innovation (Galbraith, 1975; Gertler, 1988; Harvey, 1988). The transition to the new regime was accompanied by a shift in the focus of domestic macroeconomic policy from full employment and demand management to monetary policies (low inflation even at the expense of higher unemployment); the downsizing of the public sector through privatization; and increasing emphasis on competitiveness both domestically and internationally. In industry, focus on internal economies has been replaced with that on external economies, involving a re-​agglomeration of production in selected areas (shifts from Rust to Sunbelts), with active evasion of labour pools dominated by the Fordist industry (Scott, 1988). In short, with the shift to flexible accumulation, ‘big’ business, labour, and government have all become more flexible (Figure 29.4b). While the history of post-​Fordism has been hotly disputed (Tickell and Peck, 1992), little attention has been paid the role of FABS in helping business, labour and government, (a)

Big Labour

Big Government

Big Business

(b)

Flexible Labour

Financial and Business Services Flexible Business

Flexible Government

Figure 29.4   (a) Fordism and (b) Flexible Accumulation. Source: author.

The Global Financial Networks    569 25% 20% 15% 10% 5%

1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014

0%

UK

USA

Japan

Brazil

China

India

Figure 29.5   The Share of Financial and Business Services in Total Employment. Source: author based on data from Oxford Economics.

and the whole economy become more flexible and overcome the rigidities of Fordism. To corporations FABS offered an option to outsource central coordinating functions to obtain specialist advice on business opportunities and threats in a globalizing world (Sassen, 1991; Massey, 1995). A disciplining impact of FABS on corporations came with the growing role of capital markets, putting companies across different industries and increasingly across countries in direct competition for capital (Clark and Wójcik, 2007). In relation to labour, FABS not only offered new opportunities to match highly skilled labour with employers, but also a chance for managers to outsource the burden of downsizing labour. In relation to government FABS promised a flexible and competitive way of evaluating economic performance, thus contributing to the perceived viability of self-​regulation as a substitute to regulation by government. Technology has also been a key factor elevating the role of FABS. As Wood (1991, p. 163) argued, ‘The implementation of new technologies, and of new forms of business organization, supporting what Harvey characterises as “time-​space compression”, have placed an even greater premium on expertise. This is increasingly being made available through independent organizations, in the business services sector’. In sum, FABS have been the core of technocracy in a globalizing world economy (Strange, 1986). Their growth, particularly since the 1970s, has been explosive (Figure 29.5).

Beyond the Perfect Storm Since 2008 the global financial system has experienced a perfect storm. Many financial markets have shrunk, including those for initial public offerings and securitized debt. Banks have suffered from a bad reputation, following their contribution to the subprime crisis, and investigations into market abuses, including assistance with tax evasion and money laundering, and manipulation of interest rates and foreign exchange rates. In response to the crisis,

570   Wójcik governments have been rolling out new financial regulations at a rate unseen in history, multiplying compliance costs for financial firms. At the same time, the march of new financial technologies continues, offering new forms of electronic trading, and platforms connecting borrowers and lenders directly, bypassing financial intermediaries. These forces challenge the established business models of financial firms, the global financial governance, and all elements of the GFNs. The financial services sector has lost a large number of employees in advanced economies. In the USA and the UK, the losses between 2008 and 2013 may have reached as much as 10 per cent (Wójcik and MacDonald-​Korth, 2015). Other business services have suffered less, if at all, as they are much less cyclical than finance. Accounting services are necessary in good and bad times, while demand for corporate legal services even grows with pressure to comply with mounting regulation and lawsuits. Demand for new technologies that would help financial firms reduce labour costs, has also remained buoyant. New forms of electronic or digital finance, including electronic trading, crowd-​funding, and mobile banking, imply a potential sea change, particularly in retail finance (O’Brien and Keith, 2009). This means that financial IT services are becoming more central to finance, and the whole FABS complex. It also means that IT giants like Google, Amazon, or Alibaba could expand their retail financial services, changing the structure and hierarchy of FABS. In wholesale finance, where large customers (issuers and investors) and large customized transactions, with a lot of strategic and sensitive information, are involved, technology may be less likely to cause a revolution. As a consequence, securities firms are likely to remain the elite of FABS in the near future. A shift, however, might take place within the securities industry, as investment banks, directly associated with the subprime crisis, and implicated in recent scandals, find themselves under much more regulatory pressure than asset managers. The pressures include limits on proprietary trading, higher capital requirements, separation from retail banking, and restrictions on bonuses, just to name a few. Asset management has thus far attracted much less regulatory scrutiny, has kept growing in terms of employment and revenues, and has a much better reputation, legitimized inter alia with its mission to help governments and individuals manage longevity risk. Symbolic and symptomatic in this respect is the demise of the chief executive officers (CEOs) of investment banks, and the rise of Larry Fink, the CEO of Blackrock, the world’s largest asset manager, with nearly US$5 trillion assets under management, as one of the most celebrated CEOs in the world. We might even speculate whether Blackrock is not already overtaking Goldman Sachs, as the world’s most influential financial company. Trends within FABS may translate into a changing landscape of financial centres. The rise of new financial technologies (fintech) raises questions about the location of firms carrying out these activities. In the last decade eastern New Jersey, west of Manhattan, has risen as a centre of IT infrastructure generating and executing orders in financial markets. San Francisco and Silicon Valley are a major source of fintech coming from both small and big firms. In Europe, London is the major centre of fintech, and Hong Kong makes efforts to attract the industry. India might also be seen as a prospective contestant in fintech, with leading firms such as Goldman Sachs and Blackrock already employing thousands of people in IT operations in Bangalore and Gurgaon, respectively. The shift from sell side to buy side may privilege financial centres specializing in the latter, including Boston and Munich. While new technologies make financial transactions cheaper, new regulation tends to make them more expensive. This applies particularly to cross-​border transactions, leading

The Global Financial Networks    571 to claims that the world is experiencing financial de-​globalization or balkanization (The Economist, 2014). With many financial firms focusing increasingly on domestic markets (see Royal Bank of Scotland, Barclays, ING), this raises the question about the concentration of financial activities at different scales. Early research into this issue indicates that foreign exchange trading, for example, has witnessed an acceleration of concentration in London and New York, driven mainly by growing institutional concentration of this activity in a handful of largest banks (Wójcik et al., 2017). At a national level, in the UK we have seen an unprecedented concentration of financial employment in London, at the expense of provincial financial centres, influenced by industry consolidation, and regulation creating thousands of compliance jobs in London. In contrast, Germany, with a large local and regional banking sector, experienced no significant change in the spatial distribution of financial employment (Wójcik and MacDonald-​Korth, 2015). The rise of Asian financial centres is also commonly expected, led among others by Japanese banks capitalizing on growth in Asian capital markets (Wójcik et al., 2016). Western FABS, and particularly US financial firms, however, still dominate GFNs. Chinese state-​ owned banks may sit at the top of global rankings in terms of their asset size or revenues, but they are even further from building GFNs than Chinese national champion companies in other sectors are from building GPNs (Nolan, 2012). To be sure, the internationalization of the Chinese currency may yet prove to be a catalyst of a major transformation of the GFNs. Another major catalyst may be the outcome of the UK’s European Union (EU) referendum of June 2016. London’s success as a global financial centre in the last few decades has relied to a great extent on easy access to EU markets through the ‘single-​passport’ system. Once established in London, financial firms from around the world could operate throughout the EU with no or minimal additional requirements from other member states. If Brexit makes access to EU markets more difficult, some firms may leave London for Amsterdam, Dublin, Frankfurt, Luxembourg, Paris, or other cities. As competitors are many, none is likely to overtake London as the leading European financial centre, but collectively they may certainly weaken its position. Recent events also raise important questions about the position of offshore jurisdictions. With governments raising the level of regulation and taxation, the rewards for using offshore jurisdictions increase, but so does the legal and reputational risk of getting caught and/​or blamed for unethical behaviour. While this certainly generates more work for tax lawyers and advisers, a more cautious use of offshore jurisdictions is in order. This is bad news for the most obvious tax havens like the Cayman Islands and Panama, but may be good news for countries and financial centres less associated with their functions as offshore jurisdictions, such as the Netherlands, Ireland, the UK, Hong Kong, or Singapore. Witness, for example, the recent wave of corporate inversions, whereby US corporations merge with much smaller companies, mainly in the UK, Ireland, and the Netherlands in order to move their registered offices to lower tax jurisdictions. Related to offshore jurisdictions is a recent initiative driven by civil society organizations, including Oxfam, to oblige companies to publish financial statements for each country in which they operate, in addition to consolidated statements. This could help reveal how much revenue and profit corporations book in different locations in relation to real economic activities undertaken there, thus shedding light on potential cases of tax and regulatory arbitrage, evasion, and avoidance. As of 2016, country-​ by-​country reporting in a minimal form has been applied to extractive industry in the USA and in the EU, but the battle is on and country-​by-​country reporting may be deepened and

572   Wójcik extended to other sectors (Wójcik, 2015). The introduction of country-​by-​country reporting might serve as a goldmine in researching both GFNs and GPNs. In addition to regulation, the public financial sector is becoming more important in global finance, as central banks and sovereign wealth funds account for a growing share of financial stocks and flows, and many governments took ownership stakes in financial firms. Consider the doubling of the balance sheet of the Federal Reserve since 2007 to US$4 trillion. While the size of central banks and government ownership is likely to shrink if and when the crisis is over, this will take time and sovereign wealth funds are likely to stay and grow (Clark et al., 2013). This makes more dialogue between political and economic geography in the study of GFNs urgent, and might even call for geo-​finance, a macro-​approach that combines geopolitics with finance. At a more micro-​and meso-​level we need further inquiry into the operations of individual FABS firms, their networks, and their relationships with customers (investors, governments, non-​financial firms). It is argued that in the last decades the investment chains linking ultimate investors and ultimate users of capital have grown unnecessarily long and complex, exacerbating information asymmetry, agency problems, and conflicts of interests (Dixon and Monk, 2014). New technologies linking lenders and borrowers directly are seen in this light as a chance to reduce the ‘investment miles’ (Williams, 2011). Further on the meso-​level, we need to recognize that the development of FABS remains a major force in the urbanization of emerging and developing economies. Cities in China and beyond compete fiercely to host leading FABS firms, and the impact of this process on inequality between and within cities should be of most interest to future research. In this sense, the GFN concept is largely compatible and complementary with the World City and Global City research agendas. Overall, the ambition of the GFN framework is to offer a big-​picture map of global finance and fit this map into the map of the world economy. It is a geographically focused lens to view the past, present, and future of finance.

References Arrighi, G. (1994). The Long Twentieth Century: Money, Power, and the Origins of Our Times (London: Verso). Bernstein, P.L. (2007). Capital Ideas Evolving (Hoboken, NJ: John Wiley). Braithwaite, J. and Drahos, P. (2000). Global Business Regulation (Cambridge:  Cambridge University Press). Cassis, Y. (2006). Capitals of Capital: A History of International Financial Centres, 1780–​2005 (Cambridge: Cambridge University Press). Clark, G.L. and Wójcik, D. (2007). The Geography of Finance: Corporate Governance in a Global Marketplace (Oxford: Oxford University Press). Clark, G.L., Dixon, A.D., and Monk, A.H.B. (2013). Sovereign Wealth Funds:  Legitimacy, Governance, and Global Power (Princeton, NJ: Princeton University Press). Coe, N.M., Lai, K.P.Y., and Wójcik, D. (2014). ‘Integrating finance into global production networks’. Regional Studies 48: 761–​777. de Vries, J. and van der Woude, A. (1997). The First Modern Economy. Success, Failure, and Perseverance of the Dutch Economy, 1500–​1815 (Cambridge: Cambridge University Press). Dicken, P.J. (2011). Global shift:  Mapping the Changing Contours of the World Economy (London: SAGE).

The Global Financial Networks    573 Dicken, P. and Malmberg, A. (2001). ‘Firms in territories: a relational perspective’. Economic Geography 77: 345–​363. Dixon, A.D. and Monk, A.H.B. (2014). ‘Frontier finance’. Annals of the Association of American Geographers 104: 852–​868. Ferguson, N. (1998). The House of Rothschild: Money’s Prophets 1798–​1848 (New York: Penguin). Galbraith, J.K. (1975). Money: Whence it Came, Where It Went (London: Penguin). Geisst, C.R. (1997). Wall Street: A History (Oxford: Oxford University Press). Gertler, M.S. (1988). ‘The limits to flexibility: comments on the post-​Fordist vision of production and its geography’. Transactions of the Institute of British Geographers 13: 419–​432. Glyn, A., Hughes, A., Lipietz, A., and Singh, A. (1991). ‘The Rise and Fall of the Golden Age’ in S. Marglin and J.B. Schor (eds) The Golden Age of Capitalism, pp. 39–​125 (Oxford: Clarendon Press). Haberly, D. and Wójcik, D. (2014). ‘Regional blocks and imperial legacies: mapping the global offshore FDI network’. Economic Geography 91: 251–​280. Haberly, D. and Wójcik, D. (2015). ‘Tax havens and the production of offshore FDI: an empirical analysis’. Journal of Economic Geography 15: 75–​101. Harvey, D. (1988). ‘The Geographical and Geopolitical Consequences of the Transition from Fordist to Flexible Accumulation’ in Sternlieb, G. and Hughes J.W. (eds) America’s New Market Geography, pp. 101–​134 (New Brunswick, NJ: Rutgers Center for Urban Policy Research). Henderson, J., Dicken, P., Hess, M., Coe, N.M., and Yeung, H.W.C. (2002). ‘Global production networks and the analysis of economic development’. Review of International Political Economy 9: 436–​464. Kindleberger, C.P. (1984). A Financial History of Western Europe (New York: Oxford University Press). Klagge, B. and Martin, R. (2005). ‘Decentralized versus centralized financial systems: is there a case for local capital markets?’ Journal of Economic Geography 5: 387–​421. Leyshon, A. and Thrift, N. (2007). ‘The capitalization of almost everything:  The future of finance and capitalism’. Theory, Culture & Society 24: 97–​115. Massey, D. (1995). Spatial Divisions of Labour. Social Structures and the Geography of Production (London: Macmillan). Mishkin, F.S. (2006). The Economics of Money, Banking, and Financial Markets (Boston, MA: Pearson). Morrison, A. and Wilhelm, W. (2007). Investment Banking:  Institutions, Politics, and Law (Oxford: Oxford University Press). Nolan, P. (2012). Is China Buying the World? (Cambridge: Polity). O’Brien, R. and Keith, A. (2009). ‘The geography of finance: after the storm’. Cambridge Journal of Regions, Economy and Society 2: 245–​265. Palan, R., Murphy, R., and Chavagneux, C. (2010). Tax Havens:  How Globalization Really Works (Ithaca, NY: Cornell University Press). Sassen, S. (1991). The Global City (Princeton, NJ: Princeton University Press). Scott, A.J. (1988). New Industrial Spaces (London: Pion). Sharman, J.C. (2006). Havens in a Storm:  The Struggle for Global Tax Regulation (Ithaca, NY: Cornell University Press). Shaxson, N. (2011). Treasure Islands:  Tax Havens and the Men who Stole the World (London: Bodley Head). Strange, S. (1986). Casino Capitalism (Oxford: Blackwell).

574   Wójcik Taylor, P.J. (2004). World City Network: A Global Urban Analysis (London: Routledge). The Economist (2014). ‘The balkanisation of banking’. 1 March. Tickell, A. and Peck, J.A. (1992). ‘Accumulation, regulation and the geographies of post-​ Fordism: missing links in regulationist research’. Progress in Human Geography 16: 190–​218. Vitali, S., Glattfelder, J.B., and Battiston, S. (2011). ‘The network of global corporate control’. PLoS ONE 6: e25995. Webber, M.J. and Rigby, D. (1996). The Golden Age Illusion (New York: Guilford Press). Williams, G. (2011). Slow Finance:  Why Investment Miles Matter (London:  Bloomsbury Publishing). Wójcik, D. (2011). The Global Stock Market: Issuers, Investors, and Intermediaries in an Uneven World (Oxford: Oxford University Press). Wójcik, D. (2012). ‘The end of investment bank capitalism? An economic geography of financial jobs and power’. Economic Geography 88: 345–​368. Wójcik, D. (2013a). ‘Where governance fails:  advanced business services and the offshore world’. Progress in Human Geography 37: 330–​347. Wójcik, D. (2013b). ‘The dark side of NY-​LON: financial centres and the global financial crisis’, Urban Studies 50: 2736–​2752. Wójcik, D. (2015). ‘Accounting for globalization:  evaluating the potential effectiveness of country-​by-​country reporting’. Environment and Planning C 33: 1173–​1189. Wójcik, D. and MacDonald-​Korth, D. (2015). ‘The British and the German financial sectors in the wake of the crisis:  size, structure and spatial concentration’. Journal of Economic Geography 15: 1033–​1054. Wójcik, D., Knight, E., O’Neill, P., and Pažitka, V. (2016). ‘Investment banking since 2008: the geography of shrinkage and shift’. Working Paper http://​www.fingeo.net/​wordpress/​wp-​ content/​uploads/​2017/​03/​WP3_​Investment-​banking-​since-​2008.pdf (last accessed 6 April 2017). Wójcik, D., MacDonald-​Korth, D., and Zhao, S.X. (2017). ‘The political-​economic geography of foreign exchange trading’. Journal of Economic Geography 27: 267–​286. Wood, P.A. (1991). ‘Flexible accumulation and the rise of business services’. Transactions of the Institute of British Geographers 16: 160–​172.

Chapter 30

Information Fl ows , Gl obal Financ e , a nd New Digita l  Spac e s Matthew Zook Information Flows in the Global Economy Information has long been a part of economic activity (Pred, 1977), but it is relatively recently that it has become a central object of scholarly analysis. Beginning with Bell’s (1973) theorization of the ‘post-​industrial society . . . characterized not by a labour theory but by a knowledge theory of value’, information flows and knowledge creation are now considered key variables in the configuration of global economic geographies. Although researchers have used many terms over the years—​informational society, new economy, internet economy, and so on—​all formulations emphasize the ability of information and knowledge generation to (re)shape jobs, firms, and the global geographies of economic and cultural life. In short, information flows are fundamental to the global economy and the promises of the 1990s have become the everyday experience of today, albeit often in unexpected ways. In addition to reducing dramatically the frictions associated with communications, transportation, and search (see Chapter 14), information flows have reshaped the financial industry, creating novel tactics for capital exchanges, new digital market spaces, and innovative virtual currencies for exchange. These new practices, spaces, and geographies—​manifesting within the architecture of computers and absolutely dependent upon information flows—​ are deeply embedded in the financial industry and, by extension, the entire global economy. Despite their influence, the geographies of information in finance remain relatively obscure both in their structure, but more importantly in their effect and potential power. To counter this orientation, this chapter reviews how the geography of information flows are channelled in particular and powerful ways by the Internet (and for the profit) of actors who seek advantage through the creation of new digital spaces within global finance.

576   Zook

Information, Knowledge, and Financial Geographies Castells’ (1996) theory of the network society and economy is centred on electronically mediated networks that use flows of information to organize the production, distribution, and consumption of goods and culture across space. Thus, the geography of the network economy is based upon networked nodes of information, capital, and management in which power is lodged—​what Castells calls the space of flows—​and which connect and disconnect physical places in a largely interchangeable manner. While phrased in the terminology of computer technology, Castells’ formulation is not technologically determinant but focused on how the activities and intent of actors using social networks to jump across space in particular and power-​laden ways. Although the space of flows is often conflated with advanced services and finance or political networks within regional institutions such as the European Union (EU), Castells (1998) emphasizes how these systems are also represented within underground networks within the cocaine trade that connect the nodes of the coco fields, clandestine laboratories, and secret landing strips with street-​gang distribution and money laundering through offshore financial centres. In short, networked information flows connect and disconnect actors and places in dynamic structures largely at the behest of the power lodged within the space of flows. Information, however, does not simply flow, but aggregates in particular places in the form of localized knowledge, contributing to a regional advantage (Saxenian, 1994; Markusen, 1996). Running counter to more simplistic expectations that greater informational flows would render location less meaningful (O’Brien, 1992), the ability to create and share know­ ledge has a clear spatial element. Despite the growth of information flows over the past decades, not all places or all people have equal or equally useful access to information. As Polanyi (1958) argues knowledge acquisition and sharing is difficult—​‘We know more than we can tell’—​making privileged access to information and capacity to absorb knowledge particularly important. Face-​to-​face interaction and localized social networks facilitated the exchange of more complex forms (and thus less easily transferred) of information or tacit knowledge (Leamer and Storper, 2001). Although it is overly reductionist to equate tacit knowledge with unequivocal geographical constraint (see Gertler’s 2003 critique), localized knowledge exchange represents an important way in which information flows help shape global economic geographies. These spaces of information flow and aggregation are also extremely difficult to measure. Therefore, a key project over the past decades has been mapping information to render it visible (Zook, 2000; Townsend, 2001; Malecki, 2002; Graham et al., 2015). While these mappings go beyond traditional economic subjects, the capabilities afforded by the visualized information flows are crucial to understanding changes to the geographies of the global economy. This is perhaps best seen within the logistics industry, which relies on dynamic information flows to manage global supply chains for manufacturing and retailing (Aoyama et al., 2006; Schwarz, 2006). It is precisely the tagging of product flows with corresponding information flows that has increased the efficiency and reach of logistics, that is, storing inputs ‘in transit’, making information-​based logistics a key and strategic area within the global economy. In addition to organizing the flows of material goods, information flows

Information Flows, Global Finance, and New Digital Spaces    577 about the goods themselves—​such as marketing, brand, and certification—​has emerged as a key means of creating value that surpasses any material differences between nearly identical products (Zook and Shelton, 2013). While information and knowledge are important for all industries, information flows are particularly important factors within the financial industry. Just as the expectations of perfect capital mobility are belied by spatial differences (Gertler, 1984; Clark and O’Conner, 1997; Zook, 2005), information flows are likewise constrained by space and distance creating both specialized financial centres and differences in corporate governance (Christopherson, 2002; Clark and Wójcik, 2007). While agglomeration economies and path dependency contribute to the maintenance of these clusters, the ability to access the knowledge (albeit incomplete and at times speculative) of competitors and collaborators are key factors drawing together financial actors (Clark, 2005). The spatial constraining role of information is also present in the information asymmetries in lending which can lead to both adverse selection, as investors must set terms based on less than total information, and moral hazard on the part of a loan recipient who may redirect capital to riskier ventures once a loan is secured (Gertler, 1988). As a result, local banking institutions and actors with better access to information about entrepreneurs have a knowledge advantage vis-​à-​vis those outside the region. Moreover, this local advantage can extend beyond the initial transfer of capital into strategic decisions taken by firms. Zook’s (2005) analysis of venture capital financing demonstrates that within this specialized financial sector, the transfer of knowledge and connections is as important as the actual money invested. Research has also shown that localized information and knowledge is relevant beyond initial capitalization but extends into stock markers and traders. Ivković and Weisbenner (2005) find that individual and household investors are more likely to invest in local firms which they attribute not simply to greater familiarity, but by the advantages afforded by better access to localized knowledge about these firms. Relatedly, Hau (2001) shows that proximity to firm headquarters (and the superior access to knowledge) equates to higher profits for traders, and Lo and Grote’s (2003) research highlights how evolving opinions on the prospect of firms within a localized trading community provides strong incentives to remain close by other traders despite opportunities to relocate. Related work by Clark and Monk (2013) focuses on firms’ sources of information (produced internally or relying on external sources), which can influence their profitability. These information flows and knowledge networks are each characterized by a particular set of spatial relations illustrating the fundamental role spatialized information plays in the financial industry (Wójcik, 2009, 2011).

Capital Exchanges and the Tactical Use of Information Flows In particular, capital exchanges are centres for the tactical use of information. The classic image of the trading pit with brokers shouting out buy-​and-​sell orders represents one solution to managing information flows, but the needs of traders go well beyond the physical confines of an exchange. The ability to access knowledge and insight on the finances of firms, changing commodity prices, and the activities of others all contribute to traders’ decisions to

578   Zook buy and sell securities. As a result, stock exchanges have long been a fertile site for innovations related to information access and time–​space compression (Harvey, 1989). Thus, rather than a perfect market operating in a pristine cloud of information flows, capital exchanges are deeply intertwined with material geographies and competitive, spatial strategies for information management. Over the last two decades, optimization of information acquisition and use has transferred from trading pits to the computer circuits within the matching engines of exchanges where buy-​and-​sell orders are ultimately filled today. One result of this computer trading is what is known as high-​frequency trading (HFT), which utilizes specialized computer software to pursue strategies at speeds beyond the capabilities of human beings. The tactics of HFT are multiple, but a key element is the ability to identify small price differences between stocks trading in various exchanges separated by hundreds or thousands of miles and then compressing time–​space to arbitrage these price differences slightly more quickly (in the order of several milliseconds) than other traders (or, more correctly, other trading algorithms). Concurrent to the rise of HFT has been the growth of dark trading pools (more formally known as multilateral trading facilities) that allow for anonymous trading. The combination of additional venues and anonymous trading has been key for HFT as it provides multiple networks through which traders can gain a speed advantage over other market participants (see Budish et al., 2013). Although the speed differences are minute from the perspective of humans, lower latency, that is, faster connections, between the matching engines of exchanges, has been a key source of profit for HFT. This is because ultimately what seems like continuous trading to human senses is actually composed of a series of discrete events within computers, thus providing the trader with the fastest system, opportunities for ‘purely technical arbitrage’ (Budish et al., 2013, p. 2). While there are many and complex strategies to achieve this, the ways in which HFT traders pursued this highlights how finance and capital exchanges are fundamentally geographically based practices even in an era in which stocks, orders, payments, and actants conducting the actual trade (e.g. the proprietary software overseen by human brokers) are completely digital (Zook and Grote, 2017).

High-​frequency Trading and Time–​Space Compression The algorithms used in HFT largely depend on pricing signals as they are straightforward to interpret (Toulson, 2013) and necessitate little-​to-​no human oversight. Indeed, while HFT traders monitor their algorithms during trading, there have been cases, most notably the Knight Capital Group—​which lost close to half a billion dollars in one day—​where errors in the HFT software resulted in significant loses without any direct human involvement. Given the importance of getting and acting on price information quickly, reducing latency between exchanges has been a key focus for HFT across all geographical scales. At the macro-​scale the strategy has been to create more direct pathways for fibre-​optic cable networks between major trading venues such as New York and Chicago. The original routings—​which largely followed existing right-​of-​way paths—​meandered across the landscape because it was more cost-​efficient and produced no lag detectable by human users. These new pathways follow a more direct line of sight (sometime necessitating drilling holes through mountains) and allow the HFT algorithms to send signals s a few milliseconds faster than if they travelled

Information Flows, Global Finance, and New Digital Spaces    579 via the older routes (MacKenzie et al., 2012). While undetectable by humans, this time difference provided a key speed advantage to HFT. This lead was eventually surpassed by networks of microwave transmission stations (Anthony, 2012; Laughlin et al., 2013) that could propel signals even faster as communications by fibre-​optic cables can only achieve 70 per cent the speed of light. Given this demand to save time (no matter how small the increment) it is of little surprise that optimization of latency continues at the scale of metres within the facilities in which the matching engines of exchanges are physically co-​present with the servers running the HFT algorithms. In order to ensure that no trader is ‘closer’ to the matching engine, it is now common practice to use ‘standardised fibre length’ (Deutsche Börse, 2011), as well as identical access to air conditioning as cooler computers run slightly faster. The goal is to ensure that the location within collocation centres does not differentially impact each trader’s latency. HFT even focuses on speed beyond the scale of metres, and uses various strategies to compress time–​space within the actual circuits and operating systems of their computers (Leber et al. 2011; Lockwood et al., 2012). Given the size of HFT—​estimates place it at half of all trades—​every trade within every day of exchange activity is affected either directly or indirectly by strategies and tactics that are fundamentally predicated on geographically derived differences in information flows.

Increased Risk Derived from Information Flows While the ability to compress time–​space has allowed a few HFT traders to profit tremendously, this structuring of information geography has brought with it a number of problems. The most significant issue is the fact HFT ultimately depends upon the creation of minute information inequalities rather than any measure of the health, prospects, and potential futures of firms, and thus has little to do with the ostensible purpose of financial markets, that is, allocating capital to investment opportunities in the economy. In short, HFT is little more than geographically derived speculation (see Samuelson, 1957) within financial markets. In addition, HFT has been linked to a number of price disturbances within markets that had little to do with anything happening in the overall economy. This is not particularly surprising given that HFT is primarily focused on arbitraging price differences rather than the assets, liabilities, and promise of firms and industries. Research has identified over 18,000 ‘ultrafast extreme events’ during a five-​year period in which stock prices fluctuated very quickly (see Johnson et al., 2013). The largest such disturbance was on 6 May 2010 when stock indexes dipped by 7 per cent over the course of a few minutes before returning to their earlier levels. This event was dubbed the ‘flash crash’ and HFT has been highlighted as a key element in its propagation (Kirilenko et al., 2011). This kind of vulnerability via automated feeds to trading software has not gone unnoticed; in April 2013 the Twitter account for the Associated Press (AP) was hacked and a fake tweet reporting an explosion at the White House was sent. The HFT programs ingesting the AP twitter feed (as a key pricing signal) reacted immediately, resulting in a drop of 1 per cent in stock-​market indexes (Domm, 2013). To date, these types of algorithmically induced fluctuations have been corrected quickly but the risk remains. Such a sharp drop at the close of market could easily spread globally as other markets and traders react to unexpected and unexplained declines (Cliff and Northrop, 2010).

580   Zook The case of HFT highlights the fundamental role of information flows to the financial industry and the importance placed by economic actors on reshaping the geographies of these flows to their own advantage. Information technologies allow for new tactics for manipulating information flows and creating new digital spaces in the economy offering the opportunity for those with power to leverage information in unexpected and profitable ways.

New Digital Spaces, New Virtual Currencies The new digital spaces of capital exchanges are but one example of leveraging new digital spaces. In a related trend, the emergence of synthetic worlds based in informational technologies, for example games such as World of Warcraft or Eve Online, are producing new economic spaces in which users purchase completely digital objects (tools, clothing, homes, etc.) for use in these games (Castranova, 2008). These spaces and economies are already large—​Gartner (2013) estimates the size of gaming industry as US$93 billion in 2013—​and are increasingly integrated into daily life. As Lehdonvirta and Castronova, (2014, p. 266) argue, ‘many people in the world’s richest countries today live in a more or less virtual economy, where our well-​being is no longer constrained by any material lack or need but is rather tied to the mental leisures, anxieties and status games of consumerism’. While the economies of the new digital spaces are real, they differ from material economies in that scarcity is no longer a foundational characteristic. As the music industry has seen, it is perfectly trivial to make exact copies of digital objects such as a music file, thus making the power to regulate, such as enforce intellectual property regimes, a key variable in determining economically profitable activities in the space of flows (Castells, 1996). In practice, this power largely lies with the gaming company that controls the digital space and determines how digital goods can be acquired and exchanged. Moreover, in additional to financial gain, a key incentive to create and exchange digital items is to enhance one’s social status (Lehdonvirta and Castronova, 2014) marking these digital spaces as locations for the symbolic and experiential consumption associated with cultural economies (Power and Scott, 2004). This shift from material to cultural consumption is characterized by Lehdonvirta and Castronova (2014, p.  269) as a potential (albeit partial) solution to capitalist growth. As they assert, ‘in virtual economies, we have for the first time managed to decouple economic growth from ecological impact. . . . [P]‌erhaps our consumerist status games and markers can finally become just that: compelling games that need not have any material impact on the world’. While it is easy to be skeptical of this bold claim, after all information flows consume electricity and rely on the production of electronics, it represents an intriguing argument for how capitalism’s growth imperative might be accommodated within the environmental limitations of Earth’s resources. A similar argument about how ‘immaterial information is itself a new frontier for capital’ is made by Zook and Shelton (2013), albeit less sanguinely, with their contention that digital spaces offer an ‘immaterial spatial fix’ for capital investment. To be sure, both these ideas are speculative and await further theoretical framing and empirical study, but it is clear that these new digital spaces and associated economies are real (Castronova, 2014) and of particular interest to the financial industry, and have engendered a range of new virtual currencies.

Information Flows, Global Finance, and New Digital Spaces    581

New Virtual Currencies Virtual currencies were first introduced for purchasing digital products in the new digital spaces of games but are now used to acquire material goods and services, and even act as a store of value. There are two main types of virtual currencies, those with a centralized system of control and governance, and those in which these functions are decentralized. The former, such as the Interstellar Kredit used in the Eve Online game, are generally run by gaming companies that determine monetary supply and exchange rules. While these centralized currencies are certainly novel developments, they are not without precedent given the history of company scripts or private bank notes. The second type of virtual currencies is much more innovative in that governance is decentralized via crowd-​sourced computation based on blockchain technologies, for example Bitcoin, Litecoin, and Dogecoin. At the time of writing there were over 600 of these currencies standing in stark contrast to 2013 when only Bitcoin (the largest example) was evident (European Central Bank, 2015, p. 4). The decentralized and anonymous nature of this second type of virtual currency diverges from traditional currencies backed by precious metals or fiat, and guaranteed by states or financial institutions. Without this backing it is not surprising that most suffer from issues of liquidity, uncertainty of legal obligations, and as the European Central Bank notes, ‘it is clear that they also entail risks’ (European Central Banks, 2015, p. 32). While community-​backed currencies are not without precedent, they have generally been ‘local responses to the contradictions and perceived alienation of the mainstream economy. They are underpinned by an attempt to rebuild local communities’ (Leyshon, 2004, p. 466). In contrast, virtual currencies such as Bitcoin envisage operations at a global scale backed by cryptography rather than relying on localized and embedded systems of trust. It is clear that this vision of a crowd-​ sourced monetary system would not be possible (at either its proposed scale or scope) without global information flows and related code. As Kitchin and Dodge (2011, p. 133) note, ‘software offers a growing proportion of people with a set of tools to . . . undertake tasks that were previously impossible due to issues of affordability, complexity, scale, or geographical separation’. Given the genesis of these currencies, that is, a group of loosely affiliated programmers rather than officials from a sovereign state or even a bank or gaming company, the software and ideologies behind the currencies are particularly useful to study to understand the political economies of these networked flows.

The Code of Bitcoin Bitcoin began in 2009 with the release of a vision paper and source code for the system. An introductory video describes Bitcoin simply as ‘The first decentralized digital currency’ (WeUseCoins, 2011), while a more detailed overview states that ‘Building upon the notion that money is any object, or any sort of record, accepted as payment for goods and services and repayment of debts in a given country or socio-​economic context, Bitcoin is designed around the idea of using cryptography to control the creation and transfer of money, rather than relying on central authorities’ (Bitcoin, 2012). These two quotes highlight an emphasis on decentralization and reliance upon computer algorithms that is a bedrock principle within the design of Bitcoin and derivative virtual currencies. Without a central bank (or server) confirming exchanges, Bitcoin depends upon peer-​to-​peer networks, personal

582   Zook encryption keys, and computational tasks that provide ‘proof ’ of transactions between strangers. Bitcoin is not tied to any other currency, commodity, or national economy, and exists because of and in the network of its users. The technical design of Bitcoin facilitates exchanges of Bitcoins without anything physically transferred (as would be the case with coins or bills) and without any central, authoritative entity recording the exchange (as would be the case with a bank). Instead, the network determines the validity of a transaction via the blockchain, a collective computational ‘proof-​of-​work’ task. This design aspect is critical for the functioning of Bitcoin as otherwise it would be possible for a user to spend a Bitcoin twice with the expectation that the distributed system would not know that the coin no longer belongs to that person. Without getting too deep into the specifics (interesting but not particularly relevant to this chapter) Bitcoin relies upon hashing—​an intense and irreversible computational process—​that participating users run on their machines to determine which transactions are actually valid. The key part is that the hash is designed to be solved by the Bitcoin network in about ten minutes with subsequent solutions by the network providing greater validity. Anyone submitting a false transaction, for example stealing all of the Bitcoins, would have to compete with the entire network to solve the hash and validate the transaction. As a result, false transactions are not propagated through the peer-​to-​peer network of Bitcoin, but real ones are, facilitating the smooth operation of the currency system. Bitcoins are stored by each user in an electronic ‘wallet’ located on a computer or online repository, which makes them subject to both loss or theft; however, like any information they can be backed up. As outlined earlier, all Bitcoin transactions are made public but the addresses associated with a user’s wallet are not publically available unless a user chooses to make them so. As a result, Bitcoin is a semi-​anonymous transaction system as it is difficult (albeit possible) to trace transactions back to an individual. Despite (or perhaps because) of its crowd-​sourced origin, Bitcoin has grown tremendously with a market capitalization of US$3.6 billion in June 2015 (CoinMarketCap, 2015) representing more than 80 per cent of the total capitalization of all existing decentralized virtual currencies (European Central Bank, 2015, pp. 6–​7). While there are more than 600 of these so-​called crypto-​currencies—​mostly copies of the original Bitcoin system—​the vast majority are relatively small; only ten currencies have capitalizations greater than $10 million and only forty-​two are larger than $1million (CoinMarketCap, 2015). Regardless of its size and dominance within virtual currencies, the current use of Bitcoin is relatively low, accounting for ‘around 69,000 transactions per day worldwide’, which is miniscule compared with the 274 million non-​cash, retail transactions that take place every day in the EU (European Central Bank, 2015, p. 4). Not surprisingly, a very small number of business accept Bitcoin—​the European Central Bank (2015) estimates that only three out every 10,000 businesses accepts any type of virtual currency. Some large corporations—​most notably Expedia—​have set up systems so that customers can pay for hotels with Bitcoin. However, rather than accepting the currency directly, Expedia has designed a payment flow in which Bitcoins are first converted to a standard state-​backed currency before receipt by the company (Davidson, 2015). Evidently, Expedia is utilizing Bitcoin as a medium of exchange rather than a means for the long-​term storing of value. This low level of use is a major and ongoing challenge for Bitcoin and has resulted in both currency volatility and liquidity problems (see Figure 30.1).

Information Flows, Global Finance, and New Digital Spaces    583

24h Vol

7d

1m

3m

YTD From Apr 28, 2013 All

1y

To Jun 14, 2015

$15B

$1,500

$10B

$1,000

$5B

$500.00

$0B

$0.0

Price (USD)

Market Cap

Bitcoin Charts Zoom 1d

0M Jan ‘13

Jan ‘14

Jan ‘14

2014 Market Cap

Price (USD)

Jan ‘15

2015 Price (BTC)

24h Vol

Figure 30.1   Bitcoin to US Dollar Exchange Rates, 2013–​15.

Legality of Bitcoin The legal issues surrounding the use of Bitcoin remain uncertain although Castronova (2014, p. 97) argues that ‘since no law expressly forbids virtual currencies, they appear to be completely legal’. And despite worries and a reputation for financing illegal activities, Bitcoin is used for a wide range of legal and mainstream economic activity, for example booking hotel rooms via Expedia. Indeed, one of the most promising and potentially disruptive effects of Bitcoin is facilitating transactions, particularly for cross-​border payments. While existing systems such as bank wire transfers and credit cards are the common channels for this activity today, the fees charged by the intermediaries can be relatively high. Thus, payment systems in which a company accepts ‘currency in one country, changes it into units of VCS [virtual currency schemes such as Bitcoin], transfers it via the VCS network, changes it back into currency again in the receiving country and arranges the pay-​out’ (European Central Bank, 2015, p. 14) this might be an extremely useful role for virtual currencies like Bitcoin. In other words, rather than fulfilling all the roles of money such as store of value and unit of account, Bitcoin might, in practice, primarily be used in short-​term exchange. Nevertheless, banking officials take care to point out that ‘legally, Bitcoin is not a currency, does not have the status of legal tender and/​or does not meet the definition of a financial instrument’ (European Central Bank, 2015, p. 30). Moreover, states regularly issue warnings about the risks associated with using these types of decentralized virtual currencies. Given this characterization as a risky activity, key actors within the Bitcoin community have taken explicit steps to emphasize cooperation with governments. For example, Jeff Garzik, a member of the Bitcoin development team has stated, ‘We are working with the government to

584   Zook make sure indeed the long arm of the government can reach Bitcoin . . . the only way Bitcoins are gonna [sic] be successful is working with regulation and with the government’ (Liberale et Libertaire, 2011). This statement caused considerable (and generally unfavourable) commentary among Bitcoin users as the ideologies behind the project generally reject state-​led regulation or even cooperation.

Ideologies of Bitcoin This reaction is tied to the fact that the Bitcoin vision embodies the core hacker values as outlined by Levy (1984), including calls to mistrust authority and decentralize, to provide open access to the mechanisms of the computer (a trait shared more broadly with the open-​source movement) and ultimately that computers can be used to make life better. Based on the values, Bitcoin aims to do away with (for better or worse) the modern financial system, or, more correctly, do away with particular aspects of the financial system and larger processes of regulation. It is useful to return to the original concept paper published by Satoshi Nakamoto (the creator of Bitcoin) in 2009, which states in its opening paragraph: Commerce on the Internet has come to rely almost exclusively on financial institutions serving as trusted third parties to process electronic payments. While the system works well enough for most transactions, it still suffers from the inherent weaknesses of the trust based model. Completely non-​reversible transactions are not really possible, since financial institutions cannot avoid mediating disputes. The cost of mediation increases transaction costs, limiting the minimum practical transaction size and cutting off the possibility for small casual transactions, and there is a broader cost in the loss of ability to make non-​reversible payments for non-​reversible services. With the possibility of reversal, the need for trust spreads. Merchants must be wary of their customers, hassling them for more information than they would otherwise need. A certain percentage of fraud is accepted as unavoidable. These costs and payment uncertainties can be avoided in person by using physical currency, but no mechanism exists to make payments over a communications channel without a trusted party. What is needed is an electronic payment system based on cryptographic proof instead of trust, allowing any two willing parties to transact directly with each other without the need for a trusted third party. Transactions that are computationally impractical to reverse would protect sellers from fraud, and routine escrow mechanisms could easily be implemented to protect buyers (Nakamoto, 2009, p. 1; emphasis added).

What stands out in this vision is the contrast between a ‘trust based model’ of transactions that is ‘weak’, full of ‘transaction costs’, and ‘hassling’ versus a ‘cryptographic’ model of exchange that allows ‘direct’ transactions that are ‘routine’ and would ‘protect’ both buyer and seller. In one paragraph, Nakamoto replaces the complexity of a socially embedded currency with a central regulatory power and posits an ideal where informational flows and computer code allow direct exchanges without social ties or institutions, particularly banking ones. This fundamental idea is replicated through the Bitcoin rhetoric:  ‘Compared to other alternatives, Bitcoins have a number of advantages. Bitcoins are transferred directly from person to person via the net without going through a bank or clearinghouse. This means the fees are much lower, you can use them in every

Information Flows, Global Finance, and New Digital Spaces    585 country, your account cannot be frozen and there are no pre-​requisites or arbitrary limits’ (WeUseCoins, 2011). In short, Bitcoin is a call for the disintermediation of the entire financial system, and, for good measure, socially constructed trust as well (see also Castronova, 2014, p. 175). This is a profoundly libertarian vision that is echoed within the Wiki pages for Bitcoin discussing the economic concepts of inflation and deflation. In addition to providing a much more robust summary of Hayek’s views versus Keynes on macroeconomics, as well as dismissing the labour theory of value as ‘generally accepted as false’, the Wiki provides a link to an interview with Milton Friedman where he ‘proposed to replace the central bank with a computer, and to fix the money supply growth at 4% annually’. A review of the discussion side of the Wiki (which shows the history of edits and any discussion) shows no disagreement about these statements or link. The libertarian flavour of Bitcoin comes as no surprise to anyone who has studied the culture of high technology (Parsier, 2011) and given that cultural values are reflected and codified in software (Kitchin and Dodge, 2011, p. 37), the ideologies of the founders of virtual currencies are propagated alongside the fast expansion of these system. This means that the spread of Bitcoin is not simply a new means of conducting transactions, but bound up in a larger project to replace certain values such as such as socially (and geographically) embedded systems of studied trust (Sabel, 1993) with code and cryptography. And as Lessig (2006) argues, ‘We can build, or architect, or code cyberspace to protect values that we believe are fundamental. Or we can build, or architect, or code cyberspace to allow those values to disappear. There is no middle ground. There is no choice that does not include some kind of building. Code is never found; it is only ever made, and only ever made by us’. And it is for this reason that understanding the politics behind transformative economic innovations such as Bitcoin is vitally important. After all, the rise of virtual currencies is blurring the lines between the personal, the commodity, and the state (Zook, 2013; Castronova, 2014) in novel and unexpected ways. Just as new digital spaces provide the arena for ‘real’ economic activities, decentralized virtual currencies are providing the means for ‘real’ economic transactions.

Beyond Bitcoin Whatever the outcome of the Bitcoin project, the blockchain technology that undergirds this community backed currency has already begun to exert an influence throughout the financial world. The essential element of blockchain is that every transaction is broadcasted to the entire blockchain network and through cryptographic processing a transaction (or block) is verified as valid and added to the overall chain, thus providing a transparent and unalterable set of records. While Bitcoin makes its blockchain network public, that is, anyone can join, this is not a technical requirement and other structures—​such as a closed network of trusted partners—​can also be used to make a distributed system of record keeping. This has caught the attention of the banking industry, not so much as a system of currency as Bitcoin operates, but as a way to reduce costs in routine and back-​office verification work flow. As The Economist notes (2016), blockchain technologies ‘hold out the prospect of liberation from all the dross that has accumulated in the financial system, from incompatible

586   Zook IT systems to expensive intermediaries’ and is estimated to bring up US$20 billion a year in savings. Given this potential, financial institutions have invested in a number of start-​ups—​ R3, Ripple, and the Open Ledger Project—​developing scalable, open-​source blockchain technologies for use both in international monetary exchange and inter-​organizational record keeping. Moreover, given that it is unlikely that more than one or two of these ventures will become widely adopted by even a plurality (let along a majority) of financial institutions, the competitive pressures to obtain power (and profit-​making ability) over these new digital spaces are intense (The Economist, 2016). Interest in blockchain technologies for record keeping also comes from state banks such as Russia’s Central Bank and the People’s Bank of China, which have explored the use of more centralized and private blockchain ledgers for transaction verification (Palmer, 2016). While state-​led initiatives are driven, in part, by potential cost savings, control over the new digital space engendered by a blockchain solution also has the potential advantage of providing increased visibility on financial flows and thus making tax avoidance and money laundering harder to achieve. Both examples of blockchain initiatives—​from private companies and from the state—​highlight the importance of holding power in these new digital spaces; whoever can channel information to their own advantage stands to profit, perhaps as significantly as the first high-​frequency traders. In short, implementation of blockchain technology shapes on who holds power over a particular space of flows; what was a decentralizing and devolving moment within the Bitcoin project can also be used for re-​centralization of power by corporations and states. However, once the genie (or blockchain) has escaped from the bottle it can be difficult to return earlier patterns of regulation and control. In previous decades, the industries of pornography, gambling, and email spam (as well as many other economic activities) leveraged the new digital spaces of the Internet to locate personnel, incorporations, and computer servers across country borders in an attempt to place themselves beyond state regulatory powers. While this type of geographical arbitrage was far from foolproof (Goldsmith and Wu, 2006) it did greatly complicate the efforts of authoritarian states to control speech and make it more difficult for law enforcement to combat a wide range of scams and other criminal activities (Zook, 2007). This leveraging of new digital spaces continues today and current systems designed to be beyond state control are known as the ‘dark web’ and run across the public Internet but require specialized software to access. These networks range from small, private systems to larger and more public examples such as Freenet, focused on bypassing state-​based censorship regimes, or the Tor network, which encrypts Internet protocol (IP) packets and routes them through a random set of servers to mask the identity and location of users. Given this anonymity, the Tor network has a high level of criminal activity, including child pornography and buying and sell drugs, and the preferred medium of exchange is Bitcoin (Moore and Rid, 2016). To be clear, the Tor also contains more innocuous activity as well, but when the ideologies associated with the creation of new digital space are focused on bypassing both state and corporate networks—​the case for both Bitcoin and the dark net—​the resulting space of flows becomes particularly attractive to underground and criminal activities.

Information Flows, Global Finance, and New Digital Spaces    587

The Geographies of the Information Economy The uses and flows of information remain profoundly geographical phenomena, deeply intertwined with the global economy and financial industry. The review of cases in this chapter—​HFT, new digital spaces, and virtual currencies—​does not seek to provide a comprehensive review of information in the financial sector, but instead aims to highlight key ways in which information shapes the ideologies, political economies, and geographies of finance. These examples show the fundamental ways in which strategic and tactical uses of information structure competitive advantage and opportunities for profit. This ranges from the construction of informational advantage within capital exchanges to the creation of synthetic worlds in which digital products can be consumed to dencentralized currencies and anonymous marketplaces that aim to do away with existing systems of financial regulation and legal oversight. These examples have empowered (and profited) specific groups of people—​high-​frequency traders, owners of gaming companies, computer programmers—​ who are able to use and manipulate these networks with varying degrees of authority. This also demonstrates how the ability to channel information flows can be used by relatively small groups to enact their particular vision, be it for a gaming experience or a desire to avoid scrutiny from the state. In short, the structure and effect of information flows are not simply a question of the best technological arrangements; there are any number of ways to code a trade or a transaction. Instead those with power within a particular space of flows seek to construct networks and flows to their advantage. Thus, it is fundamentally important to also assess the ideologies associated with any particular configuration of flows and to whom and how advantage is given. For example, blockchain technology is about shared record keeping and can be used as means for value exchange (as in the case of Bitcoin) or as a low-​cost means of verifying contracts (in the case of the Open Ledger Project). The economic geographies emerging from the current configuration of information flows reflect the ideologies with which they were created and the goals of their designers. And as these values are codified into their source code of information flows, they emerge as unseen yet powerful influences on the economies and geographies of the world’s economy.

References Anthony, S. (2012). ‘$1.5 billion:  the cost of cutting London-​ Tokyo latency by 60ms’. ExtremeTech, 20 March http://​www.extremetech.com/​extreme/​122989-​1-​5-​billion-​the-​ cost-​-​cutting-​london-​toyko-​latency-​by-​60ms (last accessed 7 April 2017). Aoyama, Y., Ratick, S., and Schwarz, G. (2006). ‘Organizational dynamics of the US logistics industry: an economic geography perspective’. The Professional Geographer 58: 327–​340. Bell, D. (1973). The Coming of Post-​ industrial Society; A  Venture in Social Forecasting (New York: Basic Books). Bitcoin (2012). ‘About Bitcoin’ http://​bitcoin.org/​about.html (last accessed 22 January 2012).

588   Zook Budish, E., Cramton, P., and Shim, J. (2013). ‘The high-​frequency trading arms race: frequent batch auctions as a market design response’. University of Chicago Booth School of Business http://​faculty.chicagobooth.edu/​eric.budish/​research/​HFT-​FrequentBatchAuctions.pdf (last accessed 7 April 2017). Castells, M. (1996). The Rise of the Network Society (Malden, MA and Oxford: Blackwell). Castells, M. (1998). End of Millennium, The Information Age: Economy, Society and Culture, Vol. III. (Cambridge and Oxford: Blackwell). Castronova, E. (2008). Synthetic Worlds: The Business and Culture of Online Games (Chicago, IL: University of Chicago Press). Castronova, E. (2014). Wildcat Currency: How the Virtual Money Revolution is Transforming the Economy (New Haven, CT: Yale University Press). Christopherson, S. (2002). ‘Why do national labor market practices continue to diverge in the global economy? The “missing link” of investment rules’. Economic Geography 78: 1–​20. Clark, G.L. (2005). ‘Money flows like mercury: the geography of global finance’. Geografiska Annaler: Series B, Human Geography 87: 99–​112. Clark G.L. and Monk A.H.B. (2013). ‘Financial institutions, information, and investing-​at-​a-​ distance’. Environment and Planning A 45: 1318–​1336. Clark, G. and O’Conner, K. (1997). ‘The Informational Content of Financial Products and the Spatial Structure of the Global Finance Industry’ in K.R. Cox (ed.) Spaces of Globalization: Reasserting the Power of the Local, pp. 89–​114 (New York: Guilford Press). Clark, G.L. and Wójcik, D. (2007). The Geography of Finance: Corporate Governance in the Global Marketplace (Oxford: Oxford University Press). Cliff, D. and Northrop, L. (2010). ‘The global financial markets: an ultra large scale systems perspective’. UK Office for Science https://​www.gov.uk/​government/​uploads/​system/​uploads/​ attachment_​data/​file/​289012/​11-​1223-​dr4-​global-​financial-​markets-​systems-​perspective. pdf (last accessed 7 April 2017). CoinMarketCap (2015). ‘Crypto-​currency market capitalizations’ http://​coinmarketcap.com/​ (last accessed 7 April 2017). Davidson, J. (2015). ‘No, big companies aren’t really accepting Bitcoin’ Time, 9 January http://​ time.com/​money/​3658361/​dell-​microsoft-​expedia-​bitcoin/​ (last accessed 7 April 2017). Deutsche Börse (2011). ‘Deutsche Börse offers 10 Gigabit/​s access for co-​location clients’. Press release’ http://​deutsche-​boerse.com/​dbg/​dispatch/​en/​notescontent/​dbg_​nav/​press/​10_​Latest_​ Press_​Releases/​20_​Deutsche_​Boerse/​INTEGRATE/​mr_​pressreleases?notesDoc=6D197D0 3F3943B84C1257966004C7EAA&newstitle=deutscheboerseoffers10gigabit/​&location=press (last accessed 8 July 2017). Domm, P. (2013). ‘False rumor of explosion at White House causes stocks to briefly plunge; AP confirms its Twitter feed was hacked’. CNBC, 23 April http://​www.cnbc.com/​id/​100646197 (last accessed 7 April 2017). European Central Bank (2015). ‘Virtual currency schemes—​a further analysis’ https://​www. ecb.europa.eu/​pub/​pdf/​other/​virtualcurrencyschemesen.pdf (last accessed 7 April 2017). Gartner, Inc. (2013). ‘Gartner says worldwide video game market to total $93 billion in 2013’ http://​www.gartner.com/​newsroom/​id/​2614915 (last accessed 7 April 2017). Gertler, M.S. (1984). ‘Regional capital theory’. Progress in Human Geography 8: 50–​81. Gertler, M. (1988). ‘Financial structure and aggregate economic activity: an overview’. Journal of Money, Credit and Banking 20: 559–​588. Gertler, M.S. (2003). ‘Tacit knowledge and the economic geography of context, or the undefinable tacitness of being (there)’. Journal of Economic Geography 3: 175–​199. Goldsmith, J. and Wu, T. (2006). Who Controls the Internet (New  York:  Oxford University Press).

Information Flows, Global Finance, and New Digital Spaces    589 Graham, M., Sabbata, S., and Zook, M. (2015). ‘Towards a study of information geographies:  (im)mutable augmentations and a mapping of the geographies of information’. Geo: Geography and Environment 2: 88–​105. Harvey, D. (1989). The Condition of Postmodernity: An Enquiry into the Origins of Cultural Change (Oxford: Wiley-​Blackwell) Hau, H. (2001). ‘Location matters:  an examination of trading profits’. Journal of Finance 56: 1959–​1983. Ivković, Z. and Weisbenner, S. (2005). ‘Local does as local is:  information content of the geography of individual investors’ common stock investments’. The Journal of Finance 60: 267–​306. Johnson, N., Zhao, G., Hunsader, E., Qi, H., Johnson, N., Meng, J., Tivnan, B. (2013). ‘Abrupt rise of new machine ecology beyond human response time’. Scientific Reports 3: 2627. Kirilenko A., Kyle A., Samadi, M., and Tuzun, T. (2011). ‘The flash crash: the impact of high frequency trading on an electronic market’. SSRN Scholarly Paper (Rochester, NY: Social Science Research Network) http://​papers.ssrn.com/​sol3/​papers.cfm?abstract_​id=1686004 (last accessed 7 April 2017). Kitchin, R. and Dodge, M. (2011). Code/​Space:  Software and Everyday Life (Cambridge, MA: MIT Press). Laughlin, G., Aguirre, A., and Grundfest, J. (2013). ‘Information transmission between financial markets in Chicago and New  York’ http://​arxiv-​web3.library.cornell.edu/​pdf/​ 1302.5966v1.pdf (last accessed 7 April 2017). Leamer E.E. and Storper, M. (2001). ‘The economic geography of the Internet age’. Journal of International Business Studies 32: 641–​665. Leber, C., Geib, B., and Litz, H. (2011). ‘High frequency trading acceleration using FPGAs’. Field Programmable Logic and Applications (FPL), 2011 International Conference 5–​7 September, 317–​322. Lehdonvirta, V. and Castronova, E. (2014). Virtual Economies: Design and Analysis (Cambridge, MA: MIT Press). Lessig, L. (2006). ‘Code 2.0’ http://​codev2.cc/​ (last accessed 7 April 2017). Levy, D. (1984). Hackers: Heroes of the Computer Revolution (Sebastopol, CA: O’Reilly Media). Leyshon, A. (2004). ‘The limits to capital and geographies of money’. Antipode 48: 461–​469. Liberale et Libertaire (2011). ‘Bitcoin and agorism’ http://​rulingclass.wordpress.com/​2011/​06/​ 09/​bitcoin-​and-​agorism/​ (last accessed 7 April 2017). Lo, V. and Grote, M. (2003). ‘Where Traders Go When Stock Exchanges Go Virtual—​ Concentration, Dissemination or Persistence?’ in M. Balling, F. Liermann, and A., Mullineux (eds) Technology and Finance, pp. 190–​203 (London: Routledge). Lockwood, J.W., Gupte, A., Mehta, N., Blott, M., English, T., and Vissers, K. (2012). ‘A low-​latency library in FPGA—​hardware for high-​frequency trading (HFT)’. 2012 IEEE 20th Annual Symposium on High-​Performance Interconnects https://​www.researchgate.net/​profile/​John_​ Lockwood/​publication/​262293285_​A_​low-​latency_​library_​in_​FPGA_​hardware_​for_​High-​ Frequency_​Trading_​HFT/​links/​552c2c470cf21acb0920c45c.pdf (last accessed 2 May 2017). MacKenzie, D., Beunza, D., Millo, Y., and Pardo-​Guerra, J.P. (2012). ‘Drilling through the Allegheny Mountains’. Journal of Cultural Economy 5: 279–​296. Malecki, E.J. (2002). ‘The economic geography of the Internet’s infrastructure’. Economic Geography 78: 399–​424. Markusen, A. (1996). ‘Sticky places in slippery space:  a typology of industrial districts’. Economic Geography 72: 293–​313. Moore, D. and Rid, T. (2016). ‘The darkness online: Cryptopolitik and the Darknet’. Survival: Global Politics and Strategy 58: 7–​38.

590   Zook Nakamoto, S. (2009). ‘Bitcoin:  a peer-​to-​peer electronic cash system’ http://​bitcoin.org/​bitcoin.pdf (last accessed 7 April 2017). O’Brien, R. (1992). Global Financial Integration: The End of Geography (London: Pinter). Palmer, D. (2016). ‘Russia’s central bank to study blockchain tech’ http://​www.coindesk.com/​ bank-​of-​russia-​blockchain-​working-​group/​ (last accessed 7 April 2017). Pariser, E. (2011). The Filter Bubble: What the Internet Is Hiding From You (London: Penguin). Polanyi, M. (1958). Personal Knowledge; Towards a Post-​ critical Philosophy (Chicago, IL: University of Chicago Press). Power, D. and Scott, A.J. (2004). Cultural Industries and the Production of Culture (Oxford: Routledge). Pred, A.R. (1977). City Systems in Advanced Economics: Past Growth, Present Processes, and Future Development Options (New York: Wiley). Sabel, C.F. (1993). ‘Studied trust: building new forms of cooperation in a volatile economy’. Human Relations 46: 1133–​1170. Samuelson, P.A. (1957). ‘Intertemporal price equilibrium: a prologue to the theory of speculation’. Weltwirtschaftliches Archiv 79: 181–​221. Saxenian, A. (1994). Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA: Harvard University Press). Schwarz, G. (2006). ‘Enabling global trade above the clouds: restructuring processes and information technology in the transatlantic air-​cargo industry’. Environment and Planning A 38: 1463–​1485. The Economist (2016). ‘The blockchain in finance: hype springs eternal’. 19 March http://​ www.economist.com/ ​ n ews/ ​ f inance- ​ a nd- ​ e conomics/ ​ 2 1695068- ​ d istributed- ​ l edgers-​ are-​future-​their-​advent-​will-​be-​slow-​hype-​springs (last accessed 7 April 2017). Toulson, D. (2013). ‘Do HFTs really “Game” buyside orders?’ Best Execution Spring: 38–​39. Townsend, A. (2001). ‘Network cities and the global structure of the Internet’. American Behavioral Scientist 44: 1697–​1716. Weusecoins.com (2011). ‘What is Bitcoin?’ http://​youtu.be/​Um63OQz3bjo (last accessed 22 January 2012). Wójcik, D. (2009). ‘Geography of stock markets’. Geography Compass 3: 1499–​1514. Wójcik, D. (2011). The Global Stock Market: Issuers, Investors and Intermediaries in an Uneven World (Oxford: Oxford University Press). Zook, M. (2000). ‘The web of production: the economic geography of commercial Internet content production in the United States’. Environment and Planning A 32: 411–​426. Zook, M. (2005). The Geography of the Internet Industry (Malden, MA: Blackwell). Zook, M. (2007). ‘Your urgent assistance is requested: the intersection of 419 spam and new networks of imagination’. Ethics Place and Environment 10: 65–​88. Zook, M. (2013). ‘Making currency personal:  the salutary tale of the downfall of the Domdrachma’ in R. Teigland and D. Power (eds) The Immersive Internet, pp. 238–​246 (Basingstoke: Palgrave). Zook, M. and Grote, M. (2017). ‘The microgeographies of global finance:  high frequency trading and the construction of information inequality’. Environment and Planning A 49: 121–​140. Zook, M. and Shelton, T. (2013). ‘The integration of virtual flows into material movements within the global economy’ in P. Hall and M. Hesse (eds) Cities and Flows, pp. 42–​57 (London: Routledge).

Chapter 31

‘Org anic Fina nc e ’ :  T h e Incentives i n Ou r In vestm ent Produ c ts Ashby Monk and Rajiv Sharma Introduction Estimates suggest that Earth’s population may hit ten billion by 2050, which, if true, would likely trigger a fourfold increase in natural resources consumption (Monk et al., 2015). The stress this will place on our current infrastructure will be profound. In order to avoid the effects of irreversible climate change, deepening inequality, and even military conflicts over resources, we will need to unlock large pools of long-​term capital to fund resource and infrastructure innovation. As such, it is critically important for the health of our capitalist system and, indeed, the world that the global community of long-​term investors (LTIs), which includes pension funds, sovereign funds, endowments, foundations, and family offices, begin investing in long-​term projects that will prepare us for this future state (World Bank, 2015). The Organization for Economic Cooperation and Development has indicated that the community of LTIs has more than US$100 trillion of assets under management (World Bank, 2015), which means there should be plenty of capital available for the costly economic transitions ahead. However, the mobilization of LTIs towards long-​term projects is not happening; the patient capital to support the capital-​intense, long-​development ventures and projects that could, for example, reduce greenhouse gas emissions are not there. In fact, we have widening gaps in infrastructure and energy innovation financing (McKinsey, 2012). We recognize that solving the climate crisis and fixing our national infrastructure is not an invest­ or’s job per se. Rather, LTIs such as pensions or endowments are bound by fiduciary obligations to maximize financial returns. And yet, we cannot help but believe that an investor who contributes to solving climate change and facilitates the transition to a new energy economy would be well compensated for doing so. Moreover, the impending market dislocations from these existential threats will affect the mainstream portfolios of these LTIs. All this, then, raises an important question: why can’t we find mechanisms to unlock long-​term capital for

592   Monk and Sharma long-​term projects and catalyse a more sustainable version of capitalism? The answer to this question is, on the surface at least, simple: most LTIs are not using their long time horizon, preferring instead to work through short-​term intermediaries (Barton and Wiseman, 2014). Consider that, according to New York Stock Exchange data, the average holding period of assets was eight years in 1960 and is less than half a year today. This dramatic drop in time horizon, we argue, stems from the increasing complexity and de-​localization of finance, which has allowed for an obfuscation of the fees and costs that asset managers charge to asset owners, both implicitly and explicitly. This obfuscation has, in turn, led to a distortion in the underlying incentives—​the fees and costs paid to managers are important economic incentives—​which asset owners set for the capitalist system they support with their US$100 trillion or more of assets. The complexity, de-​localization, and obfuscation create distortions to capitalism and drive an increasingly short-​term and disconnected financial world. And it is this short-​termism that creates a variety of market failures and contributes to an unhealthy form of capitalist development. To understand how all of this has come to pass requires a history lesson in the evolving logics and theories of finance, starting with the development of modern portfolio theory in the 1950s (Elton and Gruber, 1997). In the decades that followed, a plethora of academic financial theories, such as the efficient markets hypotheses and rational actor model, facilitated the mass production of finance. While mass production democratized access to investment opportunities—​often by deconstructing and repackaging assets and risks into ‘products’, ‘tranches’, ‘mandates’, and ‘allocations’—​these innovations were hard to understand. Worse, the LTIs that were targeted by such innovative products did not generally have the capabilities to assess them. Most LTIs could not—​and today cannot—​accurately describe all of the underlying fees, costs, or risks that they accepted—​or are accepting—​in their portfolios. And not understanding the fees, costs, and risks was and is tantamount to not understanding the motivations of their agents. While some gains were enjoyed in the short term thanks to this mass production, the version of finance that has emerged over the last few decades has hidden its true costs and, as it turns out, is harming modern capitalism and, thus, society. In this chapter, we consider how a more holistic version of finance-​led capitalism can emerge. At the core of our argument is the need for a more sophisticated and engaged set of LTIs that understand the ingredients in their financial products. We draw parallels to the food industry, as it has already seen a revolt against mass-​produced, and poorly understood, products. Indeed, as people have begun to understand the ingredients in their food, they have started to consume food differently—​often preferring organic foods with simple, easy-​ to-​understand ingredients. Similarly, we expect that as investors begin to understand the fees and costs hidden in their investment products, they, too, will begin to act differently—​ preferring to invest in financial products and services they understand and that are rooted in real assets in the real economy. This is a phenomenon we have taken to calling ‘organic finance’, and it serves as the key conceptual contribution of this chapter. Our chapter is a ‘call to arms’ for asset owners, institutional investors, and economic geographers to pursue a better understanding of financial products and services. We hope to help LTIs become organic investors, which means having strong internal capabilities and high levels of sophistication, which would permit a reduction in intermediaries, complexities, and abstractions. Organic investors will understand the incentives they are creating with every investment, which begins with transparency around fees, costs, and expenses. By uncovering the true cost of intermediation—​and the alignment of interests that go with

‘Organic Finance’   593 it—​organic investors will find new ways to cultivate opportunities that are more aligned with their own long-​term interests and that of the capitalist system. In the sections that follow, we use the organic finance metaphor to sustain our argument, conceptualizing the global financial capitalist system rather than reporting empirical results. The metaphor is not adequately designed to ‘represent underlying economic and social processes’ but instead be suggestive and work as an instrument of inspiration for change (Clark, 2005). Building on Ang’s (2014) work on risk factors as ingredients, we position organic finance as a means of understanding all the incentives and drivers of financial return that are within financial products and offerings. We also illustrate why the emergence of this new paradigm in finance will require the focus and attention of economic geographers.

From Food to Finance In the first half of the twentieth century, organic food was just called ‘food’. At the time, people cultivated, harvested, cooked, and ate locally; they had a clear sense of a food’s ‘place’. They consumed their food with an inherent understanding of where it came from and how it arrived on their plate. They did not require an advanced degree to understand the ingre­ dients in their food, as most of it was fresh fruits, vegetables, grains, and meats. All food was organic at that time. While seemingly a golden era, however, it came with challenges. Food was expensive to produce, difficult to grow at scale, resource intensive, and hard to transport. Food insecurity—​the threat that some people would not have access to sufficient, safe, and nutritious food to meet their needs—​was seen as a serious problem (Guthman, 2003), requiring product innovation to feed the masses. The conventional food industry thus transitioned away from an inherently local product into something that could be mass produced and mass consumed. Researchers and scientists augmented and reconstituted ‘food’ in ways that made it cheaper to grow and more durable, while still palatable. The local and idiosyncratic features that one might expect from food were replaced with generalized quality and homogenized characteristics. Most people thought this was a good thing, as these methods helped feed the masses. What few realized, however, was that widely distributing, non-​local, mass-​produced food created an entirely new set of problems for human health. The delocalization, homogenization, and, indeed, ‘productization’ of food—​in which you buy a packaged and reconstituted edible product in a warehouse and take off a wrapper before eating—​created a market with countless cheap options but with too much opacity and complexity. There was a lack of knowledge around the ingredients in these products, let alone how some of these ingredients affected people over the long term.1 Increasingly, we were eating things we did not understand. In fact, the purveyors of these products seemed to target individuals that lacked basic knowledge and ability to assess the costs and benefits of consumption.2 And so, we went from not having enough good food for people to eat to having too much bad food that everybody was eating. The obesity epidemic in the USA, and, indeed, the world, is a function of this cheap food. Many people are aware that these mass-​produced foods are bad for us, and yet we still eat them. Why? Because even for those people that realize they are eating ‘junk food’, the short-​term ease and relative low-​cost vis-​ à-​vis healthy food seems to trump any potential long-​term problems. Even if we now know

594   Monk and Sharma this food is not cheap—​and it is, in reality, not cheap because it simply shifts the costs into the future in the form of illness and malaise—​it still remains popular. In response to this, and the growing obesity epidemic globally, a new ‘organic’ food movement began to take hold, reinforced by the Food and Drug Administration’s decision to demand food companies post on their products ingredients and caloric content (Guthman, 2003). Once the ingredients were measured and displayed, people began behaving differently, focusing their consumption on healthier products (Seyfang, 2006). The organic products may have been more costly in the short term, but the long-​term costs in terms of health and environmental sustainability were finally being integrated into consumers’ decision making. We should note that people are not moving back to the old agrarian models of farming and consumption. Rather, modern tools and technologies are being used to make organic foods commercially viable. The success of Whole Foods Markets is a case in point. What does organic food have to do with finance? It turns out the financial industry has evolved in a similar manner to the food industry. Traditionally, finance was a highly personal industry based on mutual and local understanding. Bankers often put themselves at the centre of local communities, providing a service that was well understood and important. Investors, in turn, focused on a deep knowledge of specific assets and opportunities. While effective, this model of finance, as with the original model of food, was difficult to mass produce. It required a very high personal touch from local representatives that could understand in painstaking detail local circumstances. This was not only hard to do; it was nearly impossible to scale. As such, just as food security was a big policy challenge of the era, securing financing for capital-​starved industries was also difficult. Accordingly, product innovation was needed in order to transform the financial services industry into something that could be more easily accessed by all. The mass production or standardization of finance arguably stems from the early invention of securities; these included bonds in late medieval Europe (specifically in Italy) and shares in seventeenth-​century Netherlands (Dutch East India Company). These inventions facilitated asset ownership at a distance, which catalysed a new set of challenges associated with information asymmetry and principal-​agent problems (see Wójcik, 2011). The mass production of finance as we see it today, however, truly accelerated with the development of modern portfolio theory in the 1950s, which used a variety of assumptions and generalizations in order to homogenize, de-​localize, and ultimately productize finance. Financial theory told us that this de-​localization, deconstruction, disambiguation, and repackaging of risks into products facilitated investment diversification, which, in turn, allowed for the widespread distribution of financial capital and reduced the cost of capital for corporations. The ultimate financiers of our capitalist system, the asset owners or LTIs, were encouraged by these developments. For reasons we explain later, they were attracted by the ease of buying a product that purported to offer a ‘predictable return’. Accessing standardized products with return targets was easier than actually studying the underlying assets and their attendant risks, as the latter could be quite messy and idiosyncratic. But converting numerous investment risks into standardized return expectations is highly complex. It is also extremely difficult to assess ‘investment skill’ in this complex environment.3 So, while these new tools and techniques posited themselves as simple, they were anything but (Wainwright, 2011). Finance thus transitioned from an inherently local product to a global product overseen by firms in global financial centres (Lee et al., 2009). The very notion of selling financial ‘products’ (as opposed to investing in companies or assets) implied de-​localization. As

‘Organic Finance’   595 with the food industry, financial products became abstractions of ‘real’ assets. The products tranched and stripped assets of local or idiosyncratic characteristics in order that they could be sold to general consumers on exchanges with the help of rating agencies and a plethora of intermediaries. Just as in food, a financial product can, quite literally, be ‘packaged’ and ‘wrapped’ and sold to consumers looking to satisfy a specific return obligation with very little understanding of the ingredients that are meant to actually generate those returns. As with consumers of processed food, few among LTIs had the sophistication required to make smart decisions about where to consume the rapidly expanding array of financial products and services. Most did not understand the fees, costs, risks, and thus incentives being accepted either explicitly or implicitly in the bargain to move towards mass-​produced finance (see Box 31.1). And it was very hard to get ahead of the innovations in terms of knowledge. Perversely, financial education and literacy has been linked to increasing levels of obfuscation and purposeful disorientation by financial service providers intent on maintaining a knowledge gap (Carlin and Manso, 2011). In addition, the LTIs were also, oftentimes, complicit, using the high expected returns of new products as a mechanism to increase the expected return of their overall portfolio, which, thanks to actuarial mathematics, served to reduce the future financial obligations of the sponsor.4 This was particularly true of public pension funds, where aggressive return targets were often paired with under-​resourced internal investment teams, pushing them into a world of aggressive financial products they barely understood. Notwithstanding, just as the food industry was pressured to deliver organic foods to increasingly sophisticated consumers, a growing community of sophisticated LTIs is moving away from overly processed and engineered financial products and is working to invest in real assets in the real economy. To be clear, these investors are not going back to the era of ‘pioneer bankers’ living in local communities funding rural projects. Rather, these organic investors are using innovative tools that empower LTIs to take a long-​term view in their investments. They are rethinking their access points or even purchasing assets in local communities on a direct basis rather than from warehouses on Wall Street.

Box 31.1 Why Investment Fees and Costs are Overlooked Ingredients in Financial Products • Markets: There is an assumption that the market for financial services functions efficiently. It doesn’t. • Diversification: Opacity from over-​diversification creates an environment ripe for hidden fees and costs. • Priorities: Investors believe that asset allocation is the priority and that fees are not important. • Perspectives: It is difficult for investors to share fee information with peers. • Unknown Unknowns: You don’t know what you don’t know. And the truth is you don’t know a lot. • Career Risk: What happens if I uncover some overlooked fee? Won’t I look bad? Yes. • Misuse of Benchmarks: Inaccurate benchmarks can distort performance. • Overconfidence Bias: My portfolio can’t be more efficient. I am good at what I do. • Believing the Hype: Many investors feel lucky to have access to a ‘brand-​name’ manager. • Traditions: Asset-​based fees (‘AUM bps’) treat dollars like dirt, but one is easier to move than the other.

596   Monk and Sharma Similar to food, then, organic finance may seem more expensive in the short term, as it requires far greater sophistication and knowledge within the LTIs. However, in the context of the long-​term costs and misaligned incentives associated with mainstream finance, many LTIs rightly see organic finance as cheaper and more competitive over the long term. It is difficult to quantify these gains and the costs, but they undoubtedly exist, and academic research seems to confirm this emerging bias as correct (see Harris et al., 2014; Chuprinin et al., 2015; Fang et al., 2015). For example, research shows quite clearly that opaque assets cost more than transparent asset, even with the same pay-​offs (Sato, 2014). In addition, some might view the fact that organic finance accentuates home bias as a problem, but Graham et al. (2009) show that a home bias can be prudent. Indeed, local investing relative to non-​local investing can add up to 3.2 per cent per year in incremental, annual investment returns (Ivkovic and Weisbenner, 2005). The authors of these studies show that the outsized returns stem from a better understanding of the value-​rele­ vant information about the assets (Coval and Moskowitz, 2001). Clearly, there are potential problems with local investing when it is done poorly (see Hochberg and Rauh, 2013). Good governance is the key to ensuring it is done well; and, thankfully we have plenty of frameworks to ensure this is the case (Clark and Monk, 2014a, 2015). Finally, organic finance is ultimately about investing in companies and products that are sustainable. New research shows that an investment portfolio made up of companies that perform well on sustainability factors that are material to their business—​materiality is the key here—​ generates 6 per cent of alpha (Khan et al., 2015). In sum, organic finance may appear more costly, but we believe it offers a pathway to deliver sustainably higher-​risk-​adjusted returns. In our view, it is the lack of transparency about the fee and cost ingredients, which represent, quite simply, the incentives underpinning our entire investment industry, that are crippling financial capitalism today (Jennings and Payne, 2016). The models and products that purported to render finance cheaper, easier, better, and more efficient represent some of the most costly products in our society. We would even challenge the Nobel Prize-​winning theory that portfolio diversification, optimized by intermediaries, offers a ‘free lunch’ (Ibragimov et al., 2011). There is a cost to diversification, and it is the lack of understanding that comes with a portfolio that cannot be appreciated and evaluated. The organic finance movement is thus ultimately about taking the ingredients in our financial products, as with those in our food products, more seriously. In the following sections, we describe how this can be done in practice.

The Business of Institutional Investment In order to understand how the financial services industry evolved over time, it helps to start by reviewing some information about the very base of our financial system: the savers. These savers, which come in the form of pension funds, endowments, sovereign funds, foundations, and other LTIs, are the owners of assets in capitalism. Intermediaries, such as venture capitalists, hedge funds, and other asset managers, get the money they manage from these LTIs. Healthy capitalism thus demands that the asset owners (the ultimate sources of capital) set appropriate incentives for the entire system, ideally in a manner that aligns the interests of asset owners with the sponsors of projects and companies (the ultimate users of capital).

‘Organic Finance’   597 At the most basic level, LTIs exist because their sponsors decided to manage a set of future liabilities (explicit or contingent) by setting aside financial assets today and then investing those assets in financial markets (Clark and Monk, 2012a; Dixon and Monk, 2012). There are two key reasons why an asset owner would look to establish a long-​term pool of financial assets: on the one hand, prefunding ensures plan sponsors are making credible and legitimate financial promises, that is, they will actually meet their future obligations. On the other hand, it is often hoped that the pre-​funded financial assets will grow, thanks to prudent investing, at a rate faster than the liability, the sponsor, and even the overall economy. This means that future liabilities are met at a relative discount by investing in financial assets (Clark and Monk, 2012b). And, in both cases, there is an assumption that financial markets offer a reliable mechanism to manage financial assets to meet liabilities (Campbell and Viceira, 2002). In looking to meet liabilities, the sponsors of LTIs have been comfortable with the idea that their funds would be overseen and managed by a long chain of principal–​agent relationships (Dixon and Monk, 2013). LTIs are often just conduits to the for-​profit financial services industry, contracting for investment management services with external asset managers. The problem here, however, is that the contracting mechanism of most institutional investors is woefully underdeveloped (see Clark and Monk, 2014a). Additionally, most LTIs are not sufficiently resourced to recruit the people or build the systems necessary to be effective (Bertram and Zvan, 2009; Ambachtsheer, 2011; Bachher and Monk, 2012). And this fact—​that many LTIs lack internal resources to oversee properly the long chain of intermediaries and products—​is largely underappreciated and ignored in the marketplace (Neil and Warren, 2015). When people see large pools of assets under management, they assume these pools can be tapped for internal resources. That is not true, as there is often a strict firewall between the investment capital and the operating budgets of these organizations, leaving the latter woefully inadequate for the management of the former. It is for this reason that the traditional institutional investor is outsourced, rarely possessing the expertise and competencies to execute even the most basic financial transactions without the help of external advisors. Most LTIs are seen by their sponsors as government agencies or university divisions rather than what they truly are: the base of capitalism. As a result, the base of capitalism operates according to logics that are often not very capitalist. Over time, the extended chain of principal–​agent relationships and complexity that separated LTIs from productive assets became problematic. In particular, the injection of fees and costs (ultimately, new incentives and motivations) at each link of the chain served to distort the original intentions of asset owners (MacIntosh and Scheibelhut, 2012; Sharpe, 2013). The investment decisions made by asset managers often maximized the utility of the asset managers (and not the asset owners); a phenomenon known as ‘broken agency’ (see Sheffer and Levitt, 2010). Moreover, each layer of intermediation further obfuscated the risks and costs being incurred by LTIs to achieve their return objectives. As this information was lost, so, too, was oversight (Leiblein et al., 2002). The agents turned the tables on the principals, as asset managers used complexity and layers of intermediation to take advantage of under-​resourced LTIs. The current model of financial capitalism often sees agents disciplining the principals.5 This has created a nefarious culture in which asset managers believe they are ‘masters of the universe’, pursuing massive ‘free-​market’ paydays (Kinnel, 2010; Lo, 2012). The problem with this, however, is that

598   Monk and Sharma the market for asset management is not really free, nor is it efficient (Spence, 2002). Research shows quite clearly that institutional investors are ‘gamed’ on fees by their managers (Starks, 1987; Carpenter, 2000; Bebchuk and Fried, 2004; Phalippou, 2010; Foster and Young, 2010; Ellis, 2012; Robinson and Sensoy, 2013). And in cases where transparency is difficult to achieve, such as in private markets, these problems are exacerbated. Take private equity as an example: Phalippou and Gottschalg (2005) show that net of fees, fund performance adjusted for risk underperforms the S&P 500 by 6 per cent. It is also worth noting that many LTIs pay managers to develop capabilities that do not focus on long-​term value creation. There is a big difference between a manager using high fees to create long-​term value and that same manager using those fees to create and collect ‘foreknowledge’, which refers to accurate predictions of short-​term events. Foreknowledge, research tells us, offers no social value (Hirshleifer, 1971). In fact, the focus on foreknowledge is a social cost because including information into prices one week, one day, or one millisecond earlier is unlikely to lead to more efficient allocation of resources in the real economy. But many LTIs will pursue these paths because they appear, at least on the surface, to offer products that can meet aggressive return targets. When you put all of this together, it becomes clear that the sponsors of LTIs—​often government policymakers—​created an environment in which the for-​profit asset managers could extract a disproportionate share of value in the course of their business. This happened, in large part, because the sponsors could not stomach what was actually needed to build an effective and professional LTI organization (Dixon and Monk, 2013). Specifically, the sponsors of LTIs often preferred to pay low (highly transparent) salaries to the employees and extremely high (but non-​transparent) fees to external, for-​profit providers. This was the politically palatable choice, at least while the external fees remained opaque and under-​ appreciated by stakeholders. Asset managers are thus benefitting disproportionately from the politics of pension funds (Clark, 2007, 2008), as the industry receives the equivalent of a subsidy when LTI sponsors choose to under-​resource their LTIs, while setting high expected return targets. In this regard, today’s intermediaries are not all that different from the food industry’s corn farmers in that they are benefiting enormously from governments’ spending choices and priorities. Consider this: the financial industry is so reliant on the so-​called government handouts that when a government does take the step of building a professional investment organization, popular press presents it as ‘bad news for money managers and consultants hoping to grab a share of those assets’.6 What is unarguably good for a government and its citizens—​ professional management of assets—​is bad for financial intermediaries. This is a reminder that the financial industry does very well from its unsophisticated clients. Today, we are seeing signs that the pendulum may have reached its extreme position. The Securities and Exchange Commission has been investigating 400 private equity general partners and found that ‘a majority of private equity firms inflate fees and expenses charged to companies in which they hold stakes’.7 Similarly, on the asset-​owner side, major public pension funds in the USA have recently admitted that they do not even track the billions in fees they pay to asset managers.8 Many LTIs present incomplete fee pictures in their annual reports: some focus only on base fees and bury performance fees in net return numbers, while others make no attempt to quantify the implicit fees associated with holding, moving, or trading assets (despite the fact that the implicit numbers, such as spreads and transaction costs, can be very high). Is it any wonder that the financial services industry in the USA captures a third of all corporate profits and creates more billionaires than any other industry by a factor of two?

‘Organic Finance’   599 The present value of the average fees paid by asset owners to asset managers over a thirty-​year period amounts to approximately one-​third of the assets invested, which is a lot of money to pay a manager that is not, on average, going to beat passive benchmarks (Greenwood and Scharfstein, 2013). And therein is the problem: most of the highly paid professionals do not outperform the broad market indices over time. Fama and French (2010) highlight that mutual funds underperform passive benchmarks even more when you layer in fees. And even where returns are delivered above benchmarks, risk adjusting those returns for tail risk removes the alpha (Jurek and Stafford, 2011). All of this wealth flowing into a single industry that is not, in aggregate, adding much value is having a variety of negative externalities. It’s obviously luring our best and brightest minds away from socially productive uses and putting them in the business of cultivating foreknowledge (Murphy et al., 1991). It is also driving increasing short-​termism and changing the allocation of resources in our economy. The only way to remedy these problems is, in our view, to show the boards and, indeed, sponsors of these organizations the true cost of financial intermediation. Fee and cost transparency is a first step on the path towards LTI professionalization. As we show in the next section, organic finance therefore focuses on understanding the ‘hidden’ costs of finance and investment, as this will not only help us understand the incentives being created, but also demonstrate to the world the real cost of investing and trigger a massive round of LTI innovation and improvement.

Organic Finance We use the term ‘organic’ to reference any relationship in which the elements that fit together do so harmoniously and as a necessary part of a broader system of complexity. In other words, organic refers to something sustainable, healthy, and long term. In our conceptualization, organic finance refers to a form of investment that focuses investors’ attention on all the ingredients in investment products so as to appreciate the short and long-​term costs and consequences of those ingredients on value creation and performance. In essence, organic finance re-​emphasizes the importance of LTIs in the investment world and de-​emphasizes the often over-​glorified for-​profit financial services industry. The term organic has been used frequently in business or corporate contexts. For example, organic production refers to a process devoid of artificial catalysts or stimulants, rooted in ‘natural’ and thus sustainable inputs and outputs (Davidsson et al., 2006). Organic growth refers to a business growing on its own and without mergers or acquisitions (Davidsson et al., 2006). In our context, organic (finance) is about generating high investment returns without relying on artificial catalysts or opaque inputs that focus on short-​term performance with dubious consequences for long-​term commercial health and performance. Similar to how the organic farming industry avoids pesticides known to create long-​term harm to consumers, organic finance seeks to understand the effects of artificial catalysts the finance industry uses and remove those with long-​term consequences that are harmful. Organic finance is about reintroducing the notion of long-​term value creation and sustainable growth back into capitalism. Figures 31.1 and 31.2 illustrate the shift from an opaque, overly packaged form of finance towards a more transparent, professionalized version of long-​term investing—​organic finance.

600   Monk and Sharma Building Blocks of Capitalist System: Pension Funds, SWFs, Foundations, Endowments, Family Offices.

Global Set of Investment Opportunities: Public Companies, Private Companies, Equity, Debt. FEES & DISTORTIONS FEES & DISTORTIONS

FEES & DISTORTIONS

‘LOW COST’ Institutional Investors

Fund of Funds, Placement Agents, Consultant

Asset Managers

Buy Side Banks

FEES & DISTORTIONS

Sell Side Banks

Opportunity Sponsors: Companies, Government Etc.

Figure 31.1   Existing Opaque, Overly Packaged Form of Investment Management. SWF, sovereign wealth fund.

Organic Finance: Greater investor responsibility on fees and efficiency of investment access points.

‘LOW COST’ Institutional Investors

Opportunity Sponsors: Companies, Government Etc.

Figure 31.2   Organic Finance: Greater Investor Responsibility on Fees and Efficiency of Investment Access Points.

‘Organic Finance’   601 We build on the work of Ang (2014) and his focus on risk factors as ingredients in financial products. We expand it to consider the fees and costs, and their associated incentives, as additional key ingredients (and alignment factors) affecting long-​term performance. If factor investing requires looking beneath the asset class or financial product labels to the underlying risk exposures (Ang, 2014), organic finance requires looking beneath the products and mandates to the underlying incentives that are being created. An organic investor will seek to understand the most efficient and aligned access points to access risk factors, given the fees and costs (incentives). At its core, organic finance is about recognizing short-​and long-​term costs with the goal of encouraging more long-​term investing, which research shows can provide significant benefits to society (see Clark and Monk, 2015). Long-​term investments in infrastructure and green technologies are needed to help overcome some of the looming global challenges of population growth, urbanization, and climate change. Infrastructure assets, in particular, are the physical facilities that provide the building blocks of a functioning society. Institutional investment into these assets can directly support the well-​being of households, as well as production activities of enterprises at various points of the value chain (Sharma, 2012). The International Monetary Fund’s ‘World Economic Outlook’ report (2014) reveals that an increase of one percentage point of gross domestic product in public investment (used as a proxy for infrastructure investment) spending raises the level of output by about 0.4 per cent in the same year and by 1.5 per cent four years after the increase. Investments in other long-​term, private-​market asset classes can also be seen to have wider economic impacts. Venture capital investments that back entrepreneurs and new businesses, for example, have been proven to contribute to economic development (Timmons and Bygrave, 1986; Kortum and Lerner, 2000; Sampsa and Sorenson, 2011). Similarly, certain real-​estate development investments have provided economic benefit, particularly those in underdeveloped areas that could be classed as targeted investments (Hagerman et al., 2007). In our view, organic finance can help investors find aligned access points to these assets. So how do we begin to usher in this new era of organic finance? We see two keys to this phenomenon. Firstly, a key driver that must underpin the organic finance movement is a desire among LTIs to better understand the potential inputs and outputs associated with a given strategy or product. Being an organic investor requires a deep understanding of the services and products and specifically the underlying assets, risks, fees, and costs. Just as the food product ingredients placed on packaging (which helps consumers make more intelligent food choices) drove the rapid rise of the organic food industry, so, too, will a close scrutiny on financial product ingredients serve to drive the organic finance industry. Secondly, while the initial focus of organic investors will be on the risks, fees, and costs of their financial products, this is just the first step in changing behaviours. For example, the people that consume organic food, we know from the research, pay very close attention to how the food is prepared (Guthman, 2003). Put another way, the organic food movement does not stop at the purchasing of the food. The knowledge of an organic consumer shapes his or her ‘tastes’ and drives consumption behaviours towards more healthy products. Similarly, additional steps taken by LTIs to understand their products could usher in a new era of investment in which LTIs change the way they construct portfolios. It is perhaps for this reason that some have even called for an equivalent of a Food and Drug Administration for finance (Weyl and Posner, 2012). The role of regulation for increasing transparency cannot be underestimated and will be crucial for achieving the intended objectives. Indeed, the

602   Monk and Sharma knowledge that will come from a close understanding of the fees and costs of financial products will provide a deeper understanding of the financial industry, which, in turn, will catalyse greater professionalization of LTIs. Through an emphasis on transparency, driven by more professional LTIs, the understanding that comes with organic finance will spur a variety of positive externalities for the investment industry and generally speaking, the world, including:

Better Understanding Fee structures have the potential to dampen the volatility (risk) of certain asset classes, which means that asset owners can misunderstand their risk exposure. Is a university endowment with 80 per cent allocated to alternative asset managers doing a good job if it generates 10 per cent returns? We would argue that many endowments could not explain the returns generated per unit of risk and per unit of cost. The reason is that they often do not know how much risk their external managers are actually taking or how much they are really paying those managers. The authors of this study have been told by senior managers at major endowments and foundations that ‘they’d rather not know’ how much they are really paying their managers. And yet, understanding risks will help avoid preventable crises, and getting to the core of the risks means understanding the fees being paid to the managers.

Free Money For large institutional investors, even the smallest savings in costs can have a significant impact over time. This is all the more true in a low-​interest-​rate environment, where saving basis points really is important. Most LTIs are overly focused on being frugal with internal teams and systems and do not pay attention to external teams. A more holistic frugality will offer unique benefits. Also, we cannot think of anything else an investor can do to generate risk-​free returns than get a cheaper access point to the same risk exposure. Why is there no big provider of ‘implementation alpha’ today?

More Efficient Labour Market By overpaying certain asset managers (e.g. hedge funds), we as a society are telling people that you can make real money through the adroit playing of short-​term, zero-​sum games. Top business school graduates rarely work for asset owner organizations such as a public pension fund. Conversely, top PhD graduates in the pure sciences (mathematics, physics, chemistry) are lured into hedge funds. By digging into the fees and costs of finance, we can provide appropriate incentives that do not distort labour markets.

Economies of Scale When people look at in-​or outsourcing decisions, they often look only at the cost and benefits at a single point in time. They rarely consider future periods where the costs paid to

‘Organic Finance’   603 external parties allow economies of scale to accrue to those parties. By ceding capabilities to the private sector at a single point of time, an LTI may be relegating itself to a disadvantageous negotiating position in the future (Burton, 2013). Moving to an organic finance industry in which the asset owners are fully professional will allow the economies of scale to accrue more evenly among asset owners and asset managers. In sum, more professional LTIs will take a longer-​term view and bring a more harmonious version of finance-​led capitalism than the one we have today (see Hawley and Williams, 2000). That is the benefit the organic finance movement seeks.

Enlisting Economic Geographers The financial economy is often perceived as separate from the real economy (Maurer, 2008; O’Neill, 2009). Financial products have, indeed, moved investors away from real assets and risks via segments or slices of assets (Clark and Monk, 2014b). While these products ultimately rely on the performance of some underlying asset, they are de-​localized. Significantly, the removal of local or idiosyncratic risk was perceived to be a feature rather than a flaw, as it allowed for wide diversification. And yet, as we argued earlier, the layers of complexity and abstraction results in too much opacity and complexity. And it is in these grey areas, the shadows, that financial intermediaries have consolidated their power. We believe the organic finance project could generate a series of benefits for companies, project sponsors, LTIs, and even capitalism. We believe it will usher in a generation of professional and capable LTIs, which can create new incentives for long termism and sustainability. But for this vision to become a reality, a new research programme that re-​affirms the value of the ‘local’ in finance and shines light in the shadows is warranted. Accordingly, we believe there is a significant role to be played in organic finance by economic geographers. More to the point, this project could benefit greatly from the addition of social scientists that are not pure economists. The economics discipline operates at an abstract level and seeks generalizations and broad theories; it is a world of models driven by mathematics (Lazear, 2000). To an extent, these models allowed the financial services industry to operate with little accountability as to the fees and costs being charged. A bottom-​up approach that prioritizes the ‘ingredients’ in painstaking and rigorous detail—​qualitative details—​may add significantly to our understanding of financial incentives and alignment (Clark, 1998).9 This is where economic geographers can serve a useful purpose, as they focus on the rich texture of contemporary economic circumstances (Clark et al., 2003; Arnott and Wrigley, 2001). Economic geographers tend to use bottom-​up methodologies, such as case studies and fieldwork in order to conceptualize and, eventually, theorize (Clark, 1998). Geographers’ training involves a deep appreciation for outliers as well as contingencies and idiosyncrasies of all kinds of economic actions and actors. This focus on local differences has pushed the discipline towards less quantitative approaches or abstractions. In fact, economic geographers specifically seek to reconcile the unevenness of economic activity with some broader understanding of economic landscapes (Graham and Marvin, 2001; Clark and Wójcik, 2007). For example, the financialization conceptualization within economic geography tries to provide a better understanding of the machinations of financial actors and intermediaries that are reshaping the landscapes of contemporary capitalism (Faulconbridge and Muzio, 2009; Pike and Pollard, 2010; French et al., 2011).

604   Monk and Sharma We believe the community of economic geographers can thus play an important role in mapping the fees and costs of financial intermediation and institutional investment, even offering investors ‘recipes’ and ‘blueprints’ for how they might invest with better ingredients or more sturdy organizations (Wójcik 2011, 2013). Geographers can be proactive in pushing for granular and face-​to-​face conversations about the implicit costs of investing, which often require a tacit understanding of how these costs are levied (see Gertler, 2003; Storper and Venables, 2004). The data on fees and costs are also often hidden, which means public data do not exist and, as such, economists are not going to have the key input (i.e. quantitative data) in their methodological approach. It would not be the first time geographers turned their research focus to some fundamental problems in finance. Nearly thirty years ago, Botts and Patterson (1987) first argued that the economic impact of pension plans on regions was an area of inquiry necessitating further investigation and understanding. It was then Gordon Clark’s body of research on the design and governance of pension fund boards that is today guiding the manner in which these institutional investors govern their operations, from Australia and the USA to Europe and Asia (see Clark (2000) as the first book in a series by Clark). Financial intermediaries, metrics, and practices are ever-​more engrained in the economic geographies of our personal, working, and public lives through our access to and use of bank accounts, mortgages, pensions, and savings; employers’ ownership; access to capital and financing; and public infrastructure and services (O’Neill, 2009). This includes a deeper understanding of the packaging and securitization of private assets, such as urban infrastructure into financial products (Clark and O’Connor, 1997; Faulconbridge et al., 2007; Torrance, 2009; Knight and Sharma, 2015). In short, the geographical research endeavour is truly an emphasis on what sits between theory and practice and how to reconcile and, indeed, solve any issues that may arise in that hard-​to-​define ‘place’ or ‘space’ (see Bathelt and Glückler, 2011). Organic finance will require the services of economic geographers.

Conclusions Some view LTIs as if they are post-​office boxes; a place where sponsors send money before it is forwarded on to other ‘professional’ managers to deal with. But LTIs can no longer be a pass-​ through from the plan sponsor to external money managers. They have to professionalize. This seems self-​evident given they are the foundation of capitalism, but many are comfortable operating in non-​capitalist ways. But this is no longer tenable, as the financial innovations of the last decades have substantially increased the complexity of instruments and services. With all of this new complexity, the costs of financial intermediation are increasingly difficult to identify, rationalize, and minimize. Worse, these costs create new and hard-​to-​appreciate incentives that are driving short termism and unsustainable practices. This raises the question: How do we transition the finance industry towards a more ‘organic’ approach to investment? For the most part we know what we need to do to fix the problems of finance: we need the base of our capitalist system to professionalize. As we have argued, we genuinely believe that sophisticated asset owners are needed to save capitalism from endemic short termism and rent seeking and unlock the capital we require to finance critical projects. In order to professionalize, we need to improve the governance of public investment organizations, but changing governance regimes can be a difficult undertaking. Many institutional investors have

‘Organic Finance’   605 proven incapable of managing change within their organizations, allowing inertia to be a key factor influencing decision making. So the more important question today is how to catalyse the sorts of changes we need. In our view, cost and fee transparency is the most viable catalyst for innovation and change, as it demonstrates unequivocally that the status quo in finance is not sustainable. In the same way that food now comes standard with details on calories and ingredients, so too should finance come with easy-​to-​understand labels that describe the true cost of financial intermediation. The reason, then, to push fee transparency stems from the observation that we need to change fundamentally the business of asset management if we are going to have any chance of solving some of the intractable problems of our generation. Boards and sponsors of LTIs have not invested in professionalization of their investment teams because they have not yet seen the true cost of the outsourced investment model. Paying an internal investment team US$20 million may seem costly on its own. But it seems cheap when you consider the same service and performance costs US$200 million per year when it is paid to a team outside a pension fund (CEM Insights, 2010). We believe that if a pension fund or endowment board—​let alone their sponsor or general public—​saw how much LTIs were paying Wall Street for their products, asset owners would be inclined or even forced to professionalize (Ellis, 2012).10 In sum, this chapter makes two main arguments: (i) capitalism needs greater professionalism and sophistication among asset owners; (ii) the only way to develop this is to demonstrate the true cost of external professionals and the distortions these costs are creating for capitalism. It is only in recognizing the true ingredients, and their incentives, that LTIs will finally organize themselves to consume ‘healthier’ financial products. In our view, a research programme that scrutinizes not only the actors and agents of finance, but also the evolution of financial product ‘packaging’, will be required. The methods and holistic approach of economic geography will be valuable in this regard, especially compared with the traditional forms of finance and economics research. In the world of organic finance, the work of economic geographers will be critical.

Acknowledgements The authors acknowledge the direct support of Stanford University’s Global Projects Center and the members of the research consortium on institutional investment. Members of the research consortium provided access, field assistance, and partial support of staff salaries throughout this project, though members were not involved in the formation of the chapter’s content and argument. The authors would, however, like to thank Jagdeep S. Bachher, Gordon L. Clark, Elliott Donnelley II, Raymond Levitt, Derek Murphy, Duncan Sinclair, and Darek Wójcik for their insights and suggestions. Any errors or opinions herein are the authors’ own.

Notes 1. Trans fats are still common in some areas despite the fact that we know they contribute to obesity; processed meats are also still popular despite the fact that they have been shown to cause cancer (see Bouvard et al., 2015). 2. Moreover, lower-​and middle-​class families often have no other choice, as they are forced to live in ‘food deserts’ in which only fast food is available.

606   Monk and Sharma 3. Part of getting transparency may require understanding attribution and coming to a view on the debate over luck versus skill in asset management. There is an enormous literature focused on luck versus skill that we only touch upon (see Cornell, 2009 for a complete review). Asset owners should pay for skill not luck. 4. Novy-​Marx and Rauh (2009) highlight the internal conflicts of pension funds, showing that they are pushed to invest in high-​risk assets so they can justify a higher discount rate. 5. This is particularly true in the context of the endowment model of institutional investment, in which asset owners often see their most important job as getting access to top managers, no matter the cost. 6. http://​www.pionline.com/​article/​20150601/​PRINT/​306019976/​ontario-​hopes-​to-​create-​ investment-​management-​firm (last accessed 10 April 2017). 7. http://​www.bloomberg.com/​news/​articles/​2014-​04-​07/​bogus-​private-​equity-​fees-​said-​ found-​at-​200-​firms-​by-​sec (last accessed 10 April 2017). 8. http://​fortune.com/​2015/​09/​04/​calpers-​still-​cant-​get-​out-​of-​its-​own-​way-​on-​private-​ equity/​(last accessed 10 April 2017). 9. The act of going and seeing for yourself the real people or assets that are at the heart of traded financial products was emphasized with dramatic effect in Michael Lewis’s The Big Short: Inside the Doomsday Machine about the 2007–​09 subprime mortgage crisis. In the book and later film, hedge-​fund manager Mark Baum decides to travel to Miami, to investigate for himself the financial credibility of those purchasing the mortgages, and how the mortgage brokers were incentivized. The trip confirmed the suspicion of a speculative housing bubble with most big banks entrenched, ultimately leading to the collapse of the global economy. 10. And, once the true cost of financial intermediation is uncovered (Fang et al., 2015), there is a growing body of research conducted within economic geography and other disciplines that can help asset owners reorganize (Clark and Monk, 2012c; Clark et al., 2012; Bachher and Monk 2013; Dixon and Monk, 2014; Clark and Urwin, 2008, 2010).

References Ambachtsheer, K. (2011). ‘How should pension funds pay their own people’. Rotman International Journal of Pension Management 4: 18–​25. Ang, A. (2014). Asset Management—​A Systematic Approach to Factor Investing (Oxford: Oxford University Press). Arnott, R. and Wrigley, N. (2001). ‘Editorial’. Journal of Economic Geography 1: 1–​14. Bachher, J.S. and Monk, A.H.B. (2012). ‘Attracting talent to the frontiers of finance’ http://​ssrn. com/​abstract=2120167 (last accessed 10 April 2017). Bachher, J.S. and Monk, A.H.B. (2013). ‘Platforms and vehicles for institutional co-​investing’. Rotman International Journal of Pension Management 6: 64–​7 1. Barton, D. and Wiseman, M. (2014). ‘Focusing capital on the long term’. Harvard Business Review https://​hbr.org/​2014/​01/​focusing-​capital-​on-​the-​long-​term (last accessed 10 April 2017). Bathelt, H. and Glückler, J. (2011). The Relational Economy:  Geographies of Knowing and Learning (Oxford: Oxford University Press). Bebchuk, L.A. and Fried, J.M. (2004). Pay Without Performance: The Unfulfilled Promise of Executive Compensation (Cambridge, MA: Harvard University Press). Bertram, R. and Zvan, B. (2009). ‘Pension funds and incentive compensation: a story based on the Ontario teachers’ experience’. Rotman International Journal of Pension Management 2: 30–​33. Botts, H.A. and Patterson, J.G. (1987). ‘Pension fund investments: an initial geographic assessment’. The Professional Geographer 39: 416–​427.

‘Organic Finance’   607 Bouvard, V., Loomis, D., Guyton, K.Z., Grosse, Y., Ghissassi, F.E., Benbrahim-​Tallaa, L., et al. (2015). ‘Carcinogenicity of consumption of red and processed meat’. The Lancet Oncology 16: 1599–​1600. Burton G.M. (2013). ‘Asset management fees and the growth of finance’. Journal of Economic Perspectives 27: 97–​108. Campbell, J.Y. and Viceira, L.M. (2002). Strategic Asset Allocation: Portfolio Choice for Long-​ Term Investors (Oxford: Oxford University Press). Carlin, B.I. and Manso, G. (2011). ‘Obfuscation, learning, and the evolution of investor sophistication’. Review of Financial Studies 24: 754–​785. Carpenter, J.N. (2000). ‘Does option compensation increase managerial risk appetite?’ Journal of Finance 55: 2311–​2331. CEM Insights (2010). ‘Internal management does better after costs’. CEM Benchmarking Inc. http://​www.cembenchmarking.com/​Files/​Documents/​October_​2010_​CEM_​Insights.pdf (last accessed 10 April 2017). Chuprinin, O., Massa, M., and Schumacher, D. (2015). ‘Outsourcing in the international mutual fund industry: an equilibrium view’. Journal of Finance 70: 2275–​2308. Clark, G. (2000). Pension Fund Capitalism (Oxford: Oxford University Press). Clark, G.L. (1998). ‘Stylized facts and close dialogue: methodology in economic geography’. Annals of the Association of American Geographers 88: 73–​87. Clark, G.L. (2005). ‘Money flows like mercury: the geography of global finance’ Geografiska Annaler 87B: 99–​112. Clark, G.L. (2007). ‘Expertise and representation in financial institutions: UK legislation on pension fund governance and US regulation of the mutual fund industry’. Twenty-​First Century Society 2: 1–​23. Clark, G.L. (2008). ‘Governing finance: global imperatives and the challenge of reconciling community representation with expertise’. Economic Geography 84: 281–​302. Clark G.L. and Monk, A.H.B. (2012a). ‘Sovereign wealth funds: form and functions in the 21st century’. Journal of Financial Transformations 33: 17–​27. Clark G.L. and Monk, A.H.B. (2012b). ‘Modernity, imitation, and performance:  sovereign funds in the Gulf ’. Business and Politics 14: 1469–​3569. Clark, G.L. and Monk, A.H.B. (2012c). ‘The scope of financial institutions: in-​sourcing, outsourcing, and off-​shoring’ http://​ssrn.com/​abstract=2016359 (last accessed 10 April 2017). Clark, G.L. and Monk, A.H.B. (2014a). ‘Principles and policies for in-​house asset management’. Journal of Financial Perspectives 1: 39. Clark, G.L. and Monk, A.H.B. (2014b). ‘State and local pension fund governance and the process of contracting for investment services: the scope of diversity and the problem of embeddedness’. Territory, Politics, and Governance 2: 150–​172. Clark, P.B. and Monk, A.H.B. (2015). ‘Sovereign development funds:  designing high-​ performance, strategic investment institutions’ http://​ssrn.com/​abstract=2667974 (last accessed 10 April 2017). Clark, G.L. and Urwin, R. (2008). ‘Best-​practice pension fund governance’. Journal of Asset Management 9: 2–​21. Clark, G.L. and Urwin, R. (2010). ‘Innovative models of pension fund governance in the context of the global financial crisis’. Pensions 15: 62–​77. Clark, G.L. and Wójcik, D. (2007). The Geography of Finance: Corporate Governance in the Global Marketplace (Oxford: Oxford University Press). Clark, G.L. and O’Connor, K. (1997). ‘The Informational Content of Financial Products and the Spatial Structure of the Global Finance Industry’ in K. Cox (ed.) Spaces of Globalization, pp. 89–​114 (New York: Guilford).

608   Monk and Sharma Clark, G.L., Gertler, M.S., and Feldman, M.P. (2003). The Oxford Handbook of Economic Geography (Oxford: Oxford University Press). Clark, G.L., Monk, A.H.B., Orr, R., and Scott, W. (2012). ‘The new era of infrastructure investment’. Pensions 17: 103–​111. Cornell, B. (2009). ‘Luck, skill, and investment performance’. The Journal of Portfolio Management 35: 131–​134. Coval, J.D. and Moskowitz, T.J. (2001). ‘The geography of investment: informed trading and asset prices’. Journal of Political Economy 109: 811–​841. Davidsson, P., Delmar, F., and Wiklund, J. (2006). Entrepreneurship and the Growth of Firms (Cheltenham: Edward Elgar Publishing). Dixon A.D. and Monk, A.H.B. (2012). ‘Rethinking the sovereign in sovereign wealth fund’. Transactions of the Institute of British Geographers 37: 104–​117. Dixon, A. and Monk, A.H.B. (2013). ‘Reconciling transparency and long-​term investing within sovereign funds’. Journal of Sustainable Finance 2: 275–​286. Dixon, D. and Monk, A.H.B. (2014). ‘Frontier finance’. Annals of the Association of American Geographers 104: 852–​868. Ellis, C.D. (2012). ‘Investment management fees are (much) higher than you think’. Financial Analysts Journal 68: 4–​6. Elton, E.J. and Gruber, M.J. (1997). ‘Modern portfolio theory, 1950 to date’. Journal of Banking and Finance 21: 1743–​1759. Fama, E.F. and French, K.R. (2010). ‘Luck versus skill in the cross‐section of mutual fund returns’. The Journal of Finance 65: 1915–​1947. Fang, L.H., Ivashina, V., and Lerner, J. (2015). ‘The disintermediation of financial markets: direct investing in private equity’. Journal of Financial Economics 116: 160–​178. Faulconbridge, J. and Muzio, D. (2009). ‘The financialization of large law firms: situated discourses and practices of reorganization’. Journal of Economic Geography 9: 641–​661. Faulconbridge, J., Engelen, E., Hoyer, M., and Beaverstock, J. (2007). Analysing the changing landscape of European financial centres: the role of financial products and the case of Amsterdam. Growth and Change 38: 279–​303. Foster, D.P. and Young, H.P. (2010). ‘Gaming performance fees by portfolio managers’. The Quarterly Journal of Economics 125: 1435–​1458. French, S., Leyshon, A., and Wainwright, T. (2011). ‘Financializing space, spacing financialization’. Progress in Human Geography 35: 798–​819. Gertler, M.S. (2003). ‘Tacit knowledge and the economic geography of context, or the undefinable tacitness of being (there)’. Journal of Economic Geography 3: 75–​99. Graham, J.R., Harvey, C.R., and Huang, H. (2009). ‘Investor competence, trading frequency, and home bias’. Management Science V55: 1094–​1106. Graham, S. and Marvin, S. (2001). Splintering Urbanism: Networked Infrastructures, Technological Mobilities, and the Urban Condition (London: Routledge). Greenwood, R. and Scharfstein, D. (2013). ‘The growth of finance’. Journal of Economic Perspectives 27: 3–​28. Guthman, J. (2003). ‘Fast food/​organic food: reflexive tastes and the making of “yuppie chow”.’ Social & Cultural Geography 4: 45–​58. Hagerman, L.A., Clark, G.L., and Hebb, T. (2007). ‘Investment intermediaries in economic development: linking public pension funds to urban revitalization’. Community Development Investment Review 3: 45–​65. Harris, R.S., Jenkinson, T., and Kaplan, S. N. (2014). ‘Private equity performance: what do we know?’ Journal of Finance 69: 1851–​1882.

‘Organic Finance’   609 Hawley, J.P. and Williams, A.T. (2000). The Rise of Fiduciary Capitalism:  How Institutional Investors can Make Corporate America More Democratic (Philadelphia, PA: University of Pennsylvania Press). Hirshleifer, J. (1971). ‘The private and social value of information and the reward to inventive activity’. The American Economic Review 61: 561–​574. Hochberg, Y.V. and Rauh, J.D. (2013). ‘Local overweighting and underperformance: evidence from limited partner private equity investments’. Review of Financial Studies 26: 403–​451. Ibragimov. R., Jaffee, D., and Walden, J. (2011). ‘Diversification disasters’. Journal of Financial Economics 99: 333–​348. International Monetary Fund (2014). ‘World economic outlook’ http://​www.imf.org/​external/​ pubs/​ft/​weo/​2014/​01/​ (last accessed 10 April 2017). Ivkovic, Z. and Weisbenner, S. (2005). ‘Local does as local is: information content of the geography of individual investors’ common stock investments’. The Journal of Finance February: 267–​306. Jennings, W.W. and Payne, B.C. (2016). ‘Fees eat diversification’s lunch’. Financial Analysts Journal 72: 31–​40. Jurek, J.W. and Stafford, E. (2011). ‘Crashes and collateralized lending’ (No. w17422), National Bureau of Economic Research. Khan, M., Serafeim, G., and Yoon, A. (2015). ‘Corporate sustainability: first evidence on materiality’. Harvard Business School Working Paper, No. 15-​073, March 2015. Kinnel, R. (2010). ‘These funds have scary fees’. Morningstar Fund Investor 18: 1–​3. Kortum, S. and Lerner, J. (2000). ‘Assessing the contribution of venture capital to innovation’. RAND Journal of Economics 31: 674–​692. Knight, E.R.W. and Sharma, R. (2015). ‘Infrastructure as a traded product: a relational approach to finance in practice’. Journal of Economic Geography 16: 897–​916. Lazear, E.P. (2000). ‘The future of personnel economics’. The Economic Journal 110: 611–​639. Lee, R., Clark, G.L., Pollard, J., and Leyshon, A. (2009). ‘The remit of financial geography—​ before and after the crisis’. Journal of Economic Geography 9: 723–​747. Leiblein, M.J., Reuer, J.J., and Dalsace, F.E. (2002). ‘Do make or buy decisions matter? The influence of organizational governance on technological performance’. Strategic Management Journal 23: 817–​833. Lo, A. (2012). ‘Adaptive markets and the new world order’. Financial Analysts Journal 68: 18–​29. MacIntosh, J. and Scheibelhut, T. (2012). ‘How large pension funds organize themselves: findings from a unique 19-​fund survey’. Rotman International Journal of Pension Management 5: 34–​40. McKinsey (2012). ‘Infrastructure productivity. How to save $1 trillion a year’ http://​www. mckinsey.com/​industries/​capital-​projects-​and-​infrastructure/​our-​insights/​infrastructure-​ productivity (last accessed 10 April 2017). Maurer, B. (2008). ‘Re-​regulating offshore finance?’ Geography Compass 2: 155–​175. Monk, A.H.B., Kearney, S.W., Seiger, A., and Donnelley, E. (2015). ‘Energizing the US resource innovation ecosystem: the case for an aligned intermediary to accelerate GHG emissions reduction’ http://​ssrn.com/​abstract=2617816 (last accessed 10 April 2017). Murphy, K.M., Shleifer, A., and Vishny, R.W. (1991). ‘The allocation of talent: implications for growth’. The Quarterly Journal of Economics 106: 503–​530. Neil, D. and Warren, G. (2015). ‘Long-​term investing as an agency problem’. Working Paper #063/​2015, Project # T003, June. Novy-​Marx, R. and Rauh, J.D. (2009). ‘The liabilities and risks of state-​sponsored pension plans’. The Journal of Economic Perspectives 23: 191–​210.

610   Monk and Sharma O’Neill, P. (2009). ‘Infrastructure Investment and the Management of Risk’ in Clark, G., Dixon, A., and Monk, A. (eds) Managing Financial Risks: From Global to Local, pp. 163–​188 (Oxford: Oxford University Press). Phalippou, L. (2010). ‘Venture capital funds: flow-​performance relationship and performance persistence’. Journal of Banking & Finance 34: 568–​577. Phalippou, L. and Gottschalg, O. (2005). ‘Performance of private equity funds’. EFA 2005 Moscow Meetings http://​ssrn.com/​abstract=473221 (last accessed 10 April 2017). Pike, A. and Pollard, J. (2010). ‘Economic geographies of financialization’. Economic Geography 86: 29–​51. Robinson, D.T. and Sensoy, B.A. (2013). ‘Do private equity fund managers earn their fees? Compensation, ownership, and cash flow performance’. The Review of Financial Studies 26: 2760–​2797. Sampsa, S. and Sorenson, O. (2011). ‘Venture capital, entrepreneurship and economic growth’. The Review of Economics and Statistics 93: 338–​349. Sato, Y. (2014). ‘Opacity in financial markets’. Review of Financial Studies 27: 3502–​3546. Seyfang, G. (2006). ‘Ecological citizenship and sustainable consumption:  examining local organic food networks’. Journal of Rural Studies 22: 383–​395. Sharma, R. (2012). ‘Infrastructure: an emerging asset class for institutional investors’. Working paper presented at The Societal Function of Investment Asset Classes:  Implications for Responsible Investment Conference, Initiative for Responsible Investment at Harvard University. Sharpe, W.F. (2013). ‘The arithmetic of investment expenses’. Financial Analysts Journal 69: 34–​41. Sheffer, D.A. and Levitt, R.E. (2010). ‘How industry structure retards diffusion of innovations in construction: challenges and opportunities.’ CRGP Working Paper #0059, 2010. Spence, M. (2002). ‘Signaling in retrospect and the informational structure of markets’. The American Economic Review 92: 434–​459. Starks, L.T. (1987). ‘Performance incentive fees:  an agency theoretic approach’. Journal of Financial and Quantitative Analysis 22: 17–​32. Storper, M. and Venables, A.J. (2004). ‘Buzz:  face-​to-​face contact and the urban economy’. Journal of Economic Geography 4: 351–​370. Timmons, J. and Bygrave, W.D. (1986). ‘Venture capital’s role in financing innovation for economic growth’. Journal of Business Venturing 1: 161–​176. Torrance, M. (2009). ‘The rise of a global infrastructure market through relational investing’. Economic Geography 85: 75–​97. Wainwright, T. (2011). ‘Tax doesn’t have to be taxing: London’s “on shore” finance industry and the fiscal spaces of a global crisis’. Environment and Planning A 43: 1287–​1304. Weyl, E.G. and Posner, E. (2012). ‘A proposal for limiting speculation on derivatives: an FDA for financial innovation’. Coase-​Sandor Working Paper Series in Law and Economics http://​ chicagounbound.uchicago.edu/​cgi/​viewcontent.cgi?article=1012&context=law_​and_​economics (last accessed 10 April 2017). Wójcik, D. (2011). The Global Stock Market: Investors and Intermediaries in an Uneven World (Oxford: Oxford University Press). Wójcik, D. (2013). ‘Where governance fails: advanced business services and the offshore world’. Progress in Human Geography 37: 330–​347. World Bank (2015). ‘Institutional investors:  the unfulfilled $100 trillion promise’. Policy Research Presentation by Sergio Schmukler.

Chapter 32

Financiali z at i on of Every day L i fe Karen P.Y. Lai Introduction The global financial system, the influence of financial markets on corporate governance and national policies, and the broader implications of financial logics for governing economic and social lives have recently become more important in economic geography research (Hall, 2012). This is particularly evident in the traction gained by the concept of financialization among geographers and other social scientists as a way of describing the growing power of financial markets and financial institutions in economic, political, and social life (see Engelen, 2008; Pike and Pollard, 2010; French et  al., 2011). Studies range from how the finance sector dominates national political economies (Blackburn, 2006; Dore, 2008), to how firm strategies and management are increasingly beholden to the logics of finance (Williams, 2000; Froud et al., 2002; Krippner, 2005), and the ways in which households and individuals are tied into increasingly complex relationships with the international financial system (Martin, 2002; Langley, 2008a). In most conceptions, the adoption of financial logics is rendered either as an exogenous shock to political economies, as they are ‘captured’ by finance and financial actors, or as an unintended consequence of deregulation and changes in monetary policy in order to attract capital (see Streeck and Thelen, 2005; Krippner, 2012). This shift in financial logic and behaviour is also evident at the level of individuals and households. Habits of saving and financial planning have transformed over time as everyday lives and life cycles (relating to housing, consumption, health, retirement, death, etc.) are increasingly tied to the performance of financial markets, with individuals seeking market solutions for personal life goals and future security. While this chapter focuses on everyday financialization, studies at the macro-​institutional level and firm level are useful in contextualizing the processes and impacts of financialization as they unfold in the realms of everyday life.1 Institutional studies that interrogate the interdependent relations between financialization and neoliberal governments demonstrate how multiple rounds of deregulation and government support fuelling the growth of finance industries have reified financialization as the ideal technique of governance. Firm-​level

612   Lai studies from critical social accountancy also demonstrate how the increasing power of shareholders and their desire to raise corporate value has pressured management to seek out wealth creation in non-​traditional venues, such as financial and property markets, rather than through production or innovation (Froud et al., 2000, 2006). This ‘narrative of numbers’, in which key financial indicators such as shareholder value become prominent metrics of success, has resulted in distinctive changes in firm behaviour and corporate governance, away from core business activities and towards financial investments and indicators. Financial markets, instruments, and logics are increasingly framed as solutions for governments and firms in solving budgetary crises or in seeking new growth strategies (Dymski, 2009a; du Gay et al., 2012). This increasingly privileged position of finance promotes new forms of calculative and competitive economic behaviour in the contemporary neoliberal era (Duménil and Lévy, 2004; Larner, 2012). Other than analysing the processes and impacts of financialization on firms and regions (Pike and Pollard, 2010; Coe et  al., 2014), geographers have been particularly influential with respect to culturally inflected sociological research on how finance shapes everyday life within contemporary capitalist societies. The operation and impacts of financialization at the individual level reflects how increasing consumption of financial products and the growing acceptance of financial logics in the context of dwindling state-​welfare benefits normalizes risks and risk-​taking behaviour (Martin, 2002). Individuals adopt new modes of self-​governance and reflexivity to monitor their investments and consumption habits. Changing practices of borrowing and saving are also seen in both the rise of credit card and other debts, and in the channelling of savings into insurance and investment products rather than conventional bank deposits. Changing state policies, new technologies on credit scoring and securitization, and the rise of middle-​class consumers in developing economies are also changing the nature and impacts of financial consumption and financialized behaviour. While some view this as a democratization of finance and investment to a broader public (i.e. a growth market), others see it as the creation and extension of new risks with spatially uneven impacts. The financialization of everyday life involves the making of finance capitalism through particular narratives, actors, and technologies that emphasize individual responsibility, risk taking, and calculative assessment in managing personal financial security and well-​ being. The 2008 global financial crisis has highlighted the profound changes to relationships between households and global financial markets. However, through more than just predatory lending practices to vulnerable households and communities, the unfolding and impact of the subprime crisis demonstrate a much broader expansion of financial power, in which individual subjectivity, aspiration, and forms of conduct at an individual level are directly linked to global financial structures. This calls for more systematic and incisive analyses into the household and its constituent elements in the construction and mobilization of financialized behaviour and outcomes. The rest of this chapter is divided into three sections that trace the historical growth of financialization at the household level, assess the current research approach that emphasizes financial subjects and governmentality, and suggest a renewed engagement with the state as a vital and strategic actor in financialization for future research directions. The conclusion reflects on the role of cultural economy in studying the financialization of everyday life and broader issues of equitability and sustainability.

Financialization of Everyday Life    613

New Intermediaries of Finance The financialization of everyday life has been accelerated by important technological and institutional developments over the last few decades, although its beginnings could be traced further back. Since the mid-​1800s, industrial life assurance and cheque trading (known collectively as ‘doorstep finance’) were based on sales and service by door-​to-​door agents and lenders with weekly collection of contributions towards small life-​assurance policies and the issuance of credit cheques. These were established in Victorian Britain and the USA to provide the means to certain forms of consumption for the growing working class (Zelizer, 1983, 2011; McFall, 2015). In the early years, the life-​assurance industry grew out of the demand for funeral coverage, as it constituted a large expense for one of the most important ritual events in the life cycle. Over time, policies expanded to include endowment and pension plans that could be used to pay for other important life events such as weddings, anniversaries, and big-​consumption items like houses, cars, and other consumer products. While operating on a much smaller scale than industrial life assurance, cheque trading or door-​step lending also played an important role in changing the consumption patterns of individuals and households, as the credit issued (in the form of cash cheques that carried a fee plus interest) enabled poorer households to buy subsistence goods initially but was later utilized for larger consumer items. Such forms of financial services continue to operate in particular regions and neighbourhoods that have been excluded by mainstream financial systems, as those households often do not fulfil required credit ratings (Leyshon et al., 2004). The impact of information technology on the intermediation of financial products and services has been particularly influential in the growth of credit scoring. Over the last fifty years, a technocratic, statistical expertise has been gradually applied by lenders to regulate the problem of default by borrowers, which, in practice, reframes consumers and their attributes as various forms of ‘legible’ and calculable risks, such that they become amenable to new forms of government (Marron, 2007). Whereas credit used to be granted and managed by individual retailers such as large department stores and mail-​order companies, the postwar boom saw the entry of financial institutions into the profitable provision of credit for immediate personal consumption (rather than long-​term purchase with collateral such as property). The issuance of general bank credit cards represented a new form of mass consumer credit and required new ways of calculating and managing information and risks in the form of categorical, quantified data. Credit scoring and the process of constituting risk provided the lender with new means of understanding and managing individual consumers by stringing together commercial considerations of default, operational costs of the firm and standardization of credit approval procedures. Moreover, the ‘objectivity’ produced in scoring also enabled lenders to deploy statistical models as a means for refuting claims of unlawful discrimination in credit granting. The use of credit-​scoring technologies became ever more pervasive with the widespread adoption of credit cards among Anglo-​American consumers, the growth in computing power for statistical modelling and the electronic storage and management of data. These, in turn, drove lenders to pursue an ever-​larger customer base for economies of scale and contributed to increased consumption and household debt over the last two decades (McFall, 2008; Langley, 2009; Marron, 2009).

614   Lai The conceptualization and management of credit risk took on a powerful narrative as it became colonized by new credit bureaux and credit consultancy firms. These are the electronic repositories of the credit histories of almost all credit consumers in the country of jurisdiction, derived from the records of all mainstream consumer lenders (be they financial institutions or department stores), which are, in turn, used as a resource by lenders in guiding credit assessment (Leyshon and Thrift, 1999). Credit scores are thus transformed into a commodity that can be sold to lenders who, for whatever reason, would rather not formulate risk-​assessment models of their own. Concerns regarding defaults by consumers have also created new measurements for over-​indebtedness that connect individuals’ attributes and life events, such as unemployment, illness, marital changes, age, education, and number of children, to their credit ratings (Marron, 2012; Deville, 2015). The construction of individuals as quantifiable risks has also become entangled with broader uncertainties experienced by financial institutions at large as they trade entire portfolios of loans encompassing an array of consumers and credit agreements. This is done through the process of securitization, another important technology of intermediation that has become particularly prominent over the last two decades. Securitization takes an illiquid asset or groups of assets and repackages them into a tradable form of security, which could then be moved off the balance sheet of the issuing entity (thus improving its financial position and enabling new rounds of accumulation). The ways in which securitization and financial engineering created new forms of relationships between households and the larger financial system, and new risks, came under the spotlight during the 2008 subprime mortgage crisis and credit crunch (Langley, 2008b; Christophers, 2009; Aalbers, 2009a). In the past, loans were funded primarily from savings that went into financial institutions. This was seen as limiting as financial markets could provide cheaper and more available sources of funds compared with savings alone. Securitization thus enabled mortgage lenders to sell their mortgage portfolio on secondary mortgage markets to investors. In the same process, those mortgages were taken off the balance sheets of mortgage lenders, which frees up more equity for more loans. However, secondary mortgage markets are global markets, which means that a crisis of mortgage securitization soon affected institutions, investors, and economies around the world—​from Chinese sovereign wealth funds to German pension funds and from Swiss investment banks to Singaporean municipal councils (Aalbers, 2009b; Martin, 2010). Through securitization, housing has become an electronic instrument for high-​risk finance (Sassen, 2009), as well as the basis for the creation of new topographies of race and class on the urban landscapes (Dymski, 2009b; Wyly et al., 2009). The ‘calculating tools’ and ‘technical devices’ of credit scoring, computer technologies, statistical models, and securitization have therefore been crucial intermediaries of financialization in the assembly of new consumer markets, and the entanglement of daily lives, life events, and livelihoods into contemporary financial markets. Financial advisors are also key intermediaries in the financialization process. More than just connecting the supply and demand of financial products, financial advisors shape the financial knowledge and investment practices of consumers and their modes of articulation into capital markets. Finance is performed on a daily basis not only by investors, but also by financial institutions, managers, marketing professionals, and political actors (Clark et al., 2004). More than just standardized forms of technical expertise, financial advisors perform and legitimize new cultural circuits of (financialized) capitalism (Thrift, 2005) that help assemble particular kinds of investor subjects for contemporary systems of

Financialization of Everyday Life    615 accumulation. Financial advisors themselves can also be seen as knowing subjects (Larner, 2012) governed by different modes of corporate management, industry structures, remuneration structures, and incentivizing schemes that result in variegated encounters and practices with clients. While this is not the place for an in-​depth case study, an illustrative discussion of the industry structure and professional practice of the financial advisory sector in Singapore could provide some useful insights into these intermediaries of financialization. Financial advisors for the mass market in Singapore are largely divided into three groups: insurance agents based in major insurance companies; wealth managers or relationship managers employed by retail banks; and independent financial advisors affiliated with independent financial advisory firms. Other than the basic function of advising clients on financial planning, product information, and transactional services, all three types of financial advisors operate under distinctive corporate environments. Only relationship managers with banks are salaried employees, while insurance agents and independent financial advisors are self-​employed, and they are completely reliant on commission and other incentives for remuneration. The actual range of financial products that financial advisors can recommend and provide transactional services for are surprisingly limited and dependent on their affiliated companies. They range from only in-​house products to an extensive list of licensed external products. This provides highly variable motivation for tailoring financial advice and sales tactics to clients depending on product availability and incentive structures (tied to overall sales targets or product-​specific bonuses). The types of financial advisors that potential clients are likely to engage is also influenced by new or existing banking relationships (in the case of relationship managers) or family and other personal networks (more often the case with insurance agents and independent financial advisors), which, in turn, affects the financial advice (given whether more insurance-​or investment-​oriented) and the actual financial investments. Taken together, different types of financial advisors operate under distinctive forms of employment, remuneration and incentives structures, licensed products, client base, and institutional reputation. These often have a direct impact on their professional practice in terms of the financial planning process, product recommendations, and sourcing for new clientele. Instead of adhering to principles of financial protection and catering to clients’ life stages and financial goals, financial advisors may lean on established reputational effect of their companies to sell only particular types of products or they may follow existing preferences of clients to achieve quick sales, rather than deliberately ‘educating’ clients on more comprehensive financial planning and less popular products for better risk management. The financial advisory sector in Singapore hints at the complex ways in which consumers are drawn into different forms of financial relationships and their uneven access to information and resources. The financial system can be recast as a coalition of smaller constitutive ecologies, such that distinctive groupings of financial knowledge and practices emerge in different places with uneven connectivity and material outcomes. Instead of a democratization of finance, whereby financial products and services are made available to mass consumers and individuals have greater freedom to protect against the uncertainties of life through financial planning, such financial ecologies reveal the partial and uneven process of financialization through key intermediaries like financial advisors (French et al., 2011; Lai, 2016).

616   Lai

Financial Subjects and Governmentality The ways in which discourses of risk-​taking and self-​management have shaped the behaviour of individuals and households constitute one of the most vibrant areas of research in the financialization of everyday life. The shift towards financial markets for the provision of people’s daily needs and longer-​term security and well-​being is facilitated by specific narratives that emphasize individual responsibility, the normalization of risk and calculative assessment in financial management, and, by extension, the management of life stages and life goals. This concerns not only the material outcomes of financialization (in terms of new growth markets in insurance and investment products and increased financial flows), but also its impacts on the subjective understandings of one’s role within the political economy and a convergence of finance and the life cycle (Cutler and Waine, 2001; Martin, 2002; French and Kneale, 2009; Zelizer, 2011). Individuals adopt new modes of self-​governance and reflexivity to monitor their investments and consumption habits. Related studies on behavioural geographies have mapped the ways in which the wider financial environment shapes understandings of ‘rationality’ in investments and retirement strategies made by British households and individuals (Strauss, 2008; Clark, 2010; Clark, 2011). Wider processes of financialization are thus underpinned by the promotion of new forms of economic behaviour in the contemporary neoliberal era (Larner, 2012). Langley’s (2006) seminal work demonstrates how neo-​liberal governments in the USA and UK encourage citizens to participate actively and invest in financial products for retirement, ultimately legitimizing state reductions in pensions benefits (Finlayson, 2009a). This draws from Foucault’s notion of governmentality—​how states regulate behaviour ‘at a distance’ through discursive production of knowledge and techniques of self-​governance (Barnett, 2001) that motivate subjects to ascribe voluntarily to self-​disciplinary ways in order to achieve ‘rationality’. Financial planning becomes a form of biopower whereby investor subjects are mobilized to plan, calculate, and invest wisely to fulfil and secure their future well-​being (Langley, 2008a). In the process of producing these new subjects of sophisticated, calculative investors, financial risk is effectively reshaped into something that is manageable by individuals through wise and calculative ‘technologies of the self ’ (Langley, 2006). Through discourses of ‘personal responsibility’ and ‘self-​sufficiency’ produced by state-​ sponsored financial literacy programmes, individuals are normalized as responsible for their own financial well-​being (Martin, 2002). By focusing on risk taking and self-​management, scholars identify the formation of the ‘financial subject’ or ‘investor subject’ (Langley, 2006; Aitken, 2007; Langley and Leyshon, 2012) who insures himself/​herself against the risks of the life course through self-​disciplined financial practices. Neo-​liberal policies and associated banking practices, discourses, and instruments frame people as rational and responsible subjects who are expected to take care of their financial futures and assume individual responsibility for their own welfare and financial security. Technologies such as credit scoring, financial profiling, and pension fund reforms prompt consumers to internalize these market logics and to become self-​governing subjects. Financial logic thus enters everyday life through the discourses, regulations, financial instruments, and technological devices that compel people to enact financial decisions and

Financialization of Everyday Life    617 practices by allowing or disallowing particular actions or subjectivities. Risk, in the form of accepting increasing uncertainty in everyday life, is also rescaled from the state to the individual as it motivates financial subjects to take charge of and fund their future financial needs. However, the formation of financial subjects under neo-​liberal programmes of government is often contingent and contested (Erturk et al., 2007; Finlayson, 2009b). Increased anxiety and uncertainty over investments and returns may drive individuals to retreat to the safety of savings accounts, thus departing from the definition of the investor as a clearly defined and unproblematic subject position performed by rational and financially literate individuals (Langley, 2007). Investors may choose risk-​averse investment options, such as low-​performing capital-​guaranteed funds or bonds, or make investment decisions based on reputation of the issuer/​distributor and long-​standing custom with a financial institution, rather than through a calculative assessment of risk and returns (Lai, 2013). Investors could also reject financial-​market investments altogether, in favour of yields from property investment (Leyshon and French, 2009). As such, the nature of financial subject formation and of financialization itself is necessarily contested and incomplete, as financial subjects engage in different and sometimes contradictory sets of financial practices. While much of the work on financial subject formation has focused on the investor, that is, the reshaping of the passive, saver subject into an active and entrepreneurial investor subject, a smaller but growing body of work analyses the formation of biofinancial subjects, which concerns politics of care and management of the body and of life (French and Kneale, 2009). Biofinancialization, the intersection of financialization and biopolitics, refers to the ways in which contemporary processes of financialization and of the politics of life itself (Rose, 2007) intersect in new ways to produce distinctive relationships between capital and health/​bodies/​life cycles/​other bodily experiences and aspects of life and living (French and Kneale, 2012). Biofinancialization introduces a culture of (financial) valuation into everyday life, such that the worth of activities, bodies, health, and of life itself can be translated into financial evaluations, which subsequently impacts on how individuals modify their behaviour and lifestyles. In this sense, financial value espouses the primacy of investment value over other values (e.g. aesthetic, moral, ecological, cultural), such that there is future monetary profit to be gained from potentially any aspect of life and of living (Lilley and Papadopoulos, 2014). Using the frame of biopolitics and affect, French and Kneale (2009, 2012) examine the ways in which the rationale of lifestyle and habits leads to a reworking of life assurance and annuity in the UK. Biopolitical metrics such as the body mass index and alcohol units are enrolled and mobilized by the insurance industry to influence the behaviour and lifestyles of individuals as the (financial) value of their lives become bound up with these new forms of government. Markers or traces of disease or morbidity on the body, or the anticipation of such markers as in the case of lifestyle factors, become targets of calculation and discipline. Rather like the development of credit-​scoring technologies that enable the constructions of individuals as quantifiable risks, the entanglement of insurance, medical, and regulatory knowledges gives rise to the capture, sorting, and ranking of (projected) morbidity such that the (financial) value of individuals could be calculated and mobilized in terms of premiums, payouts, and exclusions. Innovations in health insurance and other forms of financial provision thus create new environments in which responsible subjects with ‘desirable’ lifestyles could be assembled (Guthman and Dupuis, 2006; French and Kneale, 2012).

618   Lai There has been some scepticism as to whether the everyday consumer possesses the level of financial literacy or even self-​awareness of their own financial status and financial goals to make informed decisions about financial planning (Erturk et al., 2007; Finlayson, 2009b). Financial education and literacy or charges of ‘irrationality’ are, however, insufficient to explain the variegated financialization of households and individual subjects. Understandings of financial freedom and financial security are not only constituted by economistic calculations set against average life expectancies; sociocultural constructs often underpin everyday attitudes towards money (Zelizer, 1993, 1994; Maurer, 2006), with value judgements being bound up in financial decisions and investment practices. The framing of ‘freedom’ and ‘security’ by financial subjects themselves requires deeper interrogation in order to explain what might appear to be irrational, passive, or contradictory financial practices, but which could well be appropriate and persuasive when viewed outside of a neoliberal governmentality frame. People ‘inhabit multiple subject positions within a financial ecology in ways that conform, diverge and subvert neoliberal versions of the responsible, financially self-​disciplined individual’ (Coppock, 2013, p. 479). The actual nature of what actually constitutes responsible and self-​disciplined financial subjects could also change over time through embodied, emotional, and socially inflected processes, rather than through rational and calculative practices (Deville, 2012, 2015). All these point to the value of a critical engagement with money cultures in examining financial subjects and practices (Gilbert, 2005; Maurer, 2006; Deville and Seigworth, 2015).

The State and/​in Financialization The analysis of state and state actors within financialization studies tends to focus on financial deregulation and its impacts on institutional change, firm behaviour, and everyday habits of savings and borrowing (van der Zwan, 2014). These approaches tend to emphasize market imperatives and neo-​liberal logics rather than consider how financialization of the economy could be a deliberate pathway sought by state actors and policymakers. The role of the state has been largely set within the context of neo-​liberalization, with market efficiency and financial logics justifying the rolling back of state functions and devolving state responsibilities for social provision to individuals and households (Dore, 2000; Cutler and Waine, 2001; Martin, 2002; Langley, 2008a). The emphasis on market imperative, however, obscures the strategic ways in which the state actively mobilizes institutions, firms, and households to adopt and enact financialization scripts for political economic purposes. More recently, some scholars have suggested that far from ‘retreating’ or ‘declining,’ the state has taken on qualitatively different roles in its relationships with financial markets, financial institutions, and non-​financial firms. This approach focuses on financialization of the state itself, as state actors and institutions turn to financial markets as solutions in the face of economic (and political) crises such as budget deficits or economic recessions (Aalbers, 2009a; Bassens et al., 2013; Hendrikse and Sidaway, 2014). Even before the 2008 global financial crises, property and mortgages have featured prominently in the financialization of everyday life. As part of efforts to roll back the postwar welfare state in the USA (and later on in the UK), residential property ownership became a core component of asset-​ based welfare under the new neo-​liberal regime. Real increases in incomes were held down

Financialization of Everyday Life    619 in order to combat inflation, leading to the demand for other forms of credit in order to sustain standards of living. Deregulation of the financial system during the 1980s encouraged financial institutions to compete in extending credit to consumers. Of particular relevance here is the practice of borrowing against the appreciation of the value of residential property, thereby bringing forward the projected future gains from the sale of an asset. Properties are thus increasingly mobilized to offset declines in real wages and to sustain lifestyles and consumption (French and Leyshon, 2012). This has created wider socio-​economic problems owing to the reluctance of governments to tackle the unsustainable housing boom and property price bubbles that have served to fuel consumer-​led growth over the past decades (Hay, 2009; Montgomerie, 2009). It might be instructive to investigate more closely the role of the state in driving financialization, instead of treating the state as a distant or reactionary actor in providing the background of deregulation amidst neo-​liberalizing pressures (i.e. conceptualizing the roles of the state and the process of financialization as interconnected but separate fields, while focusing analytical attention towards the transformation of everyday life/​living at the individual or microscale). This requires a deeper interrogation of state–​subject relations in the mobilization of financialization processes and financial subject formation and the motivations for the state in promoting financial subject formation, as these have important implications for the modes and outcomes of state-​led financialization. Individual consumers are financial subjects who not only fulfil the neo-​liberalized scripts of self-​reliant, disciplined, and responsible subjects who take care of their own financial futures; they can also be mobilized as citizen-​subjects to build a stronger and more competitive national economy through their changing financial practices, even as they benefit from greater access to financial products and services. In turn, the state is able to achieve particular developmental goals, ensure its own economic and political viability in a competitive global environment, and bolster the ruling government’s political legitimacy. This relationship between popular finance and the forging of a ‘national economy’ has been examined by Aitken (2007), who demonstrates how specific financial instruments (e.g. US Savings Bonds, New York Stock Exchange mass-​investment programme) are used in the inter-​and postwar periods for patriotic purposes. Working-​class individuals are thus enrolled into financial practices through which the national economy could be made real. Aitken’s primary objective is to deconstruct how knowledge and practices of the national economy are assembled such that the notion and operation of a ‘national’ economy’ is constructed and mobilized. His findings, however, also signal how individual financial practices play vital roles in securing the nation state through economic development. Instead of viewing financialization as primarily driven by market processes, a more developmental perspective could uncover a different set of dynamics connecting state, institutions, and individuals in the changing roles of financial logics in everyday life, and how those are embedded in broader political economic objectives. In contextualizing the transformation of saver subjects to investor subjects, the concept of financial citizenship could serve as a useful tool to analyse shifting state–​subject relations in the financialization process. The term ‘financial citizenship’ first emerged in the mid-​1990s, during a period of massive restructuring of retail banking and extensive bank branch closures in the US and UK. The decline of small British rural communities and towns, due, in part, to a lack of access to banking services, led to the rise of the concept of financial citizenship, which champions access to financial services as a basic right tied with citizenship (Leyshon et al., 2008).

620   Lai Leyshon and Thrift (1995, 1997) argue that rather than just being a matter of market adjustment to consumer demand, states need to reform national financial systems such that they are inclusionary, rather than exclusionary, in providing the basic financial services necessary for individuals’ meaningful involvement in contemporary economic life. This concept of financial citizenship was later expanded beyond the national scale to examine at a global scale how access to financial services and mobility of capital are marked by class differences and differential transaction costs between the rich and poor (Dymski and Li, 2003; Dymski, 2005). This line of research critiques the inability of households and small businesses to access a full range of depository and credit services at competitive mainstream prices, rather than through subprime or informal credit facilities, with further research focusing on the issue of financial exclusion and bifurcated markets between formal/​informal financial services for the rich/​poor (Leyshon et al., 2008; Appleyard, 2011; Coppock, 2013). Beyond addressing the politics of financial inclusion/​exclusion and spatial inequality in access to financial services, the concept of financial citizenship has the potential to deepen our understanding of the intersections between the state, institutions, and individuals in financialization by focusing on a more geopolitical reading of financial citizenship (Lai and Tan, 2015). This emphasis on the ‘citizen-​subject’ highlights the active and evolving relationship between the state and financial subjects in the financialization of everyday life. While financial subjects may be understood as knowledgeable investors (Martin, 2002) or uncertain subjects (Langley, 2007), they are also citizen-​subjects with certain (financial) rights, privileges, and duties framed within the geopolitical frame of the nation state. This conceptualization of financial citizenship explicitly places state–​citizen relations as the nexus through which financial landscapes are shaped, (re)produced, and contested through the assemblage of institutional change and financial subject formation. A ‘citizenship’ reading of financial subjects also highlights the ways in which individuals are incorporated into financial systems in ways that may fulfil the broader strategic objectives of the state to pursue economic growth and secure legitimacy. Drawing upon the British’s government responses to the 2008 financial crisis, Brassett and Vaughan-​Williams’ (2012) study utilizes this understanding of financial citizenship to demonstrate how British savers, firms, and bankers in the City of London were framed as victims requiring state assistance. The policy interventions taken by the British government were intended to safeguard the financialization processes in Britain and maintain London’s position as a leading international financial centre. Similarly, Pathak (2014) draws upon the governance of morality to investigate how the British government reframed indebtedness and responsible behaviour. Indebtedness caused by unemployment or by reckless spending were deemed individual faults that should be avoided, but indebtedness resulting from mortgages was not branded as irresponsible, as mortgages were key to the government’s asset-​based welfare policies. The financial subject position of individuals must therefore be contextualized within the state’s political economic framing of the intended roles played by individuals and households in the national economy in order to achieve particular political economic goals. While emphasizing the empirical significance and theoretical relevance of state-​led financialization for understanding financial subject formation, the above agenda has been supported by wider trends in banking over the last three decades. Erturk and Solari (2007) note how ‘interest-​based banking’ has given way to a ‘fee-​based banking’ model for both retail and investment banks in Europe and the USA, with an increased emphasis on fee-​based activities such as wealth management and the sales of financial products (see also Hardie et al.,

Financialization of Everyday Life    621 2013). In the face of banking liberalization and greater competition, banks have increasingly been shifting their business emphasis away from traditional loan mediation and transforming themselves as financial services corporations based on a wider array of fee-​generating activities and deeper participation in financial markets for capital gains. Therefore, while banks have not acted in conjunction with the state for the most part in the rolling out of ever-​more-​financial products and services for everyday consumers,2 state promotion of financialized behaviour in households and individuals has certainly aligned with the banks’ increasing business focus on financial markets and products and the enlargement of non-​ bank financial investments in insurance and investment functions. As financial citizenship focuses on the interactions between state and individuals, it avoids privileging state discourses surrounding risk, management, and responsibility in structuring and producing financial subjects (see Martin, 2002; Langley, 2008a). Focusing on state–​subject interactions also allows the constraints and agencies of both states and financial subjects to be better situated by framing their relationship as co-​constructed and negotiated through networked relations. Looking forward, this allows for a fuller and more nuanced analysis of the relational interactions between state-​driven agendas, institutional agents (e.g. pension funds, insurance firms, banks), and individual–​local narratives and practices in order to uncover the dynamics of how individuals are discursively constructed and incorporated into circuits of finance and financialization and their developmental outcomes. This has particular resonance for a broader theoretical agenda on explicating the spatialities and multiscalar impacts of global financial networks as they unfold across different scales and territories (Coe et al., 2014).

Conclusion While the nascent field of financialization has produced multiple approaches to analysing the growing significance of finance in changing capitalist modes of production, national economies, corporate strategies, and household behaviour, this chapter has focused on the diverse ways in which finance is grounded in the realms of everyday life. The preceding sections have identified key research themes and highlighted some future directions for investigating the processes and impacts of financialization on households and individuals, and how the embodied and lived experiences of the everyday citizen have enmeshed with the making of financial capitalism. Underlying the discussion is an assertion of the household as a key site from which to explore the constructions and practices of financialization. There are three specific areas of future research that could further our understanding of financialization and the ways in which it unfolds through and impacts upon everyday life. Firstly, financial subject formation occurs not only through a neo-​liberal framework of entrepreneurial investors, but also through the frame of intimacies as family and personal relationships, emotions, and care become intimately bound up in decisions about financial commitments and investment practices that concern the life course of not only individuals, but also family members and other dependants. This calls for more serious engagement with ideas of emotions, morality, and care, as they are not just legitimate but vital components to the calculation and production of financial logics and practices by (re)shaping the narratives and practices

622   Lai of households and individuals. The demand for insurance, for investment, and for credit is driven by relationships and by ties of obligation, love, and fear. Investment decisions and financial practices are steeped in the quotidian world of bodies and intimacies, inseparable from habits and routines, dispositions and moods, and the disciplining rhythms of premium payments in the enactment of individual aspirations and household responsibilities (McFall, 2008; Deville, 2015). In analysing the process and impacts of financialization in everyday life, it is therefore imperative that we bring together the financial and the mundane in the analysis of financial market development and financial subject formation. Secondly, structural factors need to be more closely interrogated in examining the variegated processes and impacts of financialization. A focus on households and individuals in financialization does not necessarily mean that structural issues are to be sidelined in favour of cultural economy analysis at the microscale (Hall, 2011).3 As financialization is extending, deepening, and normalizing the reach of financial metrics into households, new patterns of interdependencies and inequalities are being mobilized and carved out along gender, class, age, and other markers of difference with uneven socio-​spatial impacts (Pollard, 2013). A renewed sensitivity to structural issues would be in keeping with the spirit of geographical enquiry in analysing new economic agents and new spatialities of finance (Lai, 2017). One such approach could be through the framework of financial ecologies. Instead of an abstract and monolithic entity, a financial ecologies approach reframes the financial system into a coalition of smaller constitutive ecologies, such that distinctive groupings of financial knowledge, practices, and subjectivities emerge in different places with uneven connectivity and material outcomes (French et al., 2011). This brings into focus how households and individuals are changing their investment practices and are drawn into different financial relationships, as delineated by distinctive sociocultural demographics (Leyshon et al., 2004; Lai, 2013). Financial subjectivities are also being reshaped by broader shifts in financial regulation and changing corporate strategies of financial institutions to draw in everyday consumers (Lai and Tan, 2015; Lai, 2016). A research focus on the household that is situated against broader institutional and regulatory changes can thereby contribute to an economic geography that is better equipped to address broader structural issues of power and the variegated impacts of financialization across different sites of capitalist production and accumulation. Thirdly, we need to re-​engage with the role of the state in financialization, particularly through a reinvigorated focus on state–​subject relations in the mobilization of financial practices. Research into the roles of state policies, agencies, and regulatory power in reshaping the financial subject formation could generate new insights into the dynamics connecting the state, institutions, and individuals in the political economy of financialization. The concept of financial citizenship, for instance, explicitly places state–​citizen relations as the nexus through which financial landscapes are shaped, (re)produced, and contested, through the assemblage of institutional change and financial subject formation. Moreover, the very stability and health of national financial institutions and financial systems directly implicate the legitimacy of governments and their claims to power. This has become more evident after the 2008 financial crisis, with the flurry of bank bailouts in the USA, the UK, the Netherlands, and elsewhere. A more systematic treatment of the relationships between the state, firms, and individuals would enable a deeper understanding of the changing roles of financial logics in everyday life and in corporate transformation, and how those changing roles feed into strategic goals of national development or political legitimacy. This has wider resonance for research into

Financialization of Everyday Life    623 the increasingly extra-​territorial powers of state-​turned-​financial actors, such as sovereign wealth funds and pension funds, which have important implications for the future financial security of households and individuals (Clark et al., 2010; Monk, 2011; Yeung, 2011). In valourizing the role of the state and interrogating the ways in which it mobilizes and intersects with firms and individuals in financialization, there is considerable scope for bringing into focus new actors, relationships, and territories in global financial networks (Coe et al., 2014). As financial logics, institutions, and actors have become inseparable from ever more segments of economy and society (Hall, 2013), such an approach could yield valuable insights beyond the household to the topics of capitalist change, state rationalities, and regional development.

Notes 1. For more comprehensive surveys of the financialization literature, see Hall (2012) and van der Zwan (2014). 2. Although see the earlier discussion on the construction of special financial instruments in the USA during the inter-​and postwar period (Aitken, 2007) and the explicit role played by a national savings bank in Singapore in the forging of specific kinds of financial subjectivities (Lai and Tan, 2015). 3. For some critiques of how cultural economy and social studies of finance approaches have tended to overlook the political nature of financial development and market making, see Pryke and du Gay (2007) and Engelen and Faulconbridge (2009).

References Aalbers, M. (2009a). ‘The sociology and geography of mortgage markets: reflections on the financial crisis’. International Journal Of Urban And Regional Research 33: 281–​290. Aalbers, M. (2009b). ‘Geographies of the financial crisis’. Area 41: 34–​42. Aitken, R. (2007). Performing Capital: Towards A Cultural Economy Of Popular And Global Finance (Basingstoke: Palgrave Macmillan). Appleyard, L. (2011). ‘Community development finance institutions (CDFIS): geographies of financial inclusion in the US and UK’. Geoforum 42: 250–​258. Barnett, C. (2001). ‘Culture, geography, and the arts of government’. Environment And Planning D: Society And Space 19: 7–​24. Bassens, D., Van Meeteren, M., Derudder, B., and Witlox, F. (2013). ‘No more credit to Europe? Cross-​border bank lending, financial integration, and the rebirth of the national scale as a credit scorecard’. Environment And Planning A 45: 2399–​2419. Blackburn, R. (2006). ‘Finance and the fourth dimension’. New Left Review 39: 39–​70. Brassett, J. and Vaughan-​Williams, N. (2012). ‘Crisis is governance: sub-​prime, the traumatic event, and bare life’. Global Society 26: 19–​42. Christophers, B. (2009). ‘Complexity, finance and progress in human geography’. Progress in Human Geography 33: 807–​824. Clark, G.L. (2010). ‘Human nature, the environment, and behaviour:  explaining the scope and geographical scale of financial decision-​making’. Geografiska Annaler: Series B, Human Geography 92: 159–​173.

624   Lai Clark, G.L. (2011). ‘Myopia and the global financial crisis: context-​specific reasoning, market structure, and institutional governance’. Dialogues In Human Geography 1: 4–​25. Clark, G.L., Monk, A., Dixon, A., Pauly, L.W., Faulconbridge, J., Yeung, H., and Behrendt, S. (2010). ‘Symposium: sovereign fund capitalism’. Environment And Planning A 42: 2271–​2291. Clark, G.L., Thrift, N., and Tickell, A. (2004). ‘Performing finance: the industry, the media and its image’. Review of International Political Economy 11: 289–​310. Coe, N.M., Lai, K.P.Y., and Wójcik, D. (2014). ‘Integrating finance into global production networks’. Regional Studies 48: 761–​777. Coppock, S. (2013). ‘The everyday geographies of financialization: impacts, subjects and alternatives’. Cambridge Journal Of Regions, Economy And Society 6: 479–​500. Cutler, T. and Waine, B. (2001). ‘Social insecurity and the retreat from social democracy: occupational welfare in the long boom and financialization’. Review Of International Political Economy 8: 96–​117. Deville, J. (2012). ‘Regenerating market attachments: consumer credit debt collection and the capture of affect’. Journal of Cultural Economy 5: 423–​439. Deville, J. (2015). Lived Economies Of Default: Consumer Credit, Debt Collection And The Capture Of Affect (New York: Routledge). Deville, J. and Seigworth, G.J. (2015). ‘Everyday debt and credit’. Cultural Studies 29: 615–​629. Dore, R. (2000). Stock Market Capitalism:  Welfare Capitalism (Oxford:  Oxford University Press). Dore, R. (2008). ‘Financialization of the global economy’. Industrial and Corporate Change 17: 1097–​1112. Du Gay, P., Millo, Y., and Tuck, P. (2012). ‘Making government liquid: shifts in governance using financialisation as a political device’. Environment And Planning C: Government And Policy 30: 1083–​1099. Duménil, G. and Lévy, D. (2004). Capital Resurgent:  Roots Of The Neoliberal Revolution (Cambridge, MA: Harvard University Press). Dymski, G. (2005). ‘Financial globalisation, social exclusion and financial crisis’. International Review Of Applied Economics 19: 439–​457. Dymski, G. (2009a). ‘Financial Governance in the Neo-​Liberal Era’ in G.L. Clark, A. Dixon, and A.H.B. Monk (eds) Managing Financial Risk:  Frmo Global To Local, pp. 48–​ 68 (Oxford: Oxford University Press). Dymski, G. (2009b). ‘Racial exclusion and the political economy of the subprime crisis’. Historical Materialism 17: 49–​179. Dymski, G. and Li, W. (2003). ‘The macrostructure of financial exclusion: mainstream, ethnic and fringe banks in money space’. Espace, Populations, Sociétés 21: 183–​201. Engelen, E. (2008). ‘The case for financialization’. Competition & Change 12: 111–​119. Engelen, E. and Faulconbridge, J. (2009). ‘Introduction:  financial geographies—​the credit crisis as an opportunity to catch economic geography’s next boat?’ Journal Of Economic Geography 9: 587–​595. Erturk, I. and Solari, S. (2007). ‘Banks as continuous reinvention’. New Political Economy 12: 369–​388. Erturk, I., Froud, J., Johal, S., Leaver, A., and Williams, K. (2007). ‘The democratisation of finance? Promises, outcomes and conditions’. Review Of International Political Economy 14: 553–​575. Faulconbridge, J. and Muzio, D. (2009). ‘The financialisation of large law firm: situated discourses and practices of reorganisation’. Journal Of Economic Geography 9: 641–​661. Finlayson, A. (2009a). ‘Financialisation, financial literacy and asset-​based welfare’. British Journal Of Politics And International Relations 11: 400–​421.

Financialization of Everyday Life    625 Finlayson, A. (2009b). ‘Financialisation, financial literacy and asset-​based welfare’. The British Journal Of Politics & International Relations 11: 400–​421. French, S. and Kneale, J. (2009). ‘Excessive financialisation:  insuring lifestyles, enlivening subjects and everyday spaces of biosocial excess’. Environment And Planning D: Society And Space 27: 1030–​1053. French, S. and Kneale, J. (2012). ‘Speculating on careless lives: annuitising the biofinancial subject’. Journal Of Cultural Economy 5: 391–​406. French, S. and Leyshon, A. (2012). ‘Dead Pledges: Mortgaging Time and Space’ in K. Knorr Cetina and A. Preda (eds) The Oxford Handbook Of The Sociology Of Finance, pp. 357–​375 (Oxford: Oxford University Press). French, S., Leyshon, A., and Wainwright, T. (2011). ‘Financializing space: spacing financialisation’. Progress In Human Geography 35: 798–​819. Froud, J., Haslam, C., Johal, S., and Williams, K. (2000). ‘Shareholder value and financialization: consultancy promises, management moves’. Economy and Society 29: 80–​110. Froud, J., Johal, S., Leaver, A., and Williams, K. (2006). Financialization and Strategy: Narrative and Numbers (Abingdon: Routledge). Froud, J., Johal, S., and Williams, K. (2002). ‘Financialization and the coupon pool’. Capital and Class 78: 119–​151. Gilbert, E. (2005). ‘Common cents: situating money in time and place’. Economy and Society 34: 357–​388. Guthman, J. and Dupuis, M. (2006). ‘Embodying neoliberalism: economy, culture, and the politics of fat’. Environment And Planning D: Society And Space 42: 427–​488. Hall, S. (2011). ‘Geographies of money and finance I: cultural economy, politics and place’. Progress In Human Geography 35: 234–​245. Hall, S. (2012). ‘Geographies of money and finance II: financialization and financial subjects’. Progress in Human Geography 36: 403–​411. Hall, S. (2013). ‘Geographies of money and finance III: financial circuits and the “real economy” ’. Progress in Human Geography 37: 285–​292. Hall, S.M. (2016). ‘Everyday family experiences of the financial crisis’. Journal Of Economic Geography 16: 305–​330. Hardie, I., Howarth, D., Maxfield, S., and Verdun, A. (2013). ‘Banks and the false dichotomy in the comparative political economy of finance’. World Politics 65: 691–​728. Hay, C. (2009). ‘Good inflation, bad inflation: the housing boom, economic growth and the disaggregation of inflationary preferences in the UK and Ireland’. The British Journal Of Politics & International Relations 11: 461–​478. Hendrikse, R.P. and Sidaway, J.D. (2014). ‘Financial wizardry and the golden city:  tracking the financial crisis through Pforzheim, Germany’. Transactions Of The Institute Of British Geographers 39: 195–​208. Krippner, G. (2005). ‘The financialization of the American economy’. Socio-​Economic Review 3: 173–​208. Krippner, G. (2012). Capitalizing on Crisis:  The Political Origins of the Rise of Finance (Cambridge, MA: Harvard University Press). Lai, K.P.Y. (2013). ‘The Lehman minibonds crisis and financialisation of investor subjects in Singapore’. Area 45: 273–​282. Lai, K.P.Y. (2016). ‘Financial advisors, financial ecologies and the variegated financialisation of everyday investors’. Transactions of the Institute of British Geographers 41: 27–​40. Lai, K.P.Y. (2017). ‘Financial geography’ in N. Castree, M. Goodchild, W. Liu, A. Kobayashi, D. Marston, and D. Richardson (eds) The Wiley-​ Aag International Encyclopedia of

626   Lai Geography:  People, the Earth, Environment, and Technology (Hoboken, NJ:  Wiley). DOI: 10.1002/​978118786352.wbieg0656. Lai, K.P.Y. and Tan, C.H. (2015). ‘ “Neighbours first, bankers second”: mobilising financial citizenship in Singapore’. Geoforum 64: 65–​77. Langley, P. (2006). ‘The making of investor subjects in Anglo-​American pensions’. Environment And Planning D: Society And Space 24: 919–​934. Langley, P. (2007). ‘Uncertain subjects of Anglo-​American financialization’. Cultural Critique 65: 67–​91. Langley, P. (2008a). The Everyday Life of Global Finance:  Saving and Borrowing in Anglo-​ America (Oxford: Oxford University Press). Langley, P. (2008b). ‘Sub-​prime mortgage lending: a cultural economy’. Economy and Society 37: 469–​494. Langley, P. (2009). ‘Consumer credit, self discipline and risk management’ in G.L. Clark, A. Dixon, and M. Ashby (eds) Managing Financial Risks: From Global to Local, pp. 280–​300 (New York: Oxford University Press). Langley, P. and Leyshon, A. (2012). ‘Financial subjects:  culture and materiality’. Journal Of Cultural Economy 5: 369–​373. Larner, W. (2012). ‘New subjects’ in T.J. Barnes, J. Peck, and E. Sheppard (eds) The Wiley-​ Blackwell Companion to Economic Geography, pp. 358–​371 (Chichester: Wiley-​Blackwell). Leyshon, A., Burton, D., Knights, D., Alferoff, C., and Signoretta, P. (2004). ‘Towards an ecology of retail financial services: understanding the persistence of door-​to-​door credit and insurance providers’. Environment And Planning A 36: 625–​645. Leyshon, A. and French, S. (2009). ‘ “We all live in a Robbie Fowler house”: the geographies of the buy to let market in the UK’. The British Journal Of Politics And International Relations 11: 438–​460. Leyshon, A. and Thrift, N. (1995). ‘Geographies of financial exclusion:  financial abandonment in Britain and the United States’. Transactions of the Institute of British Geographers 20: 312–​341. Leyshon, A. and Thrift, N. (1997). Money/​Space:  Geographies Of Monetary Transformation (New York: Routledge). Leyshon, A. and Thrift, N. (1999). ‘Lists come alive: electronic systems of knowledge and the rise of credit-​scoring in retail banking’. Economy And Society 28: 434–​466. Leyshon, A., French, S., and Signoretta, P. (2008). ‘Financial exclusion and the geography of bank and building society branch closure in Britain’. Transactions Of The Institute Of British Geographers 33: 447–​465. Lilley, S. and Papadopoulos, D. (2014). ‘Material returns: cultures of valuation, biofinancialisation and the autonomy of politics’. Sociology 48: 972–​988. McFall, L. (2008). ‘Your Money or Your Life:  Making People Prudent’ in McFall, L., Du Gay, P., and Carter, S. (eds) Conduct:  Sociology And Social Worlds, pp. 55–​ 85 (Manchester: Manchester University Press). McFall, L. (2015). Devising Consumption: Cultural Economies of Insurance, Credit and Spending (New York: Routledge). Marron, D. (2007). ‘ “Lending by numbers”: credit scoring and the constitution of risk within American consumer credit’. Economy And Society 36: 103–​133. Marron, D. (2009). Consumer Credit in the United States: A Sociological Perspective from the 19th Century to the Present (New York: Palgrave). Marron, D. (2012). ‘Producing over-​indebtedness: risk, prudence and consumer vulnerability’. Journal Of Cultural Economy 5: 407–​421.

Financialization of Everyday Life    627 Martin, R. (2002). The Financialization of Daily Life (Philadelphia, PA:  Temple University Press). Martin, R. (2010). ‘The local geographies of the financial crisis: from the housing bubble to economic recession and beyond’. Journal Of Economic Geography 11: 587–​618. Maurer, B. (2006). ‘The anthropology of money’. Annual Review of Anthropology 35: 15–​36. Monk, A. (2011). ‘Sovereignty in the era of global capitalism: the rise of sovereign wealth funds and the power of finance’. Environment and Planning A 43: 1813–​1832. Montgomerie, J. (2009). ‘The pursuit of (past) happiness? Middle-​class indebtedness and American financialization’. New Political Economy 14: 1–​24. Pathak, P. (2014). ‘Ethnopolitics and the financial citizen’. The Sociological Review 62: 90–​116. Pike, A. and Pollard, J. (2010). ‘Economic geographies of financialization’. Economic Geography 86: 29–​51. Pollard, J. (2013). ‘Gendering capital: financial crisis, financialization and (an agenda for) economic geography’. Progress in Human Geography 37: 403–​423. Pryke, M. and Du Gay, P. (2007). ‘Take an issue: cultural economy and finance’. Economy And Society 36: 339–​354. Rose, N. (2007). The Politics of Life Itself: Biomedicine, Power, and Subjectivity in the Twenty-​ First Century (Princeton, NJ: Princeton University Press). Sassen, S. (2009). ‘When housing becomes an electronic instrument:  the global circulation of mortgages. A research note’. International Journal Of Urban And Regional Research 33: 411–​426. Strauss, K. (2008). ‘Re-​ engaging with rationality in economic geography:  behavioural approaches and the importance of context in decision-​making’. Journal Of Economic Geography 8: 137–​156. Streeck, W. and Thelen, K. (2005). ‘Introduction: Institutional Change in the Advanced Political Economies’ in W. Streeck and K. Thelen (eds) Beyond Continuity: Institutional Change in the Advanced Political Economies, pp. 1–​39 (Cambridge: Cambridge University Press). Thrift, N. (2005). Knowing Capitalism (London: SAGE). van Der Zwan, N. (2014). ‘Making sense of financialization’. Socio-​Economic Review 12: 99–​129. Williams, K. (2000). ‘From shareholder value to present-​day capitalism’. Economy And Society 29: 1–​12. Wyly, E. K., Moos, M., Kabahizi, E., and Hammel, D. (2009). ‘Cartographies of race and class:  mapping the class-​ monopoly rents of American subprime mortgage capital’. International Journal Of Urban And Regional Research 33: 332–​354. Yeung, H. (2011). ‘From national development to economic diplomacy? Governing Singapore’s sovereign wealth funds’. The Pacific Review 24: 625–​652. Zelizer, V.A. (1983). Morals and Markets: The Development of Life Insurance in the United States (New Brunswick, NJ: Transaction). Zelizer, V.A. (1993). ‘Making Multiple Monies’ in R. Swedberg (ed.) Explorations in Economic Sociology, pp. 193–​212 (New York: Russell Sage Foundation). Zelizer, V.A. (1994). The Social Meaning of Money (New York: Basic Books). Zelizer, V.A. (2011). Economic Lives (Princeton, NJ: Princeton University Press).

Chapter 33

Infrastruc t u re a nd Financ e Phillip O’Neill Introduction Infrastructure is a relatively new word and category in academic and public policy discourse (Rankin, 2009). In its simplest economic sense infrastructure is a capital asset that delivers public goods over a long period of time. By public goods we mean goods that are non-​rival and non-​exclusive. In other words, that there is no significant opportunity cost if the use of the good is extended to additional parties, and that the consumption of the good by one party does not diminish the benefits of consumption by another. The consumption of the services of a lighthouse, for example, is very different to what happens when we consume an apple. But infrastructure also has meanings beyond that of an economic category. It can refer to a suite of public works encased in a utility; a network that enables the movement of water, energy, goods, people, and information; a set of assets and resources, such as in the areas of health and education, that deliver individual and social services; and ecosystems, such as forests, wetlands, and natural riparian and coastal corridors, which supply an urban area with water, air, and drainage services. Of course, these are all legitimate uses of the word ‘infrastructure’ because in different spoken and written contexts the users of the word have important things to say; and, besides, what right do academics have to dictate what the true meaning of a word should be? The point of this opening paragraph is to show that whenever the word infrastructure is deployed a political discussion is occurring about the allocation of scarce resources—​now and into the future—​for some sort of collective venture. This chapter examines the geographical and economic dimensions of the politics of infrastructure. It commences with a selective history of how the term has been used. It then discusses the forces that have shifted the commissioning, funding, and operation of infrastructure away from public-​sector domination; and the new formats for infrastructure delivery and operation that have emerged. I discuss these in terms of the organizational, capital, and regulatory structures involved. The chapter concludes with some observations about how these arrangements are affecting the general welfare of those who live and work in our cities.

Infrastructure and Finance    629

Infrastructure’s Etymologies Adam Smith The problem of how to provide infrastructure is longstanding. The issue is elevated as a primary concern of economics in 1776 in Adam Smith’s The Wealth of Nations where Smith assigns the infrastructure task to ‘the sovereign’. This responsibility is for: . . . erecting and maintaining certain public works and certain public institutions, which it can never be the interest of any individual or small number of individuals to erect and maintain (Smith, 1976, pp. 687–​88).

It was one of only three duties that Smith thought the sovereign should undertake. Smith argued that public works were vital to the betterment of commerce, but their supply was beyond the capability of private enterprise because of their peculiar property and material characters. This assignment was an extraordinary political concession for Smith and exposes how distinctive infrastructure is as a class of object and event. An infrastructure item cannot be exclusively private because infrastructure involves crossing private domains, both geographically and functionally. Yet infrastructure is central to the commercial functions of a city. For Smith, infrastructure makes two things possible: firstly, it gives producers access to larger markets; and, secondly, it enables inputs to be drawn from a wider spatial field which, with the growth of markets, deepens specializations and divisions of labour and raises factor productivity. Infrastructure, then, is central to economic growth and capital accumulation. As a consequence, from the early days of capitalist cities, infrastructure became one of society’s great political settlements: an organizational responsibility of the state funded largely from taxation revenues, both with capital’s blessing.

Keynesian Sensibilities The twentieth century saw the urban infrastructure platform expand with rising expectations of city dwellers for common access to the provision of quality water, energy, transport, and public hygiene. By the middle of that century infrastructure became the foremost representation of the hybrid compromises of New Deal and Keynesian capitalism. Across Western nations, infrastructure steered the state into taxation and planning regimes historically resisted by private capital. Infrastructure expenditure became the counter-​cyclical tool of governments to refit capitalism with bigger markets, and more complex divisions of labour, while delivering legitimacy and reproduction outcomes through jobs creation and enduring urban betterments. In parallel, orthodox economics delivered textbook justifications for the ongoing state role. Competition theory, especially in the accounts of market structures and behaviours derived from Edward Chamberlin (e.g. 1965)  and Ronald Coase (e.g. 1970), showed the good sense of monopoly provision of a public good via a forever-​falling long-​run average cost curve. An intellectual alliance, then, between orthodox, marginal-​cost economics and the pragmatic politics of fiscal Keynesianism, legitimized the flow of public finance into the

630   O’Neill construction and operation of state-​owned infrastructure monopolies. A consequence was that infrastructure roll-​out became an uncontested political action with the utility becoming infrastructure’s organizational custodian, and the capital works budget on a state’s balance sheet its funding source.

International Politics So, in Western nations at least, the provision of a basic kit of urban infrastructure—​water, energy, drainage, roads, and public transport—​became normal. Indeed, the privileged political status of these entities was looked to enviously by other public goods domains where economic benefits were not so directly linked. Notable yearners were not only the education and health systems, but also state-​led systems that delivered art and cultural amenities and services. Today this group is called ‘soft infrastructure’. Yet there were deliberate attempts in the 1950s to broaden the meaning of infrastructure to embrace a range of social interventions to widen the benefits of capitalism in advanced nations and to underpin the process of economic development elsewhere. Here we turn to a remarkable paper by William Rankin (2009), which traces the frustrated passage of the idea of ‘social overhead capital’ into mainstream development economics and policy. Key to the story is the place of the idea of social overhead capital in Walter Rostow’s theory of the stages of economic growth.1 The intriguing thing here, of course, is that Rostow’s text Stages of Economic Growth was, in his words, a ‘non-​communist manifesto’, a pathway to economic development without resorting to the varieties of Marxist-​led revolutions being enacted at that time. Like Adam Smith, Rostow sees investment in social overhead capital as essential to capitalist take-​off in a developing nation. But, mindful of the social investments heralded in communist manifestos, Rostow’s social overheard capital includes not only railway lines and electricity grids, but also education and health systems. Rankin charts the campaign for the creation of a United Nations (UN) development arm to be known as the Special United Nations Fund for Economic Development, or SUNFED, to stand alongside the other development initiatives of the UN: the International Monetary Fund, the General Agreement on Tariffs and Trade, and the World Bank. Indeed, Rankin discovers that the 1955 SUNFED documentation actually substituted the word ‘infrastructure’ for Rostow’s ‘social overhead capital’ in an attempt to create a simpler term to refer to the wide field of asset and services investment needed to underpin a worldwide development project. While the SUNFED proposal was not adopted by the UN, a Rostovian place for social overhead investment was inserted into the World Bank’s agenda, where it remains to this day. Perhaps more significantly, the documentation and discussion surrounding SUNFED steered education and health investments permanently into the infrastructure category, albeit with the ‘soft’ modifier; and so concluded one of the more significant inter-​ textual moments in the history of public policy. Rankin’s work exposes the wider semantic process underway within the infrastructure debate. It reminds us that at the level of a single infrastructure asset no incontestable logic assigns the responsibility for provision to any entity in particular. Each asset has many potential providers each with seemingly valid claims, with variations due to political predispositions, obviously, but also due to different assumptions about the assignment of benefits through time and space. The tools of cost–​benefit analysis, including net present value, or

Infrastructure and Finance    631 discount cash-​flow analysis, have been developed as a direct consequence of intense political conflict over the desirability of different assets and the vehicles for their delivery.

Territory and Technology Building on Rankin, we can see that the infrastructure investment process codifies and reproduces public ideas about the role of the state, territorial sovereignty, and the direction of modernization. There is a feedback loop in infrastructure investment more powerful than in other economic transactions. The loop arises in combination from infrastructure’s sheer size, the expense involved, the institutional and political processes that have to be enacted, and the fine-​skills sets that have to be assembled and deployed. For one part of the world, Schipper and Schot (2011) explore this extended loop via the concept of ‘infrastructural Europeanism’. They align the progression of infrastructure conceptualization to the development of the political and economic entity of Europe (in all its changed forms). Citing Van Laak (2004, p. 247), the authors nominate 13 August 1875 as the first recorded use of the word infrastructure, from a report on a French project to describe the various ‘understructures’ (land, embankments, bridge) that enabled the installation of a railway’s ‘superstructure’ (rail lines, stations, and other overhead fixtures).2 They note, too, the long gestation of the idea such that Britain’s Winston Churchill scorned the word as the ‘jargon’ of a ‘band of intellectual highbrows’ (Van Laak, 2004, p. 247) in a House of Commons speech (as leader of the opposition) as late as June 1950. Nevertheless, the word became an unrivalled signifier for a raft of historical actions driving the extent, intensity, and speed of geographical connectedness. Infrastructure became a central instrument for the modernist projects of nation, technology, and capitalism such that investments in national and cross-​border installations became technical ventures, encased in objective, scientific logic, rather than outcomes of political argument and deliberation, even though these were inseparable. So Schipper and Schot nominate the staging of infrastructure investments in Europe as a spatial and political arrangement or pact, which they call ‘transnational infrastructure regimes’; with a first regime commencing in the first half of the nineteenth century with investments in the carriage of goods, people, and information across European borders. Important in the staged history of infrastructural Europeanism is the interplay between sectoral-​specific initiatives, such as the General Conference on Communications and Transit held in 1921 in Barcelona, Spain, and the development of wider institutional structures that consolidated and legitimized such actions. The Convention and Statute on Freedom of Transit, the legal event that arose from the Barcelona conference, is one example. There are more embracing moments like the establishment of the UN Economic Commission for Europe in 1947 and the Maastricht Treaty, which formally established the European Union (EU) in 1992 wherein the EU was charged with the generation of trans-​European networks in the areas of transport, energy, and telecommunications. We can see in all these that the growth of a distinctive institutional framework for the delivery of infrastructure, the physical growth of infrastructure investment, and the changing institutional framework of ‘Europe’ are inseparable. Earlier, Hughes (1983) had pre-​empted the infrastructure/​politics nexus by showing how the institutional systems of electricity generation and supply in Europe were more important

632   O’Neill to the roll-​out of a continental grid than the grid’s actual enabling inventions and innovations. In like manner, Lagendijk (2011) draws out the political relations essential to the evolution and expansion of a European electricity network showing how the development of a cross-​border grid became a major device for wider experimentation in European integration. An indicative action here is the process of unbundling. A consequence of the state-​ based assembly of a domestically bound, integrated electricity supply ensemble was the establishment of utilities with monopoly control of electricity production, transmission, and distribution. The establishment of a market-​based electricity supply network across Europe required the disassembly—​or unbundling—​of the vertically integrated electricity utilities. Just like the processes of assembly in the mid-​twentieth century, disassembly required integrated political action and regulatory innovation. In parallel work Laborie (2011) shows how development of international post and telecommunications services can only be read through the lens of Cold War politics with the development of the settings that enabled cross-​European and later genuinely global communications structures driven more by political negotiation than by technological advancement.

Urban Planning Finally, we reflect briefly on the development of the idea of infrastructure in the field of urban planning, a crucial component of the roll-​out of mass-​consumption twentieth-​century capitalism. Much has been written on the role of planning in post-​war, suburban Keynesian–​Fordism, but little on the role of infrastructure. Perhaps this is because infrastructure’s presence in a Keynesian capitalism was legitimized and propelled in macroeconomic domains, as discussed earlier, so its place in state funding, procurement, and management was secure. The municipal roll-​out of suburban infrastructure then proceeded as a subsumed political process and an often unseen material process. An ordinary street, for example, disguises the infrastructure corridors that carry the pipes and wires and junctions that make urban life possible. And for decades streets as infrastructure corridors generated the expect­ ation that the state would provide the regulations and finance to ensure a continuation of the enormous value-​add that infrastructure roll-​out gives to the development of land. Planners were freed up to colour the land use maps of an earth surface that arrived on their desks pre-​ fitted with the devices that enabled the movements and connections of urban life.

An Intermediate Conclusion Our discussion shows that infrastructure has always been a political rather than (exclusively) an economic, technical, social, or environmental event. We can also see that the decades following World War II were extraordinary in that they produced a remarkably settled political consensus around a particular model of infrastructure commissioning and operation. Elsewhere (O’Neill, 2014) I have called this a subsumed political process, the idea that public spending on infrastructure within a public utilities organizational model became common sense to the extent that it became instinctive with little public debate about form or method of financing. This consensus has now unravelled for reasons that will be explained.

Infrastructure and Finance    633 At this intermediate point we make these conclusions. Firstly, that infrastructure as an idea is an extraordinarily complex language event. At its core, infrastructure is a set of material and service events for securing a built environment that satisfies the needs of cap­ italism for markets and labour, while meeting the social aspirations of an urban citizenry. Reconciling the conditions for capital accumulation with the need for social legitimacy is the enduring problem of market economies (Offe, 1984; O’Neill, 1996). What fits inside the infrastructure category, however, is ever under negotiation in intensely political ways, because infrastructure is inevitably expensive and invasive. Secondly, our brief history shows how extraordinarily naïve (and therefore ineffective) the political economy of infrastructure provision over the last couple of decades has become. Across the Organisation for Economic Co-​operation and Development (OECD) nations progressive commentators argue for the retention of a utilities-​based, state-​monopoly model that is publicly owned and state-​funded. As a result, the left side of politics has little to say about the economic productivity opportunities for infrastructure, and so becomes a sideline commentator on social equity and environmental outcomes, with few viable alternative projects to offer. A return to a twentieth-​century public utilities model is proffered by default. A consequence of this political paralysis is that the next generation of infrastructure roll-​out is being designed and funded in a lop-​sided political process. Rostow (1990, p.  25) said, correctly, that infrastructure is a ‘lumpy’ historical process with ‘long periods of gestation and pay-​off ’. The lesson to be drawn from Rostow today is that the infrastructure we build this decade is the infrastructure our cities will operate for at least the first half of the twenty-​first century. Drawing on Raco, then, what we need is ‘an approach that emphasises the hybridities that characterise actually existing policies [and therefore] provides the discursive and intellectual space to develop alternative and broader ranging conceptions of development processes and practices’ (Raco, 2005, p. 344).

A Framework for Understanding Changes in Infrastructure Financing The brief review in the previous section shows that infrastructure provision today requires a targeted, deliberate framework of analysis, one that is not only informed by the economic forces at play, but also one that is cognisant of the inseparability of the commissioning and operation of large urban infrastructure assets from the politics of constructing large material entities and passageways in urban landscapes and their funding over long time spans. Large infrastructure procurement is always a major political event. What requires resolution is the difficult economic question of determining the optimal level of resources available for the construction and operation of an infrastructure asset to provide non-​rival goods in an economic system dependent (usually) on price signals to allocate resources for the production of private consumption goods. Certainly as we have seen, there are elegant expositions of the economic dimensions of the problem of assigning responsibility for supplying public goods and apportioning their rewards (see, especially, Samuelson, 1954, and Buchanan, 1965). But, in the end, infrastructural solutions are institutionally sourced rather than market driven,

634   O’Neill being bound up in political valuations requiring targeted, often bespoke, organizational and regulatory arrangements and equally considered financial packaging. R.H. Coase made this point in a not-​too-​subtle attack on his fellow economists’ empirical detachment in a famous note ‘The lighthouse in economics’: Despite the extensive use of the lighthouse example in the literature, no economist, to my knowledge, has ever made a comprehensive study of lighthouse finance and administration. The lighthouse is simply plucked out of the air to serve as an illustration. The purpose of the lighthouse example is to provide ‘corroborative detail, intended to give artistic verisimilitude to an otherwise bald and unconvincing narrative’ [from W.S. Gilbert, The Mikado]. This seems to me to be the wrong approach. I think we should try to develop generalisations which would give us guidance as to how various activities should best be organised and financed. But such generalisations are not likely to be helpful unless they are derived from studies of how such activities are actually carried out within different institutional frameworks. Such studies would enable us to discover which factors are important and which are not in determining the outcome and would lead to generalisations which have a solid base. They are also likely to serve another purpose, by showing us the richness of the social alternatives between which we can choose (1974, p. 375).

Coase had a particular interest in the nature of public and monopoly goods, writing observations about the rise of a broadcasting service (1947) and a national electricity grid (1950), the tense relationships between national regulators and utility operators (1939), the justifications for monopoly provision of national postal services (1961), the rationale for price setting in public utilities (1970), and, famously, as we have just seen, the complexities of assigning costs and apportioning benefits in the provision of lighthouse services to coastal shipping and ports operators (1974). One can see in these writings the transfer of thinking by Coase about the role of the institutional structure of an economy, in particular the ways economic transactions are framed by organizational and regulatory circumstances (e.g. Coase, 1937). Certainly, the development of the utility in Western nations in the postwar period as an organizational structure for the development of public works and their operation and maintenance can be explained in no small part by Coase’s explanation of firms as efficient sites for the internalization of vast webs of relationships that would otherwise entail tedious transactions and expensive costs. Doig’s (2001) fascinating study of the growth of the Port Authority of New York and New Jersey is a fitting illustration of the necessity for containment of the myriad of commercial, environmental, and regulatory exchanges involved in the commissioning, construction, and operation of New York’s port and transportation systems. At this juncture we note the work of Morag Torrance (2008, 2009)  who shows how institutional make-​up and obligation affect the interplay between infrastructure investors and operators. Similarly, we draw on the important contributions from Gordon Clark’s Oxford project,3 involving scholars such as Morag Torrance, Ashby Monk, Rajiv Sharma, Eric Knight, Adam Dixon, and Yin Yang in applying Coasian insights on firm behaviours, especially in management strategy, to explain the investment practices of pension funds and related institutions. Coase could never have imagined the economic and political terrain for infrastructure provisioning and operation today. Nevertheless, his work hands to us the three dimensions of a framework for studying the economic geography of urban infrastructure in the second decade of the twenty-​first century. Torrance (2009) identifies these

Infrastructure and Finance    635 dimensions as organization, capital structure, and regulation. Here we take a slightly different approach preferring to identify sets of practices under these headings rather than the sets of players referred to by Torrance (see Figure 1 in Torrance, 2009), although we see the analysis here as drawing heavily from the relational approach to infrastructure investment that permeates her article. But why these three dimensions? Firstly, they define a set of considerations that are central to the operation of infrastructure, in turn:  the rationale for behaviour (in Coasian terms) that comes from an organization’s constitution, the logic for a financing arrangement, and the possibilities that are created and constrained by a set of rules and contracts. Secondly, because they are far from discrete fields where one dimension is always partial without the presence of the other two, they force attention to historical and political associations and interactions. Thirdly, they are dimensions that incorporate tendencies towards instability because of the potential for shifts: in organizational presence and form; the nature and make-​up of circuits of capital; and the directions of social and economic regulation more generally. Stability thus comes from engineering across the dimensions of organization, finance, and regulation; a somewhat ironic situation given the lifespan of the physical objects involved. To be clear, there are public-​and private-​sector interests in each of the three dimensions and in their interplay, and, a priori, neither sector can be afforded a pre-​eminent position. What the framework provides, though, is the opportunity to monitor the roll-​out and operation of infrastructure in terms of outputs with an assessment of contributions from the public and private sectors taking place at each stage and level of the analysis. I think this leads to better, more strategic infrastructure development, and to a more discerning, robust knowledge-​generation process. We now turn to a brief discussion of the three dimensions.

Organizational Structure Chamberlin and others show how in any market, the forces of competition coexist with the anti-​competitive forces of monopolization, with enterprises framing strategy according to their possession of market power and their anticipation of how others yield their own market power.4 It is possible to see the behaviours of the giant twentieth-​century public-​sector infrastructure utilities as expressions of uncontested market control, and their high cost and poor service outcomes as consequences. Government action to break up the utilities in association with privatization of the sector occurred in the OECD nations between the 1980s and the early 2000s. Central to the break-​up was the introduction of new organizational forms. Some early-​mover acquirers of infrastructure created novel organizational forms for the capture and harvesting of returns from the newly privatized assets, with the proliferation of highly specialized (and contract-​dense) structured investment vehicles and closed private funds. Yet more stable, stakeholder-​serving organizational forms have subsequently emerged. Ongoing open and listed funds are more alert to the links between the custodianship of assets and consistent competitive returns. Highly competitive pension and sovereign wealth funds are similarly motivated, their presence impelled by growing pools of member savings. Publicly listed corporations from engineering and construction backgrounds have added infrastructure operations and investment divisions, while merchant banks and large

636   O’Neill financial and legal houses have formalized the sector as an asset class with the installation of standardized practices and specialist services, including a vocabulary of technical jargon, calculative practices, databases, a body of law, conference circuits, and so on. So there is a widening variety of organizational forms and practices in play, but with some clear trends inviting more detailed analysis.

Capital Structure The evolution and stabilization of infrastructure’s organizational structure is paralleled by a maturing of the sector’s hybrid capital structures. Three major features are observable. The first is that capital structure has become a nervous settlement between quite different interests of equity and debt capital providers. Equity providers demand control over the asset in order to diminish risk and generate above-​market returns through time while debt lenders seek assurances that returns (revenues) from the asset are managed to meet predictable obligations to debt holders and ensure final settlement. Tensions between equity and debt providers are intensified by quantitative easing—​now seen as ensuring cheap debt finance for many years to come—​with key equity providers (general partners) to a financialized infrastructure asset keen to substitute cheap debt for equity capital, and so leverage higher returns, without eroding legal control over the asset.5 Debt providers, of course, become concerned when intensified leverage heightens risk. The development of a capital structure where there is debt-​equity rapprochement, then, becomes a key process in the financialization of an infrastructure asset, with major negotiations required each time a significant re-​f inancing event takes place. This gives rise to the second feature: that both savings aggregators (typically the banks) and the savings pool managers (typically pension and sovereign wealth funds) have become confident enough in their skills and trusting enough of fellow oligopolists such that there is fluidity in membership of the debt and equity sides of the financing equation. There is much to be known about the implications of this fluidity for long-​term arrangement of infrastructure assets.6 The third feature is the emergence of user revenues as the primary source of funding for infrastructure investment. While taxation revenues, delivered direct or via availability concessions, remain significant, regulated user fees and tolling have become essential to the long-​term commercial viability of brownfields privatizations and underpin risk minimization and revenue assurance in greenfields infrastructure projects. The shift from taxation to user-​pays is attractive to governments as their balance sheets struggle, especially at the level of provincial and municipal governments where imbalance between revenues and costs are most pronounced and where debt-​raising possibilities are limited. Clearly, there is much more to be said about capital structures in the infrastructure sector, and of the extent to which their apparent stabilization will continue. For here, we can note that privatized infrastructure has been able to extricate itself from the complexities of the global financial crisis with a seemingly assured supply of finance and stable investor–​ asset relationships such that there is an expanding core of urban assets with a claim to financial and functional viability into the future. Certainly, there are equity and sustainability

Infrastructure and Finance    637 concerns arising from the nature and distribution of the infrastructure services involved; the point for now is that a stable, perhaps viable, form of capital structure outside the fiscal domains of the state seems to have emerged.

A Regulatory Regime Stability in infrastructure’s organizational and capital structures, and in the finance market more generally, come not accidently from stabilization in the sector’s regulatory environment. There is evidence, though, that the regulatory environment for the infrastructure sector has not evolved in the ways that have been typical of other product markets, as we will discuss. Certainly there are various state experiments involving, for example, templated public–​private partnership initiatives in Ontario, British Columbia, Alberta, and Quebec (Siemiatycki, n.d.), and across the UK more generally. Yet these practices have progressed little beyond their domestic territories with evidence of only modest international learning and no moves as yet towards transnational standardization. Bearing heavily on the role of regulation in the infrastructure sector, Helm and Tindall (2009) explain the need for guaranteed contractual conditions between the state and an infrastructure investor. They see each side of the investment relationship as having substantial interest in the strength of these contracts. For the state, there is the need to ensure an asset performs over its life course and that charges to consumers are fair and acceptable given the prevalence of monopoly positions for the new infrastructure owners and operators. For the investor, there is the need for adequate returns to cover operating costs and to ensure sunk costs are recoverable at a reasonable rate over the life of the infrastructure asset. There is also the need for ownership assurance, seen by the investor as a primary way of managing risk given the high level of uncertainty surrounding a large urban infrastructure asset through lengthy time periods (see also Stern, 2012). Yet given the peculiar nature of the infrastructure sector, and therefore the unlikelihood of being able to import pre-​existing regulatory devices and processes from other sectors or jurisdictions, infrastructure contracts between governments and investors have invariably been constructed on an individualized basis. Typically, these include the functions of monitoring and control, as well as prescriptions as to ownership rights, including property rights, market conditions, protection from competition, and so on. Helm and Tindall (2009, p. 149) conclude that, ‘Because of this complex interplay of political and economic factors, each privatisation [has] had its own unique characteristics and, not surprisingly the outcome of the privatisation programme as a whole [has been] a messy one’. Certainly, there has been an industry-​wide role in this continual resort to one-​off arrangements involving lucrative fees for lawyers, financial advisers, and related professional services providers. As a consequence, the standardized and transparent regulatory systems that pervade other economic sectors are absent from the infrastructure sector. These include global ISO product standards, national product regulations, international trade agreements, financial standards and practices legislation and agreements, and so on. These sorts of standardizations have yet to be created for the infrastructure investments sector, leaving the sector very much aligned to what Cutler (2010, p. 157) for the defence and security sector sees as a tendency to be:

638   O’Neill  . . . shaped by private actors, institutions and processes that operate transnationally, linking local and global orders through complex laws and regulatory arrangements. These private governance arrangements are legitimized through their claims to possess expert knowledge and authority.

Analysis of the role of privatized regulation in the government of the infrastructure sector, and therefore of its impact on the sector’s organizational and capital structures, is a topic for major research. Guidance as to its direction not only comes from Cutler (see above), but also from the regulatory capitalism literature, including works by John Braithwaite (e.g. 2008; Braithwaite and Drahos, 2000) and David Levi-​Faur (e.g. 2012; Levi-​Faur and Gilad, 2004). The Oxford Handbook of Governance captures much of this work as does scholarship in the journal Regulation and Governance. Application of this literature to the development of the privatized and financialized infrastructure sector is urgently needed.

Implications and Conclusions Clearly, there is a research agenda in infrastructure of some length and thickness. A new large commercial sector is a rare thing; and to the extent that privatized and financialized infrastructure assets are relatively novel, and given their creation and direction are subject to considerable political scrutiny and debate, it is surprising (with notable exceptions) that research output on the field is still meagre. Perhaps this comes from the way infrastructure traverses so many academic and policy disciplines involving such a level of technical and professional/​practitioner skills that academic research teams are daunted by the task. Yet there are responses with great potential. The UK government-​funded iBuild7 and ICIF8 research entities have been structured as large multidisciplinary projects with agenda across design, governance, and financing domains. Similar approaches are needed elsewhere. That said, successful research into the maturing financialized infrastructure sector needs a framework that is beyond a simple pooling of knowledge from across the discipline spectrum alongside the insights of policymakers and practitioners, however worthwhile such blended activity can be. A new framework must match the thing that infrastructure has become: a large hybrid field where there is now an imperative for heightened levels of involvement from both private and public sectors. No other economic sector can have this claim made for it. At this stage of the public sector’s history, across the world, there is insufficient fiscal capacity to maintain and re-​equip existing infrastructure and meet the demands for new assets. Moreover state agencies and the utilities are now empty of the engineering, financial, and operational management expertise to undertake these tasks, even if public sector spending capacity were available. However, the private sector also lacks major powers and capacities and so must call on the state to partner its infrastructure ventures. The list of necessary state and public-​sector contributions is lengthy: property rights, legitimacies afforded through planning and approvals processes, rights to negotiate operating and ownership conditions into the future in order to mitigate risk and deal with uncertainty without threatening the viability of the asset, creating and ensuring asset productivity through alliance with other urban assets, and so on. There is also the need for

Infrastructure and Finance    639 the state to ensure the shared outputs of infrastructure operation, which we call positive externalities, are distributed, not blocked or corralled, a process that seems in its entirety unable to be assigned to the private sector with any confidence.9 What is needed, then, is a research framework capable of dealing with all this hybridity without being trapped by the political argument that comes with the public sector–​private sector binary—​while remaining intensely political about the social and environmental outcomes that are being sought from the presence of desirable urban infrastructure. Such a framework needs close and careful preparation. This projected work in both academic research and policy development can only occur if there is a shift in the politics of infrastructure to accompany the other profound changes underway. Perhaps the major contribution of this chapter might be that by laying out the very developed, maturing, now-​monopolistic stage of the infrastructure sector the knee-​jerk opposition to substantial private-​sector presence in infrastructure provision might abate. Beyond nostalgia and history—​however much popular progressive history is based on accurate representations of the past rather than on illusion—​political effort for a restoration of almost-​exclusive public sector ownership and operation of the infrastructure sector is now futile. Certainly there is an important role of criticism. At the same time, however, there is a duty, necessitated by urgency, to undertake research that leads directly to better urban circumstances, environmentally and socially, and therefore which works from an acceptance of the hybridity we talk about, not against it. Finally, while all government portfolios have a claim to primary importance and therefore deserving of raised political attention, infrastructure’s physicality in combination with its sheer size and enormous cost mean that infrastructure is relatively unmalleable, with limited options for changing design, funding, and operation once initial procurement is completed. Retrofitting for, say, better social and environmental outcomes is very difficult with major cost and disruption hurdles to overcome. In other words, a political battle for infrastructure—​and fights over the what, where, how, and for whom questions—​may not be revisited for decades, such is the physical longevity of the assets involved. This means that activism across the many dimensions of public life must be successful in forcing the best possible infrastructure decisions in terms of economic, social, and environmental outcomes. Understanding the hybrid composition of the contemporary infrastructure sector is a prerequisite to successfully negotiated infrastructure provision. In contrast, obstinate allegiance to campaigns for a return to a mythical golden age of public ownership and operation is a diversion of energy and imagination, saying nothing of the poor analysis that lies beneath.

Acknowledgements Version of the arguments contained in this chapter have appeared in the following publications: O’Neill, P. (2014). ‘How Infrastructure Became a Structured Investment Vehicle’ in M. Roche, J. Mansvelt, R. Prince, and A. Gallagher (eds) Engaging Geographies:  Landscapes, Lifecourses and Mobilities, pp. 29–​44 (Newcastle: Cambridge Scholars). O’Neill, P.M. (2013). ‘The financialisation of infrastructure: The role of categorisation and property relations’. Cambridge Journal of Society, Economy and Regions 6: 441–​454.

640   O’Neill

Notes 1. Rostow (1990) saw infrastructure as crucial to development, especially in the transition of countries from ‘preconditions’ to ‘take off ’. Importantly, infrastructure was seen not just as a way of raising factor productivity and extending markets, but as a symbol of development with each major infrastructure venture becoming a large visual symbol of a nation’s commitment to a modernization pathway. 2. Not accidently, say Schipper and Schot (2011), there seems to be an appropriation here of the Marxist separation, popular among French philosophers at the time, of the economic base of capitalism from an overarching social superstructure. 3. See www.geog.ox.ac.uk/​staff/​glclark.html. 4. See Helm and Tindall (2009) for an exposition of the importance on understanding the infrastructure market structure as a guide to market behaviour. 5. An example is the UK water utilities where significant capital gain accrued to first movers through the simple task of replacing debt with equity capital (Armitage, 2012). 6. See Helm and Tindall (2009, p. 414): ‘These [infrastructure asset] contracts between owners, regulators, and managers are inevitably going to be incomplete, and this brings into play a crucial role for ownership in incomplete contracts . . . With unforeseen contingencies and non-​contractable managerial effort, ownership is always important because it determines who has the “residual right of control”.’ 7. See https://​research.ncl.ac.uk/​ibuild/​. 8. ICIF stands for International Centre for Infrastructure Futures. See http://​www.icif.ac.uk/​. 9. Given that outputs like decongestion, cleaner air and water, more efficient movement through the city, better public and social relations through telecommunications systems, and so on are very much internal to the functioning of infrastructure.

References Armitage, S. (2012). ‘Demand for dividends:  The case of UK water companies’. Journal of Business Finance & Accounting 39: 464–​499. Braithwaite, J. (2008). Regulatory Capitalism: How it Works, Ideas for Making it Work Better (Boston, MA: Edward Elgar Publishing). Braithwaite, J. and Drahos, P. (2000). Global Business Regulation (Melbourne:  Cambridge University Press). Buchanan, J.M. (1965). ‘An economic theory of clubs’. Economica 32: 1–​14. Chamberlin, E. (1965 [1933]). The Theory of Monopolistic Competition: A Re-​orientation of the Theory of Value (8th edition) (Cambridge, MA: Harvard University Press). Coase, R.H. (1937). ‘The nature of the firm’. Economica, New Series 4: 386–​405. Coase, R.H. (1939). ‘Review’. The Economic Journal 49: 757–​758. Coase, R.H. (1947). ‘The origin of the monopoly of broadcasting in Great Britain’. Economica, New Series 14: 189–​210. Coase, R.H. (1950). ‘The nationalization of electricity supply in Great Britain’. Land Economics 26: 1–​16. Coase, R.H. (1961). ‘The British Post Office and the messenger companies’. The Journal of Law & Economics 4: 12–​65. Coase, R.H. (1970). ‘The theory of public utility pricing and its application’. The Bell Journal of Economics and Management Science 1: 113–​128.

Infrastructure and Finance    641 Coase, R.H. (1974). ‘The lighthouse in economics’. Journal of Law and Economics 17: 357–​376. Cutler, A.C. (2010). ‘The legitimacy of private transnational governance: experts and the transnational market for force’. Socio-​Economic Review 8: 157–​185. Doig, J.W. (2001). Empire on the Hudson: Entrepreneurial Vision and Political Power at the Port of New York Authority (New York: Columbia University Press). Helm, D. and Tindall, T. (2009). ‘The evolution of infrastructure and utility ownership and its implications’. Oxford Review of Economic Policy 25: 411–​434. Hughes, T.P. (1983). Networks of Power: Electrification in Western Society, 1880–​1930 (Baltimore, MD: Johns Hopkins University Press). Laborie, L. (2011). ‘Fragile links, frozen identities: the governance of telecommunications networks and Europe’. History and Technology 27: 311–​330. Lagendijk, V. (2011). ‘ “An experience forgotten today”: examining two rounds of electricity liberalization’. History and Technology 27: 291–​310. Levi-​Faur, D. (ed.) (2012). The Oxford Handbook of Governance (Oxford and New York: Oxford University Press). Levi-​Faur, D. and Gilad, S. (2004). ‘Review: The rise of the British regulatory state: transcending the privatization debate’. Comparative Politics 37: 105–​124. Offe, C. (1984). Contradictions of the Welfare State (London: Hutchinson). O’Neill, P.M. (1996). ‘In what sense a region’s problem? The place of redistribution in Australia’s internationalization strategy’. Regional Studies 30: 401–​411. O’Neill, P. (2014). ‘How Infrastructure Became a Structured Investment Vehicle’ in M. Roche, J. Mansvelt, R. Prince, and A. Gallagher (eds) Engaging Geographies: Landscapes, Lifecourses and Mobilities, pp. 29–​44 (Newcastle: Cambridge Scholars). Raco, M. (2005). ‘Sustainable development, rolled-​out neoliberalism and sustainable communities’. Antipode 37: 324–​347. Rankin, W.J. (2009). ‘Infrastructure and the International Governance of Economic Development, 1950–​1965’ in J.F. Auger, J.J. Bouma, and R. Künneke (eds) Internationalization of Infrastructures, pp. 61–​75 (Delft: Delft University of Technology). Rostow, W.W. (1990 [1960]). The Stages of Economic Growth:  A  Non-​Communist Manifesto (Cambridge: Cambridge University Press). Samuelson, P.A. (1954). ‘The pure theory of public expenditure’. The Review of Economics and Statistics 36: 387–​389. Schipper, F. and Schot, J. (2011). ‘Infrastructural Europeanism, or the project of building Europe on infrastructures: an introduction’. History and Technology 27: 245–​265. Siemiatycki, M. (n.d.) ‘Is there a distinctive Canadian PPP model? Reflections on twenty years of practice.’ Mimeo, Department of Geography and Program in Planning, University of Toronto. Smith, A. (1976 [1776]). An Inquiry into the Nature and Causes of the Wealth of Nations (two volumes). (Oxford: Oxford University Press). Stern, J. (2012). ‘The relationship between regulation and contracts in infrastructure industries: regulation as ordered renegotiation’. Regulation & Governance 6: 474–​498. Torrance, M. (2008). ‘Forging glocal governance? Urban infrastructures as networked financial products’. International Journal of Urban and Regional Research 32: 1–​21. Torrance, M. (2009). ‘The rise of a global infrastructure markets through relational investing’. Economic Geography 85: 75–​97. Van Laak, D. (2004). Imperiale Infrastruktur: Deutsche Planungen für eine Erschliessung Afrikas 1880 bis 1960 (Paderborn: Schöningh).

Pa rt  V I I

R E S OU RC E S A N D T H E  E N V I RON M E N T

Chapter 34

The Financia l i z at i on Thesi s Revi sited:  C ommodi t i e s as an Asset  C l as s Sarah McGill Introduction Long a subject of interest in economics, commodity price dynamics attracted a great deal of interest across the social sciences beginning in the run up to the global financial crisis of 2008. Commodity markets experienced an unprecedented price boom beginning in 2003: although punctuated by a sharp drop in the wake of the global financial crisis in 2008, prices all but recovered previous highs despite weak global growth prospects. Although this recovery led to some expectation that high prices represented a ‘new normal’ for commodity markets, by 2012 prices had once again dropped significantly. The boom is now widely considered an instance of a super-​cycle: a strong and sustained rise in (primarily industrial) commodity prices driven by demand from emerging market economies, in this case led by China. For certain commodities, notably oil, price volatility was also unusually pronounced during this period. Much has been written about the super-​cycle and the ways in which it reflects short-​term, cyclical, and structural dimensions of commodity prices. While there is little dispute that the super-​cycle represented an unusually strong and long-​lasting upswing, for a great many observers, the price dynamics of that decade cannot entirely be accounted for by reference to ordinary cyclical supply and demand factors. From roughly the same time as the beginning of the upswing in the 2003–​12 super-​ cycle, financial investors such as pension funds, hedge funds, endowments, and sovereign wealth funds began injecting historically large amounts of capital into commodities, as evidenced not least (if indirectly) by the rapid growth in the volume of contracts traded on futures exchanges—​a figure that increased roughly sevenfold between 2000 and 2010 (UNCTAD, 2011, p. 15). This trend strongly suggested to many observers that price movements, if not price levels, have come to be shaped by more than supply shocks, burgeoning

646   McGill emerging-​market demand, or, indeed, any other changes in fundamentals. Instead, commodity markets were said to have become ‘financialized’. These suspicions were only heightened by dramatic commodity price spikes, most notably the apparent oil price bubble from 2006 to 2008 and food price rises in 2008 and 2010. While these two trends—​high prices and record volatility (at least in oil markets) on the one hand, and dramatic investor interest in commodities on the other—​roughly coincided, definitively proving causation between them in either direction is exceedingly difficult (UNCTAD, 2011; Bos and van der Molen, 2013, p. 4). Nonetheless, because of the actors involved and their suspected effects on prices of such critical resources as oil and basic foodstuffs, financial investment in commodities has become a political hot-​button issue. Accordingly, a large body of quantitative empirical studies as well as academic and policymaking commentary seeks various explanations for commodity price behaviour during the boom. It is only recently that something approaching a consensus has emerged on whether, let alone how, such investment shapes prices: in a word, evidence has grown that the presence of financial investors played some role in exacerbating price swings. Yet while the super-​cycle is currently in a trough, the wider questions raised during the boom concerning the intersections between financial markets and the ‘real economy’ remain highly salient. As part of the wider project on financialization, derivatives trading in general has been of long-​standing interest to market theorists. In the broadest sense, the financialization project has proposed a governing logic that has encouraged the growth of the financial sector and the spread of so-​called financial innovation and, in turn, infused firms and industries—​ not to mention individuals, households, and entire economies (French et al., 2011, p. 799). However, Muellerleile (2009) has observed that ‘On the whole, the concept of financialization as studied in the social sciences remains imbued with an almost atmospheric quality, omnipresent but absent of context or cause’, and empirical case studies have been few and far between (French et al., 2011).1 At the same time, for all that natural resources and financialization alike have been on the social science research agenda for some time, the intersection between commodities and financial markets has received less in-​depth treatment (but see Labban, 2010, for a notable exception). Now that the most recent super-​cycle can be viewed with the wisdom of hindsight, the case of commodities offers a rich and timely opportunity to match conceptual foundations with empirics. This chapter examines discourses about financialization in commodity derivatives markets as they have been rehearsed by economic geographers and economists alike. It finds that the ideological lines in these debates are drawn within what may conventionally be thought of as these two ‘camps’ as often as they are drawn between them. In particular, there are striking commonalities between the critiques of a great many geographers, heterodox political economists, and ‘mainstream’ economists. The chapter argues that, while not entirely misplaced, conventional critiques of financialization in commodity markets are problematic in several important respects. At their most crude, they reflect a wider suspicion of market agents—​whether financial or otherwise—​which, in turn, glosses over the complexities of the commodity price formation process. Such suspicion of commodity trading is far from new, being what Merton Miller (1997) has called ‘standard visceral reactions against middlemen and speculators’. Further contributing to a lack of clarity is the semantic and conceptual ambiguity evident in many other analyses. At its best, the financialization thesis recognizes the central importance of expectations in setting prices.

The Financialization Thesis Revisited    647 The chapter proceeds as follows. The next two sections outline the financialization thesis as it appears across a spectrum of scholarly opinion: while ‘The Financialization Thesis and Derivatives Trading’ focuses upon perspectives from a number of strands of political economy literature, ‘Commodity Trading and the Financialization Thesis’ sketches out the development of commodities as an asset class and outlines controversies about the impact of financial investment on commodity prices as presented in ‘mainstream’ economics and finance literature. The following two sections take the form of a ‘critique of the critique’ of the financialization thesis. The section ‘The Defence of Derivatives’ explores a number of conceptual and semantic difficulties apparent in the literature. ‘Iterated Expectations’ examines the concept of ‘speculation’ in greater detail, and based on this discussion presents an alternative picture of what precisely is meant by ‘the financialization of commodity markets’. Finally, concluding thoughts are offered.

The Financialization Thesis and Derivatives Trading Lee et al. (2009) have identified seventeen different conceptualizations—​some overlapping, some conflicting—​of ‘financialization’, all of which overtly or implicitly describe a hegemonic process, logic, or system. As articulated by Epstein (2005, p. 3), financialization refers to ‘the increasing role of financial markets, financial motives, financial actors, and financial institutions in the operation of the domestic and international economies’. At its heart, financialization is believed to be no less than the totalizing realization of neoclassical finance and economics theory, as epitomized in, for example, Merton and Bodie (2005). More than simply a pervasive logic, ‘financialization alters behaviour and values in the economy, in society and in governmental institutions and policy regimes. . . . [I]‌t has been aided and abetted by Western states driven to open up markets and globalise their economies’ (Christopherson et al., 2013, p. 352). While seldom discussed explicitly in relation to commodity production and trading (see O’Neill (2001), and Labban (2010) for exceptions), financial logic has arguably pervaded commodity production and trade at all points along the supply chain, as we shall see. Of primary interest here, however, is that strand of the literature concerning itself with the proliferation of trade in futures contracts and other financial instruments since the 1970s (Pryke and Allen, 2000; Krippner, 2005). The heterodox account of derivatives trading covers a wide spectrum of opinion but can be summarized briefly as follows. The Marxist tradition condemns outright the very existence of financial instruments in all markets as a form of ‘fictitious capital’ (Harvey, 2006, 2011). Derivatives, in this view, are conceived as spatially and temporally displaced representations of physical goods, if not a direct replacement of them. At the outset, derivatives have two strikes against them: not only does the exchange of goods in general lack intrinsic value, but derivatives themselves also have no economic worth because they are no more than abstract representations of an underlying asset. At one level, then, this school of thought offers a heterodox theory of value for derivatives, which itself has more profound systemic implications. The trading of financial derivatives represents a divorce between the ‘real economy’ and a ‘financial economy’ constituted entirely by ‘disembodied’ or fictitious

648   McGill capital. As such, financial markets have their own ‘space–​time logic’ that has no direct bearing on ‘real’ markets (Harvey, 2006; Mann, 2006 cited in Christophers, 2011; see also LiPuma and Lee, 2004). Much like credit, derivatives create financial value before the value of the underlying commodity can be realized in production and exchange (Harvey, 2006; see also Krippner, 2005). As such, their sole (de facto) purpose is believed to be a more rapid and lucrative generation of profit for an elite rentier class. In a word: ‘Finance finances itself, but does not finance investment’ (Duménil and Lévy, 2004, cited in Labban, 2010, p. 548, original emphasis). It is not difficult to see the salience of this argument to the case of commodity derivatives markets; indeed, oil futures contracts have been dismissed in more than one account as mere ‘paper barrels’ (O’Sullivan, 2009; Labban, 2010). The sheer number of derivatives traded on a given day in relation to the volume of underlying assets in productive circulation is commonly cited as proof of the ‘fictitious capital’ thesis (Labban, 2010; see also O’Sullivan, 2009). For instance, figures from UNCTAD (2011) suggest that at the end of 2010, the total number of outstanding contracts on organized exchanges alone exceeded the global total number of barrels of oil produced worldwide by a factor of 2.5. Labban (2010) takes this ‘fictitious capital’ thesis to its logical conclusion. Because commodity prices are governed by a parallel financial logic, ‘circulation of value can precede its production in the labour process and realisation in the market’ (Labban, 2010, p. 550). This, in turn, lays the foundations for a crisis: in the medium term, the bursting of an asset bubble. In the very long term, he obliquely suggests, the financialization of commodity markets may temporarily disguise the ultimate resource crisis: the depletion of oil reserves (Labban, 2010, p. 551). Implicit here is the idea that the ‘growth machine’ of capitalism, let alone finance capitalism, cannot continue indefinitely, whether due to natural limits (the ‘second contradiction of capital’ (O’Connor, 1998)) or otherwise (see Harvey, 2011). A less hard-​line approach, which both overlaps with the Marxist critique and is shared beyond the heterodox tradition, is concerned not so much with the widespread use of futures contracts per se but with their ‘abuse’ by ‘speculators’. In this account, ‘speculation’ is rarely defined explicitly but, in contradistinction to hedging, it is virtually synonymous with gambling (Bryan and Rafferty, 2006). Less dramatically, the spread of derivatives trading has fundamentally altered economic agents’ conception of risk such that their ‘focus [is] on risk-​reward ratios rather than on absolute risk’ (LiPuma and Lee, 2004, p. 45; see also Pryke and Allen, 2000; Bryan and Rafferty, 2006). However, whereas the broad-​brush approach accepts unequivocally that the term encompasses all trade in financial instruments, which are ‘targets for speculative investment’ (Harvey, 2011, p. 83) apparently by their very nature, the softer critique is somewhat hazy on precisely what types of activities constitute ‘speculation’. What is clear in this view is that, because of the dominance of finance-​led capitalism and the ‘time–​space compression’ it allows, speculation has a major destabilizing effect on commodity markets—​and the global economy writ large—​in that it contributes to both volatility and asset bubbles. Indeed, financial trading exacerbates the very volatility it is intended to smooth out. Precisely because derivatives trading lowers transaction costs, it encourages short-​termism and gives rise to price volatility, which, in turn, wastes resources and shortens the planning horizons of firms and investors (Labban, 2010, p. 546; Harvey, 2011; see also Miller, 1997). Moreover, this very volatility is what makes derivatives trading profitable (Swyngedouw, 1996, cited in Labban, 2010, p. 546). In the medium term, too, financialization accelerates and amplifies the boom–​bust cycles that characterize capitalism (Harvey, 2006).

The Financialization Thesis Revisited    649 Evident here is a condemnation of the short termism that characterizes much of contemporary financial markets. Such alarm is far from new (see Strange’s (1986) account of ‘casino capitalism’), although it has certainly been heightened in the wake of the global financial crisis (see Sinn, 2010). What is notable here is that the association between short termism and contemporary financial markets finds parallels in the ‘mainstream’ of economics and finance. As summarized by Kay (2012), short-​termist market behaviour is a twofold phenomenon, at once myopic for its failure to invest for the creation of durable economic value and hyperactive for its propensity to trade in ever-​briefer intervals. It is driven by both human behavioural biases (Clark, 2011) and a suite of wider market incentives (see Kay, 2012). Short termism in markets is typically associated with the pursuit of self-​interested gains, which disrupts the transfer of savings into investment for economic growth. This is particularly true of long-​duration projects which ‘yield the highest long-​term (private and social) returns and hence offer the biggest boost to future growth’ (Haldane and Davies, 2011, p. 14). Ultimately, therefore, to its critics financialization is not only a socially redundant activity in and of itself, but it also results in deleterious side effects in that it detracts scarce capital resources from productive investment. Even more than that, in so doing it creates a governing logic, which leads to the mispricing of assets. As we shall see in the next section, this generic argument appears to resonate perfectly in the case of commodity derivatives markets—​the archetypical bridging point between finance and the ‘real economy’.

Commodity Trading and the Financialization Thesis The Development of an Asset Class The issue of financialization in commodity derivatives trading is relatively new in and of itself, even though commodities have been regarded as a component of a well-​diversified portfolio for some time (Greer, 1978; Bodie and Rosansky, 1980, both cited in Parsons, 2010), and deep suspicion of ‘speculators’ in commodity markets dates back centuries. However, it has deeper significance in two respects. On one level, it represents one of the latest flashpoints in a decades-​long debate over causes and remedies for commodity price instability (see Radetzki, 2008). On another level, the very problems perceived to have been caused by financialization took place against the backdrop of the global financial crisis and the widespread scepticism of finance that attended it (described in the previous section). The roots of the issue can be traced back to the rise of institutional investors and ‘shareholder capitalism’ ideology beginning in the 1970s, both of which have dramatically increased both the size of financial players and the scope of their reach into new markets (Pike and Pollard, 2009). These markets were largely ‘constructed’ in order to satisfy investors’ rate of return targets, a process that can be briefly sketched out as follows. At the same time as pension funds face growing liabilities, portfolio returns have proven disappointing over the last fifteen years or more, owing to both the collapse of the equity market premium of the last fifty years and the long-​term decline in discount rates on returns (see Clark and Monk, 2013). The bursting of the equity-​market bubble in 2000 precipitated

650   McGill this process. The ensuing search for returns led investors into a range of ‘marginal’ markets, both in the sense of alternative asset classes (Leyshon and Thrift, 2007) and also of peripheral geographic regions (Clark and Monk, 2013). Along with the 2003 decision by the US Federal Reserve to allow banks to invest in physical commodity production, transport, and storage infrastructure, financial deregulation of derivatives markets during the 1990s and early 2000s—​notably the Gramm–​Leach–​Bliley Act of 1999, which partially dismantled the Glass–​Steagall Act by allowing bank holding companies’ involvement in commodity markets, and the Commodity Futures Modernization Act of 2000, which removed over-​ the-​counter (OTC) derivative trading from regulatory oversight—​helped accelerate the development of commodities as one such ‘alternative asset class’ (Clapp and Helleiner, 2012; Valiante, 2013). Market conditions at the beginning of the 2000s were thus ripe for the accelerated development of commodities as an asset class. In a narrower sense, the foundations for the current wave of investment were arguably laid in 1991with the creation of the Goldman Sachs Commodity Index (GSCI, now the S&P GSCI), an investment product that has been closely replicated many times over by other financial institutions. Beginning around 2003, however, institutional investors bought en masse into academic research (notably Gorton and Rouwenhurst, 2004), which found a negative correlation between commodity and stock returns (see Tang and Xiong, 2010, p. 9). In essence, they were subscribing to a narrative about global growth and natural resource scarcity:  the rise of emerging economies, led by China (and to a lesser extent, India, Brazil, and Russia), was creating burgeoning and unprecedented demand for commodities across the board, driving up their prices to an extent that this market offered an opportunity to earn outsize returns. Investing in commodities thus represented a direct investment in emerging markets: one that promised not only outsized returns, but also a means of portfolio diversification and a hedge against inflation. With the onset of the global financial crisis in 2008, this narrative only gained further traction, as returns in ‘traditional’ asset classes remained disappointing, while the long-​term growth prospects for the largest and most resource-​hungry economies appeared brighter than for mature markets (if less robust than they had prior to 2008). While comprehensive and reliable data on financial trading of commodities, including volumes and the number, sizes, and identities of market participants, is scarce, there exists strong indirect evidence of a dramatic increase in financial participation. For example, as already noted, the volume of both exchange-​traded and OTC derivatives has risen sharply since 2004: while the number of futures and options contracts traded on exchanges globally has increased from an initial US$15 million to over US$60 million in 2010, the notional value of outstanding OTC derivatives grew from less than US$2 million to nearly US$14 million in mid-​2008 (UNCTAD, 2011, p.15). Total financial investment in exchange-​traded commodity derivatives is also estimated to have grown markedly from US$80 million in 2005 to US$375 million in 2010 (UNCTAD, 2011, p.16) (Figure 34.1). Financial investors gain exposure to commodities primarily through four channels (UNCTAD, 2011, pp. 14–​15). Commodity indices, which attracted the lion’s share of capital (and, accordingly, controversy), are composites of futures contracts on a given range of commodities traded on exchanges. They are largely provided by investment banks: among the best known are the S&P GSCI (Standard and Poor’s Goldman Sachs Commodity Index), the Reuters/​Jeffries CRB Index, and the Dow Jones AIG Index. Investors buy into these indices by entering into a swap or other bilateral financial agreement with the index provider, which

The Financialization Thesis Revisited    651 14 12 10 8 6 4 2 0

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2008 2009 2009 2010 (June) (June) (June) Other commodities

Other precious metals

Gold

Figure  34.1  Notional Amount of Outstanding Over-​ the-​ Counter Commodity Derivatives, December 1998–​June 2010 (US$ trillion). Source: UNCTAD (2011, p. 15).

hedges its exposure through exchange-​traded commodity futures contracts. Crucially, and in contrast to more ‘conventional’ speculation on commodity futures markets, these indices comprise only long positions, which are ‘rolled over’, or replaced with longer-​dated contracts, close to the date of expiration of the contract. Index investing is considered to be a passive, longer-​term strategy, although indices are typically weighted and re-​weighted to capture prices with relatively stronger outlooks. This strategy rests on the assumption that commodities are a discrete asset class with a unique risk premium. In order to be profitable, the market in question must be in backwardation (i.e. the spot price must be higher than the futures price)—​a condition implying low inventory levels and a positive convenience yield (Domanski and Heath, 2007, p. 56; Östensson, 2012).2 A second, and more active, trading strategy entails taking both long and short positions in futures and options contracts. This is more short termist and somewhat more risky. Exchange-​traded products (ETPs), a third type of instrument, have begun to raise concerns because relatively new versions of them are backed by physical commodities, rather than using futures contracts as collateral. Recent market conditions (specifically risk aversion and low interest rates) have only enhanced the popularity of physically backed ETPs. Finally, structured products, used since 2006, ‘typically combine an underlying asset with a derivative’. Beyond these channels, some banks have entered physical markets as producers (UNCTAD, 2011, p. 42; Valiante, 2013). After the global financial crisis of 2008, investment patterns in the overall asset class also appear to have shifted. At the onset of the crisis, both financial investment in commodities and commodity prices fell in tandem, suggesting to many critics that financial

652   McGill 450 400 350 300 250 200 150 100 50 0

2005

2006

2007

2008

2009

Barcap estimates of indices

Medium-term notes

ETP-indices

ETP-industrial metals

ETP-energy

ETP-agriculture

ETP-precious metals

Figure  34.2  Financial Investment in Commodities, Assets Under Management, by Product, 2005–​11 (US$ billion). ETP, exchange-​traded product.

participation had given rise to a bubble which burst along with ‘speculative’ bubbles in other asset classes (UNCTAD, 2009, p. 55). Interest in commodities not only rebounded and even surpassed previous levels in the aftermath of the crisis, but investors also increasingly pursued more active strategies and more ‘innovative’ products. Evidence reviewed by UNCTAD (2011, p. 28) suggests not only that financial investment in commodities increased still further since mid-​2010, but also that the relative share of passive investments in indices declined from 65 to 85 per cent in 2005–​07 (the run-​up to the crisis) to 45 per cent in 2010 (Figure 34.2).

The Commodity Financialization Thesis and Antithesis In this context, therefore, the term ‘financialization’ does not denote the trading of derivatives per se, as even ‘genuine’ commercial hedgers use financial instruments (Baffes and Haniotis, 2010). Rather, it refers to the growing and significant presence and influence of participants without a direct commercial interest in a given derivatives market. These actors, which include pension funds, endowments, and hedge funds, trade commodity futures not on the basis of supply and demand fundamentals, but on short-​term considerations related to managing their own portfolios, including price movements in other asset classes such as equities and bonds (Mayer, 2009, p. 1; UNCTAD, 2011, p. 13). As well as actors on the buy side of the securities industry, entities on the sell side—​particularly investment banks—​also dramatically increased their participation in both physical and financial commodity markets during the period under discussion.3 Macroeconomic factors unrelated to commodity supply and demand have thereby come to influence decisions to invest in commodities to an unprecedented degree. Although overall portfolio allocations to commodities remain only a small percentage of total financial-​market investments,4 they are large relative to overall commodity production (Domanski and Heath, 2007, p. 53). Ultimately, however, the proportion of financial participants to commercial actors is thought to be relatively unimportant,

The Financialization Thesis Revisited    653 as even a comparatively small share of financial trades can conceivably change the behaviour of other traders and impact prices (Maugeri, 2009, pp. 159–​60; Fattouh et al., 2012, p. 6). The presence of financial investors has thus further promoted the central role played by expectations in price formation (Kemp, 2012). It must be mentioned here that the term ‘speculation’ has typically been used interchangeably with ‘financialization’, although this tendency has lessened of late. Broadly speaking, speculation has been defined as the purchase of a commodity or related derivative in anticipation of profiting from future price changes. It can take place in physical markets through hoarding and manipulation, as well as in financial markets (Baffes and Haniotis, 2010). It is, however, extremely difficult to articulate a definition of speculation that is not in some way tautological or at least redundant. Even in the narrower sense of ‘financial speculation’ for which it is often shorthand, the term loses much of its significance—​a point that will be revisited later. This point aside, what is ultimately at stake in controversies about financialization is less the definition of the term than the nature of the influence of financial trading on prices—​ if, indeed, such an influence exists at all. This is because the relationship between so-​called financial investment and futures prices is not well understood (Irwin and Sanders, 2010). Nor is the relationship between physical and financial prices well established, despite at least a half-​century’s worth of efforts to do so.5 Various econometric analyses have yielded evidence in support of both arguments (see Baffes and Haniotis, 2010, for a review). While a handful of these studies tend to be heavily cited as strong evidence (if not quite definitive proof) of one side or the other, it is often forgotten that each is making only a relatively modest claim based on a comparatively narrow set of parameters. Nonetheless, both positions can be briefly summarized as follows. On one side, many observers have found the financialization hypothesis to be counterintuitive and empirically unsound on a number of counts. They believe supply and demand fundamentals to be adequate to explain the price dynamics of the super-​cycle, given factors such as the rate of industrialization in China and long lead times on the supply side (Östensson, 2012). Moreover, they rightly point out the fundamental difference between supply and demand dynamics in futures as opposed to pure goods markets: namely that the supply and demand for futures contracts does not correspond directly to the supply and demand for the underlying commodity, and each position in a futures market must be offset against an equal and opposite position. In this instance, the large demand for futures contracts from investors via index funds must be offset by an equal volume of short positions—​ and, as such, there is no net effect of investment in futures contracts on prices. While it is true that dramatic growth in commodity index investment coincided with sharp price rises, correlation has often been uncritically accepted as a clear indicator of causation (Irwin and Sanders, 2010, p. 3), as exemplified by the highly influential US Congressional testimony of former hedge fund manager Michael Masters (2008). Nor is financial investment in and of itself a particularly convincing explanation for price volatility, as volatility has been experienced to similar degrees in commodity markets without a financial component (Frankel and Rose, 2010, p. 10). In a normative sense, too, financial investment in commodity markets is not only unobjectionable to these observers, but it also has a number of beneficial effects. Futures markets tend to function more effectively with the participation of a greater number of actors—​ provided they do not engage in herding behaviour. Absent herding, even the presence of

654   McGill many short-​term traders has a stabilizing effect on prices (Somanathan and Nageswaran 2015, p. 91). Further, many larger institutions, in particular, bring superior market intelligence to bear on their trading activities. Perhaps more importantly than market depth, financial investors also provide a degree of liquidity that had previously been lacking (Radetzki, 2008; Baffes and Haniotis, 2010; UNCTAD, 2011; Östensson, 2012; Sandor, 2012). For their part, critics of financialization assert that the enormous influx of fund activity that began in 2005–​06 has led to a breakdown in the relationship between commodity inventories and prices (Humphreys, 2009; see also Henderson et al., 2015). This has been shown in various studies to have had three effects. Firstly, price comovement—​both across commodity sectors (Baffes, 2009; Basak and Pavlova, 2016) and between commodities and other asset classes (Baffes and Haniotis, 2010; Basak and Pavlova, 2016)—​has been observed. While conventional wisdom had previously held that comovement could not completely be accounted for by fundamentals and that ‘excess’ comovement was caused by herding behaviour on the part of traders (Pindyck and Rotemberg, 1990), it now appears more likely that supply and demand variables—​most notably the greater share of energy input costs in the production of other commodities—​are a sufficient explanation (Ai et al., 2006). As for the observed closer correlation between commodity prices and other asset classes, market-​ wide changes in risk appetite simultaneously affect demand for apparently unrelated assets in the wake of the global financial crisis (Bank of Japan, 2011). Both the degree of correlation and the extent to which it is a lasting phenomenon, rather than simply a byproduct of the extraordinary events of the global financial crisis, remains a matter of dispute (see Büyükşahin, et al., 2008), although there is evidence to suggest that algorithmic trading can exacerbate correlations over short periods (Bicchetti and Maystre, 2012). Such correlation across asset classes can in principle limit investors’ ability to hedge financial risk. Secondly, financialization is alleged to have exacerbated the pronounced price volatility attendant to the recent super-​cycle. Commodity market volatility is accepted, to a certain degree, as inevitable, given that supply and demand are highly inelastic in the short run. Trade on organized exchanges generally reduces price volatility due to greater liquidity and transparency—​a point that is borne out not only in principle, but also by reference to comparisons of price data for traded and non-​traded commodities such as iron ore or molybdenum (Humphreys, 2009). However, the presence of large participants such as financial institutions or commodity funds can potentially move short-​term prices owing to their sheer weight in markets—​an effect that has been exacerbated by the increasing prevalence of high-​frequency trading (UNCTAD, 2011). Financial investment is also feared to have volatility spillover effects from other asset classes, or across different commodities (Tang and Xiong, 2010; Basak and Pavlova, 2016). Volatility is undesirable from a macroeconomic perspective because of its deleterious effects on the terms of trade and fiscal stability of commodity-​dependent economies (Cashin and McDermott, 2001, p. 25). Above a certain threshold it can also ‘contaminate’ the price discovery process. Without price stability, hedging for smaller commercial participants becomes more costly, reducing their involvement in the market. Producer and consumer planning also become subject to greater uncertainty, leading to risk aversion and lack of productive investment—​and, ultimately, further negative repercussions for macroeconomic growth. In these respects, volatility is more salient for policymakers than long-​term price trends (Baffes and Haniotis, 2010). It is, however, a difficult problem to manage: any given episode may be driven by a number of factors, including sensitivity to supply shocks, tighter

The Financialization Thesis Revisited    655 interdependencies between demand for individual commodities, and (less plausibly) inflation. Moreover, it is known that volatility in futures markets can lead to spot price volatility (Bos and van der Molen, 2013, p. 5). Finally, and most importantly, financial involvement in commodity markets is also widely feared—​by financial-​market ‘insiders’ and critics alike—​to exacerbate boom–​bust cycles in spot markets. Like volatility, bubbles are accepted as inevitable for commodities, as well as other markets. Commodity markets are characterized by various types of uncertainty, including medium-​to long-​term supply forecasts (not least due to uncertainty surrounding ultimately recoverable resources of finite mineral commodities); incomplete or unreliable inventory data; and incomplete or unreliable data on global supply and demand dynamics. As such, their participants are already subject to herd behaviour (UNCTAD, 2011, p. 60, n. 17). The presence of financial investors has been a further important factor in altering the commodity price formation process in such a way that it is driven by expectations formed in conditions of heightened uncertainty. In these circumstances investors attach greater weight to present conditions, and so futures prices will closely track spot prices (Fattouh and Scaramozzino, 2011). In more practical terms, one upshot of this breakdown in confidence is that the price formation process is driven largely by short-​termist, myopic incentives and therefore prone to herding behaviour and the formation of asset bubbles (Parsons, 2010; UNCTAD, 2011). Momentum trading has also become a common strategy, with hedge funds and others using statistical analysis to follow or even predict market movements (Parsons, 2010; UNCTAD, 2011). Since 2008, algorithmic or high-​frequency trading has also become more prevalent (Bicchetti and Maystre, 2012). The upshot is that financialization generates unreliable price signals, which causes futures markets to overshoot because financial markets can (over)react to new information more quickly than physical markets can (Dornbush, 1976, cited in Mabro, 2008). It creates greater overall market uncertainty and sends incorrect signals to producers and consumers. The end results include ‘an immense misallocation of resources’ and, from time to time, asset bubbles (Dornbush, 1976, cited in Mabro, 2008, p. 34). More problematic still, ‘an imperfect pricing system can continue to survive unchallenged for a long time until a powerful shock or a series of small shocks exposes its weaknesses and limitations and most importantly alters the balance of power (or perceived power) among the main players’ (Fattouh, 2006, p. 95).

The Defence of Derivatives Commodity futures trading has been plagued by popular suspicion at least since the founding of the Chicago Board of Trade in 1848. More recently this suspicion has been a reaction against the state’s withdrawal from commodity markets beginning in the 1970s in favour of market forces (Radetzki, 2008; Sandor, 2012). A number of reasons for this attitude have been suggested. The transparency of organized exchanges makes price volatility highly visible, while at the same time, these exchanges are perceived to hold a monopoly on information and economic power (Jacks, 2001). Yet a large body of quantitative research shows that price volatility is lower for commodities with centralized trading (see Headey and Fan, 2008; Irwin et al., 2009, cited in Irwin and Sanders, 2010).

656   McGill Perhaps more compelling is popular ‘indignation over the handsome profits generated by agents’ (Jacks, 2001), echoed in the folk assumption that trading exists, in part, to generate outsized profits for an elite rentier class (see ‘The Financialization Thesis and Derivatives Trading’ section). However, such a view confounds two separate normative issues. The implicit critique of the institutional arrangements that permit profits to grow large in the first place is certainly part of a broad-​church effort to curb financial trading profits more generally (see European Commission, 2013). In this respect, the tide has turned against Merton Miller’s (1997) once-​orthodox explanation that ‘The prospect of trading profits is the “bribe”, so to speak, that society uses to motivate the collection, and ultimately the revelation, of the dispersed information about supply and demand’. However, popular suspicion of trading also overlooks the welfare effects of futures exchanges in the form of greater ease of trade and more stable, transparent prices (Newberry and Stiglitz, 1981; Turnovsky, 1983, both cited in Pennings and Leuthold, 2000). In other words, derivatives trading transmits information to ‘promot[e]‌economically desirable adjustment of commodity stocks, thereby reducing price fluctuations’ (Working, 1953, p. 342; see also Pindyck, 2004). Contrast these assumptions with the ‘paper barrels’ view of derivatives contracts, which fail to consider them as representing a service. Indeed, Miller (1997) has dismissed wholesale critiques of derivatives as a form of ‘modern-​day Physiocracy’, rooted in an eighteenth-​ century belief that ‘the ultimate source of national wealth [is] in the production of physical commodities’. Yet contrary to the assumption underpinning the ‘fictitious capital’ and speculation theses alike, it does not follow that derivatives are a form of fictitious capital simply because the volume of futures contracts traded is orders of magnitude larger than the volume of physical commodities in ‘productive’ circulation. Unlike in markets for pure goods, the quantity of futures contracts in a given market does not match the supply of the underlying asset one-​to-​one (Radetzki, 2008; Irwin et al., 2009; Smith, 2012). Expanded trading, in part, reflects market participants’ expectations of greater price changes (Östensson, 2012, p. 21). Another reason the volume of futures contracts outstanding can grow so large is because margin requirements for exchanges are only a fraction—​typically about 5 per cent—​of a given participant’s overall position. In the case of swap markets, the nominal dollar amounts of outstanding swaps are merely ‘notional values’ and ‘bookkeeping conventions’ (Miller, 1997); and the difference between two given commodity prices is traded with swaps. In this vein, ‘cornering’ futures markets is not tantamount to cornering in physical markets (Irwin and Sanders, 2010, p. 8) as each position on an exchange must be matched by an equal and opposite position. Thus more trading does not imply more waste: on the contrary—​at least in principle—​it is a means of price discovery in that traders aggregate disparate information (Miller, 1997; Östensson, 2012, p. 21). For all these reasons, the perceived dichotomy between finance and the ‘real economy’ is, as Pike and Pollard (2009) suggest, specious. The related if slightly less hard-​line tendency is to conflate financialization with ‘speculation’, and so this latter term has typically been liberally used yet ill defined in many discourses—​even within the financial economics literature. Perhaps as a matter of convenience, econometric studies tend to adopt the Commodity Futures Trading Commission’s (CFTC) distinction between commercial and non-​commercial traders to divide hedgers from speculators, although these two categories are widely recognized as somewhat arbitrary. Serious attempts to categorize what types of trading activity properly constitute speculation as opposed to hedging typically remain unresolved, as evidenced, for example, by the

The Financialization Thesis Revisited    657 controversy surrounding the CFTC’s re-​classification in 2009 of various categories of traders (CFTC, 2009). Speculative activities may also reflect a diversity of motivations, and, accordingly, the term has both positive and normative dimensions (Fattouh et al., 2012, p. 3). It would seem that while there is a great deal of overlap between the terms ‘speculation’ and ‘financialization’, the former more often than not is meant to denote the latter. Even as a matter of convenience this is unhelpful owing to the complexities of, and blurry boundaries between, physical and financial trading: many market participants now engage in both physical production and logistics, as well as various types of trading (Baffes and Haniotis, 2010, p. 36). But more to the point, crudely equating speculation with trading in futures contracts, as much of the heterodox critique does, is not simply misleading, but it amounts to a tautology, as ‘Future pricing is by definition speculative’ (Zalik, 2010, p. 554).

Iterated Expectations In a sense, the persistent use of the term ‘speculation’ underscores the central role played by expectations about future supply and demand conditions in the price formation process in both physical (spot) and futures markets (Working, 1942, cited in Fattouh et al., 2012). Even many critics of financialization tend to acknowledge that a certain amount of speculation in this sense is beneficial to, and indeed necessary for, the proper functioning of commodity futures markets, as speculators offer both greater liquidity and more sophisticated hedging instruments. Crucially, however, this participation is beneficial only if financial investors are not simply engaging in herding behaviour or noise trading, as this creates the conditions for a speculative bubble to form (Radetzki, 2008). While such bubbles are, in turn, often mistaken for manipulation, ‘the beliefs driving a bubble can gain traction without there being any identifiable individuals behind it’ (Parsons, 2010, p. 109). In order to identify an episode of speculation with any accuracy, it would therefore be essential to identify different types of financial participants, as well as their trading strategies and tactics (Fattouh, 2010, p. 1; Sandor, 2012, p. 550). For these reasons, as Working (1963, pp. 22–​23) pointed out over half a century ago, proving or disproving that ‘speculative’ futures trading causes price instability is impossible without empirical evidence of the drivers of speculative behaviour. There is another sense in which the channels through which financial trading might influence commodity prices are opaque: real-​time data on commodity prices, as well as production, inventories, and trade, is incomplete and often of questionable accuracy. While this is problematic for studies of the possible impacts of financial investors on prices, it is also true that critics of the financialization thesis have the impossible task of proving a negative (Jarecki, 2011, p. 10), and so, in an important sense, no amount of high-​quality data will prove their case. In any event, it is doubtful whether greater volumes of more reliable data on physical markets would eliminate the problems that financialization is believed to cause. In all trading environments, ‘fixing’ information asymmetry is tantamount to creating conditions for information overload (Kay, 2012, pp. 71–​72; see also Fattouh and Allsopp, 2009, p. 2). Yet the point remains that even if individual traders had ‘sufficient’ information on supply and demand fundamentals and other traders’ motivations at the moment each trade is executed, it is their perceptions and expectations that ‘move their “animal spirits” ’ to translate this information into a given price’ (Mabro, 2008, p. 3; see also Akerlof and Shiller, 2009).

658   McGill More to the point, financial trading’s effect on commodity prices is so difficult to discern because comprehensive information about investor motivations and behaviour is simply unavailable (Radetzki, 2008; Humphreys, 2009). Yet neither these data nor econometric ‘proof ’ of financialization is necessary for financial investors to shape markets. Recall once again the centrality of expectations in commodity price formation, financialization or not. When coupled with existing institutional arrangements that incentivize short-​over long-​ term gains not only for financial investors but for ‘commercial’ producers, consumers, and traders (see Kay, 2012; Clark, 2013)—​in other words, ‘genuine hedgers’—​such an environment, in turn, exacerbates the innate human tendency towards short-​term thinking (Kay, 2012; Clark, 2013). As discussed in the ‘Commodity Trading and the Financialization Thesis’ section, this amounts to attaching outsize weight to present and near-​term conditions, where in reality commodity prices are primarily determined by cyclical (medium-​term) and structural (long-​term) supply and demand factors. Given that financial investors have been a sizeable presence in commodity markets for a decade at the time of writing, it can be expected that, as in other financial markets, greater volumes of trading activity have further enhanced this effect (Kay, 2012, p. 40). Conceivably, the very suspicion on the part of a sufficient proportion of all traders that prices reflect information from other asset classes, as well as commodity supply and demand fundamentals, will shift their short-​term expectations accordingly (UNCTAD, 2011, p. 29; Smith, 2012). This belief has, in some cases, found its way into the underpinnings of the forecasts and strategies employed by physical traders (UNCTAD, 2011; Terazono, 2012). If this is, indeed, the case, the price formation process for exchange-​traded commodities can be said to have taken on the character of Keynes’s ‘beauty contest’, or more formally Allen et al.’s (2006) ‘iterated expectations’, in which prices reflect participants’ expectations about others’ expectations ad infinitum (Allsopp and Fattouh, 2011). In this sense, the identities of trading parties is only of secondary importance at best as motivations and tactics become homogenized. Instead, their models have a more complex and intimate relationship to the very phenomenon they ostensibly do no more than describe (see Barry and Slater, 2005). The financialization of commodity markets thus resembles many similar tendencies in other securities markets (see Kay, 2012). Put differently, in the manner described by Akerlof and Shiller (2009, p.54, cited in Fattouh, 2011), participants’ beliefs or ‘stories’ have become ‘a real part of how the economy functions’, and so ‘The stories no longer merely explain the facts; they are the facts’.

Conclusions and Implications for the Financialization Thesis A close reading and synthesis of various versions of the commodity financialization thesis reveals many disagreements, most obviously over the societal value of futures trading. This is particularly true in the wake of the global financial crisis: the strong normative expectation of commodity trading on the part of society mirrors or coincides with evident societal expectations of banking and financial intermediaries. However, there is also much overlap between the heterodox and orthodox accounts: in particular, a concern that financial

The Financialization Thesis Revisited    659 investment exacerbates both commodity price volatility and boom–​bust cycles. Here it may be tempting to conclude that financialization has created both a new governing logic and an economic space more or less independent of ‘the real economy’. However, semantic and conceptual ambiguity and confusion surround commodity markets and derivatives trading, and this chapter has attempted to clarify some key issues. What can ultimately be gleaned from this exercise is that the strong presence of financial investors in these markets has helped to promote a short-​termist, expectations-​driven approach to pricing. Given the loss of faith in the price formation process stemming from these and other sources of uncertainty, it is not surprising that financial participation in commodity markets should find itself a scapegoat once more and be reviled as mere ‘speculation’. While it has long been feared that financialization might have deleterious effects on commodity prices, what is now of equal or greater concern to critics is that the growing resemblance of commodity markets to other financial markets might have made them a source of systemic financial rather than resource risk. In this respect, commodity markets are but one source of post-​crisis anxiety about macroeconomic and financial stability—​if not, indeed, nostalgia for a supposedly more stable, orderly era when financial markets satisfactorily performed their ‘proper’ functions of raising capital for productive investment, transferring risk, and safeguarding savings (Kay, 2012). Yet financialization is not necessarily a historically unprecedented, or indeed permanent, phenomenon (Krippner, 2005, p. 199). Moreover, as Kindleberger (1978), Minsky (1977, 1992), and their intellectual heirs have long observed, regulation and oversight, while going a considerable way to smooth out peaks and troughs in economic cycles, cannot entirely eliminate them. None of this is to dismiss out of hand the long-​lasting relevance of financialization in commodity markets. Even if the super-​c ycle has come to an end, certain commodities may continue to hold their relevance for investors as a proxy for the long-​term material growth prospects of emerging markets. Ultimately, however, by attaching outsized importance to the role of financial trading in setting commodity prices—​by characterizing the impact of financialization as a totalizing, transformative force—​critics have, ironically enough, lost sight of both the long-​term and the ‘real’ components of commodity markets. Coupled with faltering confidence about the sustainability and security of supply, growth in emerging economies has been more than sufficient to reshape pricing expectations. While not untouched by financialization, then, commodity markets can never entirely succumb to its logic.

Notes 1. Muellerleile (2009) himself provides one notable exception; see also Pike and Pollard (2009) as well as contributions to the November 2013 special issue of the Cambridge Journal of Regions, Economy and Society. 2. Investors can gain exposure to commodity index returns via three other financial instruments: commodity index swaps, exchange traded funds, and exchange traded notes. See Tang and Xiong (2010, p. 6, n. 9). 3. The role of the sell side was not often explicitly considered in discussions on financialization in commodity markets until relatively recently. While critically important and increasingly controversial, the changing institutional landscape of commodity trading

660   McGill constitutes another sense in which the term ‘financialization’ may be used in this context and thus lies outside the primary scope of this chapter (but see Wójcik (2012) for a general discussion of the power of investment banks). 4. In 2005, commodity futures and options accounted for only 8% of all derivatives contracts traded globally (UNCTAD, 2006). 5. Bos and van der Molen (2013, p. 4) provide a brief overview of this literature; see also Mabro (2008); Fattouh (2010, p. 5).

References Ai, C., Chatrath, A., and Song, F. (2006). ‘On the comovement of commodity prices’. American Journal of Agricultural Economics 88: 574–​588. Akerlof, G.A. and Shiller, R.L. (2009). Animal Spirits:  How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism (Princeton, NJ: Princeton University Press). Allen, F., Morris, S., and Shin, H.S. (2006). ‘Beauty contests, bubbles and iterated expectations in asset markets’. Review of Financial Studies 19: 719–​752. Allsopp, C. and Fattouh, B. (2011). ‘Oil and international energy’. Oxford Review of Economic Policy 27: 1–​32. Baffes, J. (2009). ‘More on the energy/​non-​energy commodity price link’. World Bank Working Paper No. 4982 (Washington, DC: World Bank). Baffes, J. and Haniotis, T. (2010). ‘Placing the 2006/​08 commodity boom in perspective’. World Bank Policy Research Working Paper no.5371 (Washington, DC: World Bank). Bank of Japan (2011). Bank of Japan Review: Recent Surge in Global Commodity Prices—​Impact of Financialization of Commodities and Globally Accommodative Monetary Conditions (Tokyo: Bank of Japan). Barry, A. and Slater, D. (2005). The Technological Economy (Oxford: Routledge). Basak, S. and Pavlova, A. (2016). ‘A model of financialization of commodities’. Journal of Finance 71: 1511–​1556. Bicchetti, D. and Maystre, N. (2012). ‘The synchronised and long-​lasting structural change on commodity markets:  evidence from high-​frequency data’. MPRA paper no.  37486 (Munich: Munich Personal RePEc Archive). BIS (2011). ??? Bos, J.W.B. and van der Molen, M. (2013). ‘A bitter brew? Speculation and commodity prices’. Maastricht University Working Paper RM/​12/​044. Bryan, D. and Rafferty, M. (2006). Capitalism With Derivatives:  A  Political Economy of Financial Derivatives, Capital and Class (Basingstoke: Palgrave Macmillan). Büyükşahin, B., Haigh, M.S., and Robe, M. (2008). Commodities and Equities: A ‘Market of One’? (Washington, DC: Commodity Futures Trading Commission). Cashin, P. and McDermott, C.J. (2001). ‘The long-​run behaviour of commodity prices: small trends and big variability’. IMF Working Paper no. 01/​68 (Washington, DC: IMF). CFTC (2009). ‘Disaggregated commitments of traders report: explanatory notes’ http://​www. cftc.gov/​ucm/​groups/​public/​@commitmentsoftraders/​documents/​file/​disaggregatedcotexplanatorynot.pdf (last accessed 12 April 2017). Christophers, B. (2011). ‘Follow the thing: money’. Environment and Planning D: Society and Space 29: 1068–​1084.

The Financialization Thesis Revisited    661 Christopherson, S., Martin, R., and Pollard, J. (2013). ‘Financialization: roots and repercussions’. Cambridge Journal of Regions, Economy and Society 6: 351–​357. Clapp, J. and Helleiner, E. (2012). ‘Troubled futures? The global food crisis and the politics of agricultural derivatives regulation’. Review of International Political Economy 19: 187–​207. Clark, G.L. (2011). ‘Myopia and the global financial crisis: context-​specific reasoning, market structure, and institutional governance’. Dialogues in Human Geography 1: 4–​25. Clark, G.L. (2013). ‘The Kay Review on long-​horizon investing:  a guide for the perplexed’. Rotman International Journal of Pension Management 6: 58–​63. Clark, G.L. and Monk, A.H.B. (2013). ‘Financial institutions, information, and investing-​at-​a-​ distance’. Environment and Planning A 45: 1318–​1336. Domanski, D. and Health, A. (2007). ‘Financial investors and commodity markets’. BIS Quarterly Review March. Epstein, G.A. (2005). ‘Financialization and the World Economy’ in G.A. Epstein (ed.) Financialization and the World Economy, Ch. 1 (Cheltenham: Edward Elgar). European Commission (2013). ‘Proposal for a Council Directive implementing enhanced cooperation in the area of financial transaction tax. European Commission COM(2013) 71 Final’. http://​ec.europa.eu/​taxation_​customs/​sites/​taxation/​files/​resources/​documents/​taxation/​com_​2013_​7 1_​en.pdf (last accessed 17 May 2017). Fattouh, B. (2006). ‘The Origins and Evolution of the Current International Oil Pricing System: A Critical Assessment’ in R. Mabro (ed.) Oil in the 21st Century: Issues, Challenges and Opportunities, Ch. 3 (Oxford: Oxford University Press). Fattouh, B. (2010). ‘An anatomy of the oil pricing system’. OIES Energy Comment September 2010 (Oxford: Oxford Institute for Energy Studies). Fattouh, B. (2011). ‘An anatomy of the oil pricing system’. OIES WPM 40, January 2011 (Oxford: Oxford Institute for Energy Studies). Fattouh, B. and Allsopp, C. (2009). ‘The price band and oil dynamics’. Oxford Energy Comment July 2009 (Oxford: Oxford Institute for Energy Studies). Fattouh, B. and Scaramozzino, P. (2011). ‘Uncertainty, expectations, and fundamentals: whatever happened to long-​term oil prices?’ Oxford Review of Economic Policy 27: 186–​206. Fattouh, B., Kilian, L., and Mahadeva, L. (2012). ‘The role of speculation in oil markets: what have we learned so far?’ OIES WPM 45 (Oxford: Oxford Institute for Energy Studies). Frankel, J.A. and Rose, A.K. (2010). ‘Determinants of agricultural and mineral commodity prices’. John F. Kennedy School of Government Working Paper Series rwp 10-​038 (Cambridge, MA: Harvard University Press). French, S., Leyshon, A., and Wainwright, T. (2011). ‘Financializing space, spacing financialization’. Progress in Human Geography 35: 798–​819. Gorton, G. and Rouwenhurst, K.G. (2004). ‘Facts and fantasies about commodity futures’. NBER Working Paper No. 10595 (Cambridge, MA:  National Bureau of Economic Research, Inc.). Haldane, A. and Davies, R. (2011). ‘The short long’. Speech to the 29th Société Universitaire Européenne de Recherches Financières Colloquium (Brussels, May 2011) http://​www. bankofengland.co.uk/​archive/​Documents/​historicpubs/​speeches/​2011/​speech495.pdf (last accessed 11 May 2017). Harvey, D. (2006). The Limits to Capital (London: Verso). Harvey, D. (2011). The Enigma of Capital and the Crises of Capitalism (London: Profile). Heady, D. and Fan, S. (2008). ‘Anatomy of a crisis: the causes and consequences of surging food prices’. Agricultural Economics 39: 375–​391.

662   McGill Henderson, B.J., Pearson, N.D., and Wang, L. (2015). ‘New evidence on the financialization of commodity markets’. Review of Financial Studies 28: 1285–​1311. Humphreys, D. (2009). ‘The great metals boom: a retrospective’. Resources Policy 35: 1–​13. Irwin, S.H. and Sanders, D.R. (2010). ‘The impact of index and swap funds on commodity futures markets: preliminary results’. OECD Food, Agriculture and Fisheries Working Papers, No. 27 (Paris: OECD Publishing). Irwin, S.H., Sanders, D.R., and Merrin, R.P. (2009). ‘Devil or angel? The role of speculation in the recent commodity futures price boom (and bust)’. Journal of Agricultural and Applied Economics 41: 393–​402. Jacks, D.S. (2001). ‘Populists vs. theorists:  futures markets and the volatility of prices’. Explorations in Economic History 44: 342–​362. Jarecki, H.G. (2011). ‘The relationship between commodity futures trading and physical commodity prices’. Lecture given on 5 April http://​www.cmegroup.com/​trading/​agricultural/​files/​Dr-​ Henry-​G-​Jarecki-​Lecture-​on-​Commodity-​Futures-​Trading-​and-​Prices.pdf (last accessed 5 October 2013). Kay, J. (2012). The Kay Review of UK Equity Markets and Long-​ term Decision-​ Making (London: Department for Business, Innovation and Skills). Kemp, J. (2012). ‘Are commodities at risk of de-​financialization?’ Reuters, 30 May http://​www. reuters.com/​article/​us-​column-​kemp-​commodity-​super-​cycle-​idUSBRE84T0P720120530 (last accessed 11 May 2017). Kindleberger, C. (1978). Manias, Panics, and Crashes: A History of Financial Crises (New York: Basic Books). Krippner, G.R. (2005). ‘The financialization of the world economy’. Socio-​Economic Review 3: 173–​208. Labban, M. (2010). ‘Oil in parallax: scarcity, markets, and the financialization of accumulation’. Geoforum 41: 541–​552. Lee, R., Clark, G.L., Pollard, J., and Leyshon, A. (2009). ‘The remit of financial geography—​ before and after the crisis’. Journal of Economic Geography 9: 723–​747. Leyshon, A. and Thrift, N. (2007). ‘The capitalization of almost everything:  the future of finance and capitalism’. Theory, Culture and Society 24: 94–​115. LiPuma, E. and Lee, B. (2004). Financial Derivatives and the Globalisation of Risk (Durham, NC: Duke University Press). Masters, M.W. (2008). Testimony before the Committee on Homeland Security and Governmental Affairs (Washington, DC: United States Senate). Mayer, J. (2009). The Growing Interdependence Between Financial and Commodity Markets. UNCTAD Discussion Papers No. 195 (New York and Geneva: United Nations Conference on Trade and Development). Mabro, R. (2008). ‘The oil price conundrum’. Oxford Energy Forum 74: 12–​14. Maugeri, L. (2009). ‘Understanding oil price behaviour through an analysis of a crisis’. Review of Environmental Economics and Policy 3: 147–​166. Merton, R.C. and Bodie, Z. (2005). ‘Design of financial systems: towards a synthesis of function and structure’. Journal of Investment Management 3: 1–​23. Miller, M.H. (1997). Merton Miller on Derivatives (New York: Wiley). Minsky, H.P. (1977). ‘A Theory of Systemic Fragility’ in E.I. Altman and A.W. Sametz (eds) Financial Crises:  Institutions and Markets in a Fragile Environment (New  York:  John Wiley).

The Financialization Thesis Revisited    663 Minsky, H.P. (1992). ‘The financial instability hypothesis’ SSRN http://​papers.ssrn.com/​sol3/​ papers.cfm?abstract_​id=161024 (last accessed 12 April 2017). Muellerleile, C.M. (2009). ‘Financialization takes off at Boeing’. Journal of Economic Geography 9: 663–​667. O’Connor, J. (1998). Natural Causes: Essays in Ecological Marxism (New York: Guilford Press). O’Neill, P. (2001). ‘Financial narratives of the modern corporation’. Journal of Economic Geography 1: 181–​199. Östensson, O. (2012). ‘The 2008 commodity price boom: did speculation play a role?’ Mineral Economics 25: 17–​28. O’Sullivan, D. (2009). Petromania: Black Gold, Paper Barrels and Oil Price Bubbles (Petersfield: Harriman House). Parsons, J.E. (2010). ‘Black gold and fool’s gold: speculation in the oil futures market’. Economia 10: 81–​116. Pennings, J.M.E. and Leuthold, R.M. (2000). ‘The motivation for hedging revisited’. Journal of Futures Markets 20: 865–​885. Pike, A. and Pollard, J. (2009). ‘Economic geographies of financialization’. Economic Geography 86: 29–​51. Pindyck, R.S. (2004). ‘Volatility and commodity price dynamics’. Journal of Futures Markets 24: 1029–​1047. Pindyck, R.S. and Rotemberg, J.J. (1990). ‘Do Stock Prices Move Together Too Much?’ NBER Working Papers 3324 (Cambridge, MA: National Bureau of Economic Research). Pryke, M. and Allen, J. (2000). ‘Monetised time-​space:  derivatives—​money’s “new imaginary”?’ Economy and Society 29: 264–​284. Radetzki, M. (2008). A Handbook of Primary Commodities in the Global Economy (Cambridge: Cambridge University Press). Sandor, R. (2012). Good Derivatives:  A  Story of Financial and Environmental Innovation (Hoboken, NJ: Wiley). Sinn, H. (2010). Casino Capitalism: How the Financial Crisis Came About and What Needs to be Done Now (Oxford: Oxford University Press). Smith, J.L. (2012). ‘Does speculation drive oil prices?’ Resources 181: 34–​39. Somanathan, T.V. and Nageswaran, V.A. (2015). The Economics of Derivatives (Cambridge: Cambridge University Press). Strange, S. (1986). Casino Capitalism (Oxford: Basil Blackwell). Tang, K. and Xiong, W. (2010). ‘Index investment and financialization of commodities’. NBER Working Papers 16385 (Cambridge, MA: National Bureau of Economic Research). Terazono, E. (2012). ‘Commodities traders lose physical edge’. Financial Times 2 May. UNCTAD (2006). Overview of the World’s Commodity Exchanges (New  York and Geneva: United Nations Conference on Trade and Development). UNCTAD (2009). ‘The Financialization of Commodity Markets’ in Trade and Development Report 2009, Ch. 2 (New  York and Geneva:  United Nations Conference on Trade and Development). UNCTAD (2011). Price Formation in Financialized Commodity Markets: The Role of Information (New York and Geneva: United Nations Conference on Trade and Development). Valiante, D. (2013). Price Formation in Commodities Markets:  Financialisation and Beyond (Brussels: Centre for European Policy Studies). Working, H. (1953). ‘Futures trading and hedging’. American Economic Review 43: 314–​343.

664   McGill Working, H. (1963). ‘Futures markets under renewed attack’. Food Research Institute Studies 4: 13–​24. Wójcik, D. (2012). ‘The end of investment bank capitalism? An economic geography of financial jobs and power’. Economic Geography 88: 345–​368. Zalik, A. (2010). ‘Oil “futures”: Shell’s Scenarios and the social constitution of the global oil market’. Geoforum 41: 553–​564.

Chapter 35

Vulnerable Re g i ons in a Changing C l i mat e Robin Leichenko Introduction Earth has entered the epoch of Anthropocene whereby human actions are widely recognized as an influential force in planetary biophysical and geological systems. While the 2015 Paris Agreement on reduction of greenhouse gas emissions offers hope that the most dire scenarios of climate change may be avoided, economic disruptions associated with climate-​ related extreme events and climate-​induced loss and damage are expected for the foreseeable future. For regional economies, highly visible climate events such as unprecedented flooding in Chennai, India, in 2015, which inundated 90 per cent of the city and displaced more than two million people; Typhon Haiyan (Yolanda) in 2013, which killed more than 6000 people in the Philippines; and Hurricane Sandy in 2012, which caused more than US$60 billion in property and infrastructure damage to the east coast of the USA, have focused attention on risks and vulnerabilities associated climate change. Growing awareness of the regional economic impacts of climate-​related stresses, such as declining water availability, deterioration of resource-​based livelihoods, and coastal inundation as the result of sea-​level rise, has further reinforced interest in the development of strategies to build regional climate resilience. Although economic geographers have made important contributions to the understanding of many facets of climate change, including formation of new carbon economies, evolution of energy regimes, development of carbon-​offset schemes, and determinants of regional resilience (e.g. Bumpus and Liverman, 2008; Boykoff et al., 2009; Bradshaw, 2010; Bridge, 2010; Knox-​Hayes, 2010; Essletzbicher, 2012; Bridge et al., 2013; Martin and Sunley, 2015), the field has had relatively limited engagement with study of economic impacts, vulnerability, and adaptation to climate change. Instead, most work on the economic consequences of and responses to climate disruption is being done by researchers in economics, planning, policy, and engineering. Indeed, a new field of ‘climate economics’ has emerged around exploration of how variation in weather conditions, including temperature, precipitation, and windstorms, affect local, regional, and national economies (Dell et al., 2014). Related subfields

666   Leichenko explore the effects of extreme climate events for urban and regional economic growth and development (e.g. Hallegatte et al., 2011), poverty rates (e.g. Skoufias et al., 2012), and housing and property markets (e.g. Pryce et al., 2011). While economics and allied fields have dominated study of the impacts of climate change, research on economic vulnerability to climate change is primarily happening within the interdisciplinary fields of human dimensions of climate change and disaster risk reduction, as well as in subfields of geography, including political ecology, natural hazards, environmental justice, and international development (Fuller and Pincetl, 2015). Climate vulnerability research emphasizes identification of communities, sectors, social groups, and households that are more likely than others to be harmed by climate change or are less able to adapt to climate-​related shocks and stresses. This research typically incorporates economic variables and recognizes a role for economic processes, such as globalization, in shaping vulnerabilities, yet the central focus of the work is generally directed towards understanding uneven spatial, social, and livelihood vulnerabilities, rather than towards issues that are of more central concern to economic geographers such as impacts on supply chains, regional labour markets, and future patterns of regional growth and development. Furthermore, only limited attention has been directed within economic geography to questions of how vulnerability is generated and maintained by processes such as neo-​liberalization and financialization. This chapter argues that growing public and policy attention to climate-​related economic disruption, combined with broad recognition of humanity’s role in shaping planetary systems, presents a pivotal moment for economic geographers to take a more central role in climate impact and vulnerability studies and in larger, interdisciplinary conversations about the meaning and implications of the Anthropocene. The first part of the chapter defines economic vulnerability and describes important developments in vulnerability thinking. The next section assesses the state of research on regional impacts and vulnerability, showing how ongoing work on spatial, sectoral, and household vulnerabilities provides a strong foundation for answering questions such as which local and regional economies are most vulnerable to the impacts of climate change, and which sectors and which types of workers are most susceptible to harm and least able to bounce back. The chapter then highlights emerging work on the production of economic vulnerability, which is beginning to ask questions about how market-​driven responses to climate risks are shaping vulnerabilities. The chapter concludes by identifying opportunities for economic geographers to further explore impacts and vulnerabilities and to engage with the interdisciplinary global change research community.

Defining Economic Vulnerability Vulnerability to climate change is generally understood as a propensity to be harmed by climate shocks and stresses (Intergovernmental Panel on Climate Change, 2012). Climatic shocks include extreme events, such as hurricanes and heat waves. Climatic stresses entail longer-​term changes such as sea-​level rise and gradually warming temperatures. Within the climate change literature, definitions of vulnerability typically take into account physical exposure to climate shocks and stresses, degree of sensitivity or susceptibility to harm from these events, and the ability to respond and recover.1 The highly uneven nature of

Vulnerable Regions in a Changing Climate    667 vulnerability within and across regions, communities, and households stems from differences in exposure to climate extremes, as well as variations in social, demographic economic, institutional, and technological factors that influence both susceptibility to harm and capacity to respond (Liverman, 1990; Adger, et al., 2003; Eakin and Luers, 2006; Barnett and Eakin, 2015). The notion of economic resilience incorporates consideration of how quickly regions, communities, or households might recover from climate shocks and reduce vulnerability to future events (Leichenko et al., 2015). Although vulnerability and resilience are intertwined,2 resilience is touched upon only briefly in this chapter because it is explored elsewhere in the volume (see Chapter 45). As with the concept of vulnerability, economic vulnerability has a range of different usages and interpretations (Leichenko et al., 2014). In the economics literature, economic vulnerability is typically used to specify the degree to which national economies may be more or less subject to harm from external economic shocks and perturbations (Briguglio et al., 2009). Economic shocks stem from factors such as changes in trade policy, fluctuations in exchange rates, or shifts in commodity prices. Climate change has been implicated as a source of economic shocks, including dramatic shifts in food commodity prices (Wheeler and von Braun, 2013), with significant negative implications for economic growth and poverty within and across the Global South (Ahmed et al., 2009; Hertel et al., 2010). According to this line of work, higher levels of economic vulnerability to climate change are associated with dependency on climate-​sensitive commodity sectors, such as agriculture and natural resources, in combination with lower overall income levels (Dell et al., 2012). The intersection of climatic and economic shocks is also addressed within the literature on vulnerability to multiple stressors. This work highlights the fact that climatic shocks and stresses do not happen in isolation, but are interwoven with other processes of economic, political, and social change (Leichenko and O’Brien, 2008; Casale et al., 2010; Silva et al., 2010; Jeffers, 2013; Burton and Peoples, 2014; McCubbin et al., 2015; Rhiney, 2015). Work by economic geographers demonstrates how exposure to globalization-​related economic stresses in combination with climatic shocks can reinforce existing patterns of uneven development, and also create new and unexpected vulnerabilities (Leichenko et al., 2010; Silva et al., 2010, 2015). There is also recognition that regional vulnerabilities are highly interconnected spatially and sectorally as the result of global trade linkages and foreign investment patterns, and insurance and reinsurance markets (Liverman, 2015a). The emergence and growing usage of the concept of ‘teleconnections’ within the human dimensions of global change literature (e.g. Adger et al., 2009; Seto et al., 2012; Moser and Hart, 2015) to describe how global flows of goods, services, and information may transmit vulnerabilities across space and time provides but one example of how concepts from economic geography are shaping understanding of regional climate impacts and vulnerabilities. Research also highlights the dynamic nature of economic vulnerability to climate change (Jeffers, 2013; Mechler and Bouwer, 2015). Vulnerability is continually evolving as exogenous economic and political processes such as marketization and financialization alter the capacity of regions and households to respond and adapt to climatic stresses. Work by Jeffers (2013) shows how the growing dominance of a neo-​liberal discourse of development planning shapes decision-​making processes associated with responding to economic and climatic stresses, contributing to an emphasis on technological approaches to climate risks (Jeffers, 2013). The dynamic nature of economic vulnerability is also apparent in the context of post-​disaster learning. Decision-​makers in some regions that experience extreme events

668   Leichenko are found to learn from these events and to take these lessons into account when siting new development, making locational decisions, and planning for future events, all of which shape subsequent vulnerabilities (Mechler and Bouwer, 2015). While space does not permit full consideration of critical perspectives on vulnerability (e.g. Birkenholtz, 2012; Bassett and Fogelman, 2013; Tschakert et al., 2013; Grove, 2014; Ribot, 2014), it is important to acknowledge some of the key limitations of vulnerability approaches. One important shortcoming of much vulnerability work over the last decade is that the emphasis has shifted away from ‘root causes’ of vulnerability (Ribot, 2014). Although explicit emphasis on political economy and challenges to status quo understandings of environmental hazards were key components of early vulnerability work (e.g. Liverman, 1990; Dow, 1992; Watts and Bohle, 1993; Bohle et al., 1994; Wisner et al., 1994; Cutter, 1996), this dimension is often lost in newer studies that frame the work primarily in terms of exposure to physical processes such as flooding and sea-​level rise. As a consequence, much vulnerability work focuses on documentation and quantification of locations, sectors, and populations likely to be exposed to climate shocks or stresses and who have characteristics that make them more prone to harm or less able to respond, with little or no questioning of forces putting people in harm’s way. Proposals to address vulnerability that emerge from applied studies have also been subject to critique. Because this work has little acknowledgement of underlying social processes that create vulnerability, efforts to identify vulnerable locations may inadvertently reinforce official narratives of disadvantage, as well as power structures that perpetuate these disadvantages (Yamane, 2009; Preston et al., 2011). The work has sometimes been labelled ‘post-​ political’ (e.g. Swyngedouw, 2010), in that it emphasizes technical and managerial measures, such as improvements in infrastructure that reduce physical exposure, changes in land-​use policy, and provision or expansion of early warning systems, rather than addressing power and wealth differentials that create and maintain precarity. All of these critiques reveal the need for more attention to economic processes that are shaping vulnerabilities, as well as explicit recognition of how different discourses are exercised in the application of vulnerability approaches. As discussed in the section ‘Producing Vulnerability’, economic geographers are beginning to explore many of these issues.

Studying Regional Economic Vulnerability Studies of regional economic vulnerability consider whether and how a region’s economic assets, economic activities, workers, and households might be directly or indirectly affected by climate-​related shocks and stresses. The work explores how and why regional economic vulnerability varies spatially and considers whether efforts to respond to climate change impacts may create new types of vulnerabilities. Although detailed discussion of climate mitigation policies, such as a tax on carbon, is beyond the scope of this chapter, it is important to recognize that regional economic vulnerabilities may also arise as the result of policy responses intended to limit or sequester greenhouse gas emissions. Mitigation policies can be expected to have significant consequences for many regional economies, especially for regions that produce or are highly dependent upon fossil fuels. For example, the demise of the coal industry in the USA due to a combination of pollution-​limiting policies and

Vulnerable Regions in a Changing Climate    669 growing competition from natural gas produced via hydrological fracturing, is dramatically reshaping local economies in coal-​producing regions (Vancura, 2015).

Which Economic Assets are Vulnerable? The need for identification and valuation of economic assets that are likely to be exposed to climate extremes has become a major concern for cities and regions worldwide. Research on this topic explores exposure of economic assets, including property, physical capital, and inventories that are directly exposed to prominent facets of climate change such as storm events and sea-​level rise (Leichenko at al., 2014). Studies focused on vulnerability to extreme storms generally emphasize costs associated with storm-​related damage to property and infrastructure, costs of business interruption, and secondary impacts on regional economies. Using indicators such as number of affected business establishments, taxable sales, production and employment, housing prices, and wages, this work estimates and projects damage costs associated with past and future storm events (Leichenko and Thomas, 2012). Studies of regional exposure to rising sea levels investigate projected exposure of economic assets over many decades (Bosello and De Cian, 2014). The studies typically overlay projections of sea-​level rise onto property or parcel maps in order quantify number of properties exposed, total property values, municipal tax bases, and infrastructure over various time horizons (e.g. 2020, 2050, and 2080) or for various scenarios of greenhouse emissions (e.g. Kirshen et al., 2008; Tate and Frazier, 2013; Maloney and Preston, 2014; Brady et al., 2015; Neumann et al., 2015). While studies of assets at risk are drawing public and policy attention to the potential economic costs of climate change, there are a number of areas where additional input from economic geographers might enhance this work. Most exposure studies focus on coastal regions of the Global North, particularly the US Gulf Coast, the Atlantic Seaboard, coastal Alaska, and major port cities such as London and Rotterdam. Work is beginning to appear for other areas, such as Dar es Salaam (Kebede and Nichols, 2012) and the South Pacific (Kumar and Taylor, 2015), but there is a need for more attention to highly exposed regions of the Global South, including low-​lying megacities, island nations, and ecologically important coastal zones (de Sherbinin et al., 2007; McGranahan et al., 2007; Rhiney, 2015). There is also a need for consideration of economic exposure to other types of climatic stresses, such as droughts and heat waves that may affect non-​coastal areas, and regions that depend upon climate-​ sensitive ecosystem services such as dryland agriculture or glacier-​based water supplies. Another area for potential contribution by economic geographers would be to incorporate other dimensions of ‘economic’ into estimation of exposure. In particular, there is a need for greater recognition of non-​market and non-​monetary values-​associated ecosystem services, cultural heritage sites, and assets that form the basis for informal economies (Leichenko and Thomas, 2012; Brady, 2015).

Which Regions are Vulnerable? Exploration of spatial patterns and determinants of regional economic vulnerability and resilience is another important research area. Work in this realm draws, in part, from

670   Leichenko geography’s vulnerability mapping tradition, with an emphasis on comparative assessment of exposure and susceptibility to climate and environmental hazards among cross-​ sections of states, counties, districts, or other political units (e.g. Cutter et al., 2003, 2008, 2014; O’Brien et al., 2004b). The studies often measure vulnerability via composite indices that incorporate a wide range of social, technological, institutional, environmental, and economic variables that are thought to influence susceptibility to harm or capacity to respond. While most spatial vulnerability work falls more squarely within hazards and human dimensions traditions, a number of studies adopt a more explicitly economic focus (e.g. Frazier et al., 2010; Leichenko and Solecki, 2013; Thatcher et al., 2013; Boero et al., 2015). Thatcher et al. (2013), for example, develop an index of economic vulnerability for counties in the US northern Gulf Coast that incorporates factors thought to contribute to societal risk from rising sea level, including value of residential and commercial buildings and types of infrastructure. Other studies focus on how and why regional patterns of economic vulnerability are changing over time. Preston (2013) draws on concepts including path dependency and lock-​in to investigate expected future patterns of vulnerability across US counties. Future socio-​economic conditions are projected based on extrapolation of growth trends for population, income, and earnings growth, which dictate future exposure to natural hazards of all types. The study notes that, without significant transformation of development patterns, large increases in economic losses from natural hazards are likely, even before accounting for the potential effects of climate change (Preston, 2013). The connection between climate change and spatial inequality is another topic of interest for economic geography. While the possibility that climate change will exacerbate inequalities is well recognized (Intergovernmental Panel on Climate Change, 2014), empirical studies have only begun to shed light on this issue at the regional level. A study by Silva et al. (2015), for example, explores linkages between extreme weather events, economic shocks, and regional inequalities within Mozambique. The study demonstrates that climatic and economic shocks exacerbate both income and power disparities in most regions, but there are some cases where disparities and polarization decline following climate and economic shocks. Further exploration of those unexpected cases suggested that the shocks had contributed to a so-​called ‘poverty trap’ whereby regions experienced deteriorating overall levels of income and wealth. Regions that were prone to poverty traps were found to have high dependency on agriculture and limited diversity of economic opportunity. Research also finds evidence that poverty traps may result from persistent climate stresses. In examining how sea-​level rise may affect regional economic growth and poverty dynamics over long time horizons, Hallegatte (2012) demonstrates that poverty traps may be created when loss of land, destruction of infrastructure assets, and loss of physical and social capital is followed by diversion of public resources towards costly adaptation measures such as coastal defence structures. Other researchers contest the premise that climate shocks create poverty traps, showing that shocks may, in fact, provide opportunities to enhance future resilience (Leichenko and Silva, 2014). Such was the case in Honduras, where community responses in the years after Hurricane Mitch included institutional changes that reduced vulnerability to future flooding (McSweeney and Coomes, 2011). The mixed results for these studies suggest a need for further exploration of the linkages between climate shocks, regional inequalities, and poverty traps, especially

Vulnerable Regions in a Changing Climate    671 given the expectation that shock events will become more frequent and more severe as the result of climate change.

Which Sectors are Vulnerable? Regional vulnerability is also frequently examined through a sectoral lens. Regions that depend upon climate-​sensitive sectors such as agriculture and fisheries, lumber and forestry products, outdoor recreation, and tourism are expected be more vulnerable to climate change shocks and stresses (Lal et al., 2011; Johnson et al., 2012; Morrison and Pickering, 2013; Sagoe-​Addy and Addo, 2013). In addition to direct dependence on climatic conditions such as temperature, rainfall, and snowfall, other sources of sectoral vulnerability stem from the spatial immobility of certain types of production facilities and processes. For industries such as oil and gas and mineral mining, re-​location away from flood-​vulnerable riverine areas or low-​lying coastal areas is not feasible with existing technologies (Cruz and Krausmann, 2013; Sharma and Franks, 2013). In addition to exploring how production of goods or provision of services might be directly affected by climate change, sectorally focused studies have also considered how consumer demand for tourism and recreational activities in different regions may change as a function of predicted climate changes such as loss of snow pack and warmer temperatures (Scott et al., 2008; Barrios and Ibañez, 2015). Research has also explored how economic vulnerabilities vary for producers within different sectors, attempting to specify how institutional, social, and political factors, in combination with firm characteristics such as size and assets, may interact to shape decision-​making about climate risks within different regional settings (Eakin et al., 2012; Barnett and Eakin, 2015; Vancura and Leichenko, 2015). Although investigation of sectors that are directly on the front lines of climate change, such as tourism and agriculture, is critical for illuminating regional vulnerabilities, there is a need for further investigation of how climate change may affect other sectors of the economy (Liverman and Glasmeier, 2014; Liverman, 2015a). Within the field of climate economics, integrated sectoral modelling studies are generally focused on impacts and interactions across climate-​sensitive sectors, such as water resources, agriculture, and coastal zones (e.g. Harrison et al., 2015). There remains a need for examination of climate change impacts in high-​value sectors, particularly those that are driving global economic growth and are the major sources of employment, such as chemicals, textiles, electronics, and automotives (Liverman and Glasmeier, 2014). Climate-​related disruptions of supply chains and inventories in these sectors can have long-​lasting consequences for regional economies. Flooding of production facilities in the automobile and electronics industries in Thailand in 2011, for example, resulted in severe and sustained disruptions of global supply chains with economic and political repercussions in many other regions (Stern et al., 2013; Liverman and Glasmeier, 2014). Impacts of climate change on sectors such as health care, information technology, retail, and real estate are also under-​examined (Liverman and Glasmeier, 2014). Each of these sectors may see dramatic changes in patterns of consumer demand in response to climate stresses and shocks, such as changing needs for medicine and health products, greater usage of mobile information products, shifting preferences for where to live, and changes willingness to pay or tolerance for risk.

672   Leichenko

Which Populations are Vulnerable? Differential patterns of economic vulnerability also emerge for individuals and households. Poverty is often highlighted as a key factor that increases the propensity of individuals and households to be harmed by climatic shocks and stresses (Adger et al., 2003; Füssel, 2012; Intergovernmental Panel on Climate Change, 2012; Leichenko and Silva, 2014; Mutabazi et al., 2015). Globally, poorer individuals have a greater propensity to be harmed by climate change for a variety of reasons, including fewer assets to rely on for recovery from droughts, hurricanes, and floods, dependence on livelihoods within climate-​sensitive sectors (e.g. agriculture, fishing, pastoralism), and limited access to information about climate risks (Jones et al., 2009; Skoufias et al., 2012; Barua et al., 2014). Physical health and psychological dimensions of poverty, which compound monetary disadvantage and hinder the ability to cope with external shocks, or plan for the future, also contribute to vulnerability of poor populations (Leichenko and Silva, 2014). Individual and household vulnerability are also highly variable by region. Within rural regions of the Global South, factors that contribute to vulnerability of poor households include limited land ownership, lack of options for livelihood diversification, lack of market access, and ongoing degradation of ecological resources such as forests. Growing reliance on cash crops aimed at global markets further exacerbates vulnerabilities to extreme weather and climate change, as small-​scale agriculturalists abandon traditional strategies for managing climate risks (Silva et al., 2010). In urban areas of the Global South, living and working in hazardous physical environments, in conjunction with factors such as inadequate infrastructure and weak governance, contribute to vulnerability of poor populations to climate extremes (Pelling, 2003; Douglas et al., 2008; Hardoy and Pandiella, 2009; Tanner et al., 2009; Chatterjee, 2010). Studies have also documented greater exposure to climate stresses of poor populations in wealthy countries, particularly the USA (Cutter et  al., 2003; Lal et al., 2011; Paolisso et al., 2012; Martinich et al., 2013; Maldonado et al., 2013). The economic vulnerability of relatively poor US populations is tied to factors including social isolation, limited options for affordable housing, and dependence on public transport infrastructure (Halpin, 2013; Barnes, 2015). Climate-​related transport disruptions, in particular, tend to have a disproportionate economic effect on individuals who hold low-​wage hourly positions and may not have access to private automobile transport during weather-​related shutdowns (Barnes, 2015). Research on individual and household vulnerability emphasizes that it is often the intersection of many dimensions of poverty, such as limited income, gender, ethnic or racial discrimination, lack of assets and capabilities, and failed or misguided development policies that contribute to susceptibility of poor populations (Eakin et al., 2012; Burnham et al., 2013). Emphasizing the relational nature of vulnerability, Turner (2016) suggests that social relations, including differential social obligations and opportunities, have a critical influence on individual and household vulnerability. Research by Ajibade et al. (2013) on economic vulnerability to flooding in Lagos is illustrative of these emerging intersectional and relational perspectives. The work documents greater negative impacts of flooding for low-​income women in Lagos as compared with middle-​and high-​income women, demonstrating how gender relations and gender roles, occupational status, and household structure together contribute to greater vulnerabilities for lower-​income women. Work on the gendered nature

Vulnerable Regions in a Changing Climate    673 of economic vulnerability in the Eastern Gangetic Plains of India similarly demonstrates that women from marginal farmer and tenant households are more vulnerable than other individuals, and that this vulnerability is integrally connection with gender roles and social relations which are interwoven with expectations of outmigration for economic opportunities elsewhere for men (Sugden et al., 2014).

Producing Vulnerability In addition to exploring patterns and processes driving regional vulnerability, economic geographers are also beginning to probe underlying factors that create vulnerability, including financial and governance mechanisms that put assets and people in harm’s way, produce new risks, and shape adaptation responses. Incorporating insights from a range of literatures including studies of the cultural economy, political ecology, feminist theory, and science and technology studies, work in this vein explores financialization, marketization, and commodification of climate risk, governance of corporate responses to climate change, and the emergence of the adaptation industry (e.g. Pollard et al., 2008; Pattberg, 2012; Webber, 2013; Johnson, 2014). In many cases, actions intended to reduce climate exposure or promote adaptation are found to create new and unanticipated vulnerabilities. The insurance sector plays a particularly important role in producing new vulnerabilities. Researchers have demonstrated that the emergence of index insurance, weather derivatives, and catastrophe bonds as alternative asset classes, are influencing the association of risk calculations with decisions such as where and how to build, what to grow, and how to allocate municipal finances (Pryke, 2007; Pollard et al., 2008; Johnson, 2014). Work by Johnson (2014) shows how place-​based vulnerabilities of physical assets have become a new commodity traded via the insurance-​linked securitization (ILS) market. The research reveals how climate change is increasingly understood as a business opportunity for the insurance sector because the expectation of greater future losses associated with extreme weather events allows for higher premiums. Noting that the immobility of economic assets necessitates the purchase of insurance for protection against climate-​related damage to physical capital, business interruption, and worker compensation, Johnson documents how the development of ILS markets have created the ‘ability to fashion geographic liabilities as strategic resources’ (2014, p. 173), thereby contributing to built environments that are more exposed to climatic risk. Changes in the norms and expectations around the pricing of hazard insurance have also contributed to the production of new geographies of insured risks and vulnerabilities. Researchers have long noted that subsidization of flood insurance contributes to growing climate and hazard vulnerabilities (Thomas and Leichenko, 2011), yet shifts towards risk-​or market-​based pricing of insurance are also problematic. Detailed analysis of the distributional effects of risk-​based pricing in the UK demonstrates that market-​based pricing produces new vulnerabilities within poorer communities as households forgo insurance coverage altogether (Penning-​Rowsell and Pardoe, 2015). Uneven distributional outcomes were also found in the aftermath of Hurricane Sandy in coastal New Jersey, where patterns of recovery and rebuilding varied, in part, because of differences in insurance coverage and capacity to afford higher insurance premiums for newly rebuilt properties (Leichenko et al., 2014).

674   Leichenko Other research raises more fundamental questions about the framing of climate change as an economic threat, including the mechanisms and motivations behind the emergence of the climate change risk and adaptation industries (Pattberg, 2012; Webber, 2013). Pattberg (2012) examines how climate change became a business risk for multinational corporations, showing how this risk is governed through instruments of disclosure and transparency, both of which serve the needs and interests of institutional investors. Non-​ state actors, such as the C40 Global Cities Leadership Group and the Carbon Disclosure Project, which largely operate outside the boundaries of state authority and governance, are found to play a decisive role in manufacturing corporate climate risks. Proposing the concept of performative vulnerability, Webber (2013) explores the climate change adaptation industry within the island nation of Kiribati, demonstrating how encounters between financiers and government officials produce a particular form of vulnerability. Questioning the conventional understanding of vulnerability as a latent condition, Webber argues that vulnerability is, instead, an emergent effect that is ‘produced in historical and contemporary encounters that are uneven and power laden, with meaning given by an assemblage of facts, expert actors, and objects’ (2013, p. 2722). While studies of this type are beginning to shed light on the myriad ways that climate change is articulated as an economic issue, there is need for further investigation of the production of vulnerability and adaptation within different regions and sectors.

Conclusion: Directions and Opportunities for Future Research Widespread recognition that human actions are a driving force in planetary geological systems has ignited interest across the discipline in the social, political, and economic dimensions of global change. For economic geography, the onset of the Anthropocene raises foundational questions about drivers of regional growth, spatial differences in regional economic performance, and regional responses to stresses and shocks (see Storper, 2011; Chapter 7). While geographers of all stripes have long rejected notions of environmental determinism, there is a need to come to new theoretical understandings of the role of the changing material environment for the economy. What does climate non-​stationarity mean for firm decision-​making in both the short and long term? How will gradual deterioration of environmental baselines affect regional economic growth and development trajectories? In what ways will adaptation responses to climatic and environmental stresses, such as out-​migration from inundated or drought-​prone areas, affect wages and housing markets? This chapter has highlighted a wide array of research on the regional dimensions of climate vulnerability, yet there remains a pressing need for attention to regional impacts and vulnerabilities in non-​coastal contexts, areas that depend on climate-​sensitive ecosystem services, high-​value manufacturing and service sectors, and informal economies. There is also a need for further investigation of climate impacts on regional labour markets and spatial inequalities, of intersectional and relational dimensions of economic vulnerability, and of the production and performance of vulnerability in a variety of regional contexts.

Vulnerable Regions in a Changing Climate    675 While this chapter has focused primarily on vulnerabilities to climate shocks and stresses, there is also a need for more attention to the economic consequences of both adaptation and mitigation. Adaptation decisions in response to sea-​level rise, for example, may span a continuum that ranges from expansion of physical flood defences to complete retreat from a flood-​prone area. Each decision along this continuum is likely to have significant consequences, both deliberate and unintended, that will influence economic vulnerabilities of communities, households, and firms. New flood defences may benefit protected regions but enhance flood exposure in areas that are immediately adjacent to the protective structures. Retreat will affect populations living in adjacent areas and those that are ‘left behind’, and may also affect distant locations that receive an influx of in-​migrants. Mitigation in the form of changing energy policies, such as implementation of a carbon tax, subsidization of biofuel production, or promotion of energy transitions that include widespread adoption of wind, solar, or nuclear energy, would have significant economic implications for both fossil fuel-​ producing regions and for regions that have an energy mix that is highly dependent on coal, oil, or natural gas. A number of studies suggest that unexpected vulnerabilities are emerging as the consequence of climate-​related energy policies (Marino and Ribot, 2012; Hodbod and Tomei, 2013; Venkatasubramanian, 2016), but further research is needed on the connections between climate change responses and vulnerability, particularly in light of implementation of the Paris Agreement and similar measures in the future. In addition to opportunities for theoretical and empirical work within economic geography, the study of climate change also provides a multitude of opportunities for collaboration both within and outside geography. Recognition of climate change as a material force provides avenues for engagement between economic geography and subfields such as science and technology studies and feminist geography, which are re-​thinking how economy and environment interact and seeking to identify alternative societal pathways. The need for better understanding of how environmental baselines are changing, which ecosystem disruptions are expected, and what types of non-​linearities and tipping points might arise, suggests possibilities for fruitful collaboration with researchers studying social and ecological resilience and transformation (Harden et al., 2014). With respect to methodology, economic geography’s traditional strengths in key informant and stakeholder-​based research could make important contributions to collaborative vulnerability research, where co-​production approaches are increasingly used to identify climate stresses and to ensure that vulnerability knowledge is relevant to decision-​makers (e.g. Frazier et al., 2010; Corfee-​Morlot et al., 2011; Rosenzweig et al., 2011; Leichenko et al., 2014; Brady, 2015; Ford et al., 2015). In short, study of climate change offers countless avenues for fruitful collaboration, not only with other subfields of geography, but also with the interdisciplinary global change research community (Liverman, 2015b). Economic geography has much to offer to the field of climate change and much to gain through further engagement in conversations about the Anthropocene.

Notes 1. Exposure is sometimes defined as a separate phenomenon from vulnerability: exposure results from exogenous factors that influence particular locations or sectors, while vulnerability emphasizes endogenous factors that influence susceptibility to harm from exposure and capacity to respond (Intergovernmental Panel on Climate Change, 2012).

676   Leichenko 2. The connections between vulnerability and resilience are also debated within the climate change and hazards literatures (O’Brien et al., 2004a; Cutter et al., 2008; Turner, 2010; Maru et al., 2014).

References Adger, W., Eakin, H., and Winkels, A. (2009). ‘Nested and teleconnected vulnerabilities to environmental change’. Frontiers in Ecology and the Environment 7: 150–​157. Adger, W., Huq, S., Brown, K., Conway, D., and Hulme, M. (2003). ‘Adaptation to climate change in the developing world’. Progress in Development Studies 3: 179–​195. Ahmed, S., Diffenbaugh, N., and Hertel, T. (2009). ‘Climate volatility deepens poverty vulnerability in developing countries’. Environmental Research Letters 4: 034004. Ajibade, I., McBean, G., and Bezner-​Kerr, R. (2013). ‘Urban flooding in Lagos, Nigeria: patterns of vulnerability and resilience among women’. Global Environmental Change 23: 1714–​1725. Barnes, M. (2015). ‘Transit systems and ridership under extreme weather and climate change stress: an urban transportation agenda for hazards geography’. Geography Compass 9: 604–​616. Barnett, A. and Eakin, H. (2015). ‘ “We and us, not I and me”: justice, social capital, and household vulnerability in a Nova Scotia fishery’. Applied Geography 59: 107–​116. Barrios, S. and Ibañez, J. (2015). ‘Time is of the essence: adaptation of tourism demand to climate change in Europe’. Climatic Change 132: 645–​660. Barua, A., Katyaini, S., Mili, B., and Gooch, P. (2014). ‘Climate change and poverty: building resilience of rural mountain communities in South Sikkim, Eastern Himalaya, India’. Regional Environmental Change 14: 267–​280. Bassett, T. and Fogelman, C. (2013). ‘Déjà vu or something new? The adaptation concept in the climate change literature’. Geoforum 48: 42–​53. Birkenholtz, T. (2012). ‘Network political ecology: method and theory in climate change vulnerability and adaptation research’. Progress in Human Geography 36: 295–​315. Boero, R., Bianchini, L., and Pasqualini, D. (2015). ‘Vulnerability and adaptation to severe weather events in the American Southwest’. Weather and Climate Extremes 8: 12–​25. Bohle, H., Downing, T., and Watts, M. (1994). ‘Climate change and social vulnerability: toward a sociology and geography of food insecurity’. Global Environmental Change 4: 37–​48. Bosello, F. and Cian, E. (2014). ‘Climate change, sea level rise, and coastal disasters: a review of modeling practices’. Energy Economics 46: 593–​605. Boykoff, M.T., Bumpus, A., Liverman, D., and Randalls, S. (2009). ‘Theorizing the carbon economy: introduction to the special issue’. Environmental Planning A 41: 2299–​2304. Bradshaw, M. (2010). ‘Global energy dilemmas: a geographical perspective’. The Geographical Journal 176: 275–​290. Brady, M. (2015). ‘Collaborative community hazard exposure mapping: distant early warning radar sites in Alaska’s north slope’. Poster presented at American Geophysical Union Fall Meeting, San Francisco, CA, December 14–​18. Brady, M., Leichenko, R., Auermuller, L., Lathrop, R., and Trimble, J. (2015). ‘Mapping municipal economic exposure to sea level rise in coastal New Jersey’. Poster presented at New Jersey Sea Grant Consortium Quadrennial Site Review, Monmouth University, Long Branch, New Jersey, May 13. Bridge, G. (2010). ‘Resource geographies 1: making carbon economies, old and new’. Progress in Human Geography 35: 820–​834.

Vulnerable Regions in a Changing Climate    677 Bridge, G., Bouzarovski, S., Bradshaw, M., and Eyre, N. (2013). ‘Geographies of energy transition: space, place and the low-​carbon economy’. Energy Policy 53: 331–​340. Briguglio, L., Cordina, G., Farrugia, N., and Vella, S. (2009). ‘Economic vulnerability and resilience: concepts and measurements’. Oxford Development Studies 37: 229–​247. Bumpus, A. and Liverman, D. (2008). ‘Accumulation by decarbonization and the governance of carbon offsets’. Economic Geography 84: 127–​155. Burnham, M., Radel, C., Ma, Z., and Laudati, A. (2013). ‘Extending a geographic lens towards climate justice, part  1:  climate change characterization and impacts’. Geography Compass 7: 239–​248. Burton, R. and Peoples, S. (2014). ‘Market liberalisation and drought in New Zealand: a case of “double exposure” for dryland sheep farmers?’ Journal of Rural Studies 33: 82–​94. Casale, M., Drimie, S., Quinlan, T., and Ziervogel, G. (2010). ‘Understanding vulnerability in Southern Africa: comparative findings using a multiple-​stressor approach in South Africa and Malawi’. Regional Environmental Change 10: 157–​168. Chatterjee, M. (2010). ‘Slum dwellers response to flooding events in the megacities of India’. Mitigation and Adaptation Strategies for Global Change 15: 337–​353. Corfee-​Morlot, J., Cochran, I., Hallegatte, S., and Teasdale, P.-​J. (2011). ‘Multilevel risk governance and urban adaptation policy’. Climatic Change 104: 169–​197. Cruz, A. and Krausmann, E. (2013). ‘Vulnerability of the oil and gas sector to climate change and extreme weather events’. Climatic Change 121: 41–​53. Cutter, S. (1996). ‘Vulnerability to environmental hazards’. Progress in Human Geography 20: 529–​539. Cutter, S., Ash, K., and Emrich, C. (2014). ‘The geographies of community disaster resilience’. Global Environmental Change 29: 65–​77. Cutter, S., Barnes, L., Berry, M., Burton, C., Evans, E., Tate, E., et al. (2008). ‘A place-​based model for understanding community resilience to natural disasters’. Global Environmental Change 18: 598–​606. Cutter, S., Boroff, B., and Shirley, W. (2003). ‘Social vulnerability to environmental hazards’. Social Science Quarterly 84: 242–​261. Dell, M., Jones, B., and Olken, B. (2012). ‘Temperature shocks and economic growth: evidence from the last half century’. American Economic Journal: Macroeconomics 4: 66–​95. Dell, M., Jones, B., and Olken, B. (2014). ‘What do we learn from the weather? The new climate-​ economy literature’. Journal of Economic Literature 52: 740–​798. de Sherbinin, A., Schiller, A., and Pulsipher, A. (2007). ‘The vulnerability of global cities to climate hazards’. Environment and Urbanization 19: 39–​64. Douglas, I., Alam, K., Maghenda, M., McDonnell, Y., McLean, L., and Campbell, J. (2008). ‘Unjust waters: climate change, flooding and the urban poor in Africa’. Environment and Urbanization 20: 187–​205. Dow, K. (1992). ‘Exploring differences in our common future(s): the meaning of vulnerability to global environmental change’. GeoForum 23: 417–​436. Eakin, H. and Luers, A. (2006). ‘Assessing the vulnerability of social-​environmental systems’. Annual Review of Environment and Resources 31: 365–​394. Eakin, H., Benessaiah, K., Barrera, J. F., Cruz-​Bello, G. M., and Morales, H. (2012). ‘Livelihoods and landscapes at the threshold of change: disaster and resilience in a Chiapas coffee community’. Regional Environmental Change 12: 475–​488. Essletzbichler, J. (2012). ‘Renewable energy technology and path creation:  a multi-​scalar approach to energy transition in the UK’. European Planning Studies 20: 791–​816.

678   Leichenko Ford, J., Champalle, C., Tudge, P., Riedlsperger, R., Bell, T., and Sparling, E. (2015). ‘Evaluating climate change vulnerability assessments:  a case study of research focusing on the built environment in northern Canada’. Mitigation and Adaptation Strategies for Global Change 20: 1267–​1288. Frazier, T., Wood, N., Yarnal, B., and Bauer, D. (2010). ‘Influence of potential sea level rise on societal vulnerability to hurricane storm-​surge hazards, Sarasota County, Florida’. Applied Geography 30: 490–​505. Fuller, A.T. and Pincetl, S. (2015). ‘Vulnerability studies: a bibliometric review’. The Professional Geographer 67: 319–​329. Füssel, H.-​M. (2012). ‘Vulnerability to Climate Change and Poverty’ in O. Edenhofer, J. Wallacher, H. Lotze-​Campen, M. Reder, B. Knopf, and J. Müller (eds) Climate Change, Justice and Sustainability: Linking Climate and Development Policy, pp. 9–​17 (New York: Springer Science). Grove, K. (2014). ‘Biopolitics and adaptation: governing socio-​ecological contingency through climate change disaster studies’. Geography Compass 8: 198–​210. Hallegatte, S. (2012). ‘A framework to investigate the economic growth impact of sea level rise’. Environmental Research Letters 7: 015604. Hallegatte, S., Henriet, F., and Corfee-​Morlot, J. (2011). ‘The economics of climate change impacts and policy benefits at city scale: a conceptual framework’. Climatic change 104: 51–​87. Halpin, S. (2013). The Impact of Superstorm Sandy on New Jersey Towns and Households (Newark, NJ: School of Public Affairs). Harden, C.P., Chin, A., English, M.R., Fu, R., Galvin, K.A., Gerlak, A.K., et  al. (2014). ‘Understanding human–​landscape interactions in the “Anthropocene” ’. Environmental Management 53: 4–​13. Hardoy, J. and Pandiella, G. (2009). ‘Urban poverty and vulnerability to climate change in Latin America’. Environment and Urbanization 21: 203–​224. Harrison, P., Holman, I., and Berry, P. (2015). ‘Assessing cross-​sectoral climate change impacts, vulnerability and adaptation: an introduction to the CLIMSAVE Project’. Climatic Change 128: 153–​167. Hertel, T., Burke, M., and Lobell, D. (2010). ‘The poverty implications of climate-​induced crop yield changes by 2030’. Global Environmental Change 20: 577–​585. Hodbod, J. and Tomei, J. (2013). ‘Demystifying the social impacts of biofuels at local levels: where is the evidence?’ Geography Compass 7: 478–​488. Intergovernmental Panel on Climate Change (2012). Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation (New York: Cambridge University Press). Intergovernmental Panel on Climate Change (2014). Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects (New York: Cambridge University Press). Jeffers, J. (2013). ‘Double exposures and decision making: adaptation policy and planning in Ireland’s coastal cities during a boom–​bust cycle’. Environment and Planning A 45: 1436–​1454. Johnson, K., Leichenko, R., and Major, D. (2012). ‘Assessing climate change costs and benefits for regional ecosystems’. Review of Environment, Energy and Economics (Re3): 1–​5. Johnson, L. (2014). ‘Geographies of securitized catastrophe risk and the implications of climate change’. Economic Geograhy 90: 155–​185. Jones, A., LaFleur, V., and Purvis, N. (2009). ‘Double Jeopardy:  What the Climate Crisis Means for the Poor’ in L. Brainard, A. Jones, and N. Purvis (eds) Climate Change and Global Poverty: A Billion Lives in the Balance, pp. 10–​42 (Washington, DC: The Brookings Institution).

Vulnerable Regions in a Changing Climate    679 Kebede, A. and Nicholls, R. (2012). ‘Exposure and vulnerability to climate extremes: population and asset exposure to coastal flooding in Dar es Salaam, Tanzania’. Regional Environmental Change 12: 81–​94. Kirshen, P., Watson, C., Douglas, E., Gontz, A., Lee, J., and Tian, Y. (2008). ‘Coastal flooding in the northeastern United States due to climate change’. Mitigation and Adaptation Strategies for Global Change 13: 437–​451. Knox-​Hayes, J. (2010). ‘Constructing carbon market spacetime: climate change and the onset of neo-​modernity’. Annals of the Association of American Geographers 100: 953–​962. Kumar, L. and Taylor, S. (2015). ‘Exposure of coastal built assets in the South Pacific to climate risks’. Nature Climate Change 5: 992–​966. Lal, P., Alavalapati, J., and Mercer, E. (2011). ‘Socio-​economic impacts of climate change on rural United States’. Mitigation and Adaptation Strategies for Global Change 16: 819–​844. Leichenko, R. and O’Brien, K. (2008). Environmental Change and Globalization:  Double Exposures (New York: Oxford University Press). Leichenko, R. and Silva, J. (2014). ‘Climate change and poverty: vulnerability, impacts, and alleviation strategies’. Wiley Interdisciplinary Reviews: Climate Change 5: 539–​556. Leichenko, R. and Solecki, W. (2013). ‘Climate change in suburbs: an exploration of key impacts and vulnerabilities’. Urban Climate 6: 82–​97. Leichenko, R. and Thomas, A. (2012). ‘Coastal cities and regions in a changing climate: economic impacts, risks and vulnerabilities’. Geography Compass 6: 327–​339. Leichenko, R., McDermott, M. and Bezborodko, E. (2015). ‘Barriers, limits and limitations to resilience’. Journal of Extreme Events 2: 1550002. Leichenko, R., McDermott, M., Bezborodko, E., Brady, M., and Namendorf, E. (2014). ‘Economic vulnerability to climate change in coastal New Jersey: a stakeholder-​based assessment’. Journal of Extreme Events 1: 1450003. Leichenko, R., O’Brien, K., and Solecki, W. (2010). ‘Climate change and the global financial crisis: a case of double exposure’. Annals of the Association of American Geographers 100: 963–​972. Liverman, D. (1990). ‘Vulnerability to Global Environmental Change’ in R. Kasperson (eds) Understanding Global Environmental Change:  The Contributions of Risk analysis and Management, pp. 27–​44 (Worcester, MA: The Earth Transformed Program, Clark University). Liverman, D. (2015a). ‘U.S. national climate assessment gaps and research needs: overview, the economy and the international context’. Climatic Change 135: 173–​186. Liverman, D. (2015b). ‘Remarks on “Engaging geography with global environmental change: new directions for the social sciences and humanities” ’. Panel presented at Meeting of the Association of American Geographers, Chicago, IL. Liverman, D. and Glasmeier, A. (2014). ‘What are the economic consequences of climate change?’ The Atlantic http://​www.theatlantic.com/​business/​archive/​2014/​04/​the-​economic-​ case-​for-​acting-​on-​climate-​change/​360995/​ (last accessed 14 April 2017). McCubbin, S., Smit, B., and Pearce, T. (2015). ‘Where does climate fit? Vulnerability to climate change in the context of multiple stressors in Funafuti, Tuvalu’, Global Environmental Change 30: 43–​55. McGranahan, G., Balk, D., and Anderson, B. (2007). ‘The rising tide: assessing the risks of climate change and human settlements in low elevation coastal zones’. Environment and Urbanization 19: 17–​37. McSweeney, K. and Coomes, O. T. (2011). ‘Climate-​related disaster opens a window of opportunity for rural poor in Northeastern Honduras’. PNAS 108: 5203–​5208.

680   Leichenko Maldonado, J., Shearer, C., Bronen, R., Peterson, K., and Lazrus, H. (2013). ‘The impact of climate change on tribal communities in the US: displacement, relocation, and human rights’. Climatic Change 120: 601–​614. Maloney, M. and Preston, B. (2014). ‘A geospatial dataset for U.S.  hurricane storm surge and sea-​level rise vulnerability:  development and case study applications.’ Climate Risk Management 2: 26–​41. Marino, E. and Ribot, J. (2012). ‘Special issue introduction:  adding insult to injury:  climate change and the inequities of climate intervention’. Global Environmental Change 22: 323–​328. Martin, R. and Sunley, P. (2015). ‘On the notion of regional economic resilience: conceptualization and explanation’. Journal of Economic Geography 15: 1–​42. Martinich, J., Neumann, J., Ludwig, L., and Jantarasami, L. (2013). ‘Risks of sea level rise to disadvantaged communities in the United States’. Mitigation and Adaptation Strategies for Global Change 18: 169–​185. Maru, Y., Smith, M., Sparrow, A., Pinho, P., and Dube, O. (2014). ‘A linked vulnerability and resilience framework for adaptation pathways in remote disadvantaged communities’. Global Environmental Change 28: 337–​350. Mechler, R. and Bouwer, L. (2015). ‘Understanding trends and projections of disaster losses and climate change: is vulnerability the missing link?’ Climatic Change 133: 23–​35. Morrison, C. and Pickering, C. (2013). ‘Perceptions of climate change impacts, adaptation and limits to adaption in the Australian Alps: the ski-​tourism industry and key stakeholders’. Journal of Sustainable Tourism 21: 173–​191. Moser, S. and Hart, J. (2015). ‘The long arm of climate change: societal teleconnections and the future of climate change impacts studies’. Climatic Change 129: 13–​26. Mutabazi, K., Sieber, S., Maeda, C., and Tscherning, K. (2015). ‘Assessing the determinants of poverty and vulnerability of smallholder farmers in a changing climate:  the case of Morogoro Region, Tanzania’. Regional Environmental Change 15: 1243–​1258. Neumann, J., Emanuel, K., Ravela, S., Ludwig, L., Kirshen, P., Bosma, K., et al. (2015). ‘Joint effects of storm surge and sea-​level rise on US coasts: new economic estimates of impacts, adaptation, and benefits of mitigation policy’. Climatic Change 129: 337–​349. O’Brien, K., Sygna, L., and Haugen, J. (2004a). ‘Vulnerable or resilient? A multi-​scale assessment of climate impacts and vulnerability in Norway.’ Climatic Change 64: 193–​225. O’Brien, K., Leichenko, R., Kelkar, U., Venema, H., Aandahl, G., Tompkins, H., et al. (2004b). ‘Mapping vulnerability to multiple stressors:  climate change and globalization in India’. Global Environmental Change-​Human And Policy Dimensions 14: 303–​313. Paolisso, M., Douglas, E., Enrici, A., Kirshen, P., Watson, C., and Ruth, M. (2012). ‘Climate change, justice, and adaptation among African American communities in the Chesapeake Bay region’. Weather, Climate, and Society 4: 34–​47. Pattberg, P. (2012). ‘How climate change became a business risk: analyzing nonstate agency in global climate politics’. Environment and Planning C: Government and Policy 30: 613–​626. Pelling, M. (2003). The Vulnerability of Cities:  Natural Disasters and Social Resilience (London: Earthscan). Penning-​Rowsell, E. and Pardoe, J. (2015). ‘The distributional consequences of future flood risk management in England and Wales’. Environment and Planning C: Government and Policy 33: 1301–​1321. Pollard, J., Oldfield, J., Randalls, S., and Thornes, J. (2008). ‘Firm finances, weather derivatives and geography’. Geoforum 39: 616–​624.

Vulnerable Regions in a Changing Climate    681 Preston, B. (2013). ‘Local path dependence of US socioeconomic exposure to climate extremes and the vulnerability commitment’. Global Environmental Change 23: 719–​732. Preston, B., Yuen, E., and Westaway, R. (2011). ‘Putting vulnerability to climate change on the map: a review of approaches, benefits, and risks’. Sustainability Science 6: 177–​202. Pryce, G., Chen, Y., and Galster, G. (2011). ‘The impact of floods on house prices: an imperfect information approach with myopia and amnesia’. Housing Studies 26: 259–​279. Pryke, M. (2007). ‘Geomoney: an option on frost, going long on clouds’. Geoforum 38: 576–​588. Rhiney, K. (2015). ‘Geographies of Caribbean vulnerability in a changing climate: issues and trends’. Geography Compass 9: 97–​114. Ribot, J. (2014). ‘Cause and response: vulnerability and climate in the Anthropocene’. Journal of Peasant Studies 41: 667–​705. Rosenzweig, C., Solecki, W., Blake, R., Bowman, M., Faris, C., Gornitz, V., et  al. (2011). ‘Developing coastal adaptation to climate change in the New  York city infrastructure-​ shed: process, approach, tools, and strategies’. Climatic Change 106: 93–​127. Sagoe-​Addy, K. and Addo, K. A. (2013). ‘Effect of predicted sea level rise on tourism facilities along Ghana’s Accra coast’. Journal of Coastal Conservation 17: 155–​166. Scott, D., Dawson, J., and Jones, B. (2008). ‘Climate change vulnerability of the US northeast winter recreation tourism sector’. Mitigation and Adaptation Strategies for Global Change 13: 577–​596. Seto, K., Reenberg, A., Boone, C., Fragkias, M., Haase, D., Langanke, T., et al. (2012). ‘Urban land teleconnections and sustainability’. Proceedings of the National Academy of Sciences 109: 7687–​7692. Sharma, V. and Franks, D. (2013). ‘In situ adaptation to climatic change: mineral industry responses to extreme flooding events in Queensland, Australia’. Society and Natural Resources 26: 1252–​1267. Silva, J., Eriksen, S., and Ombe, Z. (2010). ‘Double exposure in Mozambique’s Limpopo river basin’. Geographical Journal 176: 6–​24. Silva, J., Matyas, C., and Cunguara, B. (2015). ‘Regional inequality and polarization in the context of concurrent extreme weather and economic shocks’. Applied Geography 61: 105–​116. Skoufias, E., Rabassa, M., and Olivieri, S. (2012). ‘The Forecast for Poverty:  A  Review of the Evidence’ in E. Skoufias (ed.) The Poverty and Welfare Impacts of Climate Change: Quantifying the Effects, Identifying the Adaptation Strategies, pp. 17–​54 (Washington, DC: The World Bank). Stern, P., Ebi, K., Leichenko, R., Olson, R., Steinbruner, J., and Lempert, R. (2013). ‘Managing risk with climate vulnerability science’. Nature Climate Change 3: 607–​609. Storper, M. (2011). ‘Why do regions develop and change? The challenge for geography and economics’. Journal of Economic Geography 11: 333–​346. Sugden, F., Maskey, N., Clement, F., Ramesh, V., Philip, A., and Rai, A. (2014). ‘Agrarian stress and climate change in the eastern Gangetic plains:  gendered vulnerability in a stratified social formation’. Global Environmental Change 29: 258–​269. Swyngedouw, E. (2010). ‘Apocalypse forever? Post-​political populism and the spectre of climate change’. Theory, Culture & Society 27: 213–​232. Tanner, T., Mitchell, T., Polack, E., and Guenther, B. (2009). ‘Urban governance for adaptation: assessing climate change resilience in ten Asian cities.’ IDS Working Papers 2009: 1–​47. Tate, C. and Frazier, T. (2013). ‘A GIS methodology to assess exposure of coastal infrastructure to storm surge & sea-​level rise: a case study of Sarasota County, Florida’. Journal of Geography & Natural Disasters S1: 1–​12.

682   Leichenko Thatcher, C., Brock, J., and Pendleton, E. (2013). ‘Economic vulnerability to sea-​level rise along the northern US Gulf Coast’. Journal of Coastal Research Special Issue 63: 234–​243. Thomas, A. and Leichenko, R. (2011). ‘Adaptation through insurance: lessons from the NFIP’. International Journal of Climate Change Strategies and Management 3: 250–​263. Tschakert, P., van Oort, B., St. Clair, A. L., and LaMadrid, A. (2013). ‘Inequality and transformation analyses: a complementary lens for addressing vulnerability to climate change’. Climate and Development 5: 340–​350. Turner, B. (2010). ‘Vulnerability and resilience: coalescing or paralleling approaches for sustainability science?’ Global Environmental Change 20: 570–​576. Turner, M. (2016). ‘Climate vulnerability as a relational concept’. Geoforum 68: 29–​38. Vancura, P. (2015). ‘Making sense of energy transitions locally’. PhD Dissertation, Department of Geography, Rutgers University. Vancura, P. and Leichenko, R. (2015). ‘Emerging Equity and Justice Concerns for Climate Change Adaptation:  A  Case Study of New  York State’ in K. O’Brien and E. Selboe (eds) The Adaptive Challenge of Climate Change, pp. 98–​117 (New York: Cambridge University Press). Venkatasubramanian, K. (2016). ‘Examining the politics of climate change: narratives, actions, and adaptations in Gujarat, India’. PhD Dissertation, Department of Geography, Rutgers University. Watts, M. and Bohle, H. (1993). ‘The space of vulnerability: the causal structure of hunger and famine’. Progress in Human Geography 17: 43–​67. Webber, S. (2013). ‘Performative vulnerability: climate change adaptation policies and financing in Kiribati’. Environment and Planning A 45: 2717–​2733. Wheeler, T. and von Braun, J. (2013). ‘Climate change impacts on global food security’. Science 341: 508–​513. Wisner, B., Blaikie, P., Cannon, T., and Davis, I. (1994). At Risk:  Natural Hazards, People’s Vulnerability and Disasters (2nd edition) (New York: Routledge). Yamane, A. (2009). ‘Climate change and hazardscape of Sri Lanka’. Environment and Planning A 41: 2396–​2416.

Chapter 36

Carb on M arkets:  Re s ou rc e Governanc e a nd Sustainable Va luat i on Janelle Knox-​H ayes Introduction Resource extraction has long been the purview of economic geography from studies of mining, agriculture, and energy (Bakker, 2000; Clark, 2005; Michielsen, 2013) to more comprehensive analyses of resource supply chains (Bridge, 2008). Indeed, resources, whether material or energy, are essential to every economic activity. As the scale of environmental degradation and decline has increased, economic geographers are expanding their attention to include issues of environmental management, such as forestry conservation (Klooster, 2002), water management (Swyngedouw, 2009; Bakker, 2010;), and food sourcing (Ouma, 2015). Economic geographers have additionally turned their attention to the application of market-​based management to environmental challenges like climate change (Johnson, 2014). Beginning with the control of pollutants such as sulphur dioxide, market-​based approaches have spread to the management of a full range of environmental benefits or costs accrued from economic activities that are not accounted in economic transactions (positive and negative externalities), including greenhouse gas emissions, biodiversity, forest operation, and ecosystem functionality. Efforts to use market-​based approaches to account for positive and negative environmental externalities are epitomized by the marketization of carbon emissions (Böhm et al., 2012; Stripple and Bulkeley, 2013; Knox-​Hayes and Hayes, 2014). Carbon markets trace their roots to the theories of Ronald Coase and John Dales, which suggest that well-​defined property rights can be used to price unaccounted outcomes and thus overcome the problem of externalities (Coase, 1960; Dales, 2002). Drawing on this reasoning, the Kyoto Protocol attempted to translate public concern about climate change by financializing the emission of carbon into the atmosphere through greenhouse gas permits and credit prices. By drawing

684   Knox-Hayes on markets for carbon reductions, the Kyoto approach was modelled after the successful response to acid rain in the US through a sulphur dioxide cap and trading market (Lohmann, 2005). As in the sulphur dioxide markets, carbon markets operate through the establishment of a cap on the amount of carbon that can be emitted by various greenhouse gas emitters, such as power plants, heavy industrial corporations, or waste management facilities. If an entity exceeds its emissions cap, it must purchase credits to offset its emissions, thus monetizing carbon emissions for polluters. Over time, the cap would be progressively lowered, increasing the cost of carbon emissions and economically rewarding technological and operational innovation and the companies that implement those innovations. Since the ratification of the Kyoto Protocol in 2004 a variety of market systems have been set up around the world with a range of operational permutations. However, at the core of each market is the belief that climate change can be solved through the scientific measurement of greenhouse gas levels and the economic management of greenhouse gas pricing (Knight and Knox-​Hayes, 2015). Carbon markets embody a new form of climate capitalism through which economic activity is seen not only as the source of but also the solution to climate change (Perramond, 2012). The study of carbon markets opens the traditional purview of economic geography to a range of issues. For example, carbon markets draw attention to the nature and function of market creation. This is in some contrast to economic geography’s traditional focus on industry production functions across a range of scales, which take markets as given (Clark et al., 2000). In parallel with developments in disciplines like sociology that deconstruct the nature and function of markets (Callon, 1998), economic geography has taken an increased interest in how markets are constructed and enacted in local places by particular actors (Knorr-​Cetina and Preda, 2005) and how they drive economic productivity (MacKenzie, 2006). Because carbon markets are being constructed in real time (Knox-​ Hayes, 2010a), they enable economic geographers an avenue to investigate the creation of markets. The expanding focus on markets in general and carbon markets in particular allows economic geographers to expand on a long-​standing focus on the geographical scope of economies in the context of economic change, the forces driving those changes and the role of localities in global economic transformation (Clark et al., 2000). The spatially specific differences in carbon financialization also fits with the established interest within economic geography regarding distinctive geographic patterns of performance that highlight differentiation and realization as the products of ongoing economic processes that sustain long-​term spatial differentiation (Scott, 2000). Carbon markets also extend and benefit from analyses based on heterogeneity, information asymmetries, and disjoined systems of meaning. This chapter examines the underlying assumptions behind market-​based governance, and particularly the emphasis on controlling greenhouse gases through pricing. In the next section I explain the structure of carbon markets, review critical literature of carbon accounting practices, and explore the roots of the market approach to carbon governance in economic theory. Carbon markets seek to introduce a new form of governance that manages environmental resources through the pricing of positive and negative externalities. However, the mechanisms of pricing ignore issues of where externalities exist in space and time. In the third section I propose a new framework for considering the spatial and temporal dynamics of value. Such a framework links carbon markets (as an initiating mechanisms

Carbon Markets   685 of environmental finance) to the broader economy through an understanding of external value, or use value extended in time. In the fourth section I suggest that the pricing of externalities is insufficient and propose mechanisms that could better generate external value by capturing the use potential of natural resources. I conclude with some reflections on avenues of inquiry for economic geography that would add insight to the understanding of resource governance and the management of contemporary problems like climate change.

The Structure and Operation of Carbon Markets Since the Kyoto Protocol came into force in 2004, a number of regulated and voluntary carbon markets have been set up around the world aiming to achieve carbon dioxide (CO2) emission reductions, largely through cap-​and-​trade mechanisms (Michaelowa and Michaelowa, 2012). The oldest of these systems is the European Union Emissions Trading System (EU ETS), which caps the emissions of more than 10,000 industrial facilities in Europe. Additionally, regulated markets have been created in Australia, China, Japan, New Zealand, and South Korea. Finally, trial emissions markets are being established in a range of countries, including Brazil, Chile, Mexico Indonesia, and Thailand (Kossoy et al., 2014). Each system is structured with its own unique rules and procedures, but all carbon markets operate in a similar way. Regulators or market authorities in each system place a cap on the amount of carbon that can be emitted by various greenhouse gas emitters. If the carbon emitted by a capped entity exceeds its cap, the entity must purchase credits to offset its emissions. Entities that do not reach their cap can sell excess permits onto the carbon market. In theory, the cap is ratcheted down in time, and emitters either become more efficient or go out of business. Either way, the systems should reduce the total amount of emissions under the cap, as well as send a price signal through the markets that benefits carbon-​alternative fuel sources and technologies. Much of the challenge and uncertainty of the markets resides in the details of the design, as well as the enforcement of rules (Lohmann, 2009). Central to these challenges is the constructed nature of carbon credits as inverse commodities (Knox-​ Hayes, 2010b). Unlike many markets, carbon markets value absence rather than existence—​ in this case, the absence of greenhouse gases. Although there are differences across the schemes, carbon markets trade two main types of credits: allowances and offsets (Michaelowa, 2004). Both products—​which are measured in units equivalent to one tonne of CO2—​are constructed purely from information. Allowances are essentially permits that allow regulated entities to emit an amount of greenhouse gases. Offsets serve as reduction credits and mark the absence of an emissions occurrence in one location. Crucially, the materiality of the offset lies in the counterfactual: the offset is derived from a claim of emissions that would have otherwise been emitted. The counterfactual absence embodied in the offset then can be transferred to another location to allow for emissions there. Both credit types are constructed through a system of measurement that creates baselines or projection scenarios of the levels of greenhouse gas emission that would occur without intervention (see e.g. Bansal and Knox-​Hayes, 2013). As such, the reality of emissions reduction through carbon markets cannot be proven, only presented

686   Knox-Hayes through arguments of ‘additionality’ (additional greenhouse gas reductions) both within and external to each system (Mason and Plantinga, 2013). Because the markets trade commodities that are measured against otherwise-​assumed realities (i.e. a wind plant in contrast to the otherwise-​assumed thermal power plant that would provide energy in its place) it becomes very difficult to assess whether the markets are reducing emissions. A number of scholars have highlighted design flaws, failures, and pervasive weaknesses in implementation of emissions markets (Lohmann, 2009; MacKenzie, 2009a; Knight, 2011). However, arguably there have also been positive developments from the emissions markets, such as the establishment of the Carbon Disclosure Project, an initiative that asks some of the world’s largest companies to disclose voluntarily their emissions. In measuring their carbon liabilities, companies have suggested they identify inefficiencies, which helps them to not only reduce carbon emissions, but also to generate revenue (Plambeck, 2012). As a consequence of initiatives such as these, the business community is becoming increasingly aware of carbon liabilities, and policy responses to climate change are beginning to assume a profit logic (Knox-​Hayes and Levy, 2011). Often overlooked in the debates over the internal validity of carbon markets is the external interaction between markets and the environment. Specifically, a key challenge for carbon markets lies in their inability to accommodate fully the spatial and temporal scale at which CO2 and other greenhouse gases cycle within the ecosphere, and particularly the rate of actual removal from the atmosphere (Bansal and Knox-​Hayes, 2013; Knox-​Hayes, 2013). Carbon credits are defined in time frames attuned to economic rather than environmental cycles. The credits are registered on a yearly or bi-​ yearly basis and operate according to the regulatory phases (3-​5 year intervals) of emissions trading systems, whereas CO2 takes much longer to actually leave the atmosphere once released (MacKenzie, 2009b; Bansal and Knox-​Hayes, 2013). Thus, there is a fundamental mismatch between markets and the real environment. While climate change is a physical problem deeply embedded in space and time, carbon markets create inverse commodities—​ commodities absent in space and time.

Critical Studies of Carbon Markets As the prior discussion suggests, economic geographers have been particularly effective leveraging critiques against carbon governance practices. Carbon markets are complex systems of multi-​level governance that require cooperation and coordination across international, national, and regional institutions, and which ultimately blur the lines between politics and economics (Bumpus, 2011). Considerable social coordination is required to operationalize the markets (Knox-​Hayes, 2010a). Neil Smith (2008), for example, highlights the ways in which carbon markets operate through calculative practices that simultaneously aggregate social practices and individualize carbon accounting. Others note that carbon market construction co-​opts the function of the markets into existing financial institutions and financial logics (Callon, 2009; Knox-​Hayes and Levy, 2011). Ian Bailey and colleagues (2011) suggest that carbon markets are a form of ecological modernization that has unintended consequences, including governance by industry, limited geographical and political representation, and market lock-​in to experimental mechanisms with great uncertainty (Lotay, 2009).

Carbon Markets   687 Economic geographers have particularly challenged the accounting practices (Engels, 2009; Hopwood, 2009) within carbon markets, arguing that they create increasing complexity of governance process that is neither innovative nor effective (Mason and Plantinga, 2013). For example, the accounting standards that made carbon markets attractive to industry have difficulty in capturing the materiality of carbon emissions reductions (Wright, 2013). As Lansing (2011) demonstrates through a study of carbon forestry in Costa Rica, the materiality of carbon projects is lost in calculation (Lansing, 2011). Similarly Klooster (2002) finds that the accounting of emissions reductions does not adequately capture the utility of particular agricultural and conservation projects, potentially resulting in suboptimal land-​use practices (Klooster, 2002). Furthermore, projects with less environmental impact (large scale hydro projects) more easily lend themselves to commodification than projects with beneficial environmental and social impacts such as cook stoves in Honduras (Shen et al., 2013). These criticisms challenge the ways in which climate change has become framed as an economic problem, solvable with existing accounting approaches and techniques, rather than a spatial–​temporal problem of the dis-​entrainment of socio-​economic and ecological systems (Knox-​Hayes and Hayes, 2014). Carbon accounting is indifferent to where or how emissions cuts are made and discourages attention to path dependence, positive feedback, and innovation (Lovell et al., 2009). Lohmann (2009) argues that the conflation of reductions and offsets confounds probability with uncertainty, ignorance, and indeterminacy; and obstructs social thinking about long-​term directions with a focus on achieving short-​term efficiency. MacKenzie (2009b) demonstrates the complexity of commensuration (making equivalent) of different greenhouse gases by comparing the accounting CO2 emissions from a power plant in Europe with the burning of HFC-​23 (refrigerant gas) in China. He demonstrates that that assumptions and frames of accounting have a significant and often distorting influence on the evaluation of greenhouse gas impact (MacKenzie, 2009b). In the first phase of the EU ETS (the largest operating carbon market), the calculations used to commensurate HFC-​23 made the reduction of one unit of HFC-​23 11,000 times more valuable than the reduction of one unit of CO2 from a power plant. As a consequence, the markets became flooded with HFC-​23 offsets, raising questions regarding the mission of the markets to reduce emissions from industrial activity. These critiques, while essential, do not entirely engage with the materiality of carbon emissions reductions to understand why and how spatial and temporal context is abstracted through accounting. The problem is that climate change is perceived to be a failure of economics that can be understood through the framework of existing economic theories. This assumes not only that economic theories can adequately account for the interaction between human and environmental systems, but also that economic theories themselves are equipped to deal with differences in spatial and temporal scale. On the latter count, the theories upon which the markets are built are flawed: the instruments of economic valuation have not been designed to accommodate differences in spatial and temporal scale. As such, the natural environment is perpetually undervalued and the long-​term environmental consequences of actions directed at economic growth are underestimated. Climate change is not a singular problem; it is symptomatic of a much deeper crisis of value and representation at the core of how modern economies operate. Thus, by exposing the temporal and spatial limitations of economic theory, the study of carbon markets within economic geography has the potential to reshape the foundations of modern economic thought and practice.

688   Knox-Hayes

Carbon Markets from Economic Theory: Governance Through Price and Quantity Carbon markets suggest that climate change arises out of the failure to price externalities. For example, the burning of fossil fuels results in the emission of CO2 and other greenhouse gases that are not accounted for in the price of energy unless a carbon-​pricing scheme is put in place. Treating climate change as a matter of externality pricing is symbolic of ecological modernization—​the idea that markets can integrate environmental and social equity into economic instruments through the recalibration of economics. Ecological modernization takes a socio-​political problem, removes it from the realm of political discourse, and recasts it in economic, technical language (Garsten and Jacobsson, 2007). Accordingly, the market mechanism-​based economic framing of climate change situates the solutions to the problem of climate change as technocratic matters requiring only the proper implementation of economic theory (Bailey and Wilson, 2009). Consequently, ecological modernization through market-​ based governance draws heavily on microeconomic theory. Individual pursuit of self-​interest in a free-​market system can lead to the most efficient production outcomes for society (Herzog, 2013). At the core of microeconomic theory is the law of supply and demand. It was first articulated by Adam Smith in 1776 (see Smith, 1937) as the invisible hand of the market, and later developed by Alfred Marshall (1890) with the application of mathematics into a series of models for determining the movement of price and quantity of supply and demand in a market system (Figure 36.1). According to the supply-​and-​demand curve the price for a particular good will vary until it settles at a point where the quantity demanded by consumers (at the current price) will equal the quantity supplied by producers (at current prices), resulting in an economic equilibrium for price and quantity. Marshall’s principles of economics initiated the marginal revolution, the idea that the value of goods is not determined by the difficulty in acquiring them (Smith) or the labour required to produce them (Marx), but rather by the maximization of individual preference in the face of scarcity. This logic finds expression in carbon markets through the idea that once a price is introduced for carbon, supply and

Price

D

S

P

D

S 0

Q

Figure 36.1   Marshall’s Supply-​and-​Demand Curve.

Quantity

Carbon Markets   689 demand for carbon-​intensive products and industries will settle at (increasingly) lower equilibrium points. The significance of these early formulations cannot be underestimated, because modern economics is still based on the same core assumptions about value. The theories and models of neoclassical economics revolve around exchange value (the price of goods and services). From exchange value, the core functions of markets are determined by consumers and producers maximizing marginal utility to determine quantity and price. A major market failure is the inability of supply and demand (for various reasons) to provide a stable and reflective equilibrium price and socially desired quantity. The underlying logic is clear. Value is linked to price through exchange. Quantity is determined through supply and demand. This basic logic carries into every aspect of market-​ based governance, including climate change. Rather than conceptualize climate change as a failure of economies to produce at a rate the natural environment can accommodate, it is conceptualized as a failure to price externalities appropriately so as to limit the quantity of their production. The solution, markets to price greenhouse gases, carries the core logic of value through exchange. Scientists determine the appropriate quantity of greenhouse gases in the atmosphere (i.e. 450 parts per million), rationalizing levels of emissions already reached against average warming predictions. Economists calibrate the mechanisms (markets, taxes, quotas, etc.) that will establish the equilibrium price to achieve the negotiated quantity of greenhouse gas emissions. Although there is considerable complexity in the numerical modelling of economics today, the underlying logic of controlling price and quantity through supply and demand has changed very little from Smith’s time. A more considered appraisal of the interface of economic theory and the environment through carbon markets, however, suggests that the core assumptions upon which market-​based governance is built lack a full conceptualization and integration of value. Firstly, in calculations of price and quantity only one form of value (exchange) is fully theorized while the operation of the natural environment (water purification, carbon sinks) draws attention to the importance of use value. Secondly, the original conception of value misses a critical dimension of valuation: time.

Valuation and New Economic Theory: The Spatial and Temporal Distinction The first challenge to the theoretical underpinnings of market-​based governance lays in its formulation around one type of value—​exchange value. Adam Smith (1776 (see Smith, 1937)) elaborated a dichotomy of value around use and exchange: ‘The word value . . . has two different meanings, and sometimes expresses the utility of some particular object, and sometimes the power of purchasing other goods which the passion of that object conveys. The one may be called “value in use”; the other, “value in exchange” ’. Over time, the use element of value has been increasingly obscured by the modern drive by markets to create valuation through instruments of exchange. Incorporating use value pushes a reformation of economic calculation by incorporating considerations of use or utility. Take, for example, the question of how might it be possible to increase energy resources? From an exchange-​value perspective,

690   Knox-Hayes the solution focuses on changing the price of fossil fuels. Accounting for use value, however, brings in considerations of utility and with it a focus on renewability and limitations on energy use so as to increase efficiency. Use value is objective (Marx, 1867; Harvey, 1982), it is about the individual consumption of a good or service, embedded in a specific set of actors, at a determined location, performing a particular activity for a defined duration. The spatial and temporal dimensions of use value can be identified on a Cartesian grid and located in a specific frame of time. Exchange, in contrast, is value in process or circulation (McGinnis, 2003, p. 536). It is subjective: valuation that is subject to human desire and belief. For this reason economists such as Ludwig von Mises (1954) and John Maynard Keynes (2006) emphasize the subjective nature of exchange value. The distinction between objective and subjective is predominantly spatial. Space here does not refer specifically to place, but rather to a frame of reference from which relationships between subjects and objects can be understood (Lefebvre, 1991). Socio-​ economic space references socio-​material relationships, particularly relationships that are built around economic transactions. For example, a labourer paid a wage for performing a service or constructing a good is a relationship that can be framed from the standpoint of socio-​economic space. In contrast, socio-​environmental space refers to the relationship between humans and environmental resources. The felling of timber or the consumption of energy are relationships that can be framed from the standpoint of socio-​environmental space. These relationships are about physical materiality, the consumption and use of physical resources.1

Valuation With a Temporal Dimension The original dichotomy of use and exchange only goes so far, however, to address the modern failure to adequately conceptualize value. Notably, both use and exchange deal with one temporal frame—​the present possibilities of a good or service. With respect to the time, the objective–​subjective distinction can be understood through potential versus realized value. Realized value is value that exists (i.e. that can be put to use) in the present. Potential value is value that is yet to be created; it has the possibility to exist in the future. Because the future is undetermined, potential value is subjective and only arises from the fulfilment of particular conditions. As future value is not situated in an objective reality, it is not truly commensurate with value that has already come to exist. For example, a barrel of oil has both a present exchange and a use value. In 2015 a barrel of oil might at one moment in time be exchanged for US$52 and used to acquire any number of goods or services. Alternatively, the barrel could be put to use, converted to 24 gallons of gasoline,2 and used to fill the tank of an automobile that can then transport an individual 600 miles. Both of these are present values. The barrel might also be saved for use or exchange in the future. In 10 years the price of the barrel of oil may have quadrupled if demand increases and supply decreases, or it may be virtually worthless if a new technology, such as electric vehicles, eliminates demand. The future or potential value of the oil is subject to conditions that are yet to occur. The danger of commensuration through exchange arises from treating potential value (the value of a barrel of oil 10 years from now) as though it were the same, exchangeable for present value (Knox-​Hayes, 2013).

Carbon Markets   691 Taken together these distinctions suggest that use and exchange value must also have a temporal distinction, present use versus potential use; present exchange versus potential exchange. In failing to recognize these spatial and temporal distinctions, the systems and instruments of economic valuation miss a critical aspect of resource governance, namely where, when, and how value exists across space and time. The spatial and temporal distinctions of value can be illustrated through a typology (Figure 36.2). Clarifying the relationship of space and time to these forms of value elucidates the missing link (use in time) in economic valuation and market-​based governance for problems like climate change. The matrix in Figure 36.2 represents a typology of value that accounts for spatial and temporal dynamics. The vertical axis divides space into socio-​ economic and socio-​ environmental relations. The horizontal axis divides time into present (realized economy) and future (potential economy). Four distinct types of value are identified: use, exchange, derived (e.g. derivatives like wheat options or repackaged home mortgages), and external (the value of externalities, or value outside present use and exchange). Consider a common commodity like a bushel of wheat. Use value is the value acquired from using or consuming the bushel. Exchange value is the value of exchanging the commodity for something else, or the monetary price of a bushel of wheat. Derived value is the value derived from the exchange of the commodity in the future; a contract to sell a bushel of wheat at a certain price at a set time, for example. External value is the value external to the production of the commodity, for example the value of depleted soil or fertilizer run-​off from growing the bushel of wheat. The typology makes some important distinctions with respect to the value of resources. Resource value is at its core use value. The ultimate objective of any resource is use, whether now or in the future. Value is objectively embodied in the physicality of and realized through the use of the commodity. Wheat is wasted without consumption.

TIME Present (Realized)

Future (Potential)

Exchange Value (Money)

Derived Value (Derivatives)

SPACE SocioEconomic

SocioEnvironmental

M

D

C

E

Use Value (Commodities)

External Value (Externalities)

Objective Time

Subjective Time

Subjective Space

Objective Space

Figure 36.2   Typology of the Spatial and Temporal Dynamics of Value.

692   Knox-Hayes Although use value is the ultimate objective of a commodity, the only metric to account the value of the resource is exchange value. Value is measured relative to price. For this reason, carbon pricing is used to try to solve environmental problems (climate change). The challenge is that the solutions actually require a direct transformation of use (shifting energy resources) and price does not guarantee a particular use.

The Challenge of Accounting Derived and Exchange Value Derived value leverages an underlying commodity to create future, potential exchange value. External value leverages an underlying commodity to create future, potential use value. Whereas use and exchange are temporally singular—​an item can either be used or exchanged—​derived and external values present multiple alternatives for use and exchange because they both reflect subjective time frames. This creates a challenge for accounting external and derived value. Figure 36.2 illustrates both present and future or potential value with individual columns. While present value is singular (commodities and services exist in particular places and have value either in use or exchange at any given moment), future or potential value is unlimited. Figure 36.2 could therefore illustrate future value with innumerable columns moving to the right across time. Because future or potential value is subjective, within finance there exists the potential to create numerous layers of value from singular physical resources through the structuring of potential value (Knox-​Hayes, 2013). The subjective nature of future and potential value generates concerns for the ways in which financial value becomes represented in present time. Modern economic systems recognize all types of value as commensurate through pricing, regardless of qualifications of how, when, and where value actually exists. For example, derivatives contracts can reference the exchange of an underlying commodity (the bushel of wheat) across multiple time frames. These can all be represented in the present and priced accordingly. A wheat option might generate a quantitatively similar price to a bushel of wheat, but the quality of value contained within the two, as well as the consequence of their value, is considerably different. The bushel of wheat can feed hungry people. The wheat option merely provides the opportunity for a potentially advantageous economic transaction in the future. This is not to say a financial option is without value, but rather that its value is qualitatively different from the use value of the commodity itself. Similarly, external value can recognize a range of potential future use outcomes that result from the production and consumption of the underlying commodity. Negative externalities (costs) from the production of wheat might include depletion of soil or run-​off into streams. Positive externalities (benefits) might include the low-​cost provision of nutrition to poor communities. The key to the creation of external value is the idea that the ultimate objective is not only use of the resource, but also sustained use across time. Consider two forests, one of 100,000 hectares in size, the second of similar quality but only 10,000 hectares in size. Evaluating only from the standpoint of present exchange value, or price, the first is seemingly more valuable. However, the true measure of value is determined by use. The first forest is clear-​cut, harvested, and sold. The second forest has 15 per cent of its trees harvested every ten years for 100 years. Within 100 years it will have not only produced greater exchange value than the first forest (which was destroyed in a single harvest), but it will also

Carbon Markets   693 undeniably produce greater use value because it is still contains standing trees and can still be used into the future. Use, and particularly sustained use across time, is critical to the valuation of environmental resources. Valuing resources based exclusively on price, singularly in time, does not appropriately account for the value or resources. The goal of external value is not only to borrow value from the future (as with derivatives), but also to shape present consumption so as to return value to the future. The value of the forest can be extended beyond the use of the individual trees, if they are harvested in such a way as to leave the forest ecology intact. The forest will regenerate and continue to create value external to the consumption of individual trees. Each tree still contains its use value, but accounting for potential use, treated as a system and harvested slowly, they generate external value.

Environmental Finance: External Value Through Pricing or Extended Use The goal of externalities pricing is to capture external value and to build economies that recognize external value. Consider again the typology of value with the addition of various circuits of capital (Figure 36.3). This diagram illustrates the circulation of value in production and extends the basic Marxist conception of value produced through the circulation of commodities and money with a consideration of derived and external value. In each quadrant, and associated with each type of value is a distinct circuit of capital: consumption, commerce, finance, and environmental finance. The lower-​left quadrant (socio-​environmental transaction in present time) is where use value is created and consumption takes place. The upper-​left quadrant TIME Present (Realized)

Future (Potential)

Commerce (Exchange Value)

Finance (Derived Value)

SPACE SocioEconomic

Production

M C

SocioEnvironmental

Subjective Space

D E

Internalization

Consumption (Use Value)

Environmental Finance (External Value)

Objective Time

Subjective Time

Objective Space

Figure  36.3  Economies of Production and Internalization Through the Circulation of Value.

694   Knox-Hayes (present socio-​economic transaction) is where exchange value is created and commerce takes place. The upper-​right quadrant (potential socio-​economic transaction) is where derived value is created and where finance exists. Finally, the lower-​right quadrant (potential socio-​environmental transactions) is where external value is created and where environmental finance is being initiated. Finance and environmental finance derive value from the potential exchange and use of underlying resources and services (commodities and social relations). At the centre of the diagram, production is conceptualized as the transition of value between commerce, finance, and consumption according to the formula (M–​D–​C–​M’). Money is used to leverage derivative value (e.g. stocks), which is then converted invested into commodities, which are transformed into surplus capital. The key assumption to this formula in financial production is that the capital borrowed from the future (D) is less than the surplus capital (M’) generated from the production of commodities today, which enables the return on the original investment plus a profit. For this to be accurate the circuit must actually generate commodities that are useful. The establishment of new mechanisms for the pricing of externalities, such as greenhouse gases, is designed to balance the productive circuit by forcing value from finance through externalities into the creation of environmentally beneficial technologies like renewable energy. In Figure 36.3 this cycle of the ‘internalization of externalities’ (M–​E–​C–​M’) is represented in the centre of the diagram by the grey arrows moving from money to externalities (E) to commodities (C) and back to surplus value (M’). This cycle results from the creation of the fourth and hereto missing circuit of capital—​environmental finance. Internalization is considered here as the process of pricing externalities to change production decisions. For example, accounting for the price of CO2 released from the burning of fossil fuels might lead to the production of solar panels owing to new relative price competiveness of solar energy over conventional thermal energy combined with a carbon price. In theory, internalization should close the loop and return value from finance or surplus accumulation to improve the quality of the natural environment or at the least to improve socio-​environmental relations. Parallel to financial production, value is deposited into the future potential of the commodity through the priced externality, which, in turn, shapes production decisions (solar instead of thermal energy). The assumption remains that the value deposited through the externality (carbon credit) E is less than the profit accrued from the sale of the commodity M’ allowing the return on investment. For this to be true, government regulation is needed to make carbon fuel sources more expensive than alternatives, at least at the outset. In practice, there are several challenges with the ability of externality pricing to balance the social and environmental detriments of production. In order to operate effectively, environmental finance must account for externalities and generate environmentally beneficial commodities. The problem is that just as finance can short-​circuit the cycle of production, creating paths for money to generate surplus value irrespective of underlying physical commodities (e.g. 2008 US financial crisis), so too can environmental finance shortcut the cycle of externality pricing that should generate new technologies. Internalization can be short circuited if the externality is treated merely as a source of exchange value and used to leverage derivatives for the production of surplus value. As early as 2008 this was occurring in carbon markets with the creation of derivatives products like the collateralized carbon

Carbon Markets   695 obligation (Lotay, 2009). These collateralized carbon obligations undermine environmental finance as a means of rebalancing socio-​environmental relations because they derive surplus value solely from the circulation of environmental externalities through financial transactions. To be truly effective, environmental finance must generate a change in the nature of commodity production. Carbon markets seek to internalize negative externalities by creating exchange value for the absent externality. Compliance parties pay more for the carbon-​based production so as to change the use of fossil fuels. The challenge is that the value of the credit is its price (exchange value), which means it can be exchanged for other outcomes beyond just the switching of fuel sources. Price alone is insufficient to govern the creation of externalities and to change the use of resources. A more effective way to leverage external value would be to create a direct accounting of future use. There may be several paths to redirect and to maximize potential use, such as revaluing resources with spatial temporal signatures such that renewable resources have higher qualitative value. The key is to focus on the lower-​right quadrant (Figure 36.3), on the creation of external value, or the extension of use across time. This highlights the inability of economics to come to terms with value and, in particular, the absence of a theory of use value and the attendant instruments and systems of sustainable valuation. While the creation of such a theory is beyond the scope of this chapter, it is an important avenue of investigation for the discipline of economic geography. In the next section I explore various potential means of refocusing externalities on the generation of value for future use.

A Path Towards Sustainable Resource Valuation Externality pricing allows for resource exchange but does not account for temporal scale. Because externality pricing leads to resource allocation (for which temporality is central), the absence of temporal scale in externality pricing is problematic. Take the application of externality pricing to a forest. Externality pricing manages the forest by valuing the positive and negative externalities generated from various production decisions, and allowing these externalities to be exchanged. For example, a forest can serve as a water filter for a river basin. It will hold the water in the vegetation, slowing the water flow, and allowing it to percolate through the substrate, thus purifying the water before it reaches the river. Externality pricing frames this dynamic as a positive externality that through environmental finance can be packaged as a commodity, exchanged, and used to offset the water pollution of a manufacturing plant downstream. The challenge is that the service the forest provides is, in fact, geographically and temporally localized and situated within the ecology of the forest. Packaged as exchange value, it can be exchanged for pollution downstream, but physically the water purification upstream does not cancel pollution downstream. In this regard, even though the credits might be exchanged, ecosystem services are not trans-​locatable. Their physical impact lies in the places where they are located. To maximize positive externalities, the

696   Knox-Hayes instruments of exchange must recognize the ecology in which externalities are produced. If its ecology is properly maintained, a forest provides a range of externalities or ecosystem services such as the preservation of a watershed, the sinking CO2 from the atmosphere, and the support of multi-​value flora and fauna (e.g. pollinators for neighbouring crops). Furthermore, maximizing positive externalities means taking account of the temporal scale of their production. For example, the forest ecology that can support pollinators and attenuate floods requires time to become established. Thus, a more sustainable means of resource valuation accounts for use value of resources over time. To this end, several methods offer the promise of better integration of time into exchange value. Firstly, instruments of exchange value can be linked to realistic expectations of the production of use value (an issue of both leverage ratios, as well as time scale of production). Secondly, the rate of value transfer can be entrained to the rate of value production. Finally, the nature of property rights might be changed so as to better maximize use rather than exchange value. With respect to the first method of time integration, it is undoubtedly true that finance has the useful benefit of borrowing value in from the future for production today. The challenge, however, lies in the tendency to treat all value as though it is made commensurate through exchange. This, in turn, allows for distortions in the scale at which potential (future) value is represented in the present. To address this problem, regulation should focus on limiting the degree to which value can be created through derivatives. In other words, there should be limits on how far value can be temporally extended from its underlying material resource. In practice, this means that rather than a stream of cascading or multiple derivatives that increasingly distance exchange value from the material underlying, derivatives might be limited in the singular and only for specific commodities. Moreover, there should be limits placed on the leverage ratios (the amount of debt or future value, relative to equity or present productive value) that are allowed in the creation of derivatives. For natural resources, the issue of the time-​scale of resource production is attendant with problems arising from extending productive value by borrowing from the future. By factoring in timescale and rate of production, sustainable valuation becomes more likely. The second method focuses on better entraining the rate of value transfer to the rate of value production. This is particularly important for natural resources, the production and renewal of which cannot be accelerated. Allowing the exchange value of resources to be vested over longer time horizons adds recognition to the temporal scale of their production and would limit the demand for accelerated turnover. In practice, instruments for ecosystem service valuation should be long-​term investment vehicles. For example, instruments to value the production of resources such as standing forests should require long-​term ownership so that the value of the forests can actually be produced before the instruments are exchanged. If ecosystem services are eventually to be valued through financial instruments, these instruments must have controls on how quickly they can be exchanged, because natural capital can take longer to vest than financial instruments allow. This should allow for resources to be valued over greater time frames. Extending the rate of valuation should also allow for the use of resources that experience greater value over greater time scales. For example, if energy use was packaged and priced over greater time frames (yearly or even multi-​year increments instead of monthly), clean energy would experience an advantage over conventional sources.

Carbon Markets   697 Finally, the nature of property rights might be transformed to manage natural resources as service stocks rather than commodities. Consider again the example of a forest. It is a resource that has numerous uses. It can provide timber, or, if left standing, can filter water, provide a habitat for bees and biodiversity, and generate tourism. The forest’s ecological uses can magnify its value if these are accounted over greater temporal frames. So, as cutting down the forest and selling the trees for timber produces greater financial gain in the short term, over longer time frames the financial benefits of the standing forest dwarf those of the timber sale. Policies and financial practices then need to find a way to make the long and short term correspond. For example, land ownership might be charged with the creation of property services rather than rights for the natural environment. Returning again to the example of forestry, the tendency is to create a system of property rights to manage a forest system. To be protected, the forest must belong to someone because ownership ensures that an interested actor will collect the value of the conservation activity. However, to privilege the use value, the forest should be managed as a service stock that has a fixed use, rather than a commodity that can be traded and sold. That is, the forest can only derive exchange value if it is used in line with the listed use value. Thus, property granted for dedicated use would begin to accomplish the service model. The ownership is granted but not the authority to do with the resource as pleased. Rather the owner acts as a custodian that maintains of the resource. In this model, the forest is a provider of ecological services. Ownership does not convey the right to transform the forest into another use, merely to accrue the value of guaranteeing and protecting the use. The use value is thus fixed and guaranteed, rather than by proxy through exchange valuation. This approach transforms the valuation model by placing use value on level with exchange value. Challenges abound, however. Legal and economic systems would face substantial reordering, with a framework of ecological preservation at the centre in which value is conveyed only through proper use. Doing so would radically transform the way in which land is used, and would consequently have cascading effects throughout the political economy. Returning to the issue of climate change, there are implications for focusing on use value. The ultimate function of carbon markets should be to transition economies away from the use of fossil fuels. As a consequence, exchange value becomes secondary to use value. Thus, limits may be placed on the rate of trade of carbon credits, which would ensure that the carbon credits are linked to the places and timescales of carbon sinking. Additionally, there should be considerable effort dedicated to controlling the rate of exchange of environmental assets to control for temporal and spatial dislocation. At the core of economic processes focus should shift from exchange value gains to material effects. Clean energy could be made more valuable if the rate of credit maturations was extended. Moreover, revaluing resources to account for spatial and temporal effects should make clean energy more effective. Focusing on material effects within the valuation of carbon credits would shift emphasis to clean energy commodities as the basis of valuation. While this does not preclude other forms of valuation, in the end the value of carbon credits would be assessed in terms of material contributions (e.g. how much more clean energy is produced). In the spirit of capturing use value, rather than allowing credits to be exchanged for other instruments they would only yield exchange value once they transferred back into material outcomes, for example funding development of technology.

698   Knox-Hayes

Conclusion Climate change has become the pressing issue of the twenty-​first century. In an effort to confront this grave and global collective action problem, governments around the world are increasingly turning attention to the use of carbon-​pricing market mechanisms to reduce greenhouse gas emissions. Markets to price greenhouse gases have been set up in over forty-​ five countries around the world. The approach is drawn from the belief that climate change can be solved through the scientific measurement of greenhouse gas levels and the economic management of greenhouse gas pricing. Economic geographers have criticized carbon markets for a number of reasons, including the complexity of international coordination required to create the market infrastructure, the use of existing financial logics in market design, and the failure of accounting practices to capture the complexities of carbon syncing. As this chapter demonstrates, of central importance is failure of carbon markets to account for the spatial and temporal complexity of carbon reduction. In particular, carbon markets stem from a belief that resources and their attendant externalities are best managed through controls on price and quantity. In taking this approach, carbon markets are built from theories of exchange value, particularly the idea that the pricing of externalities will solve climate change. These theories, however, fail to take account of a critically important issue—​the rate and scale at which industrial production occurs relative to the ability of the natural environment to generate resources and sink waste. A more complete consideration of intersection of political economy with climate change requires a fundamental rethinking of economic value. Rather than just a spatial distinction of use and exchange, there are also temporal distinctions in the types of value operating in economic systems. Exchange value can be extended in time to derived value. Use value can likewise be extended in time to external value, the value of sustaining potential use across time. Such a reconsideration of the spatial and temporal qualities of value demonstrates that externalities are, in fact, an extension of use value rather than exchange value. Therefore, attempting to curb externalities through pricing to some extent misses the point. External value can best be captured through mechanisms that address use rather than through exchange. While this chapter does not seek to generate a new theory of use value, it does present ideas regarding the ways in which existing mechanisms could come to emphasize and generate use value. Firstly, spatial and temporal controls should be placed on the exchange of externality credits, so as to limit leverage ratios and link exchange with the material production of value. Secondly, extending time frames of the generation of exchange value would make resources such as clean energy more valuable. If the instrument of exchange is instantaneous then short-​term profits are privileged, but if the instrument is designed to accommodate greater temporal scale, profit in time must be considered. Thirdly, property rights might be transformed to manage natural resources as service stocks rather than commodities. This would allow for the maintenance of productivity of resources such as forests across time and through different owners. Carbon markets open several important lines of inquiry to economic geography. With the emphasis on context, history, and scale, economic geography can add considerable insight to the analysis of new forms of resource governance designed to address problems

Carbon Markets   699 like climate change. In addition to carbon, markets for a range of ecosystems services, including forestry, water, and biodiversity are being created. Rather than managing natural resources as conventional economic inputs, these markets attempt to manage natural resource through the potential of their externalities. These markets and systems open a range of new avenues of both theoretical and empirical investigation to the discipline of economic geography. There is a broader issue of what these markets symbolize relative to the evolution of modern capitalism. Economic geography has pushed the boundaries of economic thought with new theories of production, regional development, varieties and variegations of capitalism (Hall and Soskice, 2001; Peck and Theodore, 2007), and the evolution of financial services and markets (Clark and Wójcik, 2007). Additionally, scholars in the Marxist tradition such David Harvey (2014) have extended critiques of the function and contradictions inherent in capitalism. Yet, insufficient attention has been brought to bear on the core principles of economics, particularly the nature of value estimation that underlies the system. In particular, economic geographers could lend considerable insight into new theories of use value. What might a renewed focus on use value and its spatial characteristics lend to the understanding of industry, market creation, and resource governance? Furthermore, how might economies come to operate if they were guided by principles of use, rather than exchange? Perhaps theories of use might in time generate a more sustainable system of valuation. At the least, use value presents a new frame of analysis and a new way to come to terms with the pressing challenges the next generation of economic geographers will confront.

Notes 1. For a more comprehensive overview of the distinction between socio-​material and physical materiality see Bansal and Knox-​Hayes (2013). 2. http://​www.energy.ca.gov/​2008publications/​CEC-​180-​2008-​003/​CEC-​180-​2008-​008/​ CEC-​180-​2008-​008.PDF (last accessed 19 April 2017).

References Bailey, I. and Wilson, G.A. (2009). ‘Theorising transitional pathways in response to climate change: technocentrism, ecocentrism, and the carbon economy’. Environment and Planning A 41: 2324–​2341. Bailey, I., Gouldson, A., and Newell, P. (2011). ‘Ecological modernisation and the governance of carbon: a critical analysis’. Antipode 43: 682–​703. Bakker, K. (2000). ‘Privatizing water, producing scarcity:  the Yorkshire drought of 1995’. Economic Geography 76: 4–​27. Bakker, K. (2010). Privatizing Water: Governance Failure and the World’s Urban Water Crisis (Ithaca, NY: Cornell University Press). Bansal, P. and Knox-​Hayes, J. (2013). ‘The time and space of materiality in organizations and the natural environment’. Organization & Environment 26: 61–​82. Böhm, S., Misoczky, M.C., and Moog, S. (2012). ‘Greening capitalism? A Marxist critique of carbon markets’. Organization Studies 33: 1617–​1638.

700   Knox-Hayes Bridge, G. (2008). ‘Global production networks and the extractive sector: governing resource-​ based development’. Journal of Economic Geography 8: 389–​419. Bumpus, A.G. (2011). ‘The matter of carbon: understanding the materiality of tCO2e in carbon offsets.’ Antipode 43: 612–​638. Callon, M. (1998). The Laws of the Markets (Oxford: Blackwell Publishers). Callon, M. (2009). ‘Civilizing markets: carbon trading between in vitro and in vivo experiments’. Accounting, Organizations and Society 34: 535–​548. Clark, G.L. and Wójcik, D. (2007). The Geography of Finance: Corporate Governance in the Global Marketplace (Oxford: Oxford University Press). Clark, G.L., Feldman, M., and Gertler, M. (2000). ‘Economic geography:  transition and growth’ in G.L. Clark, M. Feldman, and M. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 3–​17 (Oxford: Oxford University Press). Clark, J.R.A. (2005). ‘The “New Associationalism” in agriculture: agro-​food diversification and multifunctional production logics’. Journal of Economic Geography 5: 475–​498. Coase, R.H. (1960). ‘The problem of social cost’. Journal of Law and Economics 3: 1–​44. Dales, J.H. (2002). Pollution, Property & Prices:  An Essay in Policy-​making and Economics (Cheltenham: Edward Elgar). Engels, A. (2009). ‘The European Emissions Trading Scheme: an exploratory study of how companies learn to account for carbon’. Accounting, Organizations and Society 34: 488–​498. Garsten, C. and Jacobsson, K. (2007). ‘Corporate globalisation, civil society and post-​political regulation: whither democracy?’ Development Dialogue 49: 143–​157. Hall, P.A. and Soskice, D. (2001). Varieties of Capitalism:  The Institutional Foundations of Comparative Advantage (Oxford: Oxford University Press). Harvey, D. (1982). The Limits to Capital (Chicago, IL: University of Chicago Press). Harvey, D. (2014). Seventeen Contradictions and the End of Capitalism (Oxford:  Oxford University Press). Herzog, L. (2013). Inventing the Market: Smith, Hegel, and Political Theory (Oxford: Oxford University Press). Hopwood, A.G. (2009). ‘Accounting and the environment’. Accounting, Organizations and Society 34: 433–​439. Johnson, L. (2014). ‘Geographies of securitized catastrophe risk and the implications of climate change’. Economic Geography 90: 155–​185. Keynes, J.M. (2006). General Theory of Employment, Interest and Money (New Delhi: Atlantic Publishers and Distributors). Klooster, D.J. (2002). ‘Toward adaptive community forest management: integrating local forest knowledge with scientific forestry’. Economic Geography 78: 43–​70. Knight, E. (2011). ‘The economic geography of European carbon market trading’. Journal of Economic Geography 11: 817–​841. Knight, E. and Knox-​Hayes, J. (2015). ‘Creating legitimate authority for environmental governance and new market creation:  a case study from Australia’. Competition & Change, 19: 36–​55. Knorr-​Cetina, K. and Preda, A. (2005). The Sociology of Financial Markets (Oxford: Oxford University Press). Knox-​Hayes, J. (2010a). ‘Creating the carbon market institution: analysis of the organizations and relationships that build the market’. Competition & Change 14: 176–​202. Knox-​Hayes, J. (2010b). ‘Constructing carbon market spacetime: climate change and the onset of neo-​modernity’. Annals of the Association of American Geographers 100: 953–​962.

Carbon Markets   701 Knox-​Hayes, J. (2013). ‘The spatial and temporal dynamics of value in financialization: analysis of the infrastructure of carbon markets’. Geoforum 50: 117–​128. Knox-​Hayes, J. and Hayes, J. (2014). ‘Technocratic norms, political culture and climate change governance’. Geografiska Annaler: Series B, Human Geography 96: 261–​276. Knox-​Hayes, J. and Levy, D.L. (2011). ‘The politics of carbon disclosure as climate governance’. Strategic Organization 9: 91–​99. Kossoy, A., Oppermann, K., Platonova-​Oquab, A., Suphachalasai, S., Höhne, N., Klein, N, et al. (2014). ‘State and trends of carbon pricing 2014’. World Bank http://​documents.worldbank. org/​curated/​en/​505431468148506727/​pdf/​882840AR0REPLA00EPI2102680Box385232.pdf (last accessed 19 April 2017). Lansing, D.M. (2011). ‘Realizing carbon’s value: discourse and calculation in the production of carbon forestry offsets in Costa Rica’. Antipode 43: 731–​753. Lefebvre, H. (1991). The Production of Space (Oxford and Cambridge, MA: Blackwell). Lohmann, L. (2005). ‘Marketing and making carbon dumps:  commodification, calculation and counterfactuals in climate change mitigation’. Science as Culture 14: 203–​235. Lohmann, L. (2009). ‘Toward a different debate in environmental accounting: the cases of carbon and cost–​benefit’. Accounting, Organizations and Society 34: 499–​534. Lotay, J.S. (2009). ‘Subprime carbon:  fashioning an appropriate regulatory and legislative response to the emerging US carbon market to avoid a repeat of history in carbon structured finance and derivative instruments’. Houston Journal of International Law 32: 459. Lovell, H., Bulkeley, H., and Liverman, D. (2009). ‘Carbon offsetting: sustaining consumption?’ Environment and Planning A 41: 2357–​2379. McGinnis, J. (2003). ‘For every time there is a season: John Philoponus on Plato’s and Aristotle’s conception of time’. KronoScope 3: 83–​111. MacKenzie, D. (2006). An Engine, not a Camera:  How Financial Models Shape Markets (Cambridge, MA, and London: MIT Press). MacKenzie, D. (2009a). Material Markets:  How Economic Agents are Constructed (Oxford: Oxford University Press). MacKenzie, D. (2009b). ‘Making things the same: gases, emission rights and the politics of carbon markets’. Accounting, Organizations and Society 34: 440–​455. Marshall, A. (1890). Principles of Political Economy (New York: Macmillan). Marx, K. (1867). Capital, volume I (Harmondsworth: Penguin/​New Left Review). Mason, C.F. and Plantinga, A.J. (2013). ‘The additionality problem with offsets:  optimal contracts for carbon sequestration in forests’. Journal of Environmental Economics and Management 66: 1–​14. Michaelowa, A. (2004). ‘CDM incentives in industrialized countries-​the long and winding road’. International Review for Environmental Strategies 5: 217–​231. Michaelowa, K. and Michaelowa, A. (2012). ‘Negotiating climate change’. Climate Policy 12: 527–​533. Michielsen, T.O. (2013). ‘The distribution of energy-​intensive sectors in the USA’. Journal of Economic Geography 13: 871–​888. Ouma, S. (2015). Assembling Export Markets:  The Making and Unmaking of Global Food Connections in West Africa (Chichester: Wiley Blackwell). Peck, J. and Theodore, N. (2007). ‘Variegated capitalism’. Progress in Human Geography 31: 731–​772. Perramond, E.P. (2012). ‘Privatizing water:  governance failure and the world’s urban water crisis—​by Karen Bakker’. Economic Geography 88: 227–​228.

702   Knox-Hayes Plambeck, E.L. (2012). ‘Reducing greenhouse gas emissions through operations and supply chain management’. Energy Economics 34: S64–​S74. Scott, A. (2000). ‘Economic Geography: The Great Half Century’ in G.L. Clark, M. Feldman, and M. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 3–​17 (Oxford: Oxford University Press). Shen, B., Wang, J., Li, M., Li, J., Price, L., and Zeng, L. (2013). ‘China’s approaches to financing sustainable development:  policies, practices, and issues’. Wiley Interdisciplinary Reviews: Energy and Environment 2: 178–​198. Smith, A. (1937 [1776]). The Wealth of Nations (New York: The Modern Library). Smith, N. (2008). Uneven Development: Nature, Capital, and the Production of Space (Athens, GA: University of Georgia Press). Stripple, J. and Bulkeley, H. (2013). Governing the Climate:  New Approaches to Rationality, Power and Politics (Cambridge: Cambridge University Press). Swyngedouw, E. (2009). ‘The political economy and political ecology of the hydro‐social cycle’. Journal of Contemporary Water Research & Education 142: 56–​60. von Mises, L. (1954). The Theory of Money and Credit (Auburn, AL: Ludwig von Mises Institute). Wright, C. (2013). ‘Global Finance and the Environment’ in R. Falkner (ed.) The Handbook of Global Climate and Environment Policy, pp. 428–​445 (Chichester: Wiley Blackwell).

Chapter 37

L ong-​r un Re s ou rc e Scarci t y Dieter Helm * Introduction For more than a century, it has been a recurring fashion to predict that key natural resources are about to run out, and that as a result economic growth will come to a shuddering halt. In the mid-​1860s, William Stanley Jevons in The Coal Question (1865) was so worried that Great Britain would run out of coal that he predicted it may contract to its former ‘littleness’. As the world’s population expanded rapidly, from around 1.7 billion in 1900 to over seven billion now, and as economic growth took off, the pressure on natural resources intensified. Almost every decade of the twentieth century has witnessed scares about running out of oil, with associated predictions of sharply rising prices and industrial dislocation. Looking forward into the twenty-​first century, given current world growth rates, the world economy will be around sixteen times bigger than it is now by 2100, and population will be anywhere between about nine billion and thirteen billion (United Nations, 2014). The scale of natural resources that will be used up to meet these extra demands, and the scale of their impacts on the wider environment, are potentially awesome. It is, therefore, an obvious question to ask: whether the rates of depletion can be sustained and therefore whether resource scarcity is a serious problem in the long run. Put another way, is economic growth sustainable in the long run? This chapter considers these questions. The next section divides up resources into renewables and non-​renewables. The section ‘Non-​renewables, Peak Oil and Resource Scarcity’ concentrates on non-​renewables and shows why concerns have been repeatedly exaggerated, using oil as an example. The ‘Renewables’ section explains why their long-​ run availability is at risk. ‘Depletion and Sustainability’ brings the two types of resources together and provides a sustainable framework within which resources can be maintained in the long run in order to ensure that the constraints do not undermine future growth, and

*Research contribution by Nevena Vlaykova is gratefully acknowledged. All errors remain mine.

704   Helm that future generations enjoy sufficient quantities of these assets. The chapter ends with the conclusions.

Different Types of Resources Nature provides resources for free:  they are natural capital.1 These resources come in a variety of shapes and sizes, and with different supply potentials. There is plenty of coal:  not so many diamonds. The claim to resource scarcity is about some resources rather than all, and therefore a series of categories need to be identified to help ascertain which ones matter. A key distinction is between renewable and non-​renewable resources (Natural Capital Committee, 2013). In some cases (renewables) nature is capable of going on providing these resources forever (renewables), while for others (non-​renewables) there is a finite stock, so that these can only be used once. In the former case, the challenge is to utilize these resources so they carry on being renewed. In the latter case, it is about who uses them, over what period, and whether any are so critical to production and hence to continuing economic activities that their depletion has to be constrained. The question of the sustainability of these resources, and of the economic growth that depends upon them, is different between renewables and non-​renewables. The ‘sustainability’ concept itself gives rise to many different interpretations, but at its core is the general idea that each generation should pass on a set of assets that is at least as good as those it itself inherited. The famous Brundtland Commission defined sustainable development as follows:2 Sustainable development is development that meets the needs of the present without compromising the ability of future generations to meet their own needs (United Nations, 1987, Chapter 2).

In the case of renewables, it is, in principle, straightforward, although there may be different compositions of an aggregate amount of renewable natural capital. In the case of non-​renewables, as the resource can only be used once, it cannot be passed onto the next generation. Instead, if the next generation is to be at least as well off in asset terms, it must inherit other types of assets (renewables, physical, and human capital) worth at least as much as the depleted non-​renewable resource. In the ‘Depletion and Sustainability’ section, these general conditions are formalized into aggregate natural capital rules.

Non-​renewables, Peak Oil, and Resource Scarcity Non-​renewables are forms of natural capital that are in finite supply. If they are used now, typically they cannot be used again (although some elements may be recycled). Most minerals fall into this category, many of which do not have substantive uses. Most of the world’s

Long-run Resource Scarcity    705 rocks have limited applications. Resource scarcity focuses on a small number of minerals that have proved core elements to the development of modern economies. Classic examples are the fossil fuels, laid down millions of years ago by natural processes. They may be created again, but they will not be restocked in any relevant human time horizon. There are several questions relevant to the long-​run scarcity of these physically limited resources: How much is left? What about substitutes? What is the impact of technical progress? Do they have significant economic value? While each has its own special characteristics, the general approach is illustrated with reference to oil. The first and most important aspect of these physically limited resources is that while they may be finite, their total is uncertain. There is physical uncertainty about the resource base and economic uncertainty about the recoverable quantities. Physical uncertainty relates to the composition of Earth’s crust. Although knowledge has advanced greatly with seismic and other surveying technological developments, the estimates of the total resource base of any of Earth’s main minerals remain primitive. Although the rock composition of Earth’s crust is reasonably well understood, the finer detail of mineral composition is absent in many areas of the planet. The US Geological Survey, for example, estimated that the Arctic Circle contains 22 per cent of the remaining undiscovered, technically recoverable resources of oil, natural gas, and natural gas liquids in the world (US Geological Survey, 2008). It is an educated guess and no more. Further, most of the resource base, even if identified, may be physically unrecoverable and even if recoverable of little value or too costly to be worth recovering with current technologies. As future technologies are not known, the recoverable resources cannot be known. The scale of misunderstanding and misrepresentation is well reflected in the various theories and empirical claims about peak oil.3 Physical estimates of the remaining oil resources have led to predictions in almost every decade of the twentieth century of imminent resource exhaustion. In the last decade, these claims were sufficient to convince policymakers that diversifying from fossil fuels to renewables (and in the British case, nuclear) is economically efficient, irrespective of the climate change concerns (see Helm, 2015b, Chapters 4 and 7, and Helm, 2017). In the 1950s, the theory of peak oil was expounded by Marion King Hubbert (1956). He analysed US oil wells, and assumed that all the significant discoveries had already been made. The total resource could be estimated by taking these oil fields, and applying to them the depletion patterns that had been observed. Thus, not only was the physical resource base known, but so too was the depletion pattern for the remaining wells. On this evidence basis it was a comparatively trivial exercise to aggregate across the known oil wells, take the rate of depletion, extrapolate, and predict that US oil production would peak around 1970, and indeed he turned out to be right about the peak, up until the recent transformation of the US oil and gas industries. Hubbert’s followers took this simple empirical projection from US data on its existing wells and generalized the results into a theory about the world’s supply of oil, and then predicted the world’s peak production date (see e.g. Campbell, 1997). Because the evidence base was less reliable (even assuming that all the big finds had been made), the estimates of the world peak varied. Demand, too, had to be assumed, and, interestingly, few, if any, of the peak oil theorists could have imagined how long fast and for how long China would grow. As a result demand was widely underestimated in China and globally. But in both the US and the global cases the peak oilers have turned out to be wrong. Hubbert had no knowledge of the potential of unconventionals, did not anticipate fracking,

706   Helm and could not have known that in the twenty-​first century, US production would push up in less than a decade to almost nine million barrels a day (Figure 37.1), holding out the prospect of eventual oil independence for North America as a whole. Hubbert and his followers took a supply-​side, physical approach and neglected the economics, and, in particular, price. As the low-​oil-​price years of the 1990s gave way to gradual rising prices in the 2000s, technical progress pushed out the boundaries of oil exploration. The combination of seismic technologies, horizontal drilling, and fracking released enormous unconventional resources, while deep-​water drilling brought the possibilities of significant offshore resources. (The subsequent boost to supply forced prices down in late 2014, and pushed marginal supplies off the market.) The Alberta tar sands never figured in the peak oiler’s estimates, nor did the deep-​ water fields in the Gulf of Mexico. Finally, the assumed depletion patterns followed from the pressure flows, and hence it was not envisaged that existing wells might be more extensively depleted, rather than being abandoned at around 50 per cent or less of their potential reserves. The key point here is that it is economics driving oil supply, not the physical constraints. As and when the limits of the physical resource are reached, the price can be expected to rise again, and this will then set off further technological advances, and encourage the search for substitutes (to which we return later). Hubbert’s peak-​oil theory was generalized to other minerals. Perhaps the most famous attempt to extrapolate the resource base was presented in the Club of Rome report in 1972—​ The Limits to Growth (Meadows et al., 1972). MIT was commissioned by the Club of Rome to run a modelling exercise to discover where economic growth had ‘limits’. The model took the usual constraints that preoccupied environmentalists: population, food production, industrial production, pollution, and consumption of non-​renewable natural resources—​and 12,000

10,000

8000

6000

4000

2000

15

10

20

05

20

00

20

95

20

90

19

85

19

80

19

75

19

70

19

65

19

60

19

55

19

50

19

45

19

40

19

35

19

30

19

25

19

19

19

20

0

Figure 37.1   US Field Production of Crude Oil (1000 Barrels per Day), 1920–​2014. Source: US Energy Information Administration (2015).

Long-run Resource Scarcity    707 left out the technology and human capital. If the five parameters experienced exponential growth, then there would inevitably be a day of reckoning. Therefore, the Club of Rome Report concluded that: If the present growth trends in world population, industrialization, pollution, food production, and resource depletion continue unchanged, the limits to growth on this planet will be reached sometime within the next one hundred years. The most probable result will be a rather sudden and uncontrollable decline in both population and industrial capacity (Meadows et al., 1972, p. 23).

The exercise had the merits of modelling the feedbacks and interactions between these five inputs, and it duly predicted global collapse sometime in the twenty-​first century. As with all such ‘garbage in–​garbage out’ exercises, the model scenario predictions are determined by the assumptions about the inputs. If these are growing exponentially there will inevitably be limits. It could not be otherwise. A stationary state would eventually materialize, and this could be achieved either by a population and capital collapse—​or by deliberating imposing limits. The Club of Rome was little more than an extension of the old Malthusian argument—​with a few extra parameters. Malthus had focused on population, and the Club of Rome was similarly concerned at a time of rapid population growth. Given the coincidence with the first OPEC (Organization of the Petroleum Exporting Countries) oil shock, and the more general economic crises in the 1970s, it was perhaps not surprising that much attention focused on the minerals section, and their depletion rates. In the Club of Rome report, the assumptions drove the conclusion that there would be serious depletion problems within the next half century. In the report, simplistic calculations were provided that purported to show that the stocks of many minerals would soon be depleted. Julian Simon took issue with the Club of Rome, and he entered into a famous bet with Paul Ehrlich, the author of The Population Bomb (Ehrlich, 1968). He asked Ehrlich to name any five minerals, and bet that the price of all of them would be lower in 1990 than it was at the time of the bet.4 He won on all counts: the 1970s, when the Club of Rome report was published, turned out to be a very high-​mineral-​ price world, on the back of the demand built up in the great post-​World War II economic boom and the Vietnam war, and for the rest of the century commodity prices refused to follow the Club of Rome script. What Hubbert and the Club of Rome got wrong was the impacts of the high prices they foresaw, particularly on technology. They could not have forecast the subsequent developments, in part because all depended on future technologies. The key mistake was to base predictions on current technologies. In the oil case, Hubbert’s followers made this mistake too, and compounded it with the assumptions that the high knowledge base about US oil wells could be extrapolated elsewhere, and that there would not be further significant discoveries. Indeed, the typical assumption of the peak-​oil theorists was that as there had not been many big discoveries in the 1980s and 1990s, there would not be any more to come. They forgot that at the low prices in the 1980s, and especially the 1990s, they were not worth prospecting for. Almost everything that could be wrong with the peak-​oil theories turned out to be so. The estimated physical reserves of oil are higher now than at any previous time. These estimates relate in large measure to conventional oil: unconventional shale and tight oil deposits are a recent addition, and it is only in the USA that the geological knowledge of these is reasonably comprehensive. Conventional resources in the Arctic, in the Antarctic, in the Sahara,

708   Helm and in much of Russia remain under-​explored. Unconventional reserves are largely a matter of guesswork outside the USA. Recent disappointing drilling experience in Poland for shale gas indicates the level of ignorance, even when promising rock formations are identified. Knowing that rocks might contain oil deposits is not the same thing as finding out whether they do, and further whether they can be extracted cost-​effectively. Finding how much is physically present and what it might cost requires holes to be drilled. An important conclusion follows: the uncertainty will remain for the foreseeable future. Predictions of exhaustion are therefore hypothetical. As the technologies for surveying Earth’s crust improve and as more holes are drilled, more information will become available. Current estimates are lower bounds, and the key question is whether at these lower bounds, plus reasonable estimates of potential additional resources, there is the prospect of scarcity. The total physical quantities are profoundly uninteresting from a scarcity perspective. In the case of oil, the lower bound is sufficiently high as to negate resource-​scarcity boundaries. Indeed, the physical supply limits will only bind if there are no alternative competing ways of capturing energy and there is no concern about carbon emissions. Neither turns out to be likely. It is worth noting that the expertise of discovering more oil as the technology and price-​ incentivized exploration carries across to gas. Until 1990, gas was regarded as a scarce premium fuel to be kept primarily for the petrochemical industry—​so scarce that it was illegal in Europe and the USA to burn it in power stations. Within a decade, it became the fuel of choice for new electricity-​generating power stations, and even before shale gas transformed the reserves position. Gas is now so physically plentiful that from a policy perceptive it is best regarded as infinite. Finally, of the fossil fuels coal is super-​abundant and super-​polluting. There is so much that is so cheap to produce that it is hardly worth trying to work out how much the total resource base might be.5 Were even a fraction of this remaining quantity to be burned, the damage to the climate would be devastating, according to current climate models. Physical limits are all about supply, whereas scarcity is about the confluence of supply and demand. The demand for fossil fuels—​and minerals generally—​depends upon the availability of substitutes. While peak-​oil theorists assume that oil is in inelastic supply, and that it is driven by transport’s dependency on oil, this is, in fact, merely a reflection that at current prices it is not economic to develop alternative substitute sources of energy supply. Yet in the transport case, oil is not essential. Indeed, the electric motor preceded the internal combustion engine, only to be out-​competed as cheap and abundant oil flowed. Vehicles can be powered by electricity, and are increasingly likely to be so. Electricity can be produced not only by fossil fuels, but also generated from nuclear, solar, geothermal, and hydro power. There is also the prospect of substitution between fossil fuels: gas can power vehicles directly, or be a fuel for the electricity displacing the internal combustion engine. For almost any manufacturing process, there is always the possibility that alternative assets might perform the same functions. In the case of energy, technical progress may yield new ways of generating electricity, and cars may convert to electricity. Consider some of the possibilities. Current-​generation solar power is very inefficient. It uses a very small fraction of the light spectrum. What would happen if more of the light spectrum was opened up, given the abundance of solar radiation the planet receives? Now consider how the solar

Long-run Resource Scarcity    709 energy is harvested. Current solar photovoltaic panels are also primitive. New technologies based upon new materials like graphene, utilizing nanotechnologies, and developing solar film might transform the generation of electricity.6 The obvious objection is that these developments are far from certain. Yet the reason they do not yet exist is not necessarily that they cannot, but rather while the price of fossil fuels is cheap, there is little incentive to do so. As resources are depleted, the price can be expected to rise, and as the price rises the incentives for R & D rise accordingly. There may be some non-​renewables for which there are no substitutes, and which play key parts in industrial processes. But this is far from clear. Iron ore as a source of steel could be described in this way, but even here new materials may displace it. Graphene is one example, following the much earlier discovery of plastics, and then carbon fibre. The key point is that the combination of necessity and scarcity cannot be just assumed, because the results of future R & D are not known. For most of the twentieth century the main non-​renewable energy sources have been remarkably abundant and remarkably cheap. For this reason they have been depleted at the expense of developing alternatives. The internal combustion engine is a century old, and both coal power stations and the basic elements in steel manufacture are even older. As and when they become scarce, the incentives change. The conclusion that follows is that the physical scarcity of non-​renewables is not the same as their economic scarcity. Physical scarcity ultimately limits these assets, but not necessarily their economic applications. The price mechanism incentivizes the rates of their depletion, and the search for substitutes and the associated R & D.

Renewables The long-​run resource scarcity problem that provides the greater threat is not the depletion of non-​renewable fossil resources, as suggested by the peak-​oil theorists and the Club of Rome would have us believe the depletion of finite resources. It is the rapid depletion of renewables, which would, if left to their own devices, be for practical purposes potentially infinite in supply. Nature keeps on reproducing them, and will go on doing so until the end of Earth’s existence, subject to how they evolve. While there is physically only one Earth in terms of finite non-​renewables, there is an open-​ended prospect with renewables. It is these natural capital assets that are in rapid decline, and at risk of long-​run scarcity. The rate of extinction is at a level that qualifies alongside the other great extinction episodes in geological history, described by E.O. Wilson as the sixth great extinction episode. As he explains, we are ‘in the midst of one of the greatest extinction spasms of geological history’ (Wilson, 1992, p. 268). Habitat and ecosystem destruction is taking place on a global scale, with serious damage being inflicted on the major rainforests in the Amazon, Congo, and the Mekong, and the gradual pollution of the oceans, alongside the smaller scale damage (Myers et al., 2000). Deserts are encroaching, farmland is increasingly characterized by monocultures, and pesticides and herbicides are increasingly killing off pests, competing plant life, and the insects and birds that depend upon these plants. Unlike non-​renewables, there is no technical progress to replace renewables—​other than evolution.

710   Helm In the case of China, the impact of rapid economic growth, conventionally measured, and renewable natural resources is starkly illustrated. China’s pollution is now global in scale. It leads the world in carbon emissions, with severe consequences for biodiversity and human well-​being. Its main rivers are biologically dead in significant parts, its agricultural land is seriously polluted with heavy metals, and its biodiversity is in serious retreat (see e.g. Economy, 2004; The World Bank and Development Research Center of the State Council, 2013). This depletion not only reduces the long-​run natural capital, but also in the process undermines the future sustainable growth paths for two reasons. First and foremost renewable natural capital is a primary factor of production. Without clean air, freshwater supplies and food supplies, human existence is not possible, and without a plentiful supply of these factor inputs, economic production is greatly constrained. These are all services that depend upon renewable natural capital. Plants mop up carbon dioxide and emit oxygen, and, indeed, it was the growth and development of plants that created the current atmosphere necessary for human existence. Fresh water is the product of a complex biology. Food supplies depend upon soil, and soil functions as it does because of its own complex biology of micro-​and macro-​organisms. Secondly, renewable natural capital yields many direct benefits. Nature is vital to human well-​being, and provides direct utility in the forms of leisure, health, and cultural and aesthetic values (Natural Capital Committee, 2014). The key feature of renewable natural capital—​that goes on providing its bounty for free—​ is challenged when they approach thresholds. These thresholds are typically both uncertain and also interdependent. The units of analysis are complex. Renewable natural capital does not come in discrete lumps of identifiable and separate assets. It comes in ecosystems, is set in habitats, and relies upon the systems’ interdependence. As a result, it is hard to provide estimates of the renewables populations, because it is not clear what the population relates to (see Mace, 2014). An example helps here. Consider Atlantic Salmon. The population is much diminished, as a result of many factors. Salmon can be counted: the total stock of the different species can be estimated. Yet the threshold below which salmon are no longer capable of renewing themselves is not adequately captured by the number of salmon. The salmon depend upon multiple ecosystems—​the feeding grounds off Greenland, the state of estuaries, river systems and their levels, spawning grounds, the existence of commercial fish farms and the associated sea lice and other parasites, and so on. The threshold beyond which salmon can no longer reproduce sustainably is made up of these multiple ecosystem features. The uncertainty about the units and the causal relations means that thresholds can only be approximately estimated. In practice, there is typically a range of uncertainty within which the renewable asset is at risk. As there is a discontinuity at the threshold, as the depletion below this level is irreversible, and as above the threshold the resource is available for free, a risk-​averse approach indicates that safe limits should be set above the mean expected threshold. But it is possible to do better than maintain the threshold. For many renewable natural resources, the benefits merit setting the optimal stock above the threshold at some target level. It is not a matter of scarcity, but rather the investment in maintaining the level of the resource that yields the highest economic value.

Long-run Resource Scarcity    711

Depletion and Sustainability In the long run, the prospects for future economic growth depend upon the depletion rules applied to resources. A sustainable growth path is one that bequeaths to the next generation a set of assets at least as good as the ones it inherited. In other words, the aggregate capital stock should be non-​declining. This general asset rule allows for substitution between the asset classes. In particular, man-​ made capital and human capital can be substitutes for depleted natural-​capital resources. A more stringent rule would require that the aggregate level of natural capital is not allowed to fall. This would still admit substitution between natural capital and other forms of capital, but within the overarching constraint. Any loss of natural capital through development has to be matched by a corresponding increase elsewhere. The justification for such a rule lies with the scale of the threats to non-​renewables noted earlier.7 The application of these asset rules differs between non-​renewable natural capital and renewable natural capital. Non-​renewables can only be used once, so depletion is a matter of who benefits. If one generation uses them, they are not available to future generations. The question is then: how should future generations be compensated for the loss? The conventional answer is that when non-​renewables are depleted, the economic rents should be reinvested in other capital assets, so that the benefits are shared with future generations. These rents are net of the costs (including normal profit) of the depletion. It is an answer that is almost always violated: very few resource-​rich countries explicitly provide for such investments. Few have sovereign wealth funds and those that do rarely follow this rule. Consider, for example, British North Sea oil and gas. The benefits from depletion were taken by the current generation. The tax yield (which accounts for most of the economic rents) went to the government of the day, and facilitated lower taxes and higher public expenditure. Some of this benefit will have fed through into investment. Some public expenditure is capital rather than current. Consumers and companies may also have invested some of the proceeds. Yet the overall impact will not have been remotely equivalent to the economic rents.8 Even the relatively straightforward North Sea example throws up a number of complicating dimensions. Technical progress has implications for the depletion rate. Suppose, as noted earlier, a substitute for oil is discovered, so that its value gradually falls towards zero. Future generations therefore place zero value on the resource. There is less need to compensate them for use now. However, suppose the depletion yields a liability for future generations—​ the carbon dioxide in the atmosphere. Compensation is then required for this liability. It remains to sort out what sort of compensating investments should be made. The aggregate natural-​capital rule dictates that the depletion of non-​renewables should fund increases in renewables. A weaker general aggregate capital version allows for a mix of capital investments. For the renewables, the aggregate rule dictates that where these assets are damaged, there should be corresponding compensation to the value at least as great. Recall that renewables can continue to provide their services for the very long run—​in effect, forever. The loss of a renewable asset, by driving the stock below the threshold, foregoes the full value of the yield in perpetuity. It is a value only bounded if a discount rate is applied.

712   Helm The stronger principle of compensating investments in other non-​renewable natural capital requires the measurement of natural capital so that assets at least as good can offset the loss (Department for Environment, Food and Rural Affairs, 2013). This turns out to be complicated: all natural assets are located in specific ecosystems, and hence a like-​for-​like compensation is potentially not going to be on offer. Suppose an ancient woodland or part of a rainforest is cut down. Recreating and restoring these assets is practically impossible. In any event there may be other, more valuable compensations in unrelated areas—​restoring a wetland, cleaning up a river, and so on. Creating a supply curve of renewable natural capital improvements is a complex undertaking. Maintaining both types of natural-​capital resources requires capital maintenance, to meet the rule. Capital maintenance is an accounting charge against revenues. It might be applied across all capital assets, or just natural capital. As a provision, its application would radically impact on national income accounts: economic performance would be net, not gross as it is currently, of capital maintenance (Mayer, 2013). This links long-​run resources to the concept of sustainable economic growth, which measures the consumption path that can be maintained over time. As future growth depends on the asset base necessary for production and consumption (resources as factor inputs, and resources for direct consumption), it follows that the measure of economic performance that meets this rule is net of capital maintenance. Currently, gross domestic product (GDP) national income accounts are gross, not net, of capital maintenance. They treat all income and all expenditure on the same—​ effectively cash—​ basis. Therefore, GDP can be increased by depleting national resources, which would mean failing to maintain capital assets. A simple example illustrates how this plays out. Suppose the government decides not to fill in the holes in the road. The costs fall, and hence the surplus of income over costs goes up. GDP goes up. But at some future date the state of the roads will require serious attention. Cost will go up and GDP will go down. The short-​term neglect increases GDP and reduces the sustainable growth rate. This simple example is being played out on a grand scale across the planet. The renewable and non-​renewable capital is being depleted, counted as positive to GDP, but with serious long-​run detrimental effects. The current generation is having a party at the expense of future generations.

Conclusions Keynes (1924) once famously remarked, ‘in the long run we are all dead’. Many environmentalists think that this may be true not only for the individuals Keynes had in mind, but also for much of the population. Long-​run resource scarcity has been viewed as a check on population and growth—​from Malthus’ on food production to Jevons on coal through to Hubbert on oil and the Club of Rome on a combination of factors. It turns out that most, if not all, of these prophets of doom are mistaken in the arena where they have been looking. There is no serious threat of long-​run resource scarcity for the sorts of non-​renewables they primarily focused upon. Not only have the non-​renewable reserves turned out to be much greater for key minerals, but technological progress has also changed

Long-run Resource Scarcity    713 the nature of the constraints. Malthus’ food scarcity has proved a weak constraint—​the seven billion people are now fed and world hunger has not increased. Where long-​run resource scarcity matters is elsewhere—​with renewable natural capital. The prospect in this century at current economic growth rates of a sixteen-​fold increase in the world economy (and the associated consumption) does not bode well for the world’s ecosystems on a business-​as-​usual basis. Renewables are largely immune from technological progress, other than by genetic manipulation. Their future bounty is only unlimited if they are kept above critical thresholds. The current rate of extinction of species—​the key long-​run resource—​will be a serious limit to growth. But it is not inevitable. Abiding by the aggregate natural capital rule set out in the chapter is the way to avoid such an outcome, and ensure economic growth is sustained.

Notes 1. See Helm (2015a) for a comprehensive treatment. 2. The Brundtland Commission was formally set up in 1984 and published its key report in 1987 (United Nations, 1987). 3. For a more extensive critique of peak oil, see Helm (2011, pp. 68–​91). 4. For his general approach, see Simon (1981). On the bet, see Ehrlich and Ehrlich (1998, pp. 100–​104). 5. The International Energy Agency predicts that the world’s proven coal reserves would last for approximately 135 years at current production levels, while actual coal resources are estimated to be twenty times larger than the known reserves (International Energy Agency, 2014). 6. For a summary of some of the battery options, see Shultz and Armstrong (2014). See also Helm (2015b, Chapter 11). 7. The two rules are formally set out in Helm (2015a, Chapter 4). 8. The depletion of non-​renewables for the benefit of the current generation is evident for both these and other assets. See Khan et al. (2014).

References Campbell, C. (1997). ‘Depletion patterns show change due for production of conventional oil’. Oil and Gas Journal 29: 33–​37. Department for Environment, Food and Rural Affairs (2013). ‘Biodiversity offsetting in England’ https://​www.gov.uk/​government/​collections/​biodiversity-​offsetting (last accessed 18 April 2017). Economy, E.C. (2004). The River Runs Black: The Environmental Challenge to China’s Future (Ithaca, NY: Cornell University Press). Ehrlich, P.R. (1968). The Population Bomb:  Population Control or Race to Oblivion? (New York: Ballantine Books). Ehrlich, P.R. and Ehrlich, A.H. (1998). Betrayal of Science and Reason: How Anti-​Environmental Rhetoric Threatens Our Future (Washington, DC: Island Press). Helm, D.R. (2011). ‘Peak oil and energy policy—​a critique’. Oxford Review of Economic Policy 27: 68–​91.

714   Helm Helm, D.R. (2015a). Natural Capital: Valuing the Planet (London: Yale University Press). Helm, D.R. (2015b). The Carbon Crunch: Revised and Updated Edition (New Haven, CT: Yale University Press). Helm, D.R. (2017). Burn Out: The Endgame for Fossil Fuels (London: Yale University Press). Hubbert, M.K. (1956). ‘Nuclear energy and the fossil fuels’. Drilling and Production Practice, American Petroleum Institute & Shell Development Co., Publication No. 95. International Energy Agency (2014). World Energy Outlook 2014 (Paris: OECD). Jevons, W.S. (1865). The Coal Question: An Enquiry Concerning the Progress of the Nation, and the Probable Exhaustion of Our Coal Mines (London: Macmillan & Co.). Keynes, J.M. (1924). A Tract on Monetary Reform (London: Macmillan). Khan, J., Greene, P., and Johnson, A. (2014). ‘UK natural capital: initial and partial monetary estimates’. Office for National Statistics http://​www.ons.gov.uk/​ons/​dcp171766_​361880.pdf (last accessed 18 April 2017). Mace, G. (2014). ‘Towards a framework for defining and measuring changes in natural capital’. Natural Capital Committee, Working Paper 1. Mayer, C. (2013). ‘Unnatural capital accounting’. Natural Capital Committee, Discussion Paper. Meadows, D.H., Meadows, D.L., Randers J., and Behrens III, W.W. (1972). The Limits to Growth:  A  Report for the Club of Rome’s Project on the Predicament of Mankind (New York: New American Library). Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B., and Kent, J. (2000). ‘Biodiversity hotspots for conservation priorities’. Nature 403: 853–​858. Natural Capital Committee (2014). ‘The state of natural capital: protecting and improving natural capital for prosperity and wellbeing’ https://​www.cbd.int/​financial/​values/​uk-​stateof-​ naturalcapital.pdf (last accessed 18 April 2017). Natural Capital Committee (2013). ‘The state of natural capital: towards a framework for measurement and valuation’ https://​www.gov.uk/​government/​publications/​natural-​capital-​ committees-​first-​state-​of-​natural-​capital-​report (last accessed 18 April 2017). Shultz, G.P. and Armstrong, R.C. (eds) (2014). Game Changers: Energy on the Move (Stanford, CA: Hoover Institution Press) Simon, J.L. (1981). The Ultimate Resource (Oxford: Martin Robertson). The World Bank and Development Research Center of the State Council (2013). ‘Seizing the Opportunity of Green Development in China’, in China 2030:  Building a Modern, Harmonious, and Creative Society (Washington, DC: World Bank Publications). United Nations (2014). ‘Probabilistic Population Projections Based on the World Population Prospects:  The 2012 Revision’. Department of Economic and Social Affairs, Population Division http://​esa.un.org/​unpd/​ppp/​ (last accessed 18 April 2017). United Nations (1987). ‘Our common future: report of the World Commission on Environment and Development’. http://​www.un-​documents.net/​our-​common-​future.pdf (last accessed 18 April 2017). US Energy Information Administration (2015). ‘US supply and disposition’ http://​www.eia. gov/​dnav/​pet/​pet_​sum_​snd_​d_​nus_​mbblpd_​m_​cur.htm (last accessed 18 April 2017). US Geological Survey (2008). ‘90 billion barrels of oil and 1,670 trillion cubic feet of natural gas assessed in the Arctic’ www.usgs.gov/​newsroom/​article.asp?ID=1980&from=rss_​home (last accessed 18 April 2017). Wilson, E.O. (1992). The Diversity of Life (Cambridge, MA: Harvard University Press).

Chapter 38

Rec oncep tua l i z i ng Re sou rce Peri ph e ri e s Caitlin A. McElroy Introduction A twelve-​hour drive on newly paved roads and dusty dirt tracks brings you from Ulaanbaatar to the booming mining region of the South Gobi Desert. Massive copper, coal, and gold extraction sites are redefining this arid landscape. The materials are quickly flowing to China and the revenues hold the promise of re-​positioning Mongolia in the global economy. Often synonymous with distant, remote, or ‘out-​there’, Mongolia is representative of the countries where substantial new resource production is taking place. These are countries and regions that have operated on the periphery of the global economy and now, owing to the combined pressures of development and environmental change, are emerging as ‘new’ resource frontiers. Re-​examining these places and our engagement with their resources reveals what resources they are bringing to the global market, as well as what resources climate change is taking away. This resource dynamic calls for a reconceptualization of how resource peripheries are understood in economic geography. Despite modern innovations and growth in the service economy, the environment still plays a critical role in determining the processes and patterns of economic activity. The production of extractive resources obtained through mining, oil and gas drilling, and forestry are some of the most visible examples of this process. Often described as the foundations of development, these resources provide the raw materials for global growth. It was an understanding of the relationships and impacts of resource extraction, production, and distribution that defined global empires and continues, in the power of its actual materiality, as well as its imagined windfalls (Weszkalnys, 2008), to shape global economic relations. The push to plunder the new periphery is now arguably different from the resource frontiers of the past. For example, the technology used for resource extraction has reduced the labour intensity of most extraction processes and opened up new places for extraction that previously presented physical limitations. The Arctic and the sea floor, in particular, pose new technical and governance challenges. More imposing yet on this new frontier is an increasingly global condition of

716   McElroy environmental constraint—​the very same confluence of conditions bringing resource extraction to new peripheries is influencing the ways this extraction occurs, and altering the approach to resource based development. Water scarcity, climate change, renewable energy demands, increasing urbanization, mobility, and even meat consumption not only drive, but also threaten and bound, resource production in ways never before experienced. It was over ten years ago that Hayter et al. argued, ‘A truly “global” economic geography cannot afford to ignore resource peripheries’ (2003, p. 21). In the intervening years there has been both a global financial crisis and a commodities ‘super-​cycle’, and both events have increased the importance of understanding the role of resource peripheries in the global economy. The points proposed in this chapter are intended as starting points to re-​examine the assumptions not just of resource peripheries, but to question and learn from the paths they are building to development and the sustainability of our environmentally based and determined economy. This chapter will present how resource peripheries have been conceptualized in economic geography, and what it is about our resource peripheries currently that is different from previous conceptualizations. Then, the chapter will move on to introduce and discuss three changes that are key in driving the transformations that are requiring our reconceptualization of resource peripheries. They are: (i) the effects of the rise and fall of the commodities super-​cycle; (ii) the increasing exposure and vulnerability of resource peripheries to climate change; and (iii) the expectations of extractive industry-​led development. These changes affect resource peripheries in multiple ways creating both opportunities for growth and development, as well as considerable vulnerabilities that affect the environmental and economic sustainability of these changes.

Geography and Resource Peripheries Core and Periphery Resource peripheries are a specific extension of general core-​periphery theory. Core and periphery are used to describe the location, state of development, and influence of states relative to each other. It refers to the idea of a central or core area of activity serving to bring together other disparate activities and places. There is often a power dynamic between the core and periphery favouring the core, or even that the power of the core is contingent on its ability to capture resources from the periphery. It is a concept that is relative to this chapter as it is applied in traditional economics, economic geography, and development (Innis, 1930, 1933; Hayter et al., 2003; Krugman, 2011). Each discipline was trying to explain uneven development, the persistent difference in economic development between different places (Krugman, 1981; Smith, 1984). Older research on core–​periphery looked at the relationship between nations. It started with the establishment of global trading firms—​such as how the East India Company in the fifteenth century was based on new networks of trade, rather than production—​and how they started to reorder the global economy (Wallerstein, 1979). This created a global

Reconceptualizing Resource Peripheries    717 trading system with nations that were core, semi-​peripheral, and peripheral to trade. Over time this system was carried forward to reflect the degree of industrialization of the nations participating in this capitalist trade system. Countries that were core were industrialized, semi-​periphery were newly industrializing, and periphery were not industrialized (Gereffi, 2005). Such classifications still resonate in terms of developed, developing, and least developed nations. For the resource peripheries we are concerned with in this chapter, these studies of how and why different spaces participate in the global economy are still critical. Another factor in the hierarchy of economic development was also a concept of ‘first nature’ or the natural resources, climate, and transportation options that define the first economic potentials of some places. Resource production advantages from these locations were used to explain some uneven development (Ottoviano and Thisse, 2004). However, other aspects of uneven development, and particularly the growth of some non-​industrial agglomerations, required further explanation. Both economic geographers and economists took up this challenge. Within nations the interest was to understand the dynamics of regional cores, cities, industrial hubs, and other forms of agglomeration. Economic geographers sought to understand agglomerations and uneven development through descriptive and qualitative research (Krugman, 2011). Then in the early 1990s economists including Fujita (1988), Venables (Fujita et al., 1999), and Krugman brought the significance of spatiality and geography to mainstream economics. In 1991 Krugman introduced ‘new economic geography’ in Geography and Trade. Here he established a new approach to core–​periphery in economics. This work focused on the effects of transport costs, economies of scale, and increased manufacturing. He looked to explain why certain places became cores, and what forces drove this agglomeration process using more tangible and comparable metrics than economic geographers. Krugman (2011) sought to model the pull of market size as a ‘centripetal’ force and the pull of natural resources from dispersed locations as a ‘centrifugal’ force. This work raised considerable debate with economic geographers regarding how ‘new’ it was and the limits and advantages of different methods (Krugman, 2011). However, it did expand the understanding of geographical significance within economics. Core–​periphery is a broad church theory that has a varied life across the history of economic development, and the scales of economic activity to which it is applied. It has also shaped and potentially reaffirmed where research is invested. As Potter (2001, p. 423) puts it: . . . that those who specialize in the core areas, and who deal with what they regard as the core topics, see work carried out in Anglo Euro-​America as theoretically more demanding and ultimately of greater societal value. In this way, Anglo-​American geography defines its own disciplinary ‘core’, whilst matters of development and development geography become part of the ‘periphery’.

Resource peripheries have been defined by the theoretical organizing of core–​periphery. They have also been researched within the biases of being peripheries and therefore of arguably less dynamism and significance. This chapter highlights how a reconceptualizing of resource peripheries is necessary to challenge the long-​standing performative effect of core–​ periphery theory, and its bias to certain research topics and global locations.

718   McElroy

Core–​Periphery and Resource Peripheries in Economic Geography Core–​periphery theory has had a particular relationship with resource peripheries within economic geography. This relationship has involved interest in uneven development and agglomerations that drive other disciplines, as well as a specific interest in nature and the environment. This interest in the environment has had important effects in directing the attention of economic geographers to and away from resource peripheries at various points in time. The early evolving programme of economic geography consistently moved away from a purposeful incorporation of resource peripheries. Economic geography started this move away from resources geography in the 1920s as part of the reaction against environmental determinism, and the work that followed from Ellen Churchill Semple (Colby, 1933). Geographers became wary and critical of the methods and implications of environmental determinism (Sauer, 1941). As a discipline, geographers moved to examine the opportunities environmental difference created rather than delimited. During this retreat from environmental determinism, and as a casualty, resource geography, the work of Canadian geographer Harold Innis was an exception. He wrote extensively on resource peripheries in his work on the Canadian ‘staples’ resource economy (Innis and Drache, 1995). His staples theory described the Canadian political economy of the 1930s (Innis, 1930, 1933). Staples were the natural resources of timber, furs, fish, and minerals Canada supplied to Europe in the dynamic of its postcolonial resource economy. The political and economic relationship between Canada and Europe was entwined in the perpetuation of Canada as a resource extraction economy. The geographical attributes of Canada, as well as its population density, bring some contextual specificity to this theory; however, it applied broadly to many postcolonial resource economies (Innis and Drache, 1995). In this theory the environment, much like labour or capital, is framed as agentive. In giving the environment this power as an agent, Innis subtly introduced a relationship between a yet-​ to-​be-​formed environmentalism and the economic production and geopolitics of resource regions or peripheries. This research defined an early theorization of the performance of resource economies and their relationships with the larger global economy. However, the conceptualization of a resource periphery, and particularly the centrality and agency of the environment, was not revisited for a long-​time in economic geography. Instead, economic geography came to focus on areas of industry and its labour and cap­ ital. Deindustrialization in large American cities was a principle research concern from the 1960s into the 1980s (Scott, 2000). Then as the discipline developed and the power of labour declined in the late twentieth century, the movement and concentrations of capital became the dominant interest (Scott, 1992). This focus on capital was biased to the activities of cores—​centres of urban growth, trade, and financialization. Resource peripheries, while a relevant factor to these developments, were largely outside of the frame of analysis. The post-​Cold War advance of globalization and acceleration in the speed of capital flow and accumulation drove global economic change, and refocused economic geography even more intensely on cores as the growing and diverse urban hubs of a new global economic system. This growth of cores through globalization created increasingly ‘sticky’ places where agglomerations occurred and the forces of uneven development processes were particularly

Reconceptualizing Resource Peripheries    719 visceral (Markusen, 1996; Hayter et al., 2003). By the early 2000s even popular media such as Thomas Friedman’s The World is Flat (2005) were framing a world with energetic cores and blunted contours of geographical difference. The significance of the heterogeneity of space was lost to the enthusiasm for the innovation produced through the consolidation of the heterogeneity of culture into globalizing cores. In contrast, peripheral areas were described as ‘slippery’ (Markusen, 1996) as they are far away and difficult to research (Potter, 2001). They were considered detached from the focus of geographers’ studies as much because of their physical locations as for their assumed minimal cultural relevance (Potter, 2001). Change, vibrancy, and innovation apparently slipped away from the peripheries to global cores. During the Cold War, in ways not too dissimilar to the structures of colonialism, resource peripheries were captured and controlled to serve one of the two competing governance and market ideology systems. The USSR developed resource infrastructures across central and south Asia and the ‘–​Stans’. Europe and North America continued to pull raw materials from both their ‘old empires’, as well as areas of influence in Latin America and Africa. The end of the Cold War created opportunities for self-​determination for these resource-​producing regions. However, redefining the resource-​ led relationship with the global economy produced conflict within these states and from external geopolitical pressures (Le Billion, 2008; Jackson, 2015). The internal conflicts of cultural identity, indigeneity, and sense of place have been researched extensively through the work of cultural anthropologists and political ecologists (Nash, 1993; Bebbington, 2010). It is this later feature of the geopolitical uncertainty of resource peripheries that is re-​emerging with new importance owing to environmental change and continued development. Many of these resource peripheries contain some of the world’s largest undeveloped resources. They are also facing the first effects of climate change and the lasting effects of poverty and contested governance. As a result, rather than a ‘peripheral’ concern, places and networks of resource production are coming to more significantly influence the dynamics of capital flow, labour power, and the continuing development of economic cores—​the central interests of economic geography.

Core–​Periphery Relative to Resource Producer and Consumer Nations With much of the economic geography research focused on the activities within cores, economists, development theorists, and anthropologists dominated research on the periphery itself and its relationship with cores. This research has added considerable information about the contestations between, and individual development of, producer and consumer nations. One of the most influential discourses about resource peripheries led by research in economics and development is the resource curse theory. Grounded in the economics of resource-​ driven development, Cordon and Neary (1982) developed the economic model of the Dutch disease. This is a process in which resource-​producing countries have a reduced manufacturing sector due to both the increased investment in extraction and the changing terms of trade that make exports very expensive. Then, in the early 1990s, Richard Auty developed what became more formally the theory of the resource curse or ‘paradox of plenty’ describing the ways mineral-​rich economies failed to achieve or maintain sustainable development

720   McElroy (Auty, 1993). Prompting a further merger of the economic and political development aspects of the resource curse in 1995, Sachs and Warner (1995) began to publish a series of quantitative economic papers that evidenced the relative development failure of resource-​rich states. These studies cumulatively created a discourse focused on the challenges of achieving development from resource wealth (Ross, 1999; Rosser, 2006). These discourses each address resource economy dynamics in countries and places that are in some way distinguishable from the mainstream global economy. The colonial and postcolonial economies of interest to Innis occupy a particular ‘otherness’ partly related to their resource production. Similarly, the resource-​rich countries focused on by the resource curse, while applicable to the USA, Australia, Canada, and Norway, are mostly developing countries. In particular, it examines those experiencing the negative or resource curse effects that may be hindering their development. There is arguably a bias in academia towards investigating resource economies as both a stage of development and as occurring in places external or periphery to the ‘mainstream’ global economy. What is sometimes left out of this narrative is the significant changing role of production networks, competition for international foreign direct investment, or the rise of state-​ owned enterprises. The networks involved in the production of mineral resources have played a considerable role in articulating the ‘peripheralness’ of resource-​based economies (Auty, 1997). These dynamics bring us back to the early core–​periphery theorizations from global trade systems. In the 1970s, advancing service economies experienced declining mineral intensity per percentage of gross domestic product (GDP). Producing nations sought to improve the capture of declining resource benefits through the creation of producer cartels and resource nationalization. However, these changes did not necessarily increase demand for resources or further integrate these producer nations into the global economy. The most accessible ores had been exhausted, bulk shipping was improved, and new discovery of resources in remote regions of Canada and Australia made up some differences (Auty, 1997). The decolonization of the Global South, and ruptures from colonial resource networks, did not necessarily improve terms of trade or capture significant benefit to improve additional economic development. Slowly multinational extractive corporations (MNCs) were reintroduced. In this new wave of international investment, MNCs were often wary of the stability, potential conflict, and governance of the countries they were operating within and became prone to create enclave operations (Auty, 1997). As we will pick up again later, environmental damage and natural capital depletion became new issues of contestation. Other scholars have observed and reacted to this other or ‘peripheralness’ in the way resource economies are analysed. Some have taken a particular interest in researching what lies within and makes up this ‘otherness’ of resource economies. Michael Watts has done extensive studies of conflicts and development surrounding Nigerian oil production (Watts, 2004). He approaches the governmentality of resource-​extraction regions as it is imposed by the nation state and oil producers, and in the response from those affected by the extraction process. This approach is specifically attuned to the significance of the local extraction zone as the frame of interest. Within this local framing the approach is holistic rather than attuned to one side of the resource binary between resource consumption and production. While research such as this plays an important role in understanding the political economy of resource production, it is bound to the micro-​level of analysis in much the same way the staples trap and resource curse have been tied to macro interstate comparisons.

Reconceptualizing Resource Peripheries    721 Finally, the processes of development and self-​determination occurring within many resource peripheries have been raised to contest the direct relationship between resource-​ driven development and a developmental curse (Wright and Czelusta, 2004; Wick and Bulte, 2006). Despite the often highly sensitive environments of resource peripheries and disproportionate effects of climate change, there are calls for the right to develop and abuse the environment in much the same way as the global economic hubs did in the previous century (Rosales, 2008). And further, perhaps the most obvious contestation that changes the power of these discourses is over the ownership, access to, and benefits from resources themselves. The struggle for peripheral self-​determination has opened up opportunities for exploitation and fights for control (Bannon and Collier, 2003; Le Billon, 2008, 2012). It is perhaps the case that in addition to considerations of long-​term development and environmental protection, the currently emerging resource peripheries have at the forefront of their own concerns the contestation over the rights to the resources and their benefits.

Exploring the New Emergent Periphery As the history of resource peripheries in economic geography reveals, certain discourses in academia, and their enactment in global trade and development policies, have helped to shape and maintain what makes resource peripheries ‘peripheral’. However, the convergence of three particular events—​changes in the resource super-​cycle; the increasing exposure of the periphery to environmental change; and expectations of extractive industry-​led development—​ presents new vulnerability and opportunities for resource peripheries. Consideration of these events has the potential to influence how resource peripheries are conceptualized moving forward.

The Growing Periphery and the Commodities Super-​cycle Taking a step back before describing the significance of the commodities super-​cycle for the re-​emergence of resource peripheries, it is important to make a point about the economic significance of resource peripheries. Resource peripheries are still engines of considerable global wealth. For example, resource revenues in 2011 contributed 8.7 per cent of South Africa’s GDP and 22.5 per cent of Russia’s (World Bank, 2015). Several of the largest firms in the world are focused on resource production such as Exxon. This Texas oil firm is the seventh biggest business in the world and competes with Apple for title of the world’s largest public company. China’s only non-​bank in the top-​ten biggest businesses in the world is also an energy company, PetroChina. While two more oil businesses make the top-​twenty biggest businesses of 2015—​Royal Dutch Shell and Chevron—​several of the mining majors such as BHPBilliton and Rio Tinto are not far behind (Forbes, 2015). Each has a market value above US$75 billion, placing them in the top-​105 largest businesses (Forbes, 2015). This scale of global business in the resources sector is not declining proportionally to the rise of the alternative sectors. Because of the size and significance of the resource business, the commodities super-​cycle—​and related to it the global financial crisis of 2007—​had important effects on resource peripheries.

722   McElroy The commodities super-​cycle began in the early 2000s and continued through the global financial crisis (Cuddington and Jerret, 2008). Commodities markets are observed to operate in cycles of growth and decline, or, if you will, periods of boom and bust that affect the whole of the industry. During a boom the prices for commodities rise, as does long-​term investment in extractive infrastructures. The boom of the 2000s was especially strong for prices and investments in the metals and minerals industry. Commodity cycles are generally marked by an upswing of ten to thirty-​five years and in twenty to seventy years have completed their full cycle (Erten and Ocampo, 2012). The demand for resources is driven by major developing countries for inputs to industrial production and urban development. The commodities boom of the 2000s was primarily driven by the rapid growth of China. The boom of the commodities markets in the 2000s was unprecedented in its growth both because of how much growth, or how high the boom of the cycle went, and in its duration (Erten and Ocampo, 2013). The persistence of this cycle through the financial crisis led to an atmosphere of investment and producer optimism that did not seem to indicate a belief that this cycle would end. Rather it came to seem like a new normal for global development and the commodity sector’s role in economic growth (Radetzki, 2013). The understanding of and belief in the commodity cycle are important because of how they influence other economic decision-​making. Commodity prices are central to policy development in resource-​dependent countries. Estimates of the duration of the cycle influence investment decisions to increase capacity. Further, commodities cycles are tied to global financial health as portfolio managers use commodities to hedge investments in other sectors (Erten and Ocampo, 2012). Understanding and trying to predict commodity super-​cycles has enchanted generations of economists throughout the twentieth century. The pattern and consequences of national economic growth in resource-​consuming nations on commodity prices, and the related growth opportunities in resource-​producing nations, is the complex advancement of early global systems of trade started in the fifteenth century that established core–​periphery dynamics. Concentration on commodity cycles started with Schumpeter in 1939, who introduced cycles as waves of variable durations. These were based on his ideas regarding creative destruction and different stages of economic development experienced simultaneously by different trading nations. Following from this was influential work from Kuznets (1940) and then the Prebish–​Singer hypothesis (Singer, 1950, 1998; Prebish, 1950). These waves or cycles have structured expectations for the duration of these booms. Such cycles indicated that the boom of the 2000s would only be as strong as growth in China or China’s ability to separate itself from the industrial world and its slow growth (Erten and Ocampo, 2012). At an even larger scale throughout the course of this boom, the geopolitics of resource security has changed. The end of the Cold War and the rise of the War on Terror has changed not only the sites of political resource significance, but made the strategy of securing new markets occur through cooperative governance and civil development instead of through resource-​providing state partnerships as was done in the Cold War years. This strategy has been seen not only by western MNCs, but also by China. Especially in Africa, Chinese resource peripheries have been new places of influence and inclusion in the larger Chinese economic system (Mohan and Lampert, 2013). Even in the absence of the jobs and total windfall expected of resource production (Weszkalnys, 2008), new infrastructure and transport are transforming small bits of African resource economies’ landscapes. The overwhelming demand from China brought exploration and investment to new places, opening

Reconceptualizing Resource Peripheries    723 up new resource peripheries. Beyond Africa and the central Asian states these places include the Arctic Ocean and its peripheral nations such as Greenland. Chinese exploration has even led to the ocean floor, where they have purchased sea-​mount claims (Petersen et al., 2016). The shared governance of these places is now tested. These new peripheries present challenges for global economic power balances, and as we will see in the next section, challenges to Earth’s environmental limits. For other countries like Mongolia, balancing China’s hunger for resources, as well as participation in free-​market capitalist democracy, has been difficult. They have exchanged the role of communist supply-​state to Russia for the global economic dominance of China. Nearly 80 per cent of Mongolia’s exports go to China, and several of its coal mines are less than 100 miles from the Chinese border. The fervent Mongolian push for expanded market options and partial ownership of the largest mines in country are part of trying to escape this dynamic, and its similarity to the captive role of a resource staple periphery as described long ago by Innis (Jackson, 2015). It is now clear in 2016 that the super-​cycle did reach its limit, as Chinese growth slowed and demand for raw materials weakened. This has brought on the collapse of the price of iron ore from a high of over US$100 a tonne in May 2014 to a low of US$50 in April of 2015 (Statista, 2015). This collapse is also visible in the oil price since late 2014 (Baffes et al., 2015). The collapse of this boom has led to an inequality of busts. The boom prices were also raised owing to supply-​side constraints, such as overall resource scarcity and lower ore grades (Erten and Ocampo, 2012). Significant investment capacity is necessary to extract these ores. Peripheries that benefited from these investments in time will ride out the downturn differently through a low level of ongoing production to service these debts. These downturn inequalities will continue to shape the resource periphery nearly as much as the boom-​time growth.

The Peripheral as Vulnerable: Environmental Change The increasing consideration of, and in effect agency of, the environment has grown with the significance of environmental change in the twenty-​first century. Emerging resource peripheries are connected to other spaces and global governance through the environment in ways that were unthinkable before the 1970s and for Innis in the 1930s. While the flight from environmental determinism pushed geographers from resource peripheries, it is a new form of environmental determinism in environmental change, fragility, and borderlessness that now pulls geographers back. As we have seen, scarcity of resources is driving exploration and development to new resource peripheries. Scarcity and damage to other essential resources such as clean water, fresh air, and productive agricultural lands is changing the way resource extraction is accepted and conducted. For example, in Mongolia, the majority of new mining developments are taking place in the arid South Gobi desert. The burden of water use for the industry is enormous compared with the traditional use by herders. Many mines have drilled boreholes to tap un-​potable aquifer water and there is great concern and uncertainty over the long-​term supply of water and the ability of the aquifers to support further industrial development (Sternberg, 2013; IFC, 2015). In some sense, water has become nearly as valuable a strategic commodity as any of the minerals in the ground. Not only will it perhaps limit

724   McElroy the development of industry and future livelihoods in the region, but it is also a geopolitical concern. Mongolia captures little of the rainwater and snows it receives, and its rivers and mountains feed much of central Asia. Among options to increase the availability of water for industry have been ideas to capture more of this water within Mongolia and therefore reduce the availability of this water downstream. The suggestion of ideas such as this has quickly raised the concern and ire of affected neighbours (Jackson, 2015). As resource extraction moves to more remote areas, it is more frequently in regions with water security concerns. This includes arid areas, as well as areas with communities dependent on fragile water supplies for their livelihoods. Resource development of all kinds from agricultural, to energy, and mineral production is now confronted by increasing water insecurity. Mining, in particular, has seen an increase in local conflict and interventions from powerful international non-​governmental organizations (NGOs) related to water use. Different from contestation over economic benefit and wealth distribution, irreparable damage to the environment as a result of resource extraction has a new urgency. Work has been stopped or delayed as a result of community protest over water contamination around the world and notably in South America. For example, Minera Yanacocha plc in Peru has suspended the exploration of additional mining areas as a result of community water security concerns (ICMM, 2015). Similarly, production was stopped in Witbank South Africa owing to concerns over water contamination, even if not directly related to mine use (ICMM, 2015). The power of local actors to stop mining, and the responsibility of ‘core’-​based extraction firms to achieve a Social Licence to Operate, has never been higher (Prno, 2013). The rise of international watchdog NGOs has made a difference in granting the environment agency that applies equally in economic cores and peripheries. More interesting for our understanding of the ways resource peripheries are changing is the rise of local environmental resistance and NGOs. They have moved resource conflict from infighting for control of rents (Le Billon, 2012) to resistance to external and foreign intervention. While global NGOs have considerable global mobility to travel to and acquire knowledge about resource peripheries to challenge MNC networks and local bases (Soyez, 2002), there is new subtle wariness of the role of these NGOs. In some ways similar to resource nationalism and the rejection of international MNCs, international NGOs bring with them a similar elite power at a distance. In response, the acceptance and use of these NGOs has become more strategic, and a greater number of smaller local NGOs are emerging. The role and power of international environmental NGOs is different in these emerging peripheries than it has been in the past. Contributing reasons for this include both the experiences of other resource peripheries, the longer contested history of governance interfering in foreign formal and soft governance in resource peripheries, and the more direct link between new resource extraction and specific development objectives. There are now at least two notable examples of major mining projects that have been curtailed because of their potential for environmental damage. One, the Pascua Lama project, was planned for high in the Andes on the border of Chile and Argentina by Barrick Gold. To conduct this project a portion of a glacier would be destroyed. Local and international protest was so extensive and organized that Barrick Gold eventually stopped pursuing the project (Li, 2016). Similarly, the Pebble Project in southern Alaska was arguably the largest new copper and gold deposit yet to be developed. The site was situated in the headwaters of the Alaskan sockeye salmon fishing industry. This multibillion dollar industry is contingent on the environmental sustainability of the fishery and is the primary source of income for the

Reconceptualizing Resource Peripheries    725 communities including the indigenous Alaskans. Advocacy from locally formed and international NGOs, and pressure on the environmental impact assessment and approvals process, slowed and delayed the project. Eventually, the invested firms abandoned the pursuit of the project and gifted the shares of the project to local organizations (Urkidi, 2010; Restino, 2014). In both instances an internationally recognized sustainable environmental asset won priority over resource extraction. The ability to overcome MNC-​driven extraction projects now stands as an example to all resource peripheries. Water is one among many environmental issues such as biodiversity loss, dust pollution, and potential for other contamination that affect the operational level of extractive projects. However, the uses of the resources extracted, such as coal, are also a serious environmental problem making its way back to the sites of production in the form of climate change. Coal-​ producing countries such as Mongolia experience the direct smog and pollution effects of burning coal, as well as some of the first consequences of global climate change (Sternberg, 2013). Mongolians, like many in resource peripheries, are stuck in a difficult position between environmental leadership and what many see as fast-​track resource-​driven development. This debate between climate change and development has changed local, national, and international perceptions of resource use and extraction. It poses the potential for the latest emerging resource periphery countries to be the first to redefine or step away from certain aspects of resource production in the name of climate change—​a potentially new political power.

The Peripheral as Opportunity: Resource-​led Development In contrast to the increasing narrative of environmental vulnerability of resource peripheries, a narrative of opportunity from resource-​led development persists. It has recently been woven together in two parts. The combination of local empowerment for self-​governance and rise of the social licence to operate has fuelled belief that the resource curse can be ‘beat’. Secondly, the global financial crisis reaffirmed the value of tangible assets, while at the same time increasing the power and size of the secondary markets, or financialization of, the commodities business. The emerging peripheral regions are engaging with global capital and political power on two new levels. At the national level, the state is taking an active role in resource ownership and production. And, locally, different community groups, especially aboriginal and first peoples, have a greater voice for their own rights (De La Cadena, 2010). For example, the Mongolian government has designated several of the largest mineral deposits in the country as of strategic interest to the nation state. Once a deposit is designated as strategic, the government has the option to have ownership of the site up to 34 per cent. The government has acted on this right and taken the full stake in the Oyu Tolgoi mine. This ownership not only has the potential to provide the country greater revenues from the site, but also requires them to play a larger role in decision-​making and capital investment. Unlike resource nationalism and the creation of state-​owned and state-​run corporations, this model draws heavily on the financial and knowledge resources of MNCs while attempting to

726   McElroy maintain both the benefits of the owner/​operator and receiving rents and royalties. Few new resource-​periphery ­countries would be unaware of the risks of the ‘resource curse’, and the stigma of institutional weakness and government corruption it indicates, if there is a failure to achieve development from resource wealth. Consequently, new ownership schemes such as in Mongolia look to provide some security against these outcomes. For local communities, the question of resource wealth for what development or for whom has become very significant. As the communities most affected by mine operations, these communities are more aware of their right to particular compensations and benefits. Often, these financial flows are negotiated directly with the operating corporations and through negotiations for investments from this wealth from the national government. The map of financial flows because of this activism is looking increasingly different in emerging resource peripheries. A key force behind this power to get access to these financial flows is the increased recognition of the long-​term costs to local communities for environmental damage (Owen and Kemp, 2013). In effect, the rise of global environmental activism and environmental constraints has opened new financial pathways from resource extraction to local development. In relation to the growing strength of local protest and social licence to operate, corporations are cautious of taking on governance-​like roles (Bebbington, 2010). The more limited labour demands of most new mines provide few jobs for local people. Most employees are fly-​in-​fly-​out experts (Storey, 2001). Therefore, the work to contribute to development has become more complicated and integrated with national and local government capacity building. The power of these firms in the areas in which they operate is purposely reduced in terms of local governance and service provision. Historically, mining firms created paternalistic governance pockets within emerging peripheries. Current activities under the banner of community relations and corporate social responsibility struggle to find the balance between providing the additional services expected of local communities, and efficiency in enabling operations and community relations, without compromising the ability for these governance capacities to develop within the resource periphery (Bebbington, 2010). The second feature reinforcing the persistence of opportunity in resource peripheries is the reaction from the global financial crisis of 2007–​08. The excessive risk taking in secondary markets and overly complex financial products undermined confidence in the financial industry (Crotty, 2009). In turn, this created many reactive responses by investors, including a flight from intangible financial products and a return to the concreteness of commodities and other resources (Silvennoinen and Thorp, 2013). Outside the crisis epicentre in Anglo-​ America and Europe, several emerging economies, including China, were experiencing relatively strong economic growth in the years following the crisis (Overholt, 2010). A critical input to their growth included both the production and consumption of the raw materials necessary to build energy infrastructure, new manufacturing centres for textiles to the latest gadgets, and transportation. For example, the import of raw materials by China was 37 per cent of the world’s in 2012, dwarfing the import of these materials by the European Union, the next largest importer at 13.6 per cent of world imports (Eurostat, 2015). Further driving investment to commodities and resource peripheries was the scrutiny of management and new regulations that tightened the market in post-​crisis countries. This created a further contrast between the speed and opportunities to do business in some emerging economies and some countries at the heart of the financial crisis.

Reconceptualizing Resource Peripheries    727 The peak of the commodities super-​cycle was aligned with the peak of increasing financialization of the mining industry. A prominent example of this is the merger of two large mining firms: Glencore and Xstrata. Unlike the past mega-​mergers of the mining sector, this merger brought the financial and trading business of Glencore together with the new acquisition-​driven management style of Xstrata. Essentially, different stages of the mining and market process were purposely becoming integrated (Aversano and Ritsatos, 2014). Where there was previously a focus of one part of the industry on the industrial operations of mining and another on the trading and sales of the commodities, these were brought closer together. This change again adjusted the global relationship between core and periphery countries.

Conclusion Framed through the history of economic geography’s engagement with resources and resource peripheries, the key difference of the peripheral spaces that are now emerging as a result of resource production is the expansion of resource development into a wide range of new peripheries, the significance of global environmental constraint and change, and the persistent and evolving purchase of the opportunity for resource-​led development. This chapter has tried to illustrate and note some of the changes to resource peripheries resulting from these events. The growth of peripheries into new semi-​national, yet uninhabited, spaces such as the Arctic and the deep-​sea floor, calls for new approaches to the collaborative governance of these places. The growth of China that has fuelled much of the last commodity boom has created a new powerful resource drawing and agglomerating core. With this China-​centred core and periphery system, its practices and processes are slightly different and worthy of significant research. Balancing these drivers to reframe the scale of our understanding of resource peripheries are new environmental concerns that question the long-​ term value and legitimacy of extractive activities. Sustainable livelihoods contingent on the protection of water and biodiversity have assumed a new political power—​a power that has finally taken local roots to connect to the buzzing discourse within affluent cores. The choice to not participate in the resource periphery dynamic has shown glimmers of possibility in a few cases. Further, the challenges of climate change unite the historical exploits of industrial development against the current ambitions of resource peripheries for similar prosperity. Resource peripheries are frequently the first spaces affected by climate change and are faced with finding both the leadership and opportunities to pursue non-​resource-​driven development pathways. And it is the allure of successfully walking this tightrope between opportunities for development and possible corruption, economic inequalities, and environmental destruction that continues to fuel optimism for resource wealth. These events together present a strong case that the time has come to reconceptualize resource peripheries within economic geography. The core-​periphery dynamic for resource peripheries must now grapple with newly significant tensions. Further research should explore how peripheries are simultaneously gaining political power, while becoming more distanced from the resource development processes, all as they become more technologically sophisticated in operations and in their financial networks. Innovations at the international level of structuring resource peripheries must also be explored, as they are used in

728   McElroy the daily lives of resource-​producing communities. The competing narratives of resource peripheries as places of hope or horror each have the power to be performative and shape future decisions. Economic geographers should take some responsibility in reconceptualizing resource peripheries, as the stakes for global development, equality, and environmental protection have rarely been higher.

References Auty, R. (1993). Sustaining Development in Mineral Economies:  The Resource Curse Thesis (London: Routledge). Auty, R. (1997). ‘Mining, Enivronment and Development: A series of papers prepared for the United Nations Conference on Trade and Development (UNCTAD)’ https://​commdev.org/​ userfiles/​files/​1416_​file_​The_​Geopolitics_​of_​Mineral_​Resources.pdf (last accessed 20 April 2017). Aversano, N. and Ritsatos, T. (2014). ‘GlencoreXstrata. . . the profitable and untethered march to global resource dominance!’ Athens Journal of Business and Economics 10: 1–​15. Baffes, J., Ayhan Kose, M., Ohnsorge, F., and Stocker, M. (2015). ‘Understanding the Plunge in Oil Prices: Sources and Implications’. World Bank, Global Economic Prospects January 2015 https://​openknowledge.worldbank.org/​bitstream/​handle/​10986/​23611/​The0great0plun0 and0policy0responses.pdf?sequence=1&isAllowed=y (last accessed 20 April 2017). Bannon, I. and Collier, S. (eds) (2003). Natural Resources and Violent Conflict (Washington, DC: The World Bank). Bebbington, A. (2010). ‘Extractive Industries and Stunted States: Conflict, Responsibility and Institutional Change in the Andes’ in R. Raman and R.D. Lipschutz (eds) Corporate Social Responsibility: Comparative Critiques, pp. 97–​116 (New York: Palgrave). Colby, C.C. (1933). ‘Ellen Churchill Semple’. Annals of the Association of American Geographers 23: 229–​240. Corden, W.M. and Neary, J.P. (1982). ‘Booming sector and de-​industrialisation in a small open economy’. The Economic Journal 92: 825–​848. Crotty, J. (2009). ‘Structural causes of the global financial crisis: a critical assessment of the “new financial architecture” ’. Cambridge Journal of Economics 33: 563–​580. Cuddington, J.T. and Jerrett, D. (2008). ‘Super cycles in real metals prices?’ IMF Staff Papers 55: 541–​565. De la Cadena, M. (2010). ‘Indigenous cosmopolitics in the Andes:  conceptual reflections beyond “politics” ’. Cultural Anthropology 25: 334–​370. Erten, B. and Ocampo, J.A. (2012). ‘Super-​cycles of commodity prices since the mid-​nineteenth century.’ DESA Working Paper No. 110 http://​www.un.org/​esa/​desa/​papers/​2012/​wp110_​ 2012.pdf (last accessed 20 April 2017). Erten, B. and Ocampo, J.A. (2013). ‘Super cycles of commodity prices since the mid-​nineteenth century’. World Development 44: 14–​30. Eurostat (2015). ‘International trade in raw materials’ ec.europa.eu/​eurostat/​statistics-​explained/​ index.php/​International_​trade_​in_​raw_​materials (last accessed 20 April 2017). Forbes (2015). ‘The world’s biggest public companies’ http://​www.forbes.com/​global2000/​list/​ #tab:overall (last accessed 7 September 2015). Friedman, T. (2005). The World is Flat:  A  Brief History of the Twenty-​ first Century (New York: Farrar).

Reconceptualizing Resource Peripheries    729 Fujita, M. (1988). ‘A monopolistic competition model of spatial agglomeration: differentiated product approach’. Regional Science and Urban Economics 18: 87–​124. Fujita M., Krugman, P., and Venables, A.J. (1999). The Spatial Economy. Cities, Regions and International Trade (Cambridge, MA: MIT Press). Gereffi, G. (2005). ‘The global economy:  organization, governance, and development’. The Handbook of Economic Sociology 2: 160–​182. Hayter, R., Barnes T., and Bradshaw, M. (2003). ‘Relocating resource peripheries to the core of economic geography theorizing: rational and agenda’. Area 35: 15–​23. ICMM (2015). A Practical Guide to Catchment-​based Water Management for the Mining and Metals Industry (London: ICMM). IFC (2015). ‘IFC helps mining companies manage south Gobi water challenges’ http:// ​ w ww.ifc.org/​ w ps/​ w cm/​ c onnect/​ d 14038804a44668984b4bf10cc70d6a1/​ South+Gobi+Water+and+Mining_​One+Pager_​ENG_​FIN.pdf?MOD=AJPERES (last accessed 28 July 2016). Innis, H. (1930). The Fur Trade in Canada (Toronto: Toronto University Press). Innis, H. (1933). The Problem of Staple Production in Canada (Toronto: Ryerson Press). Innis, H.A. and Drache, D. (1995). Staples, Markets, and Cultural Change:  Selected Essays (Montreal: McGill-​Queen’s Press-​MQUP). Jackson, S.L. (2015). ‘Imagining the mineral nation: contested nation-​building in Mongolia’. Nationalities Papers 43: 437–​456. Krugman, P. (1981). ‘Trade, accumulation, and uneven development’. Journal of Development Economics 8: 149–​161. Krugman, P. (2011). ‘The new economic geography, now middle-​aged’. Regional Studies 45: 1–​7. Kuznets, S. (1940). ‘Schumpeter’s business cycles’. American Economic Review 30: 262–​263. Le Billon, P. (ed.) (2008). The Geopolitics of Resource Wars: Resource Dependence, Governance and Violence (London: Routledge). Le Billon, P. (2012). Wars of Plunder:  Conflicts, Profits and the Politics of Resources (London: Hurst & Co). Li, F. (2016). ‘The defeat of Pascua Lama’ https://​nacla.org/​news/​2016/​03/​09/​defeat-​pascua-​ lama (last accessed 9 March 2016). Markusen A. (1996). ‘Sticky places in slippery space a typology of industrial districts’. Economic Geography 72: 293–​313. Mohan, G. and Lampert, B. (2013). ‘Negotiating china: reinserting African agency into China–​ Africa relations’. African Affairs 112: 92–​110. Nash, J.C. (1993). We Eat the Mines and the Mines Eat Us: Dependency and Exploitation in Bolivian Tin Mines (New York: Columbia University Press). Ottaviano, G. and Thisse, J.F. (2004). ‘Agglomeration and economic geography’. Handbook of Regional and Urban Economics 4: 2563–​2608. Overholt, W. (2010). ‘China in the global financial crisis: rising influence, rising challenges’. The Washington Quarterly 33: 21–​34. Owen, J.R. and Kemp, D. (2013). ‘Social licence and mining: a critical perspective’. Resources Policy 38: 29–​35. Petersen, S., Krätschell, A., and Hannington, M.D. (2016). ‘The current state of global activities related to deep-​sea mineral exploration and mining’. EAGE/​DGG Workshop on Deep Mineral Exploratio, March. Potter, R. (2001). ‘Geography and development: “core and periphery?” ’ Area 33: 422–​427.

730   McElroy Prebish, R. (1950). The Economic Development of Latin America and its Principal Problems (New York: United Nations). Prno, J. (2013). ‘An analysis of factors leading to the establishment of a social licence to operate in the mining industry’. Resources Policy 38: 577–​590. Radetzki, M. (2013). ‘The perseverance of the ongoing metal and mineral boom’. Mineral Economics 25: 83–​88. Restino, C. (2014). ‘Rio Tinto donates pebble sales’, The Bristol Bay Times, 11 April http://​www. thebristolbaytimes.com/​article/​1415rio_​tinto_​donates_​pebble_​shares (last accessed 20 April 2017). Rosales, J. (2008). ‘Economic growth, climate change, biodiversity loss: distributive justice for the global north and south’. Conservation Biology 22: 1409–​1417. Ross, M. (1999). ‘The political economy of the resource curse’. World Politics 51: 297–​322. Rosser, A. (2006). The Political Economy of the Resource Curse:  A  Literature Survey (Brighton: Institute of Development Studies). Sachs, J.D. and Warner, A.M. (1995). ‘Natural resource abundance and economic growth’. National Bureau of Economic Research working paper, revised 1997, 1999. Sauer, C.O. (1941). ‘Foreward to historical geography’. Annals of the Association of American Geographers 31: 1–​24. Schumpeter, J.A. (1939). Business Cycles (Volumes 1 and 2) (New York: McGraw Hill). Scott, A.J. (1992). ‘The Roepke Lecture in economic geography the collective order of flexible production agglomerations: lessons for local economic development policy and strategic choice’. Economic Geography 68: 219–​233. Scott, A. (2000). ‘Economic geography: the great half-​century’. Cambridge Journal of Economics 24: 483–​504. Silvennoinen, A. and Thorp, S. (2013). ‘Financialization, crisis and commodity correlation dynamics’. Journal of International Financial Markets, Institutions and Money 24: 42–​65. Singer, H.W. (1950). ‘The distribution of gains between investing and borrowing countries’. American Economic Review 40: 473–​485. Singer, H.W. (1998). ‘Beyond terms of trade: convergence/​divergence and creative/​uncreative destruction’. Zagreb International Review of Economics and Business 1: 13–​25. Smith, N. (1984). Uneven Development: Nature, Capital, and the Production of Space (Athens, GA: University of Georgia Press). Soyez, D. (2002). ‘Environmental Knowledge, The Power of Framing and Industrial Change’ in Hayter, R. and Le Heron, R.B. (eds) Knowledge, Industry and Environment: Institutions and Innovations in Territorial Perspective, pp. 187–​298 (London: Ashgate). Statista (2015). ‘Iron ore prices from May 2014 to May 2015 (in U.S. dollars per dry metric ton unit)’ http://​www.statista.com/​statistics/​300419/​monthly-​iron-​ore-​prices/​ (last accessed 20 April 2017). Sternberg, T. (2013). ‘Hazards and human-​environment systems in the Gobi Desert, Asia’. Journal of Geography & Natural Disasters 3: 106. Storey, K. (2001). ‘Fly-​in/​fly-​out and fly-​over: mining and regional development in Western Australia’. Australian Geographer 32: 133–​148. Urkidi, L. (2010). ‘A global environmental movement against gold mining: Pascua–​Lama in Chile’. Ecological Economics 70: 219–​227. Wallerstein, I. (1979). The Capitalist World-​economy (Volume 2)  (Cambridge:  Cambridge University Press).

Reconceptualizing Resource Peripheries    731 Watts, M.J. (2004). ‘Resource curse? Governmentality, oil and power in the Niger Delta, Nigeria’. Geopolitics 9: 50–​80. Weszkalnys, G. (2008). ‘Hope and oil: expectations in São Tomé e Príncipe’. Review of African Political Economy 35: 473–​482. Wick, K. and Bulte, E.H. (2006). ‘Contesting resources–​rent seeking, conflict and the natural resource curse’. Public Choice 128: 457–​476. World Bank (2015). ‘Total natural resource rents % GDP’ http://​data.worldbank.org/​indicator/​ NY.GDP.TOTL.RT.ZS (last accessed 20 April 2017). Wright, G. and Czelusta, J. (2004). ‘Why economies slow: the myth of the resource curse’. Challenge 47(2): 6–​38.

Chapter 39

Ou tside Re g i ona l Paths:  C onst ru c t i ng an Ec onom ic G e o g ra ph y of Energy Tr a nsi t i ons Susan Christopherson † Introduction: The Intersection of Evolutionary and Energy Geographies Perhaps the most influential framework for understanding change over time in energy economies (production, transportation, markets, and use patterns) builds on the concept of socio-​technical transitions and, within that, of sustainability transitions (Farla et al., 2012; Garud and Gehman, 2012; Markard et al., 2012; Raven et al., 2012). In this conceptual framework, the processes behind the transformation of patterns of energy use are social and technical:  ‘social’ in that they depend on change in individual behaviour and the adoption of new social norms regarding energy use, and ‘technical’ in requiring the adoption of new technologies for energy production and distribution. Potentially disruptive technologies coexist as niche technologies along with dominant technologies. By this way of thinking, wind power and solar power exist as niche technologies in a fossil fuel-​dominated global energy system. A socio-​technical transition will occur when the niche technology becomes dominant because of changes in behaviour and norms, and possibly as a consequence of government policy to support the adoption of the niche technology (Foxon, 2011). Energy geographers have approached the question of change over time in energy systems from a different perspective because of their concern for the role of space and place in the way energy transitions are formulated and ‘roll out’ over time (Bridge, 2008; Bradshaw, 2010, 2014; Coenen et al., 2012; Essletzbichler, 2012; Bridge et al., 2013; Patchell and Hayter, 2013). In this respect, a critical question facing energy geographers—​the processes behind change in energy economies—​is close to that drawing the attention of evolutionary † Deceased

Outside Regional Paths   733 economic geographers, that of path dependence. Path dependence refers to the ways in which previous decisions shape or constrain contemporary decisions. Economic geographers have made substantial contributions to the conceptual frameworks explaining change in economic systems by analysing its spatial dimensions (Boschma and Frenken, 2007; Boschma and Martin, 2010). One contribution emerged from the recognition of the role of ‘sunk costs’, and how their spatial logic is reflected in corporate strategies (Clark and Wrigley, 1995). Another contribution has come from theoretical and empirical work demonstrating that path dependence is place-​dependent, tied to place-​based knowledge, experience, and socio-​political relations (Clark et al., 2001; Martin and Sunley, 2006; Martin, 2012). The concept of ‘energy transitions’ is also concerned with how previous decisions influence future alternatives, describing path-​dependent influences on the process of change to new energy systems, and persistence in the use of existing energy systems. An understanding of economic and energy change as rooted in decisions made in place and space challenges the socio-​technical innovation or adaptive efficiency approaches that posit a placeless world of punctuated equilibrium and timeless logical choices between more efficient and less efficient alternatives.1 Similar to evolutionary economic geography, the field of energy geography faces theoretical problems in formulating sources of change and persistence (MacKinnon et al., 2009; Martin, 2010). Their challenge echoes that set out by Martin and Sunley (2006, p. 397):  . . . the task is not just about applying evolutionary thinking and concepts to economic geography, difficult enough though that challenge is: it is also about exploring and explicating how geography—​the role of place and space—​influences the process of economic evolution itself, and thereby how economic geography can make a contribution to the development of evolutionary economic thinking and the concepts it employs.

This chapter examines some important features of the economic geography of fossil fuel production and distribution, utilizing examples from US shale oil and gas development, the product of a recent technological innovation: high-​volume hydraulic fracturing of shale formations. The goal is to examine the role of space and time in processes of change, a focus common to both evolutionary economic geography and energy geographies (Calvert, 2016), and how the perspectives of those fields can inform one another.

How Geography Matters A close look at their common critical questions, focused on processes in space and time, indicates several potential contributions to Martin and Sunley’s (2006) project of understanding how geography enters into processes of economic evolution. Firstly, an analysis of energy resource transitions demonstrates that change may need to occur at a different scale than that at which a path-​breaking technological innovation is implemented. Secondly, it suggests why actors within regions (in this case, extraction regions) may have little impact on the processes of change if those processes are rooted in firm strategies and governance regimes emanating from outside the region. Thirdly, the persistent importance of the nation

734   Christopherson state to the profit strategies of fossil fuel energy firms is a reminder that different spatial scales continue to ‘matter’ in economic geography, especially in analysing the processes of path dependence (Dicken, 2015; Kok, 2016). Finally, it demonstrates how the material properties of a resource commodity (in this case, its relatively undifferentiated nature at extraction, and the fact that it can be obtained only in certain locations) influence the geography of firm market and non-​market strategies with respect to capital accumulation (Levy and Kolk, 2002; Bridge, 2009). For the field of energy transitions, this examination of how nation state governance constructs paths to investment returns, or eliminates the risks that are intrinsic to a highly speculative industry, illuminates sources of change and of continuity (including sources of ‘lock-​in’) that lie outside regional paths. Specifically, it illustrates that a focus on either the extraction location or the global production network alone is insufficient, and may lead us astray in our attempts to understand persistence and change in energy systems. We also need to understand the role of, and locus of, the governance infrastructure required for energy resource development.

A Particular Economic Geography Shale development is expressed in a particular economic geography, and, superficially, this geography appears to be the product of a technological change that has opened new spaces of potential extraction—​new shale ‘plays’. However, underlying the emergent geography of shale gas and oil production is a more enduring economic geography. It is rooted in the sunk costs of a powerful global firm network—​in political ties and technical knowledge, in mineral reserve investments, and in pipelines, rail lines, refineries, and other infrastructure—​ that enables oil and gas firms to achieve super profits in a risky, capital-​intensive industry. Within this persistent geography, the territorial ‘place’ most important to oil and gas firm profitability is not the economic region, but the nation state.2 The pre-​eminent and persistent role of national governance is based on the fixity of the oil and gas resource, its economic and political value to national governments, and the extensive physical infrastructure needed to move oil and gas to markets. Although relationships between oil exploration and production firms and nation states have changed over time, and a global production network led by oil-​ field service firms has emerged, national interests and capacities remain central features of the global fossil fuel energy system. It is at this scale that we see ‘lock-​in’: a global fossil fuel economy maintained through both national political alliances and continual active intervention by a network of global firms that foster and cement those alliances in the interest of maintaining their profit position. Although the national scale of governance has been downplayed in the recent literature on the rise of the region in the global economy (Farole et al., 2011; Storper, 2011) and that on global production chains (Coe and Yeung, 2015), it remains critical to understanding resource industries, and particularly oil and gas industry production and distribution (Auty, 2001; Bridge, 2014; Dicken, 2015). In the contemporary world, where oil and gas extraction is risky, difficult, and expensive, firms are ever-​more dependent on what Dicken has described as two major functions of the nation state—​infrastructure provision and favourable regulatory policy:

Outside Regional Paths   735 TNCs [transnational corporations] need states to provide the infrastructural basis for their continual existence: both physical infrastructure in the form of the built environment and also social infrastructure in the form of legal protection of private property. . . . . (2015, p. 223).

The importance of national policy and investment is both direct (e.g. regulation of pipeline construction or energy tax policy) and indirect (e.g. air-​quality regulations or trade policies that may affect energy commodities). In the case presented in this chapter—​that of the so-​ called ‘path-​breaking’ shale development in the USA—​policies governing national financial markets have also been crucial in making a very expensive, risky new technology economically feasible, at least for a time. In the exploration and extraction phase, oil and gas companies need to locate where the resource resides and then deal with the regulatory policy and infrastructure in that location. But, as will be demonstrated in what follows, the resource is not typically contained within an already established sub-​national jurisdiction. If the geographical boundaries of a resource ‘play’ do not coincide with sub-​national legal jurisdictions, oil and gas firms must move up scale to the nation state to obtain the infrastructure, regulatory permissions, and social licence they need to exploit the resource, get it to global markets, and realize its potential profitability. Indeed, with the exception of the headquarters ‘agglomerations’ in Harris County, Texas, USA or Aberdeen, Scotland (MacKinnon et al., 2004), the economic region as a geographical ‘place’ is a secondary consideration for the oil and gas industry. Oil and gas company decisions, even those of ‘operating’ companies in the extraction area, are determined by governance and corporate interventions from outside the extraction region. It is striking that the oil and gas production and distribution industry remains rooted in national territories, even in an era of global networks (Christopherson and Clark 2007; de Graaff, 2011, 2012; Coe and Yeung, 2015). By contrast with conventional notions of the transnational corporation as strategically ‘footloose’, oil and gas industry firms remain tied to national spaces because of the political and economic resources instituted by those ties. This is particularly true of nationalized oil and gas resources, such as those in Venezuela, China, and, to some extent, The Russian Federation. But national interdependence is also characteristic of major private oil and gas companies, such as Exxon or BP, because of their need for the political, economic, and regulatory resources of the state (de Graaff, 2011, 2012; Bradshaw, 2014; Coll, 2012; Dicken, 2015). These resources provide the global oil and gas company not only with the opportunity to extract and profit from the resource itself, but also with financial tools and fiscal and regulatory policies that enable it to weather the boom–​ bust cycles inherent to resource industries. The ability of US oil and gas firms to defer tax payments over long periods, taking tax write-​offs during periods of low profitability, is only one example of the industry’s ability to use national policy to address risk and volatility. So, in the case of shale development in the US, national-​scale financial market governance and regulatory conditions were critical to the speed of shale development, as discussed in the following sections. In the next section, the processes operating on the ‘extraction play’ and its regional context are counter-​posed with taken-​for-​granted ideas about inward investment and regional economic development. This difference is best illustrated by reference to the concept of the ‘resource curse’ as it is manifested in shale gas and oil development regions. Rather than increasing capacity for economic development, the integration of shale investment spaces

736   Christopherson into the global oil commodity production chain depletes regional economic development capacity. Finally, the case of shale development in the US is used to illuminate some key ways in which national governance created a framework for change in the fossil fuel regime, enabling US operating companies to move rapidly to exploit this risky and expensive fossil fuel.

Differentiating the Subsurface ‘Play’ from the Surface ‘Region’ The development of fossil fuel energy resources is viewed by key actors, especially the oil and gas industry, in spatial frames that have little relationship to economic regions above ground. The development of shale oil and gas in the USA, for example, is discussed with reference to two geographies. The primary geography is the shale ‘play’. The ‘play’ refers to a subterranean geological formation within which shale gas and/​or oil are expected to be found, and, by extension, the surface territory above it that has been targeted for investment. The term ‘play’ also implies the speculative nature of exploration and drilling (Energy Information Administration, 2015). On shale maps of the USA, shale plays are large ink-​blot territories, sometimes with graduated colouring to indicate relative investment potential and profitability (Energy Information Administration, 2015). The most potentially profitable sites are described as ‘sweet spots’ or ‘fairways’. Surface features, including cities, nature reserves, watersheds, and historically important landscapes are invisible, as are potentially conflicting uses of the land under which the shale resources lie (Elden, 2013). For example, Rahm (2011) describes the territory of shale-​gas development in the USA in terms of shale ‘plays’—​subsurface areas within which shale development may be carried out profitably using the available technology—​and simply identifies the states within which the plays are located; for example, Texas has five shale plays. To the oil and gas industry, ‘the play’—​a concept evoking both geophysical characteristics and profit potential—​is the only relevant geography, an area (both horizontal and vertical) in which investment takes place. Disregard for surface features of the extraction landscape is illustrated by some models of potential shale development that plot wells per acre uniformly across the region, without regard to surface features, housing, public facilities, or already developed industries there (Weinstein and Clower, 2009; Christopherson and Rightor, 2012). Surface population concentrations or the existing economic activity of regions within shale plays are secondary, if not extraneous, to perceptions of ‘the play’. Instead, the ‘local’ scale within a play is that of the extraction site, the well, and the profit potential of a shale play is measured in optimum number of wells per acre and their potential productivity, with little regard for what exists on the surface landscape, whether cities or wilderness. Because the spatial frame for shale gas and oil extraction is not the surface economic region but the subsurface ‘play’, consideration of regional externalities are typically disconnected from the extraction process. These include impacts on housing prices, retail services, traffic congestion, other industries (particularly tourism and agriculture), and crime rates (Christopherson and Rightor, 2015).

Outside Regional Paths   737 The second geography relevant to development of this energy resource is that of whatever political jurisdiction governs extraction and distribution. In the UK and throughout much of the world, this is exclusively the nation state. In the USA, it is a combination of the national government and sub-​national states, which regulate different aspects of the energy development process. For example, permission to extract is regulated at the state level, except on federal lands. However, the infrastructure required to transport oil or gas—​ by pipeline, rail, barge, tanker, or truck—​is governed at the national level. National regulation also governs critical environmental arenas, such as groundwater, surface water, and air quality, that may be affected by the extraction process. Notably, local or regional jurisdictions have been excluded from this governance regime in the USA. This has been based on a rationale that shale development occurs across local and regional jurisdictional boundaries (Christopherson and Rightor, 2015; Finkel, 2015), and allowing local or regional regulation may impede the benefit to (and profit from) national and global markets lying outside the extraction region. By contrast, the public debate over shale development in the USA has been focused almost exclusively on the extraction site and adjacent community, expressing both the fears of environmental hazard and the promise of economic benefit. Yet, the economic region within which the extraction site lies is a geography that is only peripherally relevant to industry decision-​making or policymaker concern. Meanwhile, the national scale has been under-​examined in accounts of the shale revolution, despite its central role in facilitating the industry’s supply chain or moving its commodity to market (Christopherson, 2015).

The Play as a Complex Investment Space The concept of the shale extraction area as an investment ‘play’ affects the time frame and spaces of the resource development process, and how oil and gas industry firms think about creating a profit strategy (Kok, 2016). For example, because by definition shale plays are arenas for speculative investment, shale operators and their investors may purchase significant amounts of land on the surface of the play. These land holdings, as well as leased subsurface mineral rights, provide an additional form of investment to complement actual production of oil and gas, and ‘flipping’ real estate and mineral rights holdings—​selling them for more than the purchase price—​has become an accepted dimension of shale oil and gas investment. Indeed, the ability to purchase and trade in land and mineral rights for subsurface assets was central to many firms’ profitability strategies during the initial phase of US shale development, between approximately 2009 and 2013. Such non-​extraction-​based investment strategies may be time constrained, and produce unexpected consequences. For example, because mineral rights leases usually specify that the lease becomes void unless drilling and extraction take place within a specified time period, maintaining control of the value of those mineral rights requires shale operators to pursue a ‘drill-​and-​hold’ strategy. Even the development of a single well on each leased parcel, intended to secure those contracts, can produce a drilling boom big enough to glut the market and suppress the price of gas or oil. This dimension of development has become more visible in the US shale ‘revolution’ since 2013:

738   Christopherson Firms that made billions in the early stage of the boom by buying exploration rights for gas fields and then flipping them are increasingly forced to move into the more costly business of extracting oil, natural gas and gas liquids like propane and butane, after increased supply depressed prices for dry gas (Carey, 2014).

The first phase of development, that of reaping returns from the potential of reserves, is one in which a minimal government role can enhance the speed and volume of private returns on investment. Once those investment strategies run their course and extraction becomes a reality, however, national and regional government support become essential to investment returns. Access to surface storage and transportation infrastructure—​gathering lines, pipelines, pumping or compressor stations, roads, and rail links—​become critical to supplying the drilling operations and moving the commodity to markets where it can achieve sufficient profits to justify the expensive, capital-​intensive, and speculative investments at the well site. Thus, favourable national regulation—​of financial markets, enabling rapid returns from investment in reserves, and in infrastructure, to move the commodity to markets or hold it awaiting higher prices—​creates the context within which oil and gas firms can achieve profits, even in this high-risk enterprise. While the surface economic region is secondary to the space in which economic investment in fossil fuels is conducted, this does not mean the surface region is irrelevant to oil and gas firms. What ‘secondary’ does mean is that the time frame and spatial frame for investment based on a resource below the surface are fundamentally disconnected from regional development on the surface. The presence of an urban economy, or an agricultural economy, or tourism-​related activities above the shale ‘play’ is insignificant to the investment decision. This is the source of the resource curse that has historically plagued resource extraction regions.

The ‘Resource Curse’ Revisited The disconnect between how oil and gas firms decide where to invest (which lies in speculative investment and in the need to transport the subsurface resource to global markets), and the economic fate of extraction regions, is reflected in a literature that describes the extraction region as a ‘sacrifice’ zone or a ‘victim’ of extraction, and in the literature on the ‘resource curse’ (Humphreys et al., 2007; Bridge, 2011). As Bridge (2008) has noted, blame for negative economic outcomes in extraction regions is frequently placed on governance failures, rather than linked to the inherent unpredictability and lack of sustained regional investment that is intrinsic to oil and gas extraction. In the usual resource curse narrative, the way the oil and gas firms derive profits from places is downplayed in favour of a focus on the accountability of government officials. In the geography of global oil and gas production and distribution, extraction regions are occupied temporarily; they are not the focus of long-​term investment based on their economic assets (other than those below ground). Research on the US Marcellus shale play indicates that the areas where extraction has occurred experience loss of population, and end up with less diverse economies (Christopherson and Rightor, 2014; Pennsylvania State Data Center, 2015). The volatility of extraction activity undermines opportunities for other,

Outside Regional Paths   739 longer-​term investments in the surface regions above the ‘plays’, and narrows rather than expands their opportunities for regional economic development. Whenever the resource is depleted, or if prices decline and the commodity becomes too expensive to extract profitably, drilling firms rapidly exit the region. While contemporary analysis of the resource curse does recognize the role of governance in the failure of extraction regions to benefit from the presence of the commodity, that failure is portrayed almost exclusively as a consequence of government corruption or ineptitude. Questions of government regulatory capacity or variations in their territorial control are missing from the conventional depictions of the resource curse; the implications of financial speculation in natural resource assets, or the importance of firm–​nation interdependence in profiting from natural resource assets, are rarely part of the resource curse discussion. In the final section, I examine how state–​firm interdependence is central to firms whose past, and chosen future, lie in the continued primacy of fossil fuels as an energy source. The story of the early entry, boom cycle, and current ‘bust’ in US shale development illustrates the continuing significance of national territorial governance—​in this case, to support higher-​risk, smaller-​scale oil and gas production.

Why was the USA an Early Adopter of Shale Development? By many accounts, the US is an exceptional context for the production of gas and oil from shale formations. According to Sandrea (2014, p. 2), the emergence of the shale oil and gas boom in the USA is attributable not only to global price increases, but also to a set of conditions that are facilitated by US financial market governance and regulation of the oil and gas industry. They include: (i) the availability of capital and credit in the US through the high-​ yield or ‘junk’ bond market; (ii) private ownership of mineral rights; (iii) enabling regulations governing the leasing of those rights and the oil and gas production process; (iv) the easy availability and inexpensiveness of critical inputs, particularly freshwater; (v) hydraulic fracturing rig availability; and (vi) authorization to drill a large number of wells quickly, and to use the experimental technology of high-​volume hydraulic fracturing. Most of these conditions, which explain why shale development was able to take place so rapidly in the USA, point to the continued importance of the state to the oil and gas industry. Many US states where high-​volume hydraulic fracturing was approved very quickly (exemplified by Texas, Louisiana, and Pennsylvania) have limited regulatory requirements to assess its environmental implications, or to test the technology before applying it at scale. The approach taken in the USA was one of ‘we will learn what works and doesn’t work as we drill’. Private ownership of mineral rights made acquisition of those rights fast and easy, and quelled opposition to the new technology. Private ownership of mineral rights also made it possible to invest in shale reserve ‘futures’ and to achieve investment returns by trading in those futures. Finally, the presence of operating companies such as Chesapeake Energy, and oil-​field service companies such as Halliburton, which were well versed in investment plays and well

740   Christopherson equipped by the drilling of conventional wells that employ hydraulic fracturing, allowed rapid development of the new shale oil and gas sites. By contrast, other countries with shale reserves have been slow to exploit them. Capital is difficult to obtain to support high-​risk oil-​field company expansion, and regulations governing oil fields are strenuous. The high-​volume hydraulic fracturing technology is banned in France and Germany because of strong popular opposition. Even the exceptional circumstances that would allow hydraulic fracturing in Germany are covered by regulatory protections that make utilization of the technology prohibitively expensive. In the UK, high-​volume hydraulic fracturing is legal, but it is subject to extensive testing and environmental impact analysis that slows the process of well construction. Shale play development in the USA has demonstrated that, to be profitable, a shale play requires multiple wells drilled in a short period, so time delays for well construction constitute a major barrier to shale development. In addition, mineral rights in the UK are owned by the nation. While this would appear to be an advantage to extraction companies because they have only one entity with which to bargain, state-​controlled mineral rights solidify opposition in those communities where drilling would take place. Communities have little to gain from local drilling, and potentially much to lose. And, a major profit strategy that drove rapid shale development in the USA is missing in the UK and other countries: the opportunity to ‘flip’ property holdings on the premise that the mineral rights underlying the property will produce a high ​investment return for the land purchaser. UK development relies solely on extraction for profit, and that is a riskier proposition than pre-​extraction land speculation. In addition to enabling speculative investment in real estate through private property markets, US governance of financial markets played another important role in the US shale boom. The easy availability of capital for shale drilling and oil-​field service companies through the high-​yield or ‘junk’ bond market enabled rapid expansion of these companies at the onset of the boom (Denning, 2015). The USA has the world’s largest debt market (accounting for approximately half of the global market), and the most active market in high-​risk bonds (Smart and Gitman, 2015). These bonds are highly attractive to investors, including international investors through the ‘Yankee’ market for foreign bonds issued in US dollars. Although high risk, the value of most of these bonds is transparent because they are regulated by the US Securities and Exchange Commission (SEC), and are rated by American ratings agencies such as Moody’s Investors Service and Standard & Poor’s. However, because of an SEC exemption, a significant portion of the industry’s high-​risk debt issues are largely unregulated, easily obtainable, and not scrutinized by ratings agencies. The role of high-​risk debt in fuelling the shale boom has become more evident with the decline in oil and gas prices, and the consequent bankruptcy of highly indebted oil-​field service companies (Sider, 2015). Oil and gas company debt constitutes nearly a quarter of the US $1.2 trillion high-​yield and leveraged loan market, and is widely blamed for increasing the stress on that market (Fitzgerald, 2015; Linnane, 2015). Faced with the prospect of a crisis in the financial markets brought on by high-​risk debt related to shale development, the US government has stepped in to support the oil and gas industry, both directly and indirectly. Support has included purchase of oil for federal strategic reserves, and policies to prop up the high-​risk bond market. In the most recent example, to help alleviate the effects of declining oil prices in North America, US policy was changed in 2015 to allow oil exports to higher-​priced markets. The entire ‘Big Oil’ network promoted this change because, although they compete with each other, they have a shared interest in

Outside Regional Paths   741 such a state policy. In this way, the global production network led by oil-​field service firms such as Halliburton and Schlumberger intersects with national governance of oil and gas production and distribution. In the USA, it was market governance enabling speculative short-​term investment that made smaller-scale extraction and service operations viable, even when the large global production firms (e.g. Exxon, BP, and Shell) would not invest in risky exploration and production at a small scale. At the same time, the emergence of a sophisticated global network of oil-​field service firms gave these small ‘operating firms’ access to the high-​quality technologies and equipment they needed to make smaller reserve extraction a profitable proposition. And, because of the political power of the oil and gas industry nationally, there has been considerable protection against market risks for actors at the top of the global oil and gas production network.

About Energy Transition Processes In this chapter, I have attempted to take Martin and Sunley’s (2006) prescription for a more geographically informed understanding of economic change seriously. Firstly, I suggest that some geographical scales of action are more important than others. The significance of a particular scale depends on how an industry uses spatially distributed resources, and what it requires to exploit those resources. In the case of the oil and gas industry, conventional regional concentrations of resources are not as significant as control over spatially extensive resources. The rapid development of shale in the USA also illustrates the important role of national financial market governance, in this case by enabling access to flexible financing that can adapt to changing costs and risks. The highly variable progress of shale development across countries suggests that it is at the national scale that the most decisive interactions between the global oil and gas production network and regulatory institutions take shape. To obtain legitimacy, influence regulation over key resources such as transportation networks, and reduce financial risks for high-​risk ventures, firms in the network rely on non-​market strategies. In the US case, these non-​market strategies combine: (i) lobbying and campaign funding to effect political influence on the national regulatory environment; and (ii) the exercise of soft power to influence the national narrative about the role of energy in the economy and the trade-​offs between environmental protection and energy resource development. An analysis of the oil and gas industry as a variant of path dependent social, economic, and technological transitions can inform and qualify theories of path dependency in evolutionary economic geography in several respects. Most prominently, assumptions about production spaces and economic regions are up-​ended. In oil and gas production, the subterranean landscape of investment is disconnected from the surface spaces of already existing economic regions. And, because of their focus on what might be termed the extraction and distribution landscape, oil and gas firms focus on governance at the national scale. It is at that scale that firms, including self-​defined global firms, devise both market and non-​market strategies to influence how the state governs their industry, and hence, constructs transition paths. What this tells us is that industry specificity—​what is being produced and how it is being produced—​varies the geographical story of economic change.

742   Christopherson

Notes 1. The concept of energy transitions has taken on a normative dimension among researchers who analyse transitions towards sustainable renewable energy systems (Pasqualetti, 2011; Markard et al., 2012). 2. In oil and gas production, the role of national governance could be described in terms of a spatial ‘fix’ referring to the territorial basis for capital accumulation. A spatial fix can be expressed in long-​term territorial investments that underlie capital accumulation (to enable mobile capital at other geographical scales). This may include investments in transportation and communication or in a regional finance infrastructure. Spatial fix can also refer to temporary movements across geographic spaces to resolve accumulation crises (see Harvey, 2001; Jessop, 2006).

References Auty, R. (2001). Resource Abundance and Economic Development (Oxford: Oxford University Press). Boschma, R., and Frenken, K. (2007). ‘Applications of Evolutionary Economic Geography’ in K. Frenken (ed.) Applied Evolutionary Economics and Economic Geography, pp. 1–​25 (Cheltenham, and Northampton, MA: Edward Elgar). Boschma, R. and Martin, R. (eds). (2010). The Handbook of Evolutionary Economic Geography (Cheltenham, and Northampton, MA: Edward Elgar). Bradshaw, M.J. (2010). ‘Global energy dilemmas: a geographical perspective’. The Geographical Journal 176: 275–​290. Bradshaw, M.J. (2014). Global Energy Dilemmas: Energy Security, Globalization and Climate Change (New York: Wiley). Bridge, G. (2008). ‘Global production networks and the extractive sector: governing resource development’. Journal of Economic Geography 8: 389–​419. Bridge, G. (2009). ‘Material worlds: natural resources, resource geography and the material economy’. Compass 3: 1217–​1244. Bridge, G. (2011). ‘Past Peak Oil: Political Economy of Energy Crises’ in R. Peet, P. Robbins, and M. Watts (eds) Global Political Ecology, pp. 307–​324 (New York: Routledge). Bridge, G. (2014). ‘Resource geographies II:  the resource-​state nexus’. Progress in Human Geography 38(1): 118–​130. Bridge, G., Bouzarovski, S., Bradshaw, M., and Eyre, N. (2013). ‘Geographies of energy transition: space, place and the low carbon economy’. Energy Policy 53: 331–​340. Calvert, K. (2016). ‘From energy geography to energy geographies’. Progress in Human Geography 40(1): 105–​125. Carey, D. (2014). ‘Private equity shifts shale strategy as land grab ends’. Bloomberg News, 21 March http://​f uelfix.com/​blog/​2014/​03/​2 1/​private-​e quity-​shifts-​shale-​strategy-​as-​l andgrab-​ ends/​#9854101=0 (last accessed 14 December 2015). Christopherson, S. (2015). ‘Risks Beyond the Well Pad:  The Economic Footprint of Shale Gas Development in the US’ in M.L. Finkel (ed.) The Human and Environmental Impact of Fracking:  How Fracturing Shale for Gas Affects Us and Our World, pp. 115–​130 (Oxford: Praeger).

Outside Regional Paths   743 Christopherson, S. and Clark, J. (2007). Remaking Regional Economies: Power, Labor, and Firm Strategies in the Knowledge Economy (London: Routledge). Christopherson, S. and Rightor, N. (2012). ‘How shale gas extraction affects drilling localities: lessons for regional and city policy makers’. Journal of Town and City Management 2: 350–​368. Christopherson, S. and Rightor, N. (2014). ‘NIMBYs or concerned citizens: responding to shale oil and gas development’. Progressive Planning 198: 32–​35. Christopherson, S. and Rightor, N. (2015). ‘Confronting an Uncertain Future: How US Communities are Responding to Shale Gas and Oil Development’ in D.E. Albrecht (ed.) Our Energy Future: Socioeconomic Implications and Policy Options for Rural America, pp. 40–​60 (Oxford and New York: Routledge). Clark, G.L. and Wrigley, N. (1995). ‘Sunk costs:  a framework for economic geography’. Transactions of the Institute of British Geographers 20(2): 204–​223. Clark, G.L., Tracey, P., and Lawton Smith, H. (2001). ‘Agents, endowments, and path-​ dependence:  a model of multi-​ jurisdictionalregional development’. Geographische Zeitschrift 89: 166–​181. Coe, N.M. and Yeung, H. (2015). Global Production Networks (Oxford:  Oxford University Press). Coenen, L., Benneworth, P., and Truffer, B. (2012). ‘Toward a spatial perspective on sustainability transitions’. Research Policy 41(6): 968–​979. Coll, S. (2012). Private Empire: Exxon Mobil and American Power (New York: Penguin Press). de Graaff, N. (2011). ‘A global energy network? The expansion and integration of non-​triad national oil companies’. Global Networks 11(2): 262–​283. de Graaff, N. (2012). ‘Oil elite networks in a transforming global oil market’. International Journal of Comparative Sociology 53(4): 275–​297. Denning, L. (2015). ‘Beware energy’s junk debt army’. Bloomberg.com, 14 December http://​ www.bloomberg.com/​gadfly/​articles/​2015-​12-​14/​junk-​bond-​market-​dominated-​by-​tiny-​ commodities-​issuers-​gadfly (last accessed 15 January 2016). Dicken, P. (2015). Global Shift: Mapping the Changing Contours of the World Economy (7th edition) (New York: Guilford). Elden, S. (2013). ‘Secure the volume:  vertical geopolitics and the depth of power’. Political Geography 34: 35–​51. Energy Information Administration (2015, 17 April). ‘New maps highlight geologic characteristics of US tight oil, shale plays’ http://​www.eia.gov/​todayinenergy/​detail.cfm?id=20852 (last accessed 21 January 2016). Essletzbichler, J. (2012). ‘Renewable energy technology and path creation:  a multi-​scalar approach to energy transition in the UK’. European Planning Studies 20(5): 791–​816. Farla, J., Markard, J., Raven, R., and Coenen, L. (2012). ‘Sustainability transitions in the making: a closer look at actors, strategies and resources’. Technological Forecasting and Social Change 79(6): 991–​998. Farole T., Rodríguez-​Pose, A., and Storper, M. (2011). ‘Human geography and the institutions that underlie economic growth’. Progress in Human Geography 35(1): 58–​80. Finkel, M.L. (ed.) (2015). The Human and Environmental Impact of Fracking: How Fracturing Shale for Gas Affects Us and Our World (Oxford: Praeger). Fitzgerald, W. (2015). ‘How crisis in the energy sector could spark a repeat of the subprime bust’. Forbes.com, 22 April http://​www.forbes.com/​sites/​christopherhelman/​2015/​04/​22/​how-​ crisis-​in-​the-​energy-​sector-​could-​spark-​a-​repeat-​of-​the-​subprime-​bust/​#3e6224e1b7e7 (last accessed 15 March 2015).

744   Christopherson Foxon, T.J. (2011). ‘A coevolutionary framework for analysing a transition to a sustainable low carbon economy’. Ecological Economics 70(12): 2258–​2267. Garud, R. and Gehman, J. (2012). ‘Metatheoretical perspectives on sustainability journeys: evolutionary, relational and durational’. Research Policy 41(6): 980–​995. Harvey, D. (2001). Spaces of Capital:  Towards a Critical Geography (Edinburgh:  Edinburgh University Press; New York: Routledge). Humphreys, M., Sachs, J.D., and Stiglitz, J.E. eds (2007). Escaping the Resource Curse (New York: Columbia University Press). Jessop, B. (2006). ‘Spatial Fixes, Temporal Fixes, and Spatio-​temporal Fixes’ in N. Castree and D. Gregory (eds) David Harvey: A Critical Reader, pp. 142–​166 (Oxford: Blackwell). Kok, I. (2016). ‘Politics of transparency: contested spaces of corporate responsibility, science and regulation in shale gas projects of the UK and US’. Thesis Submitted for the Degree of Doctor of Philosophy, Oxford University. School of Geography and the Environment. Levy, D. and Kolk, A. (2002). ‘Strategic responses to global climate change: conflicting pressures on multinationals in the oil industry’. Business and Politics 4(3): 275–​300. Linnane, C. (2015). ‘Oil’s slump played havoc in US junk bond market in September’. Barron’s Market Watch, 2 October http://​www.marketwatch.com/​story/​oils-​slump-​played-​havoc-​ in-​us-​junk-​bond-​market-​in-​september-​2015-​10-​02 (last accessed 30 January 2016). MacKinnon, D., Chapman, K., and Cumbers, A. (2004). ‘Networking, trust and embeddedness among SMEs in the Aberdeen oil complex’. Entrepreneurship & Regional Development: An International Journal 16(2): 87–​106. MacKinnon, D., Cumbers, A., Pike, A., Birch, K., and McMaster, R. (2009). ‘Evolution in economic geography:  institutions, political economy, and adaptation’. Economic Geography 85(2): 129–​150. Markard, J., Raven, R., and Truffer, B. (2012). ‘Sustainability transitions: an emerging field of research and its prospects’. Research Policy 41(6): 955–​967. Martin, R. (2010). ‘Roepke Lecture in Economic Geography: rethinking regional path dependence: beyond lock-​in to evolution’. Economic Geography 86(1): 1–​27. Martin, R. (2012). ‘(Re)Placing path dependence:  a response to the debate’. International Journal of Urban and Regional Research 36(1): 179–​192. Martin, R. and Sunley, P. (2006). ‘Path dependence and regional economic evolution’. Journal of Economic Geography 6(4): 395–​437. Pasqualetti, M.J. (2011). ‘Social barriers to renewable energy landscapes’. Geographical Review 101(2): 201–​223. Patchell, J. and Hayter, R. (2013). ‘Environmental and evolutionary economic geography: time for EEG2?’. Geografiska Annaler 95(2): 111–​130. Pennsylvania State Data Center (2015, 26 March). ‘U.S. Census Bureau releases 2014 county population estimates: Cumberland is the fastest growing county since 2010’ http://​pasdc. hbg.psu.edu/​sdc/​pasdc_​files/​researchbriefs/​2014_​County_​Estimates_​RB.pdf (last accessed 30 January 2016). Rahm, D. (2011). ‘Regulating hydraulic fracturing in shale gas plays: the case of Texas’. Energy Policy 39(5): 2974–​2981. Raven, R., Schot, J., and Berkhout, F. (2012). ‘Space and scale in socio-​technical transitions’. Environmental Innovation and Societal Transitions 4: 63–​78. Sandrea, I. (2014). ‘US shale gas and tight oil industry performance: challenges and opportunities’ http://​www.oxfordenergy.org/​2014/​03/​us-​shale-​gas-​and-​tight-​oil-​industry-​performance-​ challenges-​and-​opportunities/​ (last accessed 8 February 2016).

Outside Regional Paths   745 Sider, A. (2015). ‘Fracking firms that drove oil boom struggle to survive’. Wall Street Journal, 23 September http://​www.wsj.com/​articles/​fracking-​firms-​that-​drove-​oil-​boom-​struggle-​to-​ survive-​1443053791 (last accessed 30 January 2016). Smart, S. and Gitman, L. (2015). Fundamentals of Investing (12th edition). Pearson Series in Finance http://​www.downage.org/​ebooks-​list/​fundamentals-​of-​investing-​12th-​edition-​pearson-​ series-​in-​finance_​2ue77.html (last accessed 30 January 2016). Storper, M. (2011). ‘Why do regions develop and change?: the challenge for geography and economics’. Journal of Economic Geography 11(2): 333–​346. Weinstein, B.L. and Clower, T. L. (2009). ‘Potential economic and fiscal impacts from natural gas production in Broome County, New York’. Report prepared for Broome County, NY http://​ www.gobroomecounty.com/​countyexec/​broome-​county-​releases-​natural-​gas-​economic-​ impact-​study (last accessed 14 July 2010).

Pa rt  V I I I

ST R AT E G I E S F OR  DE V E L OP M E N T

Chapter 40

Green Grow t h Cameron Hepburn, Alexander Pfeiffer, and Alexander Teytelboym Introduction If economic growth is to continue for another century or more, it will, in some sense, have to be ‘green’ to avoid undermining its very foundations. Many international organizations, including the World Bank (World Bank, 2012), Organisation for Economic Co-​operation and Development (OECD, 2011), United Nations Environment Programme (UNEP, 2011), Asian Development Bank (ADB, 2013), European Bank for Reconstruction and Development (EBRD, 2011), and the African Development Bank (AfDB, 2013), have recognized this in a multitude of reports. Several initiatives, including the Global Green Growth Institute and Global Commission on the Economy and Climate, are devoting themselves entirely to understanding green growth. Green growth is a cousin of the even more vague and once more popular term ‘sustainable development’ (Hepburn and Bowen, 2013; Bowen and Hepburn, 2015; Smulders et al., 2014). The notion of ‘sustainability’ extends well beyond environmental issues to a variety of long-​term concerns, and the notion of ‘development’ includes health, education, culture, and, potentially, various other aspects of social and economic policy. By contrast, the term ‘green’ tends to reflect a more focused emphasis on the environment and natural capital and ‘growth’ is considerably narrower than ‘development’, in most cases it is boiled down to the convenient and politically appealing short-​term metric of the rate of increase in gross domestic product (GDP). Thus ‘green growth’ is a more narrow and focused concept than ‘sustainable development’. As broader issues of inequality, stability, culture, and institutions are not explicitly captured by the notion of ‘green growth’, these tend to be brought under the umbrella by the various international institutions noted above by extended phrases such as ‘inclusive green growth’. While the precise definition is contested, this chapter follows Bowen and Hepburn (2015) in defining ‘green growth’ as old-​fashioned growth in per capita GDP that is accompanied by natural capital levels that are non-​decreasing. We distinguish two variants. ‘Strong green growth’ involves no trade-​off between growth and natural capital protection, even in the short term. Under ‘weak green growth’, natural capital protection may reduce growth rates

750   Hepburn et al. below business-​as-​usual growth in the short term, but growth will be higher in the long term. The strong variant of ‘green growth’ particularly pleases politicians and policymakers; the absence of a short-​run trade-​off between environmental protection and economic growth enables proponents to claim to have it all. For instance, Hallegate et al. (2012) argue that ‘green growth’ does not even imply any slowing of economic growth compared with business-​as-​usual in the short term because of immediate positive effects on the economy, such as co-​benefits (e.g. reduced local pollution), growth in new ‘green’ sectors, and less energy price volatility via reduced dependence on fossil fuel imports. Naturally, the appeal to politicians is strong, as exemplified by Ed Davey, the former British Secretary of State for Energy and Climate Change who, at the Green Growth Summit in October 2014, remarked: ‘We are not sacrificing our economies to deal with climate change. Quite the opposite—​going green means going for growth’.1 In contrast, many conventional economic analyses appear to consider ‘weak green growth’ to be more plausible. The Stern Review (Stern, 2007) argued that we should incur relatively minor economic costs over the current generation to reduce greenhouse gas emissions and mitigate climate change, in order for future generations to accrue much greater benefits in the future. In short, although Stern made the case that the long-​term interests ought to dominate, there is, nevertheless, a trade-​off between the short-​term and the long-​term. There are two positions against some form of green growth. In our view, neither is plausible. The first is that there are no serious environmental constraints, and that we can largely ignore the impacts on natural capital and continue to grow into the indefinite future. While our analysis of the data, presented in the subsection ‘Limited Mineral and Energy Resources?’, suggests that some concerns of environmentalists are not well founded, it now also appears that we are approaching ‘planetary boundaries’ on a number of dimensions that will derail economic growth if left unaddressed (Rockström et al., 2009). In short, a ‘dirty growth’ position is not plausible in the long run—​the challenges of protecting natural capital cannot merely be ignored. The second competing position comes from the opposite end of the spectrum—​it is argued that these environmental pressures are so severe that we need to stop GDP growth altogether, and instead focus on ‘prosperity without growth’ (Jackson, 2011). There are three reasons to be sceptical of this position. Firstly, a billion people remain in extreme poverty. Growth in per capita GDP at a reasonable rate over coming decades is necessary, desirable, and likely in many developing countries to lift people out of poverty. Secondly, the technological progress required to solve many environmental problems, not least climate change, will require innovation and investment. This innovation and investment becomes practically difficult, if not impossible, in an economy with zero growth (Hepburn and Bowen, 2013). Thirdly, protecting natural capital by reducing GDP is probably the most expensive solution possible.2 In conclusion, the no-​growth position is neither necessary nor desirable. Consequently, this chapter advances the position that green growth—​continued growth in GDP alongside the protection of natural capital—​is almost trivially necessary. The real challenge, as we see it, is to flesh out precisely what this means and what policies might work in practice. This chapter is structured as follows. The next section briefly examines the evidence for the environmental constraints and so-​called ‘planetary boundaries’ that undermine a dirty growth trajectory in the long run and motivated the focus on green growth. The section

Green Growth   751 ‘Economic Theory of Green Growth’ reviews the economic theory of green growth, the different conceptualizations of green growth, the feasibility and optimality of theoretical growth trajectories, the role of innovation and capital substitutability, and the relationship to concepts such as weightless growth. ‘Policies for Green Growth’ considers the set of policies that might support green growth, and argues that any green growth strategy must recognize a host of political and social constraints. The chapter ends with a conclusion.

Natural Capital and the Real Environmental Constraints Defining Natural Capital Given that we have defined ‘green growth’ as GDP growth that preserves or enhances natural capital, it is important to be able to define ‘natural capital’. Natural capital is one of the six capitals that comprise our stock of wealth (Hamilton and Hepburn, 2014), but it is not entirely straightforward to provide a tight and uncontested definition. Conceptually, natural capital is the stock (not the flow) of resources in the natural environment that are able to produce value for humans. For instance, the stock of natural capital includes our reserves and resources of coal, oil, gas, minerals, and other subsoil assets, along with functioning ecosystems, fisheries, forests, biodiversity, and a stable climate. Precisely what is included, and what is not, is not the focus of this chapter, but a good place to begin is with the international System of Environmental-​Economic Accounting, developed jointly by organizations including the United Nations and the OECD, and as discussed and employed by the World Bank in their work on national accounts (Hamilton, 2006; World Bank, 2011). Services from these assets are provided by nature for free. Without human intervention, for instance in the form of private property rights, markets, or taxes, the flows of services from the stock of natural capital tend to be excessively consumed (Pigou, 1920; Coase, 1960). Humans have intervened with some forms of natural capital—​such as oil, coal, gas, and minerals—​and created property rights to the land that contains these assets, or, indeed, to the assets themselves, from which functioning markets and resource prices have emerged. However, other sorts of natural capital—​such as biodiversity and a stable climate—​tend not to have appropriate rights, markets, and prices. As we shall see, it is the latter that are the cause for concern.

Limited Mineral and Energy Resources? Many forms of natural capital, such as fossil fuels, are best described as ‘exhaustible’ or ‘non-​renewable’ because they do not renew themselves on human timescales.3 As these exhaustible resources can appear to be essential inputs in many production processes, it is not unreasonable to ask what will happen when we run out of them. These concerns about resource limits and scarcity fuelled publications such as the Limits to Growth (Meadows et al., 1972). It also motivated the now-​famous bet between Ehrlich and Simon about whether

752   Hepburn et al. the price of a bundle of natural resources would fall or rise over a corresponding decade (The Core Team, 2017). Figure 40.1 shows that over the relevant time period the proponent of abundance, Simon won the bet as prices fell. However, commodity prices fluctuate around with all sorts of shifts and shocks, and aggregate price movements over any decade of the last century do not provide an indication of any strong trend. More recent concerns about peak minerals have been somewhat more nuanced. For instance, the latest report to the Club of Rome by Bardi (2014) stresses that civilization is at risk because of increasing extraction costs, rather than because of imminent depletion of mineral resources. As a result of these increasing costs, the claim is that production of many mineral commodities is on the verge of decline. Debates have been ongoing for decades, particularly in relation to oil (Maugeri, 2004; Ehrenfeld, 2005). Serious analysis suggests that real resource constraints would, in principle, bite, if it is assumed that (i) all nine billion people impose material demands at the level of currently rich countries; and (ii) that these demands are met without any technological progress—​that is, using current technologies (Gordon et al., 2006). Using a similar technologically static lens, various commentators have suggested we might be in peak coal, peak copper, peak phosphorus, peak gas, peak uranium, and peak oil.4 The core concern to economists of these hypotheses about resource scarcity is that as resources become scarcer and prices increase, economic activity will slow down, potentially asymptotically to zero (Stiglitz, 1974) other things being equal. However, other things are very far from being equal. The very fact that the scarcity of exhaustible natural resources is reflected in their market price helps ensure that such resources are not exhausted. This is because higher resource prices induce firms to find substitutes and to invest more in exploring for more resources. A  classic example is the

A: End of WWI: 1918

B: Start of Great Depression: 1929

C: End of WWII: 1945

E: Second oil shock: 1979–80

F: Dissolution of the Soviet Union: 1990

G: Start of global financial crisis: 2008 SimonEhrlich bet

260

McKinsey Commodity Price Index

240 220

A

B

C

D

D: First oil shock: 1973–74

E

F

G

200 180 160 140 120 100 80 60 1900

1910 1920

1930 1940

1950

1960 1970

1980

1990 2000

2010

Figure 40.1   Commodity Prices Over the Last 100 Years. WWI, World War I; WWII, World War II. Source: The CORE Project (2015).

Green Growth   753 substitution from copper wire in telecommunications to fibre optic, which is both cheaper and provides a higher-​quality set of services.5 Is there any harder evidence upon which to base this optimism that prices and markets will work to trigger the desired responses? We analyse publicly available data from the United States Geological Survey (USGS), along with further historical data requested from USGS, which provides (annual) US prices, US consumption, and the best estimates of worldwide production and economically extractible reserves/​resources since 1957. Focusing on the minerals at risk of depletion, we analysed the top-​ten economically important minerals that, based on present production trends, would be depleted within the next century. For these minerals, we have a near complete time series. The data were harmonized to ensure that all units are consistent across time. For each given past year, a simple linear trend of production was calculated based on all previous year’s data.6 Based on this production rate, the projected year of exhaustion of each mineral was estimated. Results of the analysis are shown in Figure 40.2. The horizontal axis shows the year of analysis, beginning in 1957. For this and each subsequent year, a forecast year of exhaustion of each mineral is determined, shown on the vertical axis. The forecast year of exhaustion is calculated by taking current consumption, projecting this into the future using the consumption trend, and combining this forecast of future consumption with the reserves as known in the year of analysis. Figure 40.2 shows that the forecast exhaustion dates largely follow a trend with angle to the origin of forty-​five degrees. That is to say, for every year that passes, enough additional reserves are discovered (or, with better technology, more

2060

Forecasted exhaustion year

2040

2020

2000

1980

1960 1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

2005

2010

Year of forecast Gold Tin Fluorspar

Lead Cadmium Copper

Barite Diamond 10 years reserve

Zinc Silver 50 years reserve

Figure 40.2   Forecast Exhaustion Dates of Minerals Extend into the Future.

2015

754   Hepburn et al. resources become more economically extractible) to push the forecast date of exhaustion one year into the future. While the USGS represent the most reliable, comprehensive, and consistent source of mineral information over time, the estimates should, nevertheless, be taken with some caution. Reserves data can be very unreliable, especially when the minerals are mined in conflict regions for which data may not be available for some years. Before the collapse of the Soviet Union, it was also difficult to verify reserves in former socialist countries. Nevertheless, the clear result of this analysis is that past concerns about minerals depletion have not been well founded. While production may decline, there is little evidence that we are running out of anything. Moreover, as production of most minerals is growing, our estimates do not support any ‘peak mineral’ theories. Of course, it is true that the persistent over-​pessimism of past forecasts does not necessarily imply that current and future pessimistic forecasts will not turn out to be correct. And it is logically true that indefinite economic growth for hundreds or thousands of years will eventually exhaust any finite stock of reserves. However, humans rarely appropriate minerals more than several kilometres into Earth’s crust (which is five to seventy kilometres deep depending on the precise location), and the evidence strongly suggests that market and price dynamics have provided, and will continue to provide, incentives for innovation and substitution. Concerns about natural resource depletion in the near future appear misplaced. To be clear, the existence of market prices on minerals and energy sources such as coal, oil, and gas does not mean that such resources are properly priced, that is, in a way that is efficient from a societal point of view. Many such resources involve externalities generated by production and consumption that are not fully internalized in the price, leading to excessive consumption. For instance, consumption of fossil fuels produces carbon dioxide (CO2), which is inadequately priced around the world—​the consequence of the failure to internalize this externality is that fossil fuels are too cheap. But it does not follow from this that we are at risk of running out of fossil fuels. Instead, we are at risk of running out of atmosphere—​ or, more precisely, the exhaustible resource of the atmosphere’s ability to accumulate greenhouse gases without giving rise to an unsafe climate for human civilization.

Natural Capital and ‘Planetary Boundaries’ Environmental problems are often most severe when there are absent, inadequate, or unclear property rights, whether private or communal rights. Where natural capital is open access—​it can be used by anyone—​it is highly likely to be overused with environmental problems the result. On large scales, such degradation of natural capital can threaten green growth, and even just growth itself if degradation undermines the conditions for economic productivity. The mere fact that a natural resource is renewable does not, of course, mean that there is no need for concern. For instance, fish stocks can replenish themselves when their relative and absolute population is sufficiently high, and yet global fish stocks are now massively overharvested (Costello et al., 2008). Biodiversity is another example of renewable natural capital that is under threat (Butchart et al., 2010). Indeed, it is perhaps ironic that often the natural capital at risk is renewable natural resources rather than the exhaustible resources.7 By the same logic is that, in the previous section, just as the presence of a market price provides an incentive for innovation and exploration on the supply side and economizing

Green Growth   755 behaviours and resource substitution on the demand side, the absence of prices implies the absence of these incentives. Sometimes, otherwise ‘common-​pool’ or ‘open-​access’ resources do have partial implicit or explicit prices, created by community or government intervention, but the prices concerned are too low, leading to socially suboptimally excessive consumption. As noted above, the most significant greenhouse gas (CO2) is not appropriately priced because each nation has a partial (short-​sighted) incentive to free ride on the efforts of others, absent a multilateral agreement. And some forms of natural capital—​such as biodiversity—​frequently are not priced at all. While biodiversity has value as an input into production (e.g. bees produce honey) and as input in our utility functions (e.g. the pleasure derived from experiencing a bio-​diverse ecosystem), it is nevertheless very hard to measure and to create markets or other instruments that would give rise to a price. As such, governments have largely failed to create biodiversity prices. And when natural capital is unpriced it is particularly vulnerable to overexploitation. These various isolated examples of threats to natural capital are, in fact, symptoms of a broader set of very important environmental problems. With the explosion of the human population from one billion people to over seven billion people over the last century, there are now few places on Earth that are untouched by human hands. Geologists have suggested that we have now moved geological epoch from the Holocene to the Anthropocene, because the evidence of humanity is now to be found in the geological record (Crutzen, 2006). The so-​called ‘planetary boundaries’ that the human species is bumping up against are multifaceted. Rockström et al. (2009) proposed nine such quantitative boundaries, each one permitting some margin for error and the recognition that, in reality, scientists are unable to provide certainty about the location of system thresholds.8 They concluded that three such boundaries had already been crossed, as shown in Figure 40.3. While this may be not necessarily imply disaster—​each of these thresholds includes a margin for error—​there is no doubt that the serious environmental challenges lie in the domain of these unpriced, often global or regional, domains of natural capital. While these are genuinely serious challenges—​unlike the natural resource limitations discussed earlier—​they do not by necessity imply that further growth in the value of goods and services that we provide each other is impossible. Indeed, it is arguable that further economic growth is required to address these challenges (Hepburn and Bowen, 2013). However,

Earth-system

Parameters

Pre-industrial

Proposed

Status

value

boundary

(as of 2009)

(i) Atmospheric CO2 concentration (ppmv)

280

350

387i

(ii) Change in radiative forcing (W/m2)

0

1

1.5

process Climate change Biodiversity loss

Species extinct per million species per year

0.1-1

10

>100

Nitrogen cycle

N2 removed from atmosphere for human use

0

35

121

i

See http://www.esrl.noaa.gov/gmd/aggi/ (Accessed 18 August 2015) for updated estimates of CO2 concentrations – now above 400 ppm – and radiative forcing.

Figure 40.3   Selected Planetary Boundaries Already Breached. Source: Rockström et al. (2009).

756   Hepburn et al. the notion that economic growth in and of itself will lead to us resolving these problems seems highly complacent at the global level.9 Further theory is required.

Economic Theory of Green Growth The previous section noted that although economic growth increases pressure on natural resources and the environment, many of the concerns about peak resources are misplaced. We are not running out of resources or energy in general. Rather, concern should be focused on the specific natural capital that is improperly priced—​in particular, assets such as biodiversity, fisheries, climate stability, and so on. This section reviews the relevant economic theory of green growth, with a view to identifying guidance for tackling these challenges. Much of the theoretical research in the area has been focused on the feasibility of economic growth over an infinite horizon. The sun is expected to last only another five billion years,10 and a degree in mathematics is not required to observe that this is significantly less than infinity. In short, to some degree the question of infinite growth has a trivial answer—​it is not possible, at least in our solar system, as we know it. More practically, it is questionable whether humanity will see out another thousand years given various risks, including many that we are creating ourselves (Rees, 2003). And, if humans continue to exist over that time frame, it is far from impossible that we will have been fundamentally changed by technological developments (Bostrom, 2014). So we do not seek to examine closely whether economic growth is possible through an infinite horizon (Daly, 1997; Stiglitz, 1997) as is often the focus of the economic models. Instead we seek to tease out the implications of the models for the possibility of sustained economic growth over the next century or so.

The Simple Economics of Green Growth with Exhaustible Resources A simple formulation of the problem of achieving green growth can be set out as a straightforward optimization problem in the spirit of Ramsey (1928). The objective in such models is to maximize the discounted flow of utilities of a representative agent subject to some production, environmental, and budget constraints. Early analysis considered the management of an economy with an exhaustible resource (Hotelling, 1931). In this model, a resource owner decides whether to leave the resource in the ground or to extract and sell it. With the obviously incorrect assumptions of no technological progress or substitution, it emerges that the price of the resource would be expected to rise at the rate of interest. Observed prices have not risen at the rate of interest, but instead been largely trendless, revealing something about the accuracy of those assumptions (Lin and Wagner, 2007). Nevertheless, the Hotelling model provides the base case. In the model, extraction occurs at the socially optimal rate and consumption of the resource falls over time. The core conclusion of the model is very simple: an agent who only lives off a finite resource is doomed to an eventual decline. While unrealistic in several important ways,

Green Growth   757 Hotelling’s formalization of the dynamics of optimal use of a finite resource laid a foundation for much of subsequent green growth literature. The picture changes dramatically when technological progress is introduced into a model with production requiring an exhaustible resource. Triggered by the first oil crisis, economists (Dasgupta and Heal, 1974; Solow, 1974; Stiglitz, 1974) extended Hotelling’s model to incorporate labour supply and saving decisions, as well as (exogenous) technological progress and population growth. A brighter future is then possible. Stiglitz (1974) shows that there is a steady state where consumption grows at a constant rate, the saving rate is constant and the rate of natural resource use declines at a constant rate. Man-​made capital gradually substitutes for the resource as the driving force of production over time. Moreover, a high saving rate leads to lower short-​term consumption but higher long-​run growth rates. Therefore, unlike the traditional Ramsey model, the economy has multiple steady states and the government (modelled as a putative ‘social planner’) can, in principle, implement any of them. Increasing population competes with production scale and technological progress to guarantee non-​decreasing consumption. However, the exhaustible natural resource is, indeed, gradually exhausted along any feasible growth path. In such models, although continued economic growth is possible, green growth is impossible, simply because natural capital is assumed to be entirely exhaustible, rather than partially renewable. In models in which the exhaustible natural capital stock appears directly into the utility function, in addition to the production function, the agent’s willingness to preserve the exhaustible resource is increased. The severity of the conclusion is attenuated, but the fundamental result does not change: continued growth is feasible (and optimal if there is sufficient technological progress), but green growth is not—​it is also optimal to gradually draw down the natural capital stock (Krautkraemer, 1985). Incorporating the natural capital stock in both the production and the utility function is plausible for much natural capital. Climate stability is, perhaps, the canonical example—​changes to the climate will harm both production and consumption. And, as the ‘the greatest market failure the world has seen’ (Stern, 2007), the scale of this particular challenge to green growth is not trivial. It is precisely the incorporation of potentially substantial environmental damages directly in the utility function, as well as the production function, that differentiates and complicates modern models of green growth under climate change. The production function in some economic growth models also includes ‘energy’ as a factor, along with labour and capital (and occasionally materials).11 The argument for explicitly incorporating energy as a factor of production is that, for the most part, it cannot easily be substituted for by other inputs (Stern, 2011). Some economic growth models that include energy further differentiate between renewable energy (e.g. solar and wind) or fossil fuels generating greenhouse gases. Green growth under these conditions typically ensures that the total stock of CO2 in the atmosphere is below a certain threshold, such as one trillion tonnes of carbon (Allen et al., 2009). The optimal transition path depends on the initial stock of capital (and the corresponding marginal utility of consumption), as well as on the supply of fossil fuels relative to renewables (Smulders et al., 2014). In order to achieve green growth in such models, consumption typically has to fall in the short term, relative to business-​as-​usual, in order to accumulate investment in clean energy capital. This intuition is similar to the steady-​state growth in Stiglitz (1974). Hence, there is a trade-​off between the consumption of the present and future generations (Nordhaus, 2008). Because of the externality of climate change, environmental policy, such as carbon pricing,

758   Hepburn et al. is necessary to implement the optimal transition (Golosov et al., 2014). The rate at which present and future consumption changes optimally depends on the expected damages of climate changes, as well as on a host of normative factors, such as the social discount factor and elasticity of inter-​temporal substitution (Sælen et al., 2009; Heal and Millner, 2014). Many economists subscribe to the ‘weak green growth’ paradigm and this appears consistent with the outcomes of most integrated assessment models (Nordhaus and Yang, 1996; Tol, 1997; Hope, 2006), although these have come in for significant criticism (Pindyck, 2013; Stern, 2013; Weitzman, 2013).

Strong Green Growth As Broome (2010) points out, the ‘very most important thing about climate change is … that the problem of climate change can be solved without anyone making any sacrifice’. As climate change is an externality, the society must be in an inefficient outcome therefore it should be possible to move to a Pareto-​superior outcome making everyone (every generation) weakly better off. This suggests that the sacrifice of present generation’s consumption in order to guarantee green growth for future generation (see earlier) may not be necessary. If the government is able to borrow, it can make the necessary climate investments today without reducing the consumption of the present generation (Foley, 2007). The following generations will also be better off (despite having to repay the debt) because they will benefit from the climate investments and reduced emissions. This optimistic vision of ‘green growth’ has been called ‘strong green growth’ (Jacobs, 2013). Policies advocated by ‘strong green growth’ proponents usually involve green subsidies, usually because the first-​best of carbon tax is infeasible. This kind of ‘Keynesian’ stimulus may form a part of responsive fiscal policy in a recession, while reducing the (long-​run) costs of renewables and mitigating climate change. However, green subsidies in absence of a carbon tax may create perverse incentives for fossil fuel producers. As they expect the demand for fossil fuels to drop in the future (they are concerned about their assets becoming stranded), they extract more fossil fuel compared with the no-​subsidy baseline and emit more CO2 causing the so-​called ‘green paradox’ (Van der Ploeg and Withagen, 2012).

Technological Change and Green Growth Irrespective of the specific model of green growth, a major feature is the requirement for some combination of technological change and substitutability between man-​made capital and finite natural resources in production function. Understanding these further is an important area of research. As Hepburn and Bowen (2013) observe, the emissions intensity of GDP would need to decrease by roughly 7 per cent annually in order to achieve an ‘absolute decoupling’ of economic growth and CO2 emissions—​that is, an absolute fall in CO2 emissions despite increases in population and output per person. Whether human societies will choose to deliver these innovations at reasonably low cost, and in time to prevent the worst damages, remains unclear.

Green Growth   759 In order to understand how green innovation responds to market forces, it is crucial to take into account endogenous technological change (Romer, 1990; Aghion and Howitt, 1992). Knowledge creates positive spillovers and intuitively, like any public good, it is likely to be underprovided by the free market. Significant theoretical work has been done to understand endogenous green innovation (Pittel, 2002) and many integrated assessment models now incorporate some element of endogenous technological change (Gillingham et al., 2008), although such modelling approaches are rarely as compelling as they might be (Farmer et al., 2015). Acemoglu et al. (2012) provide a very general formulation of the green innovation problem. In their model, there are clean and dirty energy sectors, both of which are subject to endogenous innovation. Workers choose which sector they want to work in. There are two externalities: the environmental externality in the form of CO2 emissions and the knowledge externality with sector-​specific productivity spillovers. The authors show that the optimal policy is to subsidize green innovation and tax carbon for a limited period of time until the clean-​energy sector can compete for research talent with the dirty-​energy sector by offering sufficiency high wages. Once that point is reached, no more subsidies or taxes will be necessary and the environmental disaster will be averted. Interestingly, the results of this model rely, as the earlier Stiglitz–​Solow–​Dasgupta–​Heal models do, on the substitutability between clean and dirty inputs of production. If these inputs are complementary, and hence dirty inputs are absolutely necessary for production, the environmental disaster cannot be avoided. The crucial point in the innovation approach to green growth is that the delay is costly because as time passes the dirty sector gains some additional advantage over the clean sector. Certainly, innovation in the fossil fuel sector continues at present, with the advent of shale oil and gas and notable recent declines in costs. This provides a justification for the role of government as a long-​term planner, especially in infrastructure investment, in order to avoid dirty infrastructure lock-​in. Green innovation gives some hope to the ‘strong green growth’ proponents. Aghion et al. (2012) show that spillovers from innovation are up to 40 per cent greater in the clean-​energy sectors.

More Mind, Less Matter Economic growth is usually associated with greater use of material resources. However, GDP incorporates both material and non-​material products and services. Hepburn and Bowen (2013) advocate for a shift towards materially lighter, knowledge-​based or labour-​ demanding sectors. Even if environmental damages affect material production to such an extent that we can only reach a steady state in the material throughput of the economy (Stokey, 1998), it should be possible to continue producing non-​material goods and services, thereby increasing GDP. A structural change has already been observed in many advanced economies since the beginning of the Internet age (Quah, 1997, 1999), leading to the notion of a ‘weightless’ economy. More weightless economies may even be more productive. Indeed, using US data, Baptist and Hepburn (2013) show that firms and sectors with lower material and higher labour intensity have higher total factor productivity. This conceptual vision of green growth probably applies only to very advanced economies, which do not face immediate needs of increasing material consumption of food,

760   Hepburn et al. energy, and shelter, and where material recycling rates are edging upwards as discussion of the ‘circular economy’ increases. Yet, even for advanced economies it is not obvious to what extent material and immaterial factors of production are substitutable.

Policies for Green Growth Substitutability, Scales, and Time Horizons Some economists would consider that the role of government is relatively minimal, perhaps only to ‘getting prices right’, provided, of course, that the costs of intervening to improve prices are outweighed by the benefits. In this view, there is no role for government in setting overall economic direction and in determining appropriate growth trajectories. While delivering ‘correct’ prices is absolutely central, prices at the margin can only be judged to be correct or otherwise with reference to a broader path of economic development. The standard marginal (partial equilibrium) cost–​benefit analysis that assumes that prices are fixed after policy interventions may also prove less useful for large-​scale interventions to protect natural capital, such as mitigating climate change (Dietz and Hepburn, 2013). As such, this simultaneously omnipotent and removed view of the government is potentially misleading in understanding how green growth is likely to be brought about. There are three central elements for policy to consider:

(i) Substitutability:  the ultimate objective is to substitute away from dirty energy towards clean energy while preserving natural capital. (ii) Scale: the scope of policy should be commensurate to the geographical and political scale of the challenges. (iii) Time horizons: policies and institutions will not succeed in meeting the objectives without credibility over long time frames. We consider each of the three elements in turn. Firstly, any economic model of green growth rests on the assumed ability to substitute, over time, for the gradual depletion of exhaustible resources, while simultaneously protecting remaining critical renewable resources (Ekins, et al., 2003). For instance, managing climate change is technically feasible because renewable energy, combined with other technologies (grid interconnection and balancing, storage, and demand-​side responses), can eventually serve as a substitute for non-​renewable energy. Similarly, weightless green growth can be achieved if non-​material factors are substitutes for material factors. More generally, green growth can be achieved if innovation delivers clean inputs that serve as satisfactory substitutes for dirty inputs. Therefore, a key policy prescription for green growth is to set incentives to support the innovation and deployment of clean factors or reduce production dirty factors. Secondly, the geographical scale of different environmental constraints has drastically different implications for the possibility of different policy responses. Pressures on natural capital arise on different scales; institutional solutions of appropriate scales are also required (Ostrom, 2009). For example, a Hotelling-​type solution might be applied to local challenges of optimal natural resource extraction, such as the extraction of a locally abundant

Green Growth   761 commodity. Biodiversity is a problem at a regional scale and it can be addressed with national or international interventions (Hepburn and Helm, 2014). Carbon emissions and climate change are, of course, archetypical global problems, and climate change is just one of several ‘planetary boundaries’ that need to be defined and addressed (Rockström et al., 2009). Thirdly, the costs of different environmental problems and the benefits of solving them are different at different points in time. Cost–​benefit analyses of green growth policies will, in some cases, be very sensitive to time horizons and scope of policy. For example, in a case study in Brazil Vogt-​Schilb et al. (2014) show that metro, rail, and bullet trains are not sensible mitigation options if the goal is to reduce emissions by 10 per cent by 2020—​indeed, they have no hope of achieving such a goal. However, they form a critical part of an optimal emissions mitigation plan if the target is a 20 per cent reduction in emissions by 2030. Such large infrastructure programmes take a long time, but have large effects, and hence are appropriate to meet longer-​term objectives.

Changing the Game It is clear that policymakers have faced substantial challenges in getting prices in place on natural capital such as biodiversity and the climate, where the issues are global and long term. Carbon prices are too low, and even when economically sensible systems have been put in place, there is no guarantee that the political conditions will enable them to be sustained, as was experienced in Australia (de Lemos and Aydos, 2014). Biodiversity prices barely exist at all, and there is relatively little evidence of what policy interventions actually work and why (Miteva et al., 2012; Hepburn and Helm, 2014). In many cases, these problems are encountered because the problems are global and policymakers find themselves in a game somewhat resembling the prisoners’ dilemma. While in reality the game is not quite as bad—​unlike the prisoners’ dilemma there are repeat interactions, communication is possible, and side deals can be constructed—​it is clear that our collective record at playing global natural capital games is relatively poor. Economic reasoning and analysis suggests, however, that our record need not be quite so poor. Rather than hoping that the players will ‘see sense’, or that superior negotiators will forge a deal, or that political pressure will come to the rescue, it may be more valuable to step back and ask how the underlying structure of the game might be changed. Changing the structure of the game is likely to lead to different strategies being adopted by the players and different outcomes. Along these lines, Nordhaus (2015) suggests the construction of a ‘climate club’—​the simple idea is that members should derive more benefit from joining the club than from remaining outside. Club members would impose trade sanctions on non-​members, similar to the proposal of Helm et al. (2012), except that Nordhaus’ sanctions would be a uniform percentage tariff on all imports into the club region. A major benefit of club membership is therefore to avoid such trade sanctions. This benefit needs to be high enough to outweigh the cost of joining, which would be the requirement that members impose a minimum carbon price, and Nordhaus (2015) suggests US$25–​50/​total CO2. A somewhat different, but very promising, line of reasoning emerges from research by Barrett and Dannenberg (2014). They run experiments in which the players can choose the game before they play it. In their set up, players can play a coordination game or a prisoners’

762   Hepburn et al. dilemma. The maximum possible pay-​offs individually and collectively are potentially higher in the prisoners’ dilemma than in the coordination game, but cooperation is much harder to sustain and rapidly breaks down. In comparison, the (second-​best) coordination game delivers, in practice, better behaviour and substantially improved overall performance. Despite this, experimental results suggest that the players start by aiming for the ideal, and then fail to achieve it. The disappointing experience of their failing in the prisoners’ dilemma (a ‘push’ factor) and the expectation of better results in the coordination game (the ‘pull’ factor) leads players to change the game. Both lines of research suggest that changing the game to a ‘club’ holds out better prospects for the protection of global natural capital, and hence green growth, because it helps overcome problems of free-​riding. However, Barrett and Dannenberg (2014) emphasize the importance of simplicity and enforcement in their approach, and hence focus on a club built around technology standards, rather than prices, that are easier to monitor. An example of such an approach in the climate-​change context might be a ban on coal-​fired power generation (Collier and Venables, 2013), or an emissions performance standard on power plants (e.g. the beleaguered Clean Power Plan proposed by the Environmental Protection Agency in the USA). In the context of biodiversity, approaches to be considered include production standards set by creative coalitions of nation states and corporations, mandating that consumer demand is met by supply chains that do not harm to specific habitats and species. The challenge with such standards is to design them in such a way as to discourage rent seeking and government capture (Helm, 2010; Hepburn, 2010).

Sensitive Intervention Points The economic theory of green growth explored in the ‘Economic Theory of Green Growth’ section focuses on first-​best policies to deliver optimal outcomes. In such utopias, social planners can adjust prices to correct for all the market failures. In reality, of course, governments are unable to commit to environmental policies (Helm et  al., 2003; Brunner et al., 2012). Behavioural inertia, lock-​in, and path dependence may prevent straightforward implementation of policies (Aghionet al., 2014). Key institutions lack critical scientific information, including about vital thresholds (Barrett and Dannenberg, 2014). Political consensus is elusive. Government failure is not uncommon for national problems (Helm, 2010; Hepburn, 2010), and the free-​riding problems discussed in the previous section are prevalent at the international level. Even apparently plausible solutions, such as the climate club of Nordhaus (2015), face significant uphill struggles. Enforcement problems make implementation of green growth policies difficult. Simply ‘getting the prices right’ is often not remotely possible, or it is at least very difficult, and may not even be a sufficient condition for success. In reality, therefore, governments might be well advised to think less in terms of optimality, and more in terms of system change, seeking intervention points within the system that are sensitive and could deliver magnified results. Such ‘sensitive intervention points’ require going beyond second-​best policies (Hallegatte et al., 2011), and thinking through unintended consequences, such as the ‘green paradox’ (Van der Ploeg and Withagen, 2012). One example is the quite remarkable social and economic impacts of rapid technological change. In particular, the cost of solar photovoltaic modules has dropped in price

Green Growth   763 by a factor of over 2000 since 1956. As shown in Figure 40.4, costs have fallen at around 10 per cent per annum for the last thirty years (Farmer and Lafond, 2015), with even greater cost declines in recent years. It is not far-​fetched that increases in global public spending on clean-​energy research and development, from trivial levels of US$6 billion per annum (King et al., 2015), could have a significant effect on our ability to tackle climate change at low cost. Consideration of the intervention points that are ‘sensitive’ in this manner is aided by an awareness of the potential shifts in political and business decisions arising from new communication technologies. Heimans and Timms (2014) claim that the conjunction of new digital and communication technologies is creating new forms of political, social, and economic power that are more distributed and less hierarchical. For our purposes there are two key dimensions. Firstly, innovation is moving away from closed research laboratories and into patent-​free spaces, where crowd sourcing can play a large role. Innovation prizes spur teams to compete with one another to solve socially important problems. In 2014, Tesla set a precedent (and a wise business decision) by making its patents freely available to the public in order to spur the development of electric vehicle infrastructure and innovation. In early 2015, Toyota followed Tesla’s example. Secondly, technologies are enabling people to coalesce more quickly and easily around particular ideals, such that incremental individual decisions can rapidly become large-​scale social movements. Combined with new distributed energy technologies, and peer-​to-​peer financing models, solar and wind power technologies are, at a very small scale, turning households from consumers to producers of electricity.

50.0 All prices in 2011 U.S. dollars

Solar energy

5.0 1e+00 0.17$/kWh = 2013 price in $/Wp Nuclear DOE SunShot Target = coal elec. price 1e–02

0.5

Coal fuel price for electricity generation 1900

1950

price in $/Wp

price in $/kWh

1e+02

2000

1.0 2050

Photovoltaic module prices U.S. nuclear electrity prices U.K. Hinkley Point price in 2023

Figure 40.4   Progress in Solar versus Coal and Nuclear. Source: Farmer and Lafond (2016). ‘How Predictable is Technological Progress?’, Research Policy, 45:3, p647-​665. This image is distributed under the terms of the Creative Commons Attribution 4.0 International Licence (CC BY 4.0). https://​creativecommons.org/​licenses/​by/​4.0/​. For permission to reuse outside of these licence terms, please contact the rights holder.

764   Hepburn et al. Additionally, huge amounts of behaviour data allow firms to apply machine-​learning tools to help consumers save energy and optimize individual resource use. The list can go on. The key point is that these new technologies and trends are shifting the points of leverage in systems that have appeared stubbornly resistant to efforts to protect natural capital and transition to green growth.

Conclusion This chapter began by dismissing the two positions against green growth as implausible. The notion that growth does not need to be ‘green’—​that there are no real environmental constraints—​is not borne out by the data on climate change, biodiversity, fisheries, and other unpriced natural capital. But, equally, these constraints do not imply that we must halt economic growth. On the contrary, further innovation and technological progress, along with careful navigation of the politics, will be required to overcome environmental challenges. This only becomes more difficult in a recessionary or zero-​growth environment. As zero-​growth is neither necessary nor desirable, green growth is trivially required. But the fact that green growth is required for long-​run human prosperity does not make it inevitable. Economic theory suggests that provided we have scope for continued technological progress and a degree of substitutability between dirty and clean inputs, green growth is feasible and in the long-​run is even optimal. Such theory prescribes ‘getting the prices right’ to ensure that the value to be generated from natural capital is maximized as economic growth continues. Empirically, however, the trends are adverse, and despite the considerable effort of policymakers to slow them, very little has been achieved for key environmental challenges. This is because reality diverges rather dramatically from optimal green economic growth models. The problems are manifold, but perhaps the most fundamental underlying problem is well understood: the incentives of the individual actors (whether nation states, corporations, or individuals) are at odds with social good. But although the problem is well understood, compelling solutions have not, thus far, been forthcoming. What might work? The two most promising avenues towards green growth lie outside the optimality paradigm of economics. The first starts from the recognition that for international natural capital challenges, the problem is the underlying structure of the political game itself. It is unrealistic suddenly to expect the players to play the same game in a different way. Interventions would seek to change the structure of the game, from prisoners’ dilemmas to coordination games, focusing on highly pragmatic policies that are simple and easy to enforce. The second avenue is a systems-​based approach that involves seeking ‘sensitive intervention points’. For instance, rather than attempting to protect natural capital directly by bans, prices, or other such interventions, policies might accelerate the development of appropriate clean technology, triggering greater system change. While challenges of government capture and failure need to be kept in mind, so too do the emerging methods of political and social transformation, themselves enabled by new digital technologies.

Green Growth   765 In sum, there is no compelling alternative for our economic systems other than a transition to green growth. It is necessary, feasible, and desirable. But green growth policies for the Anthropocene are likely to require innovative policies that harness technology, markets, and prices, coupled with a realistic assessment of political economy constraints, and an acceptance of pragmatic, second-​best solutions.

Notes 1. See https://​www.gov.uk/​government/​speeches/​address-​to-​the-​green-​growth-​summit (last accessed 13 August 2015). 2. Consider reducing greenhouse gas emissions, for instance. The set of current abatement technologies includes some that are very cheap—​indeed nearly free—​and others that are much more expensive at around $1200/​total carbon dioxide (tCO2), these all look remarkably cheap when compared with cutting GDP. Current global GDP is, very roughly, US$80 trillion per annum and global emissions are roughly 40 billion tCO2 per annum. On average, therefore, each tonne of CO2 was associated with around US$2000 of output. An abatement technology that cost US$2,000/​tCO2 would not represent good value for money. 3. Fossil fuels do renew, in the sense that the processes that convert solar energy to biomass and then to coal, oil, and natural gas do continue to occur, but the timescales involves millions of years. Effectively, these resources are non-​renewable, or exhaustible, on timescales relevant to humans. 4. Peak minerals (achieving a maximum rate of production) should not be confused with mineral resource depletion (elimination of the stock of resources). The former is a flow concept, the latter refers to the state of the stock. Peak minerals could, in theory, sit alongside steady-​state resources, if discoveries equal production. 5. Note that overall demand for copper has continued to increase because of new uses for copper, including in air conditioning and motor vehicle electronics (Gordon et al., 2006). 6. We have also fitted exponential (growth) trends and the results are largely unchanged. 7. There are exceptions: forests are a renewable resource that, in developed countries, are now increasing in area, arguably because of the merits of good governance and property regimes. In contrast, the atmospheric ‘sink’ for carbon dioxide is now understood to be an exhaustible resource on human timescales (Allen et al., 2009) that is clearly under threat because it is an open-​access resource. 8. The ‘proposed’ boundaries are best viewed as high-​quality early estimates. For instance, for climate change it now appears that damage is a function of peak warming, which is a function of cumulative carbon emissions, rather than atmospheric concentrations of carbon dioxide (CO2) or CO2 equivalent and radiative forcing (Allen et al., 2009). 9. Country’s per capita GDP and its level of environmental degradation—​a so-​called ‘environmental Kuznets curve’ (Stokey, 1998; Brock and Taylor, 2010). We think that a robust relationship that often emerged from theoretical models is unlikely to hold or serve as a useful guide to policy (Hepburn and Bowen, 2013). 10. See http://​www-​istp.gsfc.nasa.gov/​istp/​outreach/​workshop/​thompson/​facts.html (last accessed 21 April 2017). 11. Such models are commonly referred to as having a KLEM (Capital (K), Labour (L), Energy (E), and Materials (M)) production function. See, for instance, Berndt and Khaled (1979) and Baptist and Hepburn (2013) for an application.

766   Hepburn et al.

References Acemoglu, D., Aghion, P., Bursztyn, L., and Hemous, D. (2012). The environment and directed technical change. American Economic Review 102: 131–​166. ADB (2013). ‘Low-​carbon green growth in Asia: policies and practices’ https://​www.adb.org/​publications/​low-​carbon-​green-​growth-​asia-​policies-​and-​practices (last accessed 21 April 2017). AfDB (2013). ‘African development report: towards green growth in Africa’ https://​www.afdb. org/​fileadmin/​uploads/​afdb/​Documents/​Publications/​African_​Development_​Report_​ 2012.pdf (last accessed 21 April 2017). Aghion, P. and Howitt, P. (1992). ‘A model of growth through creative destruction’. Econometrica 60: 323–​351. Aghion, P., Dechezleprêtre, A., Hemous, D., Martin, R., and van Reenen, J. (2012). ‘Carbon taxes , path dependency and directed technical change: evidence from the auto industry’. CEP Discussion Paper No 1178 November 2012. Aghion, P., Hepburn, C., Teytelboym, A., and Zenghelis, D. (2014). ‘Path dependence, innovation and the economics of climate change’. Policy paper November 2014 http://​www.lse. ac.uk/​GranthamInstitute/​wp-​content/​uploads/​2014/​11/​Aghion_​et_​a l_​p olicy_​p aper_​ Nov20141.pdf (last accessed 21 April 2017). Allen, M., Frame, D.J., Huntingford, C., Jones, C.D., Lowe, J.A., Meinshausen, M., and Meinshausen, N. (2009). ‘Warming caused by cumulative carbon emissions towards the trillionth tonne’. Nature 458: 1163–​1166. Baptist, S. and Hepburn, C. (2013). ‘Intermediate inputs and economic productivity’. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 371: 20110565. Bardi, U. (2014). Extracted: How the Quest for Mineral Wealth is Plundering the Planet (White River Junction, VT: Chelsea Green Publishing). Barrett, S. and Dannenberg, A. (2014). ‘Sensitivity of collective action to uncertainty about climate tipping points’. Nature Climate Change 4: 36–​39. Berndt, E. and Khaled, M. (1979). ‘Parametric productivity measurement and choice among flexible functional forms’. Journal of Political Economy 87: 1220–​1245. Bostrom, N. (2014). Superintelligence: Paths, Dangers, Strategies (Oxford: Oxford University Press). Bowen, A. and Hepburn, C. (2015). ‘Green growth: an aseessment’. Oxford Review of Economic Policy 30: 407–​422. Brock, W.A. and Taylor, M.S. (2010). ‘The green Solow model’. Journal of Economic Growth 15: 127–​153. Broome, J. (2010). ‘The Most Important Thing About Climate Change’ in J. Boston, A. Bradstock, and D. Eng (eds) Public Policy: Why Ethics Matters, pp. 101–​116 (Canberra: ANU E Press). Brunner, S., Flachsland, C., and Marschinski, R. (2012). ‘Credible commitment in carbon policy’. Climate Policy 12: 255–​271. Butchart, S.H.M., Walpole, M., Collen, B., van Strien, A., Scharlemann, J. P.  W., Almond, R.E.A., et al. (2010). ‘Global biodiversity: indicators of recent declines’. Science 328: 1164–​1168. Coase, R.H. (1960). ‘The problem of social cost’. Journal of Law and Economics 3: 1–​44. Collier, P. and Venables, A.J. (2013). ‘Closing coal: economic and moral incentives’. Preliminary version, September https://​www.economics.ox.ac.uk/​materials/​papers/​13823/​paper132.pdf (last accessed 10 May 2017). The Core Team (2017). The Economy: Economics for a Changing World (1st ed) (Oxford University Press), (Chapter 20, Economics of the Environment).

Green Growth   767 Costello, C., Gaines, S.D., and Lynham, J. (2008). ‘Can catch shares prevent fisheries collapse?’ Science 321: 1678–​1681. Crutzen, P.J. (2006). ‘The anthropocene’. Earth System Science in the Anthropocene, pp. 13–​18, http://​doi.org/​10.1007/​3-​540-​26590-​2_​3 (last accessed 10 May 2017). Daly, H.E. (1997). ‘Georgescu-​Roegen versus Solow/​Stiglitz’. Ecological Economics 22: 261–​266. Dasgupta, P. and Heal, G. (1974). ‘The optimal depletion of exhaustible resources’. The Review of Economic Studies 41: 3–​28. Dietz, S. and Hepburn, C. (2013). ‘Benefit-​cost analysis of non-​marginal climate and energy projects’. Energy Economics 40: 61–​7 1. EBRD (2011). ‘Special report on climate change: The low carbon transition’ http://​www.ebrd. com/ ​ news/ ​ publications/​ special-​ reports/​ special-​ report-​ on-​ climate-​ change-​ t he-​ l ow-​ carbon-​transition.html (last accessed 21 April 2017). Ehrenfeld, D. (2005). ‘Recalculating future oil reserves’. Science 309: 54 LP-​56. Ekins, P., Simon, S., Deutsch, L., Folke, C., & De Groot, R. (2003). ‘A framework for the practical application of the concepts of critical natural capital and strong sustainability’. Ecological Economics 44(2): 165–185. Farmer, J.D. and Lafond, F. (2016). ‘How predictable is technological progress?’ Research Policy 45: 647–​665. Farmer, J.D., Hepburn, C., Mealy, P., and Teytelboym, A. (2015). ‘A third wave in the economics of climate change’. Environmental and Resource Economics 62: 329–​357. Foley, D.K. (2007). ‘The economic fundamentals of global warming’ http://​www.santafe.edu/​ research/​working-​papers/​abstract/​fc322939d1f8116ac9f8212840abb86e/​ (last accessed 21 April 2017). Gillingham, K., Newell, R.G., and Pizer, W.A. (2008). ‘Modeling endogenous technological change for climate policy analysis’. Energy Economics 30: 2734–​2753. Golosov, M., Hassler, J., Krusell, P., and Tsyvinski, A. (2014). ‘Optimal taxes on fossil fuel in general equilibrium’. Econometrica 82: 41–​88. Gordon, R.B., Bertram, M., and Graedel, T.E. (2006). ‘Metal stocks and sustainability’. Proceedings of the National Academy of Sciences 103: 1209–​1214. Hallegatte, S., Heal, G., and Fay, M. (2011). ‘From growth to green growth—​a framework’. Policy Research Working Paper Series, 5872(November), 1–​38. Hallegatte, S., Heal, G., Fay, M., and Treguer, D. (2012). ‘From growth to green growth—​a framework’ http://​www.nber.org/​papers/​w17841 (last accessed 21 April 2017). Hamilton, K. (2006). Where is the Wealth of Nations? Measuring Capital for the 21st Century (Washington, DC: World Bank Publications). Hamilton, K. and Hepburn, C. (2014). ‘Wealth’. Oxford Review of Economic Policy 30: 1–​20. Heal, G. and Millner, A. (2014). ‘Uncertainty and decision making in climate change economics’. Review of Environmental Economics and Policy 8: 120–​137. Heimans, J. and Timms, H. (2014). ‘Understanding “new power” ’. Harvard Business Review https://​hbr.org/​2014/​12/​understanding-​new-​power (last accessed 21 April 2017). Helm, D. (2010). ‘Government failure, rent-​seeking, and capture: the design of climate change policy’. Oxford Review of Economic Policy 26: 182–​196. Helm, D., Hepburn, C., and Mash, R. (2003). ‘Credible carbon policy’. Oxford Review of Economic Policy 19: 438–​450. Helm, D., Hepburn, C., and Ruta, G. (2012). ‘Trade, climate change, and the political game theory of border carbon adjustments’. Oxford Review of Economic Policy 28: 368–​394. Hepburn, C. (2010). ‘Environmental policy, government, and the market’. Oxford Review of Economic Policy 26: 117–​136.

768   Hepburn et al. Hepburn, C. and Bowen, A. (2013). ‘Prosperity with Growth:  Economic Growth, Climate Change and Environmental Limits’, in R. Fouquet (ed.) Handbook on Energy and Climate Change, pp. 617–​638 (Cheltenham: Edward Elgar Publishing). Hepburn, C. and Helm, D. (2014). Nature in the Balance:  The Economics of Biodiversity (Oxford: Oxford University Press). Hope, C. (2006). ‘The marginal impact of CO2 from PAGE2002:  an integrated assessment model incorporating the IPCC’s five reasons for concern’. Integrated Assessment Journal 6: 19–​56. Hotelling, H. (1931). ‘The economics of exhaustible resources’. Journal of Political Economy 39: 137–​175. Jackson, T. (2011). Prosperity Without Growth: Economics for a Finite Planet (London: Routledge). Jacobs, M. (2013). ‘Green Growth’ in R. Falkner (ed.) The Hanbook of Global Climate and Environment Policy, pp. 197–​214 (Oxford: Wiley-​Blackwell). King, D., Browne, J., Layard, R., Donnell, G.O., Rees, M., Stern, N., and Turner, A. (2015). ‘A global Apollo programme to combat Climate change’ http://​cep.lse.ac.uk/​pubs/​download/​special/​Global_​Apollo_​Programme_​Report.pdf (last accessed 10 May 2017). Krautkraemer, J. (1985). ‘Optimal growth, resource amenities and the preservation of natural environments’. Review of Economic Studies 52: 153–​169. de Lemos, E. and Aydos, P. (2014). ‘What went wrong? lessons from a short-​lived carbon price in Australia’ https://​papers.ssrn.com/​sol3/​papers.cfm?abstract_​id=2535301 (last accessed 10 May 2017). Lin, C.-​Y. and Wagner, G. (2007). Steady-​state growth in a Hotelling model of resource extraction. Journal of Environmental Economics and Management 54: 68–​83. Maugeri, L. (2004). ‘Oil:  never cry wolf—​why the petroleum age is far from over’. Science 304: 1114–​1115. Meadows, D.H., Meadows, D.L., Randers, J., and Behrens, W.W. (1972). The Limits to Growth (London: Earth Island). Miteva, D.A., Pattanayak, S.K., and Ferraro, P.J. (2012). ‘Evaluation of biodiversity policy instruments: what works and what doesn’t?’ Oxford Review of Economic Policy 28: 69–​92. Nordhaus, W.D. (2008). A Question of Balance:  Weighing the Options on Global Warming Policies (New Haven, CT: Yale University Press). Nordhaus, W.D. (2015). ‘Climate clubs: overcoming free-​riding in international climate policy’. American Economic Review 105: 1339–​1370. Nordhaus, W.D. and Yang, Z. (1996). ‘A regional dynamic general-​equilibrium model of alternative climate-​change strategies’. American Economic Review 86: 741. OECD (2011). ‘Towards green growth’ https://​www.oecd.org/​greengrowth/​48012345.pdf (last accessed 21 April 2017). Ostrom, E. (2009). Understanding Institutional Diversity (Princeton, NJ: Princeton University Press). Pigou, A.C. (1920). The Economics of Welfare (London: Macmillan). Pindyck, R.S. (2013). ‘Climate change policy: what do the models tell us?’ Journal of Economic Literature 51: 860–​872. Pittel, K. (2002). Sustainability and Endogenous Growth (Cheltenham:  Edward Elgar Publishing). Quah, D. (1997). ‘Increasingly weightless economies’. Bank of England Quarterly Bulletin 37: 49–​56. Quah, D. (1999). ‘The weightless economy in economic development’. Centre for Economic Performance Discussion Paper No. 417.

Green Growth   769 Ramsey, F.P. (1928). ‘A mathematical theory of saving’. EconomicJournal 138: 543–​559. Rees, M. (2003). Our Final Century: Will the Human Race Survive the Twenty-​first Century? (London: William Heinemann). Rockström, J., Steffen, W., Noone, K., Persson, A., Chapin, F.S., Lambin, E.F., et al. (2009). ‘A safe operating space for humanity’. Nature 461: 472–​475. Romer, P.M. (1990). ‘Endogeneous technological change’. Journal of Political Economy 98: 71–​102. Sælen, H., Dietz, S., Hepburn, C., Helgeson, J., and Atkinson, G. (2009). ‘Siblings, not triplets: social preferences for risk, inequality and time in discounting climate change’. Political Science 3. Smulders, S., Toman, M., and Withagen, C. (2014). ‘Growth theory and “green growth” ’. OxCarre Research Paper 135. Solow, R.M. (1974). ‘Intergenerational equity and exhaustible resources’. Review of Economic Studies 41: 29. Stern, N. (2013). ‘The structure of economic modeling of the potential impacts of climate change: grafting gross underestimation of risk onto already narrow science Models’. Journal of Economic Literature 51: 838–​859. Stern, N.H. (2007). Stern Review: The Economics of Climate Change (Cambridge: Cambridge University Press). Stern, D. (2011). ‘The role of energy in economic growth’. Ecological Economics Reviews 1219: 26–​51. Stiglitz, J.E. (1974). ‘Growth with exhaustible natural resources: efficient and optimal growth paths’. Review of Economic Studies 41: 123–​137. Stiglitz, J.E. (1997). ‘Georgescu-​Roegen versus Solow/​Stiglitz’. Ecological Economics 22: 269–​270. Stokey, N.L. (1998). ‘Are there limits to growth?’ International Economic Review 39: 1–​31. The Core Team (2017). The Economy: Economics for a Changing World, 1st ed. (Oxford: Oxford University Press) (Chapter 20, Economics of the Environment). Tol, R.S.J. (1997). ‘On the optimal control of carbon dioxide emissions:  an application of FUND’. Environmental Modeling and Assessment 2: 151–​163. UNEP (2011). ‘Towards a green economy: pathways to sustainable development and poverty eradication’ https://​web.unep.org/​greeneconomy/​sites/​unep.org.greeneconomy/​files/​field/​ image/​green_​economyreport_​final_​dec2011.pdf (last accessed 21 April 2017). Van Der Ploeg, F. (2011). ‘Macroeconomics of sustainability transitions: second-​best climate policy, Green Paradox, and renewables subsidies’. Environmental Innovation and Societal Transitions 1: 130–​134. Van der Ploeg, F. and Withagen, C. (2012). ‘Is there really a Green Paradox?’ Journal of Environmental Economics and Management 64: 342–​363. Vogt-​Schilb, A., Hallegatte, S., and de Gouvello, C. (2014). ‘Long-​term mitigation strategies and marginal abatement cost curves:  a case study on Brazil (English)’. Policy Research Working Paper http://​documents.worldbank.org/​curated/​en/​570361468020976042/​Long-​ term-​mitigation-​strategies-​and-​marginal-​abatement-​cost-​curves-​a-​case-​study-​on-​Brazil (last accessed 21 April 2017). Weitzman, M.L. (2013). ‘Tail-​hedge discounting and the social cost of carbon’. Journal of Economic Literature 51: 873–​882. World Bank (2011). ‘The changing wealth of nations: measuring sustainable development in the new millennium’ http://​elibrary.worldbank.org/​doi/​abs/​10.1596/​978-​0-​8213-​8488-​6 (last accessed 10 May 2017). World Bank (2012). Inclusive Green Growth:  The Pathway to Sustainable Development (Washington, DC: World Bank Publications).

Chapter 41

Pu rsuing E qu i ta bl e Ec onom ic G row t h in the Gl oba l  S ou t h Andrés Rodríguez-​P ose and Callum Wilkie Introduction Achieving economic growth has always been and remains of utmost importance to individuals and policymakers at all territorial scales. It is only recently, however, that this interest in economic growth and the ‘quantitative dimensions of development’1 has been tempered by a concern for more ‘qualitative development’ and for the character of the economic growth being achieved. That is, policymakers and researchers alike are becoming increasingly attuned to the notions of equity and inequality, as well as to the frequent failure of economic growth to enhance the livelihoods of all members of a given society (e.g. Sala-​ i-​Martin, 2002; Dunford, 2005; Kanbur and Venables, 2005; Milanovic, 2005, 2009, 2011; Organisation for Economic Co-​operation and Development, 2011, 2015; Asian Development Bank, 2012; Ranieri and Ramos, 2013; Bourguignon, 2015). A broad consensus is emerging around the need not only to tackle inequality, but also to address the root of the problem and make economic growth itself more inclusive and equit­ able (e.g. Asian Development Bank, 2008; World Bank, 2009a; European Commission, 2010; African Development Bank Group, 2013; Organisation for Economic Co-​operation and Development, 2014, 2015). That said, much work remains to be done to advance our understanding of what more equitable economic growth entails and, more importantly, how it can be achieved. The contribution of this chapter is to provide a multifaceted exploration of the notion of equitable economic growth with a specific focus on the Global South. More specifically, the chapter pursues a twofold goal. Firstly, it develops a conceptualization of ‘equitable economic growth’. Secondly, it provides insights into how equitable economic growth may be achieved in the Global South, focusing specifically on the suitability of territorially specific economic growth strategies.

Pursuing Equitable Economic Growth in the Global South    771 The remainder of the chapter proceeds as follows: the next section, ‘Understanding the Need for “Equitable Economic Growth” ’ introduces and explores the need for a more equitable brand of economic growth in the Global South. The section ‘Conceptualizing “Equitable Economic Growth” ’ conceptualizes the notion of equitable economic growth. ‘Operationalizing “Equitable Economic Growth” in the Global South’ explores issues of rele­ vance to the achievement of equitable economic growth in developing contexts. The final section concludes by introducing avenues to be pursued in future research concerning equitable economic growth.

Understanding the Need for ‘Equitable Economic Growth’ Before formulating a robust definition for ‘equitable economic growth’, the immediacy of the need for a more equitable brand of economic growth has to be highlighted. While delivering a detailed exploration of economic inequality is beyond the scope of the chapter, we aim to provide the foundations for a discussion of the need to conceptualize and operationalize equitable economic growth. The necessity of a more equitable form of economic growth—​in the Global South, in particular—​is a product of at least three factors, all closely correlated. Firstly, economic growth in the absence of a concern for equity has a documented propensity to disproportionately benefit certain geographies and certain segments of society while leaving others unaffected or even worse off. Secondly, the inherent biases of economic growth contribute to interpersonal and territorial inequality that is especially pervasive across the Global South and in their urban environments in particular. Thirdly, inequitable economic growth and, more precisely, the economic inequality it breeds have a host of implications, the most prominent of which is the effect they have on sustained economic growth and dynamism.

The Inherent Inequity of Economic Growth The relationship between economic growth and equity has been subject to extensive explor­ ation and debate. On the one hand, it has been stressed that the benefits of economic growth will, in time, ‘trickle down’ and through the entirety of the socio-​economic spectrum. In turn, the trickle-​down effect of economic growth will contribute to the alleviation of poverty and the enhancement of the welfare of society—​including its poorest and most disadvantaged members and territories (e.g. Kuznets, 1955; Hirsch, 1980; Aghion and Bolton, 1997). On the other hand, however, concerns have been raised about the extent to which benefits may actually be realized by the bottom of the socio-​economic pyramid (e.g. Thornton et al., 1988; Arndt, 1983). Increasingly, theoretical and empirical contributions linking growth to notions of equity and inequality report that economic growth is neither an automati­ cally equitable process nor does it deliver benefits that permeate all territories and all tiers of society (e.g. Organisation for Economic Co-​operation and Development, 2011; Asian Development Bank, 2012; Cingano, 2014).

772   Rodríguez-Pose and Wilkie Certain groups of society are more consistently capable of reaping the benefits of economic growth (e.g. Atkinson et al., 2011; Organisation for Economic Co-​operation and Development, 2011; Cingano, 2014). Wealthy individuals, as Kakwani and Pernia (2000, p. 3) highlight, tend to benefit the most from economic growth. The inherent advantages associated with wealth, including higher levels of education and skills, as well as access to capital, facilitate the accumulation of wealth in what seems to be an increasingly limited number of hands (Piketty, 2014). Meanwhile, poorer people are not only more frequently confronted by social, political, and institutional biases and impediments that bar them from realizing the benefits from growth, but also suffer as a consequence of the development and implementation of pro-​rich policies (Kakwani and Pernia, 2000, p. 4). The same can be said of territories. Venables (2005, p. 4) observes that certain—​often urban—​regions are more favourably positioned to both achieve and benefit from economic growth as a result of their ‘first and second nature geographies’ and the advantages and externalities that arise from them. Examples of the inequitable tendencies of economic growth from the developing world are numerous. Growing interpersonal and territorial inequality in China and Thailand provide evidence of the failure of even robust, sustained, and seemingly durable economic growth to benefit or enhance the well-​being of the entire country. China has achieved almost unprecedented economic growth in recent decades. However, not all members of society have benefited from aggregate economic growth. A rising layer of rich individuals, as well as coastal China, have been the winners of this process. But the majority of the Chinese population and those living in inland China, in particular, have not profited in equal measure (Rozelle, 1994; Jian et al., 1996; Kanbur and Zhang, 1999, 2005; Yang, 2002; Fan et al., 2011; Knight, 2014). This failure of rapid economic growth to yield more equal opportunities and ultimately equal benefit across the country raises questions about the sustainability of the pursuit of pure economic growth without consider­ ation for equity or distribution. Similarly, Thailand enjoyed a long period of relatively high growth based on the dynamism of its capital, Bangkok, which hosts more than 20 per cent of the country’s population. However, this economic growth has been accompanied by a level of polarization, which on an interpersonal and especially on a territorial level, makes Thailand one of the most economically unequal countries in the world (Cistulli et al., 2014). Inequitable development has played a significant role in the political stalemate affecting the country since 2006 and which is now seriously undermining its growth performance and potential. In short, it would seem that economic growth is not necessarily an inherently equitable process. It is, as a result of both socio-​economic and institutional factors and influences, predisposed to benefiting certain segments of society to a disproportionately greater extent, while leaving those who are in greater need unaffected or relatively worse off.

Inequitable Growth and Inequality in the Global South One of the most immediate outcomes of the inequitable character of economic growth is the entrenchment or worsening of inequality at various territorial scales. While economic inequality exists in both developed and developing countries, interpersonal and territorial inequalities are often especially pronounced and pervasive in developing contexts.

Pursuing Equitable Economic Growth in the Global South    773 An indicative examination of Gini coefficients across countries confirms the pervasiveness of interpersonal inequality in the Global South.2 Figure 41.1 depicts the Gini coefficients for a selection of developing countries. The average Gini coefficient of the generally wealthier Organisation for Economic Co-​operation and Development (OECD) countries is provided for comparison on the left-​hand side of Figure 41.1. It is evident that developing countries are, on the whole, more unequal than their more developed counterparts. Countries in Latin America, including Honduras (with a Gini coefficient of 0.537 in 2013), Colombia (0.535 in 2013), Brazil (0.529 in 2013), and Guatemala (0.524 in 2011), and Africa, including South Africa (0.634 in 2011), Namibia (0.61 in 2009), and Botswana (0.605 in 2009), stand out as especially unequal. In much of the Global South, levels of interpersonal inequality (proxied by a country’s respective Gini coefficient) have remained more or less constant at or around the levels indicated in Figure 41.1, despite relatively high levels of growth between the mid-​1990s and the early 2010s. Sustained increases in levels of interpersonal inequality have, however, been the norm in recent years and decades. The most notable of these is perhaps China. Between 1987 and 2010 China’s Gini coefficient increased from 0.299 to 0.421, despite rapid modernization, industrialization, and robust economic growth. Sizeable increases in interpersonal inequality have also been observed in, among other countries, South Africa (0.593 in 1993 to 0.634 in 2011); Nigeria (0.387 in 1985 to 0.43 in 2009); Bangladesh (0.269 in 1985 to 0.32 in 2010); and Indonesia (0.293 in 1987 to 0.356 in 2010). Territorial inequalities are equally pervasive across the Global South. Figure 41.2 depicts levels of territorial income inequality—​measured using the second Theil index—​across a

0.7

0.6 0.5

0.4

0.3 0.2

0

OECD Average (late 2000s) Cambodia (2012) Bangladesh (2010) Ethiopia (2010) India (2009) Sierra Leone (2011) Sudan (2009) Indonesia (2010) Tanzania (2011) Sri Lanka (2012) Bhutan (2012) Vietnam (2012) Thailand (2012) Burkina Faso (2009) Congo, Rep. (2011) Senegal (2011) Uruguay (2013) China (2010) Argentina (2013) Uganda (2012) Nigeria (2009) Philippines (2012) Chad (2011) Benin (2011) El Salvador (2013) Peru (2013) Mozambique (2008) Nicaragua (2009) Togo (2011) Malawi (2010) Malaysia (2009) Dominican Republic (2013) Ecuador (2013) Bolivia (2013) Mexico (2012) Paraguay (2013) Costa Rica (2013) Chile (2013) Rwanda (2010) Panama (2013) Guaetemala (2011) Brazil (2013) Colombia (2013) Honduras (2013) Lesotho (2010) Zambia (2010) Botswana (2009) Namibia (2009) South Africa (2011)

0.1

Figure 41.1   Gini Coefficients Among Selected Developing Countries. Note: ‘OECD’ Average drawn from Organisation for Economic Co-​operation and Development (2011). Source: World Bank Gini estimates, World Development Indicators Database.

774   Rodríguez-Pose and Wilkie 0.25

0.2

0.15

0.1

0.05

AUS NZL JPN NLD DNK USA CAN FIN CHE SWE IRL KOR POL LTU ESP AUT VEN SVN GRC NOR ROM FRA PRT DEU BOL CZE GBR ITA BEL HUN SVK ARG EST BUL CHL COL IND TUR LVA CHN ZAF VNM BRA KAZ RUS MEX KGZ PER ECU PHL GTM IDN THA

0

Figure  41.2   Differences in Interpersonal and Territorial Income Inequalities Among Selected Countries (Second Theil Index) (Author’s Elaboration and Calculation). Source: Organisation for Economic Co-​operation and Development and various National Statistical Offices; data for 2010 or closest year available.

sample of developed and developing countries. Developed countries—​represented in darker shades—​are generally more territorially equal, with the USA, Australia, and Canada being among the most so. Developing countries, by way of comparison, display significantly higher levels of territorial inequality. This is a sign that, like for certain layers of society, certain territories within a country benefit to a disproportionately greater extent from broader processes of economic growth than others. As pervasive as inequality may seem at the national level across the Global South, nowhere is it more pronounced than in developing urban environments. Cities of all sizes across the industrialized and developing world have, in recent years, evolved into especially dynamic environments (Dobbs et al., 2011). Ample evidence, however, suggests that urban economic growth, in particular, is far from equitable (Sankhe et al., 2010; Zhuang, 2010; Obeng-​ Odoom, 2012; Thorat and Dubey; 2012; Gustafsson and Quheng, 2013; Gustafsson and Sai, 2013; Tripathi, 2013; Turok and McGranahan; 2013) and that this inequality jeopardizes any potential economic benefits derived from urbanization (Sankhe et al., 2010). Cities thus remain characterized by high levels of inequality (income and otherwise), poverty, and social exclusion. This is especially true for cities in the Global South, where the emergence and growth of urban areas has, in many cases, been accompanied by the growth of (and worsening conditions in) informal settlements and slums. This has given rise to a host of societal challenges, as well as issues related to hunger, food security, health, physical well-​being, and access to education and opportunity, all of which ultimately stem from pervasive poverty and economic and social inequality (Baker, 2008; Moreno et al., 2010). Urban economic growth in the Global South has been pervasively unequal, benefiting those at the top of the pyramid to a much greater extent than others and has given rise to galloping

Pursuing Equitable Economic Growth in the Global South    775 inequality often associated with unequal access to basic services, as well as with social exclusion and societal polarization (Moreno et al., 2010). The most pressing question now relates to why, beyond moral imperatives, should we be especially concerned with the pervasive and often increasing inequality?

The Socio-​economic Implications of Inequality The inequitable tendencies of economic growth are associated with a number of implications and consequences. In the shorter term, inequality imposes serious social, political, and environmental costs on society. Economic inequality can also jeopardize the mid-​and longer-​term economic potential and dynamism of a given territory (Berg et al., 2008; Sankhe et al., 2010; Berg and Ostry, 2011; Ostry et al., 2014; Cingano, 2014; Organisation for Economic Co-​operation and Development, 2015). At least two recent empirical investigations have found robust evidence to suggest that inequality can compromise future economic growth (Cingano, 2014; Ostry et  al., 2014). Ostry et al. (2014, p. 25) examine the relationship between inequality and growth across a large sample of industrialized and developing countries and reach the conclusion that it ‘would still be a mistake to focus on growth and let inequality take care of itself not only because inequality may be ethically undesirable but also because the resulting growth may be low and unsustainable’. Cingano’s (2014, p. 28) investigation of OECD countries over the last thirty years also revealed a ‘sizeable and statistically significant’ negative relationship between inequality and growth. It is on the basis of this finding that Cingano (2014, p. 28) asserts that ‘focusing exclusively on growth and assuming that its benefits will automatically trickle down to different segments of the population may undermine growth in the long run inasmuch as inequality actually increases. Moreover … reversing the long-​run rise in inequality would not only make societies less unfair, but also richer’. It is therefore plausibly inferred that the rampant interpersonal and territorial inequality plaguing much of the Global South could evolve into yet another prominent impediment to the achievement of the growth and, more importantly, socio-​economic development that these countries are struggling to achieve. While jeopardized economic dynamism is a concern the world over and at all territorial scales, the prospect of slowing or compromised urban growth due to economic inequality is of particular concern in the context of the Global South. Cities are increasingly perceived as the engines of economic growth and development (Quigley, 1998; Duranton, 2000; Fujita and Thisse, 2002; Glaeser, 2011). While this assertion is by no means universally accepted (Bryceson et al., 2009; Turok and McGranahan, 2013; Fay and Opal, 1999; Brückner, 2012) and there is some disagreement as to both the extent to which cities and the types of cities have the capacity to drive economic growth (OECD, 2012; Parkinson et al., 2012; Roberts, 2014), few dispute that cities of all sizes assume a critically important role in the functioning of the global economy and do, in fact, influence the growth and development prospects of the regions and counties within which they are situated. It is not surprising, then, that cities are often awarded privileged positions in economic growth and development strategies, be they regional, national, or even supranational (see e.g. the World Bank’s (2009b) World Development Report 2009). The implications of inequality in developing urban contexts therefore extend well beyond the more intuitive immediate social costs. The failure to

776   Rodríguez-Pose and Wilkie address and rectify the aforementioned rampant inequality in developing urban contexts could thus conceivably threaten future economic growth and socio-​economic development not only in cities, but also across regions and countries given the ‘catalytic’ role cities play in broader processes of economic growth. Taken together, the frequent failure of economic growth to deliver proportionate benefits across the entirety of the socio-​economic spectrum; the often widespread inequality it contributes to; and, most importantly, the adverse effect of inequality on future economic growth, give rise to an immediate need for economic growth of a more equitable nature and for policies and strategies geared explicitly towards equity across the Global South.

Conceptualizing ‘Equitable Economic Growth’ The formulation of a robust conceptualization of equitable economic growth is an essential first step not only for future scholarly pursuits in its direction, but also for the development and implementation of the policies that are and will increasingly be necessary to promote and achieve economic growth that is more inclusive and ultimately beneficial to a greater proportion of society. In an effort to provide such a definition, the following subsection first explores the ways in which researchers and policymakers have contemplated and assessed the equality and inclusiveness of economic growth through a brief examination of pro-​poor growth—​a concept that is understood to serve as the foundation for much of the thinking of equitable (Ranieri and Ramos, 2013)—​and inclusive growth. Drawing upon the reviewed literature, the section then proposes a functional conceptualization of equitable economic growth.

Pro-​poor growth The most suitable point of departure for an effort to conceptualize equitable economic growth is pro-​poor growth. The notion of pro-​poor growth—​or, more specifically, an interest in pro-​poor growth strategies—​arose from the recognition that economic growth disproportionately benefits certain groups within a population. In the simplest sense, pro-​poor growth is understood to be ‘[economic] growth that benefits the poor’ (Ranieri and Ramos, 2013, p. 5). More nuanced conceptualizations of pro-​poor growth have, however, drawn a distinction between growth that benefits the poor and growth that benefits the poor to a greater extent than it benefits the rest of society. Grosse et al. (2008) termed the former ‘absolute pro-​poor growth’ and the latter ‘relative pro-​poor growth’. While definitions inevitably vary in their specificities, distinguishing between absolute pro-​poor growth and relative pro-​poor growth is based upon an evaluation of the distributional outcomes of economic growth. Absolute pro-​poor growth can conceivably occur in the absence of an effect on inequality or the distribution of income in society. According to conceptualizations of pro-​poor growth adopting an absolute perspective, growth is understood to be pro-​poor if the poor benefit to any extent.

Pursuing Equitable Economic Growth in the Global South    777 Relative pro-​poor growth, however, can only occur if growth disproportionally benefits the poor—​an outcome that is in opposition to the innate distributional tendencies of economic growth addressed in the preceding section. As Ranieri and Ramos (2013, p. 5) assert, ‘poor people’s income [must] grow more than wealthier people’s income’ for growth to be classified as pro-​poor in the relative sense. This distinction is particularly important in a discussion of equitable economic growth. Generally speaking, conceptualizations of inclusive economic growth align (explicitly or implicitly) with either absolute or relative pro-​poor growth, in accordance with their view of levels of and changes in inequality. The conceptualization of equitable economic growth proposed later in this chapter is associated with a modified conceptualization of relative pro-​ poor growth.

Inclusive Growth Much of the literature on inclusive economic growth assesses what might be considered as three dimensions of inclusiveness.3 Firstly, a relevant dimension of the inclusiveness of growth relates to who benefits from (and, as discussed later, participates in) episodes of economic growth. Pro-​poor growth focuses on outcomes specifically for the poorest and most marginalized population groups and emphasizes the relationship between poverty and economic growth. While some conceptualizations of inclusive growth have retained this prioritization of poverty, inclusive economic growth is increasingly understood to both benefit and engage all of society—​ regardless of the initial level of income—​rather than only its most marginalized citizens. Klasen (2010, p. 2), for example, pointed out that inclusive growth benefits ‘all stripes of society, including the poor, near-​poor, middle-​income groups and even the rich’. The acceptance of inclusiveness referring to all of society is reflected in the recent definitions of inclusive growth adopted by the Asian Development Bank, the African Development Bank, and the OECD. In addition to a demographic dimension, conceptualizations of inclusive economic growth assess it in terms of outcomes or, more accurately, the distribution of its outcomes. Outcomes are often income related. For example, inclusive growth according to Habito (2009, p. 1) is economic growth (with gross domestic product its proxy) that directly benefits the poor and results in decreases in the level of poverty. Inclusiveness in this context implies that economic growth benefits not only those who are better off, but also the poor and marginalized of society. Retaining a focus on outcomes, yet defining outcomes in terms of opportunity rather than income specifically,4 Ali and Zhuang (2007) indicate that inclusive economic growth is fundamentally about the creation of opportunities—​opportunities in this context referring to employment opportunities—​and facilitation of widespread access to those opportunities. The inclusiveness of economic growth may be measured via a ‘social opportunity function’, as proposed by Ali and Son (2007). This function reflects both the average number of opportunities available to a population and the distribution of opportunities. In retaining a focus on the outcome of economic growth but emphasizing opportunity rather than income, Ali and Zhuang’s (2007) conceptualization of inclusive economic growth represents a significant departure from more traditional pro-​poor growth approaches (e.g. Ranieri and Ramos, 2013).

778   Rodríguez-Pose and Wilkie In addition to assessing the inclusiveness of growth in terms of its outcomes, several conceptualizations also consider the inclusiveness of the growth processes. Ianchovichina and Lundstrom (2009, p. 2) suggest that inclusive growth processes are marked by their inclusion of and contributions from ‘a large part of the country’s labour force’. Further reinforcing the importance of the inclusiveness of the growth process, the authors also note that achieving inclusiveness through redistribution does not constitute inclusive economic growth. It thus becomes immediately apparent from a cursory review of the various conceptualizations of both pro-​poor growth and inclusive economic growth that there are several factors that must be taken into consideration in the formulation of a robust, functional definition of equitable economic growth.

Defining Equitable Economic Growth In the simplest, most intuitive sense, equitable economic growth is economic growth from which all members, or certainly more members of society, accrue some benefit. The growth is equitable in the sense that the outcomes of growth are more evenly distributed across society. Unlike pro-​poor growth, which focuses exclusively on poverty and the poor and marginalized segments of society, equitable economic growth—​like more recent conceptualizations of inclusive economic growth (Klasen, 2010)—​is concerned with the outcomes of growth as they pertain to society in general. This is not to say that special consideration should not be made for the most marginalized groups of the population in light of the inherent disadvantages associated with marginalization and the prevalence of structural impediments inhibiting their capacity to benefit from economic growth (Kakwani and Pernia, 2000, p. 3). Equitable growth must permeate all layers of society without harming overall economic performance. Merely asserting that equitable economic growth is growth, the outcomes of which are realized by a broader portion of the population, offers little in the way of how this ambitious outcome is achieved. That is, it does not address the mechanisms through which benefits are delivered. Benefit, in the context of economic growth, refers to an increase in economic well-​being or income. The capacity to benefit directly from economic activity is understood to be contingent upon capacity to participate in income-​generating activities.5 Participation in economic activity refers to employment. Thus, if someone is employed and participating in an economic activity, they are receiving income and thereby benefiting from it. Equitable economic growth, hence, refers not only to the growth of the economy, but also to the creation of new employment opportunities, which allow more individuals to capture directly the benefits of increased economic activity, and the expansion of an economy. The creation of economic opportunity is not, however, sufficient to constitute equitable economic growth. Equitable economic growth, following the conceptualization of inclusive growth forwarded by Ali and Son (2007), must also involve facilitating widespread access to those opportunities. The quality of the employment opportunities has also been highlighted as particularly important in several conceptualizations of inclusive economic growth. Ianchovichina and Lundstrom (2009) emphasize the importance of generating both employment and

Pursuing Equitable Economic Growth in the Global South    779 productive employment. Similarly, Ali and Zhuang (2007) stress ‘decent employment’, citing problems of underemployment, especially in developing country contexts, as well as the link between decent work and productivity. Closely related to the notion of decent employment is that of formality. The acknowledgement of and differentiation between formal and informal employment is critically important for developing an ‘employment-​oriented’ conceptualization of equitable economic growth that is readily applicable to the Global South, given the prevalence of the informal sector in the developing world (Flodman Becker, 2004; Kessides, 2006). Informal employment is not commonly considered to be or associated with decent employment (International Labour Organization, 2002). Accordingly, advocacy for the creation of decent employment as the mechanism through which equitable economic growth is achieved would imply a focus on formal employment opportunities and the formal sector more broadly. Job creation in the informal sector in developing contexts, however, cannot be disregarded, given its magnitude and prominence. Kessides (2006, p. 22) observes that ‘the informal economy workforce is estimated to account for 78 per cent of non-​agricultural employment in Africa, 93 per cent of all new jobs created and 61 per cent of urban employment’. This prevalence, coupled with the absence of suitable regulatory frameworks and institutions to promote the formalization of the informal economy in many developing environments (Chen, 2012), gives rise to the realization that focusing solely on the creation of formal employment in the pursuit of equitable economic growth in developing urban contexts is not sufficient and neglects economic and social realities. The promotion of equitable economic growth must then reflect an awareness of informality and must include explicit measures to both empower individuals who are active in and reliant on the informal sector, as well as to promote institutional reform and upgrading to enable the formalization of informal economic activity. Together, such efforts will help to not only harness the economic potential of the informal economy, but also permit individuals to transition from informal employment into formal, decent employment and reap the socio-​economic benefits associated with doing so. It may then be concluded that equitable economic growth refers to more than the creation of employment opportunities. More precisely, it is the creation of decent, productive employment opportunities in both the formal and informal sectors, as well as the gradual formalization of informal economic activity via institutional reform. There are then, as Ali and Zhuang (2007) observe in their conceptualization of inclusive economic growth, two requisites for equitable economic growth. The first is long-​term sustainable economic growth. ‘[H]‌igh and sustainable growth is key to creating productive and decent employment opportunities’ (Ali and Zhuang, 2007, p. 12). The second is what they term ‘social inclusion’ (Ali and Zhuang, 2007, p. 13), referring to efforts to ensure that any potential barriers inhibiting individuals’ access to opportunities created via economic growth are alleviated. Barriers may relate to attributes and abilities of an individual that would make them either more or less able to realize a newly created opportunity (these are addressed principally through investment in education, health, and social services) or they may be more fundamental institutional or structural barriers (Ali and Zhuang, 2007). The creation of employment opportunities that enable more of the population to participate in economic activities ensures that continued economic growth is inclusive not only in the sense that it benefits a greater proportion of society, but also that the growth is more

780   Rodríguez-Pose and Wilkie participatory in nature (the inclusiveness of the outcomes and the process are inevitably intertwined). Additionally, it is perhaps more sustainable because increased employment reflects a more efficient use of local resources (Ali and Zhuang, 2007). A final point relates to our conceptualization’s alignment with absolute or relative pro-​ poor growth or, rather, its view of inequality. In the most literal sense, equitable economic growth implies that each member of society benefits identically from economic growth. Such an outcome would however, as Ranieri and Ramos (2013) point out, leave inequality unaffected and effectively perpetuate and reinforce the patterns of inequality in society. Inequality and exclusion are, as addressed, widespread in the Global South and in cities, in particular, presenting a host of challenges and potentially compromising the future growth of cities and countries. The reduction of inequality must therefore be a feature of equitable economic growth. Our conceptualization of equitable economic growth is thus aligned with relative pro-​ poor growth. However, as mentioned previously, benefits to the poorest and most marginalized must not come at the expense of the remainder of the population—​gains for one group at the expense of another would suggest possible inefficiency. Growth must raise all boats, although it will raise some to a greater extent than others. This stipulation is critical. Building on previous conceptualizations of inclusive economic growth,6 specifically their emphasis on broad societal engagement and benefit and on the creation of opportunity in the form of employment, we define equitable economic growth as: Long-​term sustainable economic growth that creates economic opportunity in the form of decent and productive employment in both the formal and informal sectors that may be accessed by all of society regardless of economic status, gender, or ethnicity, thus enabling all of society to both benefit directly from and participate in economic activity and future growth.

Equitable economic growth should also benefit the poorest and most marginalized segments in a society to a disproportionally greater extent and thus reduce inequality and exclusion, although not at the expense of the rest of the society. An advantage of this conceptualization of equitable economic growth is that it lends itself well to quantification. By prioritizing the creation and subsequent accessibility of employment opportunities, both of which are easily measured and monitored, the equity of economic growth can be assessed more or less quantitatively and objectively. Quantifying equitable economic growth as per the proposed definition will require the collection of data for various statistical indicators, including indicators of economic performance and growth (e.g. gross domestic product, gross national product, or gross value added); aggregate and industry-​or sector-​specific employment-​and labour force-​related indicators (long-​term, short-​term, and cyclical unemployment or employment, labour force turnover, and job creation); wage and income distribution-​related indicators; and absolute and relative poverty indicators and indicators of accessibility to basic services (e.g. health care, education, or sanitation). Finally, there are three additional points of considerable relevance to our conceptualization of equitable economic growth that must be emphasized. The first relates to its temporal dimension. The second relates to notions of sustainability. The third to the tensions and challenges associated with the pursuit of equitable economic growth.

Pursuing Equitable Economic Growth in the Global South    781 Efforts to promote equitable economic growth must adopt a longer-​term perspective. The time frame for equitable economic growth efforts will vary in accordance with the contextually tailored policies and strategies developed in light of the opportunities, challenges, and overall conditions that characterize heterogeneous environments (Department for International Development, 2014). Time frames should trend towards the longer rather than the shorter term.7 A longer-​term orientation is necessary because achieving equitable economic growth entails sustained economic growth, societal transformations, and more fundamental structural changes, as opposed to ‘quick fixes’. These more profound changes can only occur in the medium-​to-​long terms through a sustained and conscious effort led by governments, policymakers, and, often in many parts of the Global South, international organizations (Rodríguez-​Pose, 2013; Department for International Development, 2014). The outcomes of equitable economic growth-​oriented efforts should therefore only be reasonably expected and assessed with an awareness of the longer-​term nature of equitable economic growth. The failure to adopt a medium-​to-​long-​term perspective can result in the misdirection of efforts and resources, as well as the formulation of unrealistic expectations, both of which can undermine the achievement of equitable economic growth. Additionally, sustainability must be taken into account. Both urbanization and urban economic growth can impose considerable costs on the future potential for development of territories if a concerted effort is not made to address issues of sustainability and a holistic perspective that evaluates not only economic, but also social and environmental outcomes is not adopted (Turok and McGranahan, 2013, p. 479). Approaches and efforts designed to promote equitable economic growth must factor in sustainability to ensure that economic growth is viable over the longer term (Department for International Development, 2014) and that, following the Brundtland Commission (World Commission on Environment and Development, 1987), it is not achieved at the expense of the future economic growth and socio-​economic well-​being. Finally, any discussion of equitable or inclusive economic growth inevitably raises questions concerning the compatibility of the pursuits of equity and economic growth. More specifically, it is entirely appropriate to consider whether equitable economic outcomes come at the expense of economic growth or whether there is a trade-​off between equity and growth. It is conceivable, depending on the approach adopted, that equity and growth are not automatically or immediately reconcilable. Accordingly, policymakers must be aware of any potential tension between these two objectives and design and implement policies and strategies that reflect this awareness. Within our definition, which stresses the creation of and subsequent facilitation of access to economic opportunity, equity and growth are not only compatible, but are actually complementary. In fact, the sought-​after equitable outcomes cannot be attained without robust economic growth (Ali and Zhuang, 2007, p. 12). This perspective is premised primarily on the acceptance that equity is not achieved via the ex post redistribution of economic growth outcomes. Rather, equity is achieved via the mobilization of available (human) resources—​ in itself consistent with the pursuit of greater economic efficiency—​and by ensuring that opportunity exists (and is readily accessible) for individuals to participate in and therefore directly benefit from and even make a substantive contribution to economic growth. Increased labour force participation—​ the main mechanism through which equity is achieved—​ultimately supports and could act as a catalyst for continued, more economically sustainable growth.

782   Rodríguez-Pose and Wilkie

Operationalizing ‘Equitable Economic Growth’ in the Global South With the conceptualization and definition presented in the preceding section in mind, our focus must now shift towards the operationalization of equitable economic growth in developing and emerging contexts. There are two points of particular relevance here. The first concerns the suitability of certain types of strategic approaches for the pursuit of equitable economic growth. The second relates to challenges associated with the development, implementation, and execution of equitable economic development approaches in the Global South.

Spatially Targeted Approaches for the Pursuit of Equitable Economic Growth While differences inevitably exist across the specific policies and strategies adopted in the pursuit of economic growth and development in heterogeneous territories, most of these approaches may be broadly classified as either ‘place-​based’ (spatially targeted) or ‘spatially blind’ (Barca et al., 2012).8 Both place-​based and spatially blind approaches represent a departure from the generally top-​down, supply-​side-​oriented policy approaches favoured in the past (Barca et  al., 2012). That is, however, where the similarities between the two paradigms stop. Spatially blind approaches are based on the notion that the processes that shape growth and development are more or less uniform across space and that identifying the most viable opportunities for growth and development is a challenging if not impossible exercise. This suggests that resources (financial and otherwise) are better spent ensuring that individuals, regardless of where they live, have the capacity to both seek out and ultimately capitalize upon economic opportunities, wherever they may be. Place-​based approaches, however, focus specifically on territories (and, by extension, the individuals that occupy them) and are founded on the belief that there is a need to tailor policies and strategies to address unique contextual conditions and reflect the opportunities, challenges, and resources that characterize a given territory’s ability to induce growth and development that benefits its residents (Barca et al., 2012).9 Either of these policy approaches could be employed in the pursuit of equitable economic growth—​strong cases can be made for the utility of both, as evidenced by the endorsement of both practices by various international organizations. That said, spatially targeted approaches may be the most suitable avenue for the achievement of equitable economic growth in developing and emerging contexts. The advocacy for territorially specific policies in the pursuit of equitable economic growth in the Global South is fundamentally based on the heterogeneity of developing environments and the relevance of unique contextual conditions to the outcomes of development strategies and policies (Pike et al., 2006; Ascani et al., 2012). Developing countries and regions vary tremendously across a host of dimensions (e.g. economic, social, political, and institutional),

Pursuing Equitable Economic Growth in the Global South    783 and no two environments are identical. If contextual conditions shape processes of growth and development—​an assertion that is validated by the incapacity of traditional, aspatial policies to deliver widespread success across diverse socio-​economic contexts (Pike et al., 2006; Ascani et al., 2012)—​policies seeking to impel growth and development, equitable or otherwise, must account for and, indeed, be tailored to the contextual conditions of every place. In that respect, it would seem as though spatially targeted approaches are the only viable option for the pursuit of equitable economic growth and for coping with the diversity of contextual conditions that exist across the Global South. Within the broader classification of spatially targeted approaches, local economic development (LED) practices enjoy frequent and increasingly widespread support. There is ample evidence that confirms that LED approaches have the capacity to propel both economic growth and more holistic socio-​economic development across developing countries (see e.g. Potter et al., 1999; European Commission, 2008; Rodríguez-​Pose and Palavicini-​Corona, 2013). It is important to note that LED approaches, like any development strategy, are by no means assured of success. That said, there is cause for cautious optimism about their potential to affect change and deliver growth and development. Rodríguez-​Pose and Tijmstra (2009, p. 112) capture this view, stating that although LED approaches ‘may not be the panacea for development, an increasing number of cases are showing that they may lead to a greater adaptability and sustainability in changing economic conditions’. There is no universally agreed upon definition of LED (Rogerson and Rogerson, 2010; Akudugu and Laube, 2013).10 That said, the various conceptualizations of LED tend to share some similarities, the most notable of which are an implicit or explicit emphasis on the engagement and participation of stakeholders from all sectors; local leadership and ownership; the mobilization and (where relevant) sustainable exploitation of local resources and a marked territorial orientation (Rogerson and Rogerson, 2010). These defining features lend LED a distinctly different character to both the top-​down approaches relied upon in the past and to more modern spatially blind policies and make them more amenable to delivering equitable economic growth in a diversity of developing contexts than alternative approaches. From a theoretical perspective, the most prominent reason for the endorsement of LED strategies, specifically in the context of equitable urban economic growth in developing environments, is that they mobilize and capitalize upon local potential to create a competitive advantage upon which a given territory, and the firms that occupy it, may rely in increasingly open and competitive regional, national, and global economies (Vázquez-​Barquero, 1999; Pike et al., 2006). The centrality of local potential implies that, as Vázquez-​Barquero (1999, p. 85) notes, ‘economic development does not necessarily have to be polarized and focused in large cities’ or, more broadly, focused only in a handful of select territories. Because LED approaches rely on the mobilization of local potential and the development of approaches that are reflective of both this local potential and other local contextual conditions, they are conceivably viable across territories of various levels of ex ante favourability—​developed and less developed, urban and non-​urban. Moreover, it has been suggested that outcomes of LED approaches may also be more economically sustainable or enduring as a result of their embedding of economic activity rather than simply attracting it in a given territory and the extent to which they target, and work within the confines of, fundamental contextual conditions with a view to ‘[improve] the productive context’ (Vázquez-​Barquero, 1999, p. 84).

784   Rodríguez-Pose and Wilkie

Executing Equitable Economic Growth Strategies in the Global South: Mitigating Challenges and Overcoming Obstacles The pursuit of equitable economic growth in developing contexts faces considerable challenges in the design and implementation of policies. The nature of the barriers faced by economic growth and development strategies will vary greatly depending on the starting conditions of the territories in which they are implemented. There are, however, two related pitfalls to which strategic approaches in developing contexts seem especially susceptible—​ technical capacity constraints and financial constraints. There is ample evidence to suggest that technical capacity constraints, that is, a lack of expertise, practical experience, and knowledge, and financial constraints, that is, a lack of financial resources attributable to perhaps insufficient political commitment or the failure to establish partnerships in an effort to mobilize financial resources, can undermine the implementation and, in some cases, even the initial formulation of economic growth approaches across the Global South. The African experience perhaps best exemplifies this argument. In South Africa, for example, the ‘mixed or uneven’ (Rogerson, 2014, p. 215) outcomes of urban growth strategies employed across the country are commonly attributed to debilitating constraints in certain contexts and sufficient availability of necessary resources in others (Nel and Rogerson, 2007; Nel et al., 2009). In other words, sufficiently capacitated urban areas, such as Cape Town, Johannesburg, or Durban, have been able to develop and then implement pro-​growth, competitiveness-​oriented LED approaches that resulted in the achievement of ‘a level of systemic competitiveness that is relatively high, even compared with other middle income countries’ (Meyer-​Stammer, 2008, p. 15). By contrast, as Nel and Rogerson (2007, p. 6) observe, ‘success appears “sporadic” ’ outside of these sufficiently capacitated environments. The detrimental effects of various constraints on the outcomes of urban policies are equally visible outside of South Africa. In parts of Ghana, the absence of personnel with sufficient policy knowledge and experience and a lack of funding and inadequate control over the expenditure of the minimal financial resources available, coupled with challenges associated with aligning the interests of various parties and establishing a single vision, has severely hampered policy implementation (Akudugu and Laube, 2013). The consequences of financial constraints are also evident in Zambia, where a lack of financial resources (as well as technical capacity constraints) has inhibited the implementation of local development strategies and action (Hampwaye, 2008; Hampwaye and Rogerson, 2010). It is important to note that these constraints can be overcome and that there are several options for doing so. These include the formation of public–​private partnerships that permit both the distribution of the financial burden of development projects or initiatives across more, perhaps better-​resourced, actors, as well as the sourcing of technical expertise from the private sector where skills tend to be more abundant (see e.g. Gibb and Nel, 2007); the establishment of local economic development agencies composed of sufficiently trained personnel with the capabilities and expertise necessary to design and execute development strategies (Rogerson, 2011; Lawrence, 2013); or capacity-​building initiatives that

Pursuing Equitable Economic Growth in the Global South    785 provide those tasked with overseeing development strategies with the skills necessary to do so (Rodríguez-​Pose and Tijmstra, 2007). Consequently, the identification of and consider­ ation for contextually imposed constraints as an element of the policy planning and design process, and the incorporation of initiatives to address them, is essential in order to ensure that what may otherwise be a sound strategy is not derailed by what would seem relatively easily foreseen challenges.

Concluding Remarks: Where Do We Go From Here? This chapter forays into the development of an understanding of equitable economic growth. It has aimed to provide a holistic, employment-​oriented conceptualization of equitable economic growth on the basis of previous research, as well as basic insights into the operationalization of equitable economic growth focusing specifically on the suitability of different strategic approaches for its pursuit and eventual achievement. In doing so, it has further opened the door for subsequent research and exploration. The chapter also reflects the greater interest in pursuing equitable and/​or inclusive economic growth displayed by various regional institutions and international organizations. The African Development Bank Group, the Asian Development Bank, the Development Bank of Latin America, and the European Commission, for example, have all made a concerted effort to conceptualize inclusive economic growth and have prioritized its pursuit in their respective strategic plans. Similarly, the OECD, the World Bank, and the United Nations Development Programme, among others, have engaged substantively with the notion of inclusive economic growth and have directed time and resources to its study and promotion. There remains, however, a need for research and investigations of a more academic nature to complement the robust, but not yet sufficient, work of the aforementioned policy-​oriented institutions. More precisely, further theoretical engagement with the notion of equitable economic growth will be necessary in order to add greater depth and nuance to our collective understanding of economic development across the world and particularly in the Global South. Subsequent research should follow the work of Rauniyar and Kanbur (2010), McKinley (2010) and Klasen (2010), and consider ways in which equitable growth may be linked to or could promote more equitable or inclusive economic development. Future research must also adopt a conceptualization of equitable economic growth that targets specific goals and leads to viable strategies and measurable outcomes. The conceptualization provided in this chapter represents a starting-​up point and can serve as a base for future empirical approaches to explore the dynamics of, and the multitude of factors that shape, processes of equitable economic growth across a diversity of geographic and socio-​economic contexts. Further theoretical and empirical research of the sorts highlighted here will be absolutely essential not only to advance our scholarly understanding of equitable economic growth, but also, and more importantly, for the eventual formulation of evidence-​based policies that promote a more equitable brand of economic growth in the Global South and beyond.

786   Rodríguez-Pose and Wilkie

Acknowledgements This chapter draws heavily on the research on equitable economic growth in cities conducted by both authors at the request of the Cities Alliance. The findings of this research can be found in Rodríguez-​Pose and Wilkie (2015).

Notes 1. Pike et al. (2007, p. 1260) distinguish between quantitative and qualitative dimensions of development. The former ‘concern numeric measures, for example, a per capita growth rate of gross domestic product’. The latter ‘relates to the nature of local and regional development, for example the sustainability (economic, social, environmental) and forms of growth, the type and “quality” of jobs, the embeddedness and sustainability of investments, and the growth potential, sectoral mix and social diversity of new firms’. 2. Higher Gini coefficients correspond to higher levels of inequality. The Gini coefficient is by no means a perfect indicator of interpersonal inequality. That said, it is a useful tool for shaping indicative insights into patterns of inequality to inform the remainder of this chapter). 3. Recent definitions of equitable, inclusive, and/​or sustainable growth have tended to prefer the term ‘inclusive’. 4. Opportunity and income, however, are intimately related, as addressed in the next sections. 5. One may benefit more indirectly from economic growth through income redistribution. The achievement of a more equitable or inclusive distribution of the outcomes of economic growth through redistribution does not, however, constitute inclusive growth (Ianchovichina and Lundstrom, 2009). 6. We have limited the analysis to equitable economic growth rather than adopting a more holistic perspective that considers equitable economic development. A conceptualization of equitable economic development, derived from our conceptualization of equitable economic growth, follows Rauniyar and Kanbur (2010), McKinley (2010), and Klasen (2010) in incorporating ‘non-​income dimensions’ (Ranieri and Ramos, 2013, p. 8), including widespread access to basic services. A more robust engagement with the notion of equitable economic development is beyond the scope of this chapter but will likely be an integral part of future equitable economic growth-​related efforts and exercises. 7. The Department for International Development (2014) acknowledges that there are exceptional contexts and circumstances where a short-​term orientation may be appropriate. It is implied, however, that the short-​term focus should evolve into a longer-​term one as conditions improve. 8. Place-​based policies are most commonly associated with the Barca (2009) report entitled ‘An Agenda for a Reformed Cohesion Policy’ and the OECD (2009) report How Regions Grow, whereas spatially blind policies are often associated with The World Bank’s World Development Report—​Reshaping Economic Geography (2009b). 9. For a concise, comprehensive description and comparison of placed-​based and spatially blind approaches, see Barca et al. (2012).

Pursuing Equitable Economic Growth in the Global South    787 10. The International Labour Organization (2006, p. 2) defines LED as a ‘participatory development process that encourages partnership arrangements between the main private and public stakeholders of a defined territory, enabling the joint design and implementation of a common development strategy, by making use of the local resources and competitive advantage in a global context, with the final objective of creating decent jobs and stimulating economic activity’.

References African Development Bank Group (2013). At the Center of Africa’s Transformation (Abidjan: African Development Bank Group). Aghion, P. and Bolton, P. (1997). ‘A theory of trickle-​down growth and development’. The Review of Economic Studies 64: 151–​172. Akudugu, J.A. and Laube, W. (2013). ‘Implementing local economic development in Ghana: multiple actors and rationalities’. SEF Working Paper Series 113 (Bonn: Centre for Development Research, University of Bonn). Ali, I. and Son, H.H. (2007). ‘Measuring inclusive growth’. Asian Development Review 24: 11–​31. Ali, I. and Zhuang, J. (2007). ‘Inclusive growth toward a prosperous Asia: policy implications’. ERD Working Paper Series 97 (Manila: Asian Development Bank). Arndt, H.W. (1983). ‘The “trickle-​down” myth’. Economic Development and Cultural Change 32: 1–​10. Ascani, A., Crescenzi, R., and Iammarino, S. (2012). ‘Regional economic development:  a review’. SEARCH Working Paper WP1/​03. Asian Development Bank (2008). Strategy 2020—​The Long-​Term Strategic Framework of the Asian Development Bank (Manila: Asian Development Bank). Asian Development Bank (2012). Asian Development Outlook 2012, Confronting Rising Inequality in Asia. Asian Development Outlook (Mandaluyong City: Asian Development Bank) Atkinson, A.B., Piketty, T., and Saez, E. (2011). ‘Top incomes in the long run of history’. Journal of Economic Literature 49: 3–​7 1. Baker, J.L. (2008). ‘Urban poverty:  a global view’. Urban Papers 5 (Washington, DC:  The World Bank). Barca, F. (2009). ‘An agenda for a reformed cohesion policy: a place-​based approach to meeting European Union challenges and expectations’. Independent report prepared at the request of Danuta Hübner, Commissioner for Regional Policy (Brussels: European Commission). Barca, F., McCann, P., and Rodríguez-​Pose, A. (2012). ‘The case for regional development intervention:  Place-​based versus place-​neutral approaches’. Journal of Regional Science 52: 134–​152. Berg, A. and Ostry, J.D. (2011). ‘Inequality and Unsustainable Growth: Two Sides of the Same Coin?’ International Monetary Fund Staff Discussion Note http://​www.imf.org/​external/​ pubs/​ft/​sdn/​2011/​sdn1108.pdf (last accessed 24 April 2017). Berg, A., Ostry, J.D., and Zettelmeyer, J. (2008). ‘What makes growth sustained?’ IMF Working Papers No. WP/​08/​59. Bourguignon, F. (2015). The Globalisation of Inequality (Princeton, NJ: Princeton University Press). Brückner, M. (2012). ‘Economic growth, size of the agricultural sector, and urbanization in Africa’. Journal of Urban Economics 71: 26–​36.

788   Rodríguez-Pose and Wilkie Bryceson, D.F., Gough, K.V., Rigg, J., and Agergaard, J. (2009). ‘Critical commentary. The World Development Report 2009’. Urban Studies 46: 723–​738. Chen, M.A. (2012). ‘The informal economy:  definitions, theories and policies’. WIEGO Working Papers No. 1 (Cambridge, MA: WIEGO). Cingano, F. (2014). ‘Trends in income inequality and its impact on economic growth’. OECD Social, Employment and Migration Working Papers (Paris:  Organisation for Economic Co-​operation and Development). Cistulli, V., Rodríguez-​Pose, A., Escobar, G., Marta, S., and Schejtman, A. (2014). ‘Addressing food security and nutrition by means of a territorial approach’. Food Security 6: 879–​894. Department for International Development (2014). Defining and Delivering Inclusive Growth (London: DFID). Dobbs, R., Smit, S., Remes, J., Manyika, J., Roxburgh, C., and Restrepo, A. (2011). Urban World:  Mapping the Economic Power of Cities (San Francisco, CA:  McKinsey Global Institute). Dunford, M. (2005). ‘Growth, inequality and cohesion:  a comment on the Sapir report’. Regional Studies 39: 972–​978. Duranton, G. (2000). ‘Urbanization, Urban Structure, and Growth’ in J.M. Huriot and J.F. Thisse (eds) Economics of Cities: Theoretical Perspectives, pp 290–​317 (Cambridge: Cambridge University Press). European Commission (2008). Changing the World Locally—​25 Success Stories of Development Cooperation at Local Level (Brussels: European Commission). European Commission (2010). Communication from the Commission Europe 2020—​A Strategy for Smart, Sustainable and Inclusive Growth (Brussels: European Commission). Fan, S., Kanbur, R., and Zhang, X. (2011). ‘China’s regional disparities: experience and policy’. Review of Development Finance 1: 47–​56. Fay, M. and Opal, C. (1999). ‘Urbanization without growth: a not-​so-​uncommon phenomenon’. Policy Research Working Papers (Washington, DC: The World Bank). Flodman Becker, K. (2004). The Informal Economy (Stockholm:  Swedish International Development Cooperation Agency). Fujita, M. and Thisse, J.F. (2002). Economics of Agglomeration (Cambridge:  Cambridge University Press). Gibb, M. and Nel, E. (2007). ‘Small town redevelopment: the benefits and costs of local economic development in Alicedale’. Urban Forum 18: 69–​84. Glaeser, E. (2011). Triumph of the City: How Our Greatest Invention Makes Us Richer, Smarter, Greener, Healthier and Happier (New York: Penguin Press). Grosse, M., Harttgen, K., and Klasen, S. (2008). ‘Measuring pro-​poor growth in non-​income dimensions’. World Development 36: 1021–​1047. Gustafsson, B. and Quheng, D. (2013). ‘A New Episode of Increased Urban Income Inequality in China’ in S. Li, H. Sato, and T. Sicular (eds) Rising Inequality in China, pp. 255–​288 (Cambridge: Cambridge University Press). Gustafsson, B. and Sai, D. (2013). ‘Unemployment and the Rising Number of Nonworkers in Urban China: Causes and Distributional Consequences’ in S. Li, H. Sato, and T. Sicular (eds) Rising Inequality in China, pp. 289–​331 (Cambridge: Cambridge University Press). Habito, C.F. (2009). ‘Patterns of inclusive growth in developing Asia:  insights from an enhanced growth-​poverty elasticity analysis’. ADBI Working Paper Series 145 (Tokyo: Asian Development Bank Institute). Hampwaye, G. (2008). ‘Local economic development in the city of Lusaka, Zambia’. Urban Forum 19: 187–​204.

Pursuing Equitable Economic Growth in the Global South    789 Hampwaye, G. and Rogerson, C.M. (2010). ‘Economic restructuring in the Zambian Copperbelt: local responses in Ndola’. Urban Forum 21: 387–​403. Hirsch, B.T. (1980). ‘Poverty and economic growth: has trickle down petered out?’ Economic Inquiry XVII: 151–​158. Ianchovichina, E. and Lundstrom, S. (2009). ‘Inclusive growth analytics—​framework and application’. Policy Research Working Paper 4851 (Washington, DC: The World Bank). International Labour Organization (2002). Report VI—​Decent Work and the Informal Economy (Geneva: ILO). International Labour Organization (2006). A Local Economic Development Manual for China (Geneva: ILO). Jian, T., Sachs, J.D., and Warner, A.M. (1996). ‘Trends in regional inequality in China’. China Economic Review 7: 1–​21. Kakwani, N. and Pernia, E.M. (2000). ‘What is pro-​poor growth?’ Asian Development Review 18: 1–​16. Kanbur, R. and Venables, A.J. (2005). Spatial Inequality and Development (Oxford:  Oxford University Press). Kanbur, R. and Zhang, X. (1999). ‘Which regional inequality? The evolution of rural–​urban and inland–​coastal inequality in China from 1983 to 1995’. Journal of Comparative Economics 27: 686–​701. Kanbur, R. and Zhang, X. (2005). ‘Fifty years of regional inequality in china: a journey through central planning, reform, and openness’. Review of Development Economics 9: 87–​106. Kessides, C. (2006). The Urban Transition in Sub-​Saharan Africa—​Implications for Economic Growth and Poverty Reduction (Washington, DC: Cities Alliance). Klasen, S. (2010). ‘Measuring and monitoring inclusive growth:  multiple definitions, open questions, and some constructive proposals’. ABD Sustainable Development Working Paper Series 14 (Manila: Asian Development Bank). Knight, J. (2014). ‘Inequality in China: an overview’. World Bank Research Observer 29: 1–​19. Kuznets, S. (1955). ‘Economic growth and income inequality’. The American Economic Review 45: 1–​28. Lawrence, F. (2013). ‘The role of local economic development agencies in the South African local economic development landscape’. Urban Forum 24: 523–​541. McKinley, T. (2010). ‘Inclusive growth criteria and indicators: an inclusive growth index for diagnosis of country progress’. ABD Sustainable Development Working Paper Series 14 (Manila: Asian Development Bank). Meyer-​Stammer, J. (2008). ‘Systemic competitiveness and local economic development’ www. meyer-​stamer.de/​2008/​Systemic+LED_​SouthAfrica.pdf (last accessed 24 April 2017). Milanovic, B. (2005). Worlds Apart: Measuring International and Global Inequality (Princeton, NJ: Princeton University Press). Milanovic, B. (2009). ‘Global inequality recalculated’. Policy Research Paper 5061, September (Washington, DC: World Bank). Milanovic, B. (2011). The Haves and the Have Nots (New York: Basic Books). Moreno, E.L., Oyeyinka, O., and Mboup, G. (2010). State of the World’s Cities 2010/​2011—​ Bridging the Urban Divide (Nairobi: United Nations Human Settlements Programme). Nel, E., Binns, T., and Bek, D. (2009). ‘Misplaced expectations? The experience of applied local economic development in Post-​Apartheid South Africa’. Local Economy 24: 224–​237. Nel, E. and Rogerson, C.M. (2007). ‘Evolving local economic development policy and practice in South Africa with special reference to smaller urban centres’. Urban Forum 18: 1–​11.

790   Rodríguez-Pose and Wilkie Obeng-​Odoom, F. (2012). ‘Neoliberalism and the urban economy in Ghana: urban employment, inequality, and poverty’. Growth and Change 43: 85–​109. Organisation for Economic Co-​operation and Development (2009). How Regions Grow (Paris: OECD Publishing). Organisation for Economic Co-​operation and Development (2011). Divided We Stand—​Why Inequality Keeps Rising (Paris: OECD Publishing). Organisation for Economic Co-​operation and Development (2012). Promoting Growth in all Regions (Paris: OECD). Organisation for Economic Co-​operation and Development (2014). Making Inclusive Growth Happen (Paris: OECD). Organisation for Economic Co-​operation and Development (2015). In it Together: Why Less Inequality Benefits All (Vol. 84) (Paris: OECD Publishing). Ostry, J.D., Berg, A., and Tsangarides, C.G. (2014). ‘Redistribution, inequality and growth’. IMF Staff Discussion Notes (Washington, DC: International Monetary Fund). Parkinson, M., Meegan, R., Karecha, J., Evans, R., Jones, G., Tosics, I., and Hall, P. (2012). Second Tier Cities in Europe:  In an Age of Austerity Why Invest Beyond the Capitals (Liverpool: ESPON and Institute of Urban Affairs, Liverpool John Moores University). Pike, A. Rodríguez-​Pose, A., and Tomaney, J. (2006). Local and Regional Development (Abingdon: Routledge). Pike, A., Rodríguez-​Pose, A., and Tomaney, J. (2007). ‘What kind of local and regional development and for whom?’ Regional Studies 41: 1253–​1269. Piketty, T. (2014). Capital in the 21st Century (Cambridge, MA: Harvard University Press). Potter, J., Walsh, J., De Varine, H., and Barreiro, F. (1999). Best Practices in Local Development (Paris: Local Employment and Economic Development Programme, OECD). Quigley, J.M. (1998). ‘Urban diversity and economic growth’. The Journal of Economic Perspectives 12: 127–​138. Ranieri, R. and Ramos, R.A. (2013). ‘Inclusive growth: building up a concept’. Working Paper 104 (Brasilia: International Policy Centre for Inclusive Growth). Rauniyar, G. and Kanbur, R. (2009). ‘Inclusive growth and inclusive development: a review and synthesis of Asian Development Bank literature’. ADB Occasional Papers 8 (Manila: Asian Development Bank). Rauniyar, G. and Kanbur, R. (2010). Inclusive Development: Two Papers on Conceptualization, Application and the ADB Perspective (Manila: Asian Development Bank). Roberts, B.H. (2014) Managing Systems of Secondary Cities—​Policy Responses in International Development (Brussels: Cities Alliance and United Nations Office for Project Services). Rodríguez-​Pose, A. (2013) ‘Do institutions matter for regional development?’ Regional Studies 47: 1034–​1047. Rodríguez-​Pose, A. and Palavicini-​Corona, E.I. (2013). ‘Does local economic development really work? Assessing LED across Mexican municipalities’. Geoforum 44: 303–​315. Rodríguez-​Pose, A. and Tijmstra, S. (2007). ‘Local economic development in sub-​Saharan Africa’. Environment and Planning C: Government and Policy 25: 516–​536. Rodríguez-​Pose, A. and Tijmstra, S. (2009). ‘On the emergence and significance of local economic development strategies’. CAF Working Papers 2009/​07 (Caracas: CAF). Rodríguez-​Pose, A. and Wilkie, C. (2015). ‘Conceptualising equitable economic growth in cities’. Cities Alliance Discussion Paper No. 2. Rogerson, C.M. (2011). ‘Tracking local economic development policy and practice in South Africa, 1994–​2009’. Urban Forum 22: 149–​168.

Pursuing Equitable Economic Growth in the Global South    791 Rogerson, C.M. (2014). ‘Reframing place-​based economic development in South Africa: the example of local economic development’. Bulletin of Geography, Socio-​Economic Series 24: 203–​218. Rogerson, C.M. and Rogerson, J.M. (2010). ‘Local economic development in Africa: Global context and research directions’. Development Southern Africa 27: 465–​480. Rozelle, S. (1994). ‘Rural industrialisation and increasing inequality:  emerging patterns in China’s reforming economy’. Journal of Comparative Economics 19: 362–​391. Sala-​i-​Martin, X. (2002). ‘The disturbing “rise” of global income inequality’. NBER Working Paper, 72 (August 2001). Sankhe, S., Vittal, I., Dobbs, R., Mohan, A., and Gulati, A. (2010). India’s Urban Awakening: Building Inclusive Cities, Sustaining Economic Growth (Delhi: McKinsey Global Institute). Thorat, S. and Dubey, A. (2012). ‘Has growth been socially inclusive during 1993-​94–​2009-​10?’ Economic & Political Weekly 47: 43–​53. Thornton, J.R., Agnello, R.J., and Link, C.R. (1978). ‘Poverty and economic growth:  trickle down has petered out’. Economic Inquiry 18: 159–​163. Tripathi, S. (2013). ‘Is urban economic growth inclusive in India?’ The Journal of Applied Economic Research 7: 507–​539. Turok, I. and McGranahan, G. (2013). ‘Urbanization and economic growth: the arguments and evidence for Africa and Asia’. Environment and Urbanization 25: 465–​482. Vázquez-​Barquero, A. (1999). ‘Inward investment and endogenous development. The convergence of the strategies of large firms and territories?’ Entrepreneurship and Regional Development 11: 79–​83. Venables, A.J. (2005). ‘Spatial disparities in developing countries: cities, regions, and international trade’. Journal of Economic Geography 5: 3–​21. World Bank (2009a). What is inclusive growth? (Washington, DC: The World Bank). World Bank (2009b). World Development Report 2009—​Reshaping Economic Geography (Washington, DC: The World Bank). World Commission on Environment and Development (1987). Our Common Future (Oxford: Oxford University Press). Yang, D.T. (2002). ‘What has caused regional inequality in China?’ China Economic Review 13: 331–​334. Zhuang, J. (ed.) (2010). Poverty, Inequality, and Inclusive Growth in Asia: Measurement, Policy Issues, and Country Studies (Manila: Asian Development Bank).

Chapter 42

J ust Grow th: St rat e g i e s f or Grow th wi t h  E qu i t y Karen Chapple The Emergence of Strategies for Growth with Equity Against a backdrop of increasing income inequality, accompanied by the ongoing rescaling of the state, recent decades have seen the rise of a dizzying array of strategies to achieve more equitable and inclusive local and regional growth. In response to a changing federal role—​most obviously in North America, but also in the European Union, Latin America, and Asia—​actors from civil society, government, and business, working either through consensus or political conflict, have enacted more inclusive policies and programmes. In some cases, these actors are essentially advancing an agenda for city and regional competitiveness with the idea that benefits may trickle down to the less advantaged (Brenner and Wachsmuth, 2012). In others, strategies for inclusion are explicit and intentional, with efforts often initiated from the grass roots. This development has in some ways caught economic geography off guard. Through much of the twentieth century, theories of regional economic development sought to explain why some regions prosper and others falter. Although points of contention remain, few would disagree that three keys to growth are a competitive export base, entrepreneurship and innovation, and urban agglomeration. Yet, these understandings crystallized during a bygone period of global development, the post-​World War II decades of great prosperity. In the USA in particular, with the labour-​management bargain in place providing a ‘productivity dividend’ for workers, growth meant a burgeoning middle class. But in today’s context of rising inequality, across both advanced industrialized and developing countries, these theories fall short in their implied prescriptions for action. While exports, entrepreneurship, and agglomeration all grow the pie or increased incomes, they have struggled at incorporating the disadvantaged into the mainstream. This, then, has challenged the field of economic geography itself, which is continually pushed by critical geography to acknowledge these marginalized voices.

Just Growth   793 This chapter describes two basic approaches to growth with equity—​defined here, as in most of the literature, as economic inclusion (rather than racial or social justice)—​that cities and regions are adopting, depending on the regional and national context. The first type focuses on regional economic growth: first growing the pie, and then redistributing the benefits. Strategies for inclusion might seek ways of linking disadvantaged groups to growth sectors. For example, regions might develop business clusters connected to workforce development programmes, or facilitate real-​estate development, but with community benefits such as construction jobs. Countries guided, in part, by neo-​liberal ideology, like the USA and the UK, tend to follow this model at both the national and regional scales. The second type attempts to reduce inequality directly, with the expectation of ‘just growth’: the idea is to create a floor for equity by providing a safety net and/​or building new capabilities, which will increase growth at the same time (Pastor et al., 2000; Eberts et al., 2006; Benner and Pastor, 2012). Inclusion here typically builds on a base of local participation, supported by a strong national welfare state. These strategies may supplement or distort markets, for instance by supporting local entrepreneurs through government purchasing set-​asides, or creating a minimum wage floor. Countries in Europe, particularly Scandinavia, and Latin America tend to adopt this approach, but, increasingly, regions within advanced industrial countries do as well (Benner and Pastor, 2012). Although regional competitiveness matters—​stronger markets can afford to be more inclusive—​the strategies adopted depend on the unique characteristics that each region develops over time based on its economic strengths and institutional environment (Storper, 2013). Interaction processes within the economy—​networks, labour markets, knowledge transfers, and so forth—​create a complex web of institutions that shape choices about inclusion. Increasingly, heightened pressures on disadvantaged communities from global climate change are also affecting how regions pursue inclusion. The following first examines the overall context for the emergence of strategies for equit­ able growth, that is, the rise of inequality. After providing a brief overview of theories of regional and local economic development, the chapter describes the two types of strategies. A conclusion examines the significance for economic geography and the likely evolution of the field.

Context: Inequality and Growth Intentional strategies for growth with equity began emerging in developed countries with deindustrialization, as economies experienced a new surge in inequality. The rise of low-​ wage work exacerbated inequality, with uneven effects across regions. With the rescaling of the state leading to new openings for governance, many cities and regions began experimenting with new approaches to equitable growth. Ironically, the model that many pointed to came from developing or newly industrialized countries, particularly in Latin America but also South East Asia, where high growth rates were coupled with a strong welfare state (Benner and Pastor, 2012). The following looks first at the context for rising inequality, and then explores the underpinnings of the ‘just growth’ paradigm.

794   Chapple

Restructuring and Inequality As the global economy continually reinvents itself, cities, and regions restructure themselves, shifting to one economic base from another. When regions succeed in restructuring from manufacturing to services, they tend to experience increasing income inequality (Harrison and Bluestone, 1988). This pattern is particularly pronounced in countries such as the USA, Canada, and the UK, which have also experienced the retrenchment of government in the neo-​liberal era. Deindustrialization first hit the leading manufacturing regions of the USA and other industrialized countries in the 1970s, and now seems to be affecting other areas such as Latin America, as well, albeit for different reasons (Bluestone and Harrison, 1982; Kollmeyer, 2009; Brady et al., 2011). By the end of the twentieth century, a new phase of restructuring had also led to the information age, with a new informational mode of development (Castells, 1996). This meant a shift in job quality as well, as regions have experienced a net gain of low-​ wage work. Scholars have invoked technological, as well as institutional, explanations for the shift. The skill-​biased technological change perspective argues that the introduction of new technologies (e.g. computers) resulted in rising demand for college-​educated workers—​ and real-​wage declines for less educated workers (Katz and Murphy, 1992). The institutional explanation focuses on the transformation of the governance of the labour market. The so-​ called Fordist era, roughly 1946–​1973, had brought rising real wages, productivity growth, oligopolistic competition among large firms, and relative labour peace (Osterman, 1999). Accompanying this was a set of norms shared by employers and a largely unionized industrial workforce that embraced workplace practices such as internal job ladders and a productivity ‘dividend’ to workers. When these institutions broke down in the 1970s, following the fall of the Bretton-​Woods system, income inequality increased. Corporations began regularly shedding large portions of their core workforce, increasingly relying upon a flexible, contingent pool of workers whose earnings are forced down by stiff competition, often from abroad (Appelbaum et al., 2003). Most economists did not anticipate the rise of income inequality. According to the Kuznets curve, inequality should rise as economic growth accelerates, and then will decrease as average incomes begin gaining from the new prosperity. From the perspective of Simon Kuznets—​one from the developed world in the 1950s—​this seemed like a reasonable hypothesis, and many countries in continental Europe (as well as Japan) have shown declining income inequality as expected (Alvaredo et al., 2013). But, in reality, countries have followed a variety of development paths (Stern, 2004). Although the twenty-​first century has brought about declining inequality in many countries with high growth rates, high levels of inequality have persisted in Latin America, Southern Africa, and, most recently, China (Economic Commission for Latin America and the Caribbean, 2010; The Economist, 2013). Yet in the countries of East Asia, a growing middle class is benefitting from the reinvestment of capital into programmes that support equality, such as universal education (Stiglitz, 1996), while the number living in poverty has decreased. Institutions play a powerful role in shaping income inequality, accounting for the different rates of inequality among high-​income countries with similar development trajectories (Alvaredo et al., 2013). In the USA and UK, four factors account for high inequality: favourable tax rates for the rich, generous executive pay, inherited wealth, and capital (instead of earned) income (Alvaredo et  al., 2013). Because of inherited privilege, there is less

Just Growth   795 intergenerational mobility in the USA than in most advanced industrialized countries, that is, the children of the rich tend to stay rich, while children of the poor stay poor (Mishel et al., 2006). This rising inequality has a price. As income shifts from the bottom to the top, consumption declines, as higher-​income households save more of their income. This then curtails government spending, just as additional investment in education, technology, and infrastructure is needed in order to facilitate upward mobility. As a result of the growth in poverty and loss of the middle class, economic efficiency and growth decline, and societies become more unstable (Stiglitz, 2012). Economic restructuring benefits some regions and affects others less able to retool their economies. Uneven development is a key feature of capitalist industrialization, as capital seeks out greater profits (Storper and Walker, 1989). In the last half of the twentieth century, plant closings and high unemployment in areas like the Rustbelt in the USA helped accelerate a shift in population and employment to the South and West, while new command-​ and-​control centres for the global economy emerged in select metropolitan regions (Sassen, 1991). Although both technological change and global trade influenced these shifts, the two forces have affected regions differently, with information technology transforming jobs in larger cities like Chicago and trade affecting the regions with more labour-​intensive manufacturing plants (Autor et al., 2013). Uneven development is also resulting from the growth of innovation sectors, such as information technology or biotechnology, in select metropolitan areas—what Moretti (2012) calls ‘innovation hubs’. Innovation might be described as an epistemological transformation, or new knowledge, that combines resources to create new products or services that reach markets. Hubs emerge not just because of the classic pattern of urban agglomeration—​firms clustering in order to benefit from shared knowledge, markets, and inputs—​but also because of the ‘creative class’ dynamic: innovative places attract the very highly educated workers who are in short supply, in turn attracting more firms (Florida, 2002). This type of cluster creates particularly high multiplier effects—the growth of tradable export sectors creates new demand for inputs and also goods and services for the workers and their households (Moretti, 2012). The overall effect of hubs is to grow the local economy rapidly, driving up local housing prices and wages. But the metropolitan regions without this innovative upward spiral are left further and further behind, with negative impacts on regional opportunity structures. For instance, intergenerational mobility declines with income inequality, lackluster growth, and poor social capital (Chetty et al., 2013). Thus, in many regions of the world, increasing inequality accompanies rapid growth. Yet, most agree that growth itself is necessary for economic inclusion. This creates a new political imperative to ensure that some of the benefits of growth accrue to the less advantaged. How do we design strategies for growth with equity? A better understanding of the relationship between inequality and growth gives us some preliminary conclusions.

Inequality, Growth, and the Role of Policy Research has yielded mixed results on how inequality affects growth, with some finding a negative relationship and others a positive. Lopez and Servén (2009) suggest that conflicting

796   Chapple findings may be due to the use of cross-​sectional versus panel data, different national contexts, or a non-​linear relationship between the two. Despite these uncertainties, there are indications that adopting strategies to reduce inequality will have a positive effect on growth. A growing body of evidence from World Bank and International Monetary Fund research on developing countries suggests that lower levels of inequality, supported by investment in education, increase the amount and dur­ ation of growth (Birdsall et al., 1995; Berg et al., 2012). It seems that poverty alone—​even controlling for inequality—slows growth, mostly by deterring investment (Lopez and Sérven, 2009). This raises the intriguing possibility that reducing inequality and/​or poverty results in higher economic growth at the regional level as well. Only a few studies have looked at the efficacy of reducing inequality as a growth strategy for regions—but the preliminary findings are positive. Firstly, higher levels of inequality mean less per capita income growth, and, secondly, as inequality decreases, higher growth results (Pastor et al., 2000; Eberts et al., 2006; Benner and Pastor, 2012). Writing from the national perspective, Stiglitz (2012) points to Brazil under Presidents Enrique Cardoso and Luiz Inácio Lula da Silva as an example of how to reduce inequality while still increasing growth. Underlying the transformation in Brazil were cash transfers to the poor (the Bolsa Familia), aid to alleviate hunger, and state investments in education, particularly for the poor. Stiglitz makes the case for an array of intentional policies to reduce inequality, with public investment to help those at the bottom reach their potential. At the regional level, Benner and Pastor (2012) have led research on strategies for growth with equity, which they define in terms of economic inclusion. They identify four structural factors behind growth with equity:  political consolidation (among cities or between city and county), economic diversity, public-​sector employment, and a strong Black/​Latino middle class. With these factors in place, implementation of equitable strategies occurs via the formation of epistemic communities, processes (or organizations) that convene diverse stakeholders in a dialogue that builds understanding of diverse perspectives and priorities, and, ultimately, a shared vision for the region. The regions that are most effective at growth with equity will be those with a stable employment base that form these epistemic communities, and typically with stable leadership. The question, then, is which strategies should they pursue? We turn next to the toolkit of local and regional economic development strategies.

Theories and Strategies for Inclusive Growth The two basic approaches to growth with equity are increasing regional economic growth and then redistributing the benefits, or reducing inequality directly by providing a safety net and/​or building new capabilities. The first approach finds its theoretical underpinnings in regional economic development, while the second grows more from local and community economic development.

Just Growth   797

Regional Economic Development: Growing the Pie Theories about regional economic development abound, with considerable consensus about how regions grow. Yet, there is still considerable uncertainty about how to grow equitably, that is, to produce high-​quality jobs and connect a region’s less advantaged residents to opportunity. Of the many factors behind regional economic development, this section looks at three—​exports, agglomeration, and entrepreneurship—​and their relationship to equity.

Exports As Douglass North argued in 1955, regions grow through new exogenous demand for exports, which spurs a cycle of growth. To meet demand, jobs and production must increase, from which new sales bring revenue to pay the workers and proprietors. That income translates into more local consumption of goods and services from ‘residentiary’ industries (North, 1955). Meanwhile, the growth of the export sector spurs the development of related local industries, such as production inputs. Regions can influence how growth occurs by helping to reduce production costs (e.g. through investment in transportation or human capital). Export strategies are not a simple recipe. Some regions never become successful exporters, while others export for centuries but become over-​specialized, or never manage to diversify and industrialize. The impact of exports is much weaker for larger regions, which can grow on the basis of internal trade (e.g. New York City) (Tiebout, 1956). Supply factors in a region—such as government expenditures, non-​economic migration, changing consumption patterns, or local industry structure—​will shape how effectively it responds to export growth (Tiebout, 1956; Chinitz, 1961). Even if export-​driven strategies succeed, there is little guarantee of job quality or fit for disadvantaged residents. In fact, regions driven by export demand are particularly susceptible to competition from other low-​cost regions around the world, the challenges of keeping pace with technological change, and, in the case of resource-​based economies, the depletion of natural resources (Jacobs, 1984).

Agglomeration As regions grow, external economies develop to improve their competitive position. With a number of firms in the same industry, or localization economies, suppliers begin locating nearby to provide ancillary services, such as marketing, access to credit, and legal services. Workers with specific skills migrate to the area. The proximity of this cluster of firms leads to knowledge sharing and spillovers, an idea Porter (1998) later adapted to become his clusters, or geographical concentrations of interconnected companies and institutions that create the specializations that drive regional competitive advantage and growth. Another form of external economies is urbanization economies, in which the higher volume of economic activity in an urban area (i.e. urban size) helps to drive down costs as urban services and infrastructure—as well as the market—​expand. By relying on outside suppliers and sharing facilities, businesses can cope with uncertainty.

798   Chapple Localization economies increase specialization, while urbanization economies increase diversity (Jacobs, 1969). The debate remains whether specialization or diversity is optimal for economic inclusion; no research has examined how diversity and specialization reshape the income distribution and income inequality (Dissart, 2003). The evidence suggests that regional economic diversity increases job growth (Glaeser, 2000; Dissart, 2003; Feser et al., 2008; but see Porter, 2003). However, the preliminary evidence on per capita income growth points to specialization as the means to improving income levels (Dissart, 2003; Porter, 2003; Pede, 2013). Thus, urban agglomerations are able to grow the pie and raise regional incomes. In terms of equitable growth, diverse regions may provide more job opportunities, while specialized regions may succeed better at raising incomes. The higher levels of accessibility and interaction that proximity provides may also reduce the barriers that disadvantaged workers face in competing for jobs. At the same time, the downside of agglomeration is the higher cost of living; as labour and housing costs rise and congestion increases, chances for inclusion may diminish.

Entrepreneurship A related factor in regional economic growth is entrepreneurship. When firms organize to allow for rapid learning, experimentation, and adjustment, knowledge spillovers—​and thus agglomeration—​increase. The paradigmatic example of Silicon Valley shows how loosely organized networks of actors—​buyers, suppliers, competitors, and related institutions—​ can respond to economic changes, allowing the transfer of skills and technology. Even more importantly, their economic action is not based on individualistic calculation of costs and benefits, but is embedded into larger institutional and social frameworks (Saxenian, 1994). It is not clear, however, how inclusive entrepreneurial networks are. The horizontal networks of entrepreneurial firms in the Silicon Valley story (and also described by Michael Piore and Charles Sabel (1984) in the district of Emilia-​Romagna) are, indeed, comprised mostly of small firms, and new firm start-​ups and young firms account for a disproportionate share of job creation (Haltiwanger et al., 2011; Neumark et al., 2011). Whether large or small, entrepreneurial firms clearly create jobs. But these opportunities may not be the most equitable. Start-​up firms tend to be undercapitalized and pay relatively low wages; this is particularly true among immigrant communities who may count on low-​cost labour from network contacts (Bates, 1997; Ong and Loukaitou-​Sideris, 2006). If they are high-​tech, they may not offer many jobs to low-​wage workers. Also, there are multiple barriers to entry for lower-​income entrepreneurs, including financial insecurity, poor access to credit, low appetite for risk, and low education and training. In particular, lack of inherited wealth and family assets create disadvantages for entrepreneurship and self-​employment (Quadrini, 2000). Whether looking at the strength of the export base, or the region’s diversity, or entrepreneurship, as the source of regional growth, there remains one outstanding issue: Who will get the jobs? Job growth in a region creates a chain of job shifts within and across regions. For every 100 new jobs in a metropolitan area, about eighty new workers move in from outside the region; the higher the skills demanded and wages paid in the new job, the more in-​migration will occur (Bartik, 1993). The twenty remaining jobs will be taken by local residents, but they are more likely to be underemployed (in terms of hours, wages, or challenge) than the long-​term unemployed. As the workers from both within and outside the

Just Growth   799 region leave their current jobs, they create openings for others, and, presumably, down the job chain, the most disadvantaged (or unemployed) worker will gain an opportunity. But chances are that that opportunity will be in another region, because of the attraction of in-​migrants to high-​growth regions. Because of this job-​chain effect, the most equit­ able approach is to focus on increasing regional per capita earnings instead of job creation (Bartik, 2011).

Strategies These theories of regional economic development have shaped the strategies used in practice. Strategies that attempt equitable growth by growing the pie and then connecting the disadvantaged to opportunity typically rely on either business attraction or endogenous development. Both approaches may help to increase exports, agglomeration, and clustering, while endogenous strategies also facilitate entrepreneurship. In business attraction strategies, cities and states try to attract firms to relocate, typically via tax incentives, subsidized loans, or simply marketing (Eisinger, 1995; Bradshaw and Blakely, 1999). In practice, cities, counties, and states still rely heavily on the use of tax incentives to attract businesses, despite evidence of mixed effects on business location decision-​ making and relatively low job creation compared with existing businesses (Peters and Fisher, 2004; Kolko and Neumark, 2007). However, business attraction efforts do tend to target relatively high-​wage firms, such as high-​tech manufacturers and exporters generally. Policymakers typically work very closely with the businesses, in the process often developing strategies for inclusion such as customized training and hiring processes that can ensure that local residents are prepared for jobs. With the realization that globalization was diminishing the opportunities for new ‘smokestacks’ to chase, cities and states began emphasizing more endogenous development (Teitz, 1994). This autonomous or demand-​side economic development applies the idea of ‘development from within’, targeting businesses sometimes in specific neighbourhoods but more commonly throughout a city or region (Eisinger, 1988). Endogenous development focuses attention on the entrepreneurial potential of existing firms, often embedded within clusters. Thus, these strategies focus on business start-​ups, expansion, and retention, typically via revolving loan funds, business incubators, business management assistance, and R & D/​ innovation support. The importance of start-​ups, young businesses, and rapid firm expansions in job creation suggests that endogenous development strategies will be particularly fruitful for economic inclusion. In fact, Porter (1995) extended his work on clusters to argue that disadvantaged communities in the inner city could compete endogenously by capitalizing on local market demand and resources. In theory, endogenous development strategies should result in wage increases, as they increase productivity. But, in practice, results are likely mixed, as younger firms, in particular, lack the stability to remunerate workers. There is also no guarantee that local firms will hire locally, and, in fact, locals may have prejudices about hiring from certain areas that lead to discrimination (Dewar, 2013). In terms of the potential for inner-​city competitiveness, there is little evidence that core locations with concentrations of disadvantage have connected successfully to regional clusters, as Porter suggested they could, with the possible exception of health care (Coyle, 2007).

800   Chapple Another set of strategies for inclusion target specific places, with the idea either that the benefits will trickle down to local disadvantaged communities, or that the communities can capture some of the benefits deliberately through development agreements. Results are mixed, or even negative. For instance, enterprise zone programmes target tax credits to disadvantaged areas, but have had no greater success in reducing unemployment or poverty or creating jobs and businesses than comparison areas without zone designation (Greenbaum and Landers, 2009; U. S. Government Accountability Office, 2006). They tend to underperform in expected job creation and effects disappear quickly (Greenbaum and Landers, 2009; Cray et al., 2011). Another form of place targeting is commercial real-​estate development projects by the public sector, often working in partnership with the private sector. The public sector provides direct financing or various tax incentives with the idea that local communities will capture the benefits; however, studies to date have not demonstrated significant equity benefits of public-​led commercial development and market-​led retail attraction strategies (Chapple and Jacobus, 2009). Real-​estate development can be more directly inclusive via community benefits agreements (CBAs). These CBAs may complement linkage policies, which require compensating side payments for housing and social needs from large-​scale commercial development, and local hiring ordinances, which reserve construction jobs in development receiving public funds for local residents (Molina, 1998; Salkin, 2007). Legally enforceable contracts negotiated between a prospective developer and community representatives, the CBA exacts community economic development commitments from the developer (e.g. local hires at living wages, affordable housing units, neighbourhood amenities) in exchange for the community’s support of its proposed development (Gross et al., 2005). The CBA tends to focus on job quality, perhaps, in part, because of the involvement of labour unions. Thus, traditional theories fail to address inequality, and related strategies have attempted inclusion with limited success. The next subsection turns to strategies that further inclusion directly, with the idea of generating growth.

Community Economic Development: Building Capacity Since the urban fiscal crises of the 1970s, caused by global economic restructuring and government devolution, cities and regions have actively tried to increase the level of local economic activity by increasing the capacity of residents, and, in turn, their wealth. Similar dynamics have led to the rise of local and community economic development around the globe, although in countries with a strong safety net, the pursuit of inclusive growth is likely less intense. The advent of neo-​liberalism in many countries has meant a growing role for private and non-​profit-​sector actors in these economic development strategies, in essence substituting for government; however, in many developing countries and social democracies, the public sector maintains a dominant role (Anheier and Salamon, 2006). The idea of reducing inequality and poverty as a growth strategy stems not just from the work of World Bank economists, but also more broadly from the idea that societies will benefit broadly from fostering the capabilities of those at the bottom of the economic spectrum (Sen, 1999). Thus, Stiglitz (2012) recommends intentional policies to reduce inequality that, in addition to limiting the accumulation of wealth through rent seeking and tax

Just Growth   801 favouritism, level the playing field by investing in an array of social policies such as education, health care, asset building, and full employment. Where national governments fail to support such social investments consistently, and functions have devolved to the local level, actors at the city and regional level are organizing to ensure that local economic development strategies are inclusive. In the USA, this has taken the form of community and regional organizing for economic justice, in what Pastor et al. (2009) call a new ‘social movement regionalism’. Decades of community battles over redevelopment projects and gentrification have built local organizing capacity and political savvy that have helped to connect marginalized groups and neighbourhoods to economic opportunity. The new economic justice movement evolved from a response to the thirty-​year advent of globalization, capital mobility, devolution, and neo-​liberalism that was facilitated by local community development capacity. Deindustrialization had weakened the labour movement, and a search for new approaches resulted in new union coalitions with faith-​based and other community organizations. At the local level, these coalitions pressured cities to adopt linkage policies and developers to sign CBAs that would ensure benefits for the local community—​that is, guaranteeing that some growth would trickle down. But the strategy with the greatest potential for reducing inequality has been the minimum or living-​wage movement, which tackles inequality directly with the idea that higher wages will, at a minimum, not impede growth (Dube et al., 2010). The struggle to improve job quality began via living-​wage coalitions that passed laws in over 140 cities, and has continued via organizing to raise the minimum wage at the state level (Fairris et al., 2005; Dube, 2014). The proponents of these economic justice movements are using the region as their strategic arena to organize their constituents (Pastor et  al., 2009). Over time, strategies are spreading across the country through the regional networks of national organizations, many faith based, such as the Gamaliel Foundation, the PICO National Network, and the Industrial Areas Foundation. Their success at creating community economic development seems ultimately to depend on their ability to exercise power at higher levels of government—​for instance, by obtaining federal authorization for local-​hire ordinances in infrastructure projects (Weir and Rongerude, 2007; Swanstrom and Banks, 2009). In the absence of national initiatives, this kind of multi-​level organizing underpins many strategies for growth with equity. In cities and regions, initiatives may be fortified if they are supported by diverse stakeholders joining in epistemic communities (Benner and Pastor, 2012). Specific strategies that address inequality and poverty in order to foster growth fall into three categories: education and workforce development, investing in people through place, and access to capital.

Education and Workforce Development Improving educational attainment is fundamental for economic inclusion; the returns to education are well established, particularly for low-​and middle-​income countries (Card and Krueger, 1990; Psacharopoulos and Patrinos, 2004). But both the declining returns from working and the job instability that characterize the economy today mean that education alone is not enough to guarantee upward mobility (Bernhardt et al., 2001). Upward mobility also depends on local labour market conditions, corporate culture, union pressure, and other factors (Appelbaum et al., 2003).

802   Chapple Educational reform strategies have long focused on improving the quality of education, as well as access to education, at levels from primary school to university (Cooper et al., 2014). Also, a growing number of strategies are trying to increase labour market participation through workforce development. The best programmes are workforce intermediaries that adopt sector initiatives, targeting specific industry sectors in order to create a win–​win situation by restructuring employment practices in a way that is beneficial to both employers and low-​wage workers (Marano and Tarr, 2004). Some of these programmes also try to address long-​term upward mobility through career-​ladder initiatives, from increasing the pay of existing jobs to creating new tiers within occupations to using education and training to advance workers into occupations with better pay (Fitzgerald, 2006).

Investing in People Through Place Another local strategy for intervening in equity is ‘investing in people through place’ via strategies that improve neighbourhood economies and quality of life (Giloth, 2007). Investment in public goods and services in a place makes economic benefits more accessible and improves life chances for neighbourhood residents. Community reinvestment builds up neighbourhood markets and land values, and thus helps to protect and grow family assets. Lowering crime, improving schools, and supporting home improvements all contribute to stimulating asset appreciation and neighbourhood markets. In the USA, in particular, a system of community development intermediaries—such as community development corporations and community development financial institutions—​support revitalization work such as commercial district revitalization, business assistance, social services, and job training (Yin, 1998). Community-​based financial services provide an alternative credit system for both businesses and individuals, such as savings and checking accounts, home loans, micro-​ loans, and business loans. Another opportunity to support people in place is leveraging the resources of large anchor institutions (e.g. universities and hospitals), as well as government purchasing. Anchor institutions are fixed assets in communities and are unlikely to relocate out of regions. They purchase an array of goods and services—​such as food, laundering, record disposal, and recycling—​which open up opportunities for entrepreneurship. Government purchasing is another opportunity: in the USA, almost US$100 billion of federal procurement dollars goes to small businesses each year, by government mandate (US Small Business Administration, 2010). Economic inclusion strategies that shift government spending to disadvantaged enterprises do not create more jobs, but may provide a net benefit by reducing the need for expenditure on public services (Treuhaft and Rubin, 2013).

Access to Capital Chronic asset poverty—​lacking the accumulated wealth to help subsidize education, housing, or other costs of living—​makes it difficult to save and creates a drag on growth. Data on asset poverty suggest that more than one-​fifth of households in the USA—​and a disproportionate share of minority families—​are asset poor, double the rate of income poverty (Corporation for Enterprise Development, 2005). Dealing with wealth inequality directly, rather just accommodating the labour market through work supports, is key to addressing the root causes of income inequality (Oliver and Shapiro, 1995).

Just Growth   803 Individual development account programmes have emerged as the most direct way to build assets, by offering matched savings accounts that can be used to develop wealth via purchasing a home, obtaining higher education, or starting a small business (Corporation for Enterprise Development, 2005). When new home or business ownership leads to improvements (e.g. occupying a vacant structure, renovating a building, or building a successful business), positive externalities result: the changes attract new investment and improve the quality of life for existing residents (if they are able to cope with rising housing costs and stay in place). To keep communities stable, asset-​building programmes can be combined with shared-​equity models, which cap the gains from selling an asset, or require reinvestment of profit in the community (Weber and Smith, 2003). Another set of strategies for inclusive growth focus on the entrepreneurial capacities of low-​income communities, with the idea that enterprise can supplement the benefits of more formal jobs (a process often called ‘income patching’) (Edgcomb and Armington, 2013). Although there are mixed results from around the world, some of the programmes have been startlingly successful, in terms of programme return on investment, business survival rates, and job growth (about half of microenterprises create jobs) (Banerjee et al., 2013; Edgcomb and Thetford, 2013). Likewise, minority entrepreneurship programmes more generally can support growth with equity. Despite concerns that such businesses are overly prone to failure or exploitation, there is some evidence that these businesses hire disadvantaged local residents who otherwise might not be employed, and particularly if accompanied by some higher education, can lead to upward mobility (Bates, 1997; Boston, 2006). Just as the strategies that attempt to foster inclusion through growth may fall short, so may the strategies that focus on equity in the hopes of increasing growth. Experimentation with workforce development, revitalization, and access to capital is ongoing, and research is only just beginning to establish what works. Even if strategies such as regional organizing, government purchasing set-​asides, community development intermediaries, and micro-​ entrepreneurship succeed in creating new markets and jobs, the growth may not be net new. At best, these strategies may sustain and redistribute growth.

Implications for Economic Geography Although many countries have succeeded at slowing or reversing the growth of inequality, increasing globalization and technological change continue to put downward pressure on wages, as the elite benefit. Even where an institutional structure supportive of inclusion is in place, an economic downswing can erase its effects. Growing inequality at a national scale tends to exacerbate patterns of uneven development within a country. Rapid urbanization may also reinforce patterns of inequality, for instance as is occurring in the Global South where governments are accommodating global capital in increasingly privatized cities (Mitra, 2014). Inequality—​and fostering inclusive growth—​is not only a key challenge of the twenty-​first century, but is likely to become even more salient as climate change advances. Reducing greenhouse gas emissions will heighten conflicts between growth and equity, and global warming itself could mean increasing instability, particularly in developing countries (Intergovernmental Panel on Climate Change, 2014).

804   Chapple Even though the need for more inclusive growth is here to stay, there is still considerable uncertainty about how to achieve it. Investments in education, infrastructure, and social welfare clearly pay off at the national level, but effects at the local and regional level are complex. Strong market regions with competitive advantages may be best positioned for growth in theory, but in practice the extent of inclusion will depend on how production has been organized historically, and the resultant local institutions and networks (Storper, 2013). More inclusive political processes may enable community participation in governance even in weaker markets. Yet, more inclusive growth at the regional level may not mean net new growth in the aggregate: one community or region may succeed at the expense of another. These regional variations mean that there is no simple recipe for inclusive growth—​in somewhat of a shift from the halcyon days of regional science (and economic geography), when the question was simply how to generate growth in regions with different industry structures and resource constraints. Researchers will struggle to identify intentional policies behind inclusive growth through systematic analysis of the processes of change across a large number of regions, because each region is unique. Pursuing economic inclusion means developing a better understanding not just of local economic structure and institutions, but also of variations in human capabilities and aspirations. This then requires the sort of deep observation of specific regions that occurs via comparative studies adopting an exceptionalist or historicist perspective. Understanding growth with equity, and the intentional strategies that can support inclusion, may, indeed, be a full employment plan for economic geographers.

References Alvaredo, F., Atkinson, A.B., Piketty, T., and Saez, E. (2013). ‘The top 1 percent in international and historical perspective’. Journal of Economic Perspectives 27: 3–​20. Anheier, H.K. and Salamon, L.M. (2006). ‘The Nonprofit Sector in Comparative Perspective’ in W.W. Powell and R. Steinberg (eds) The Nonprofit Sector: A Research Handbook, pp. 89–​114 (New Haven, CT: Yale University Press). Appelbaum, E., Bernhardt, A., and Murnane, R.J. (eds). (2003). Low-​Wage America: How Employers are Reshaping Opportunity in the Workplace (New York, NY: Russell Sage Foundation). Autor, D.H., Dorn, D., and Hanson, G.H. (2013). ‘The geography of trade and technology shocks in the United States’. American Economic Review 103: 220–​225. Banerjee, A.V., Duflo, E., Glennerster, R., and Kinnan, C. (2013). ‘The miracle of microfinance? Evidence from a randomized evaluation’. MIT Department of Economics Working Paper No. 13-​09 http://​ssrn.com/​abstract=2250500 (last accessed 25 April 2017). Bartik, T.J. (1993). ‘Who benefits from local job growth: migrants or the original residents?’ Regional Studies 27: 297–​311. Bartik, T.J. (2011). ‘What works in job creation and economic development’. Paper presented at Transforming Communities Conference of the National Employment Law Project, Flint, MI. Bates, T. (1997). ‘Entrepreneurship as a Route to Upward Mobility among the Disadvantaged’ in T. Bates (ed.) Race, Self-​Employment, and Upward Mobility: An Illusive American Dream, pp. 207–​224 (Baltimore, MD: The Johns Hopkins University Press).

Just Growth   805 Benner, C. and Pastor, P. (2012). Just Growth: Inclusion and Prosperity in America’s Metropolitan Regions (Abingdon: Taylor & Francis). Berg, A., Ostry, J.D., and Zettelmeyer, J. (2012). ‘What makes growth sustained?’ Journal of Development Economics 98: 149–​166. Bernhardt, A., Morris, M., Handcock, M.S., and Scott, M.A. (2001). Divergent Paths: Economic Mobility in the New American Labor Market (Manhattan, NY: Russell Sage Foundation). Birdsall, N., Ross, D., and Sabot, R. (1995). ‘Inequality and growth reconsidered: lessons from East Asia’. The World Bank Economic Review 9: 477–​508. Bluestone, B. and Harrison, B. (1982). The Deindustrialization of America:  Plant Closings, Community Abandonment, and the Dismantling of Basic Industry (New  York, NY:  Basic Books). Boston, T.D. (2006). ‘The Role of Black-​owned Businesses in Black Community Development’ in P. Ong (ed.) Jobs and Economic Development in Minority Communities:  Realities, Challenges, and Innovation, pp. 11–​75 (Philadelphia, PA: Temple University Press). Bradshaw, T.K. and Blakely, E.J. (1999). ‘What are “third-​wave” state economic development efforts? From incentives to industrial policy’. Economic Development Quarterly 13: 229–​244. Brady, D., Kaya, Y., and Gereffi, G. (2011). ‘Stagnating industrial employment in Latin America’. Work and Occupations 38: 179–​220. Brenner, N. and Wachsmuth, D. (2012). ‘Territorial Competitiveness:  Lineages, Practices, Ideologies’ in B. Sanyal, L.J. Vale, and C. Rosan (eds) Planning Ideas That Matter: Livability, Territoriality, Governance and Reflective Practice, pp. 179–​204 (Cambridge, MA: MIT Press). Card, D. and Krueger, A. (1990). ‘Does school quality matter? Returns to education and the characteristics of public schools in the United States’. No. w3358, National Bureau of Economic Research. Castells, M. (1996). The Rise of the Network Society: The Information Age: Economy, Society and Culture (Volume I) (Malden, MA, and Oxford: Blackwell). Chapple, K. and Jacobus, R. (2009). ‘Retail Trade as a Route to Neighborhood Revitalization’ in M.A. Turner, H. Wial, and N. Pindus (eds) Urban and Regional Policy and its Effects, Volume II, pp. 19–​68 (Washington, DC: Brookings Institution and Urban Land Institute). Chetty, R., Hendren, N., Kline, P., and Saez, E. (2013). ‘The economic impacts of tax expenditures:  evidence from spatial variation across the US’ https://​journalistsresource.org/​ studies/​economics/​inequality/​economic-​impacts-​of-​tax-​expenditures-​evidence-​from-​ spatial-​variation-​across-​the-​u-​s (last accessed 25 April 2017). Chinitz,. (1961). ‘Contrasts in agglomeration: New York and Pittsburgh’. American Economic Review: Papers and Proceedings 51: 279–​289. Cooper, B.S., Cibulka, J.G. and Fusarelli, L.D. (eds) (2014). Handbook of Education Politics and Policy (London: Routledge). Corporation for Enterprise Development (2005) ‘Assets: an update for innovators’. Newsletter no.  2 http://​www.cfed.org/​think.m?id=113&groupid=assets&clusterid=1 (last accessed 25 April 2017). Coyle, D.M. (2007). ‘Realizing the inner city retail opportunity: progress and new directions, an analysis of retail markets in America’s inner cities’. Economic Development Journal 6: 6–​14. Cray, A., Nguyen, T., Pranka, C., Schildt, C., Sheu, J., and Rincon Whitcomb, E. (2011). Job Creation: A Review of Policies and Strategies (Berkeley, CA: Institute for Research on Labor and Employment). Dewar, M. (2013). ‘Paying employers to hire local workers in distressed places’. Economic Development Quarterly 27: 284–​300.

806   Chapple Dissart, J.C. (2003). ‘Regional economic diversity and regional economic stability: research results and agenda’. International Regional Science Review 26: 423–​446. Dube, A. (2014). ‘Proposal 13: Designing Thoughtful Minimum Wage Policy at the State and Local Levels’ in M.S. Kearney and B.H. Harris (eds) Policies to Address Poverty in America, pp. 137–​146 (Washington, DC: Brookings). Dube, A., Lester, T.W., and Reich, M. (2010). ‘Minimum wage effects across state borders:  estimates using contiguous counties’. The Review of Economics and Statistics 92.4: 945–​964. Eberts, R., Erickcek, G., and Kleinhenz, J. (2006). ‘Dashboard indicators for the northeast Ohio economy:  prepared for the Fund for Our Economic Future’. Working Paper 06-​05 (Cleveland, OH: The Federal Reserve Bank of Cleveland). Economic Commission for Latin America and the Caribbean (2010) Time for Equality: Closing Gaps, Opening Trails (Santiago: ECLAC). Edgcomb, E. and Armington, M.M. (2013). The Informal Economy: Latino Enterprises at the Margins (Washington, DC: Microenterprise Fund for Innovation, Effectiveness, Learning and Dissemination, Aspen Institute). Edgcomb, E. and Thetford, T. (2013). Microenterprise Development as Job Creation (Washington, DC: Microenterprise Fund for Innovation, Effectiveness, Learning and Dissemination, Aspen Institute). Eisinger, P.B. (1988). The Rise of the Entrepreneurial State: State and Local Economic Development Policy in the United States (Madison, WI: University of Wisconsin Press). Eisinger, P. (1995). ‘State economic development in the 1990s:  politics and policy learning’. Economic Development Quarterly 9: 146–​158. Fairris, D., Runsten, D., Briones, C., and Goodheart, J. (2005). Examining the Evidence: The Impact of The Los Angeles Living Wage Ordinance on Workers and Businesses (Los Angeles, CA: Los Angeles Alliance for a New Economy). Feser, E., Renski, H., and Goldstein, H. (2008). ‘Clusters and economic development outcomes:  an analysis of the link between clustering and industry growth’. Economic Development Quarterly 22: 324–​344. Fitzgerald, J. (2006). Moving Up in the New Economy: Career Ladders for U.S. Workers (Ithaca, NY: Cornell University Press). Florida, R. (2002). The Rise of the Creative Class, and How it’s Transforming Work, Leisure and Everyday Life (New York: Basic Books). Giloth, R. (2007). ‘Investing in Equity: Targeted Economic Development for Neighborhoods and Cities’ in M.I.J. Bennett and R. Giloth (eds) Economic Development in American Cities: The Pursuit of an Equity Agenda, 23–​50 (Albany, NY: State University of New York Press). Glaeser, E.L. (2000). ‘The New Economics of Urban and Regional Growth’ in G.L. Clark, M.P. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 83–​98 (Oxford: Oxford University Press). Greenbaum, R.T. and Landers, J. (2009). ‘Why are state policy makers still proponents of enterprise zones? What explains their action in the face of a preponderance of the research?’ International Regional Science Review 32: 468. Gross, J., LeRoy, G., and Janis-​Aparicio, M. (2005). ‘Community benefits agreements: making development projects accountable’. Good Jobs First http://​www.goodjobsfirst.org/​sites/​ default/​files/​docs/​pdf/​cba2005final.pdf (last accessed 25 April 2017).

Just Growth   807 Haltiwanger, J., Jamin, R.S. and Javier, M. (2011). ‘Who creates jobs? Small vs. large vs. young’. NBER Working Paper No. 16300 (Cambridge, MA: National Bureau of Economic Research). Harrison, B. and Bluestone, B. (1988). The Great U-​turn:  Corporate Restructuring and the Polarizing of America (New York: Basic Books). Intergovernmental Panel on Climate Change (2014). Fifth Assessment Report (New  York: United Nations). Jacobs, J. (1969). The Economy of Cities (New York: Random House). Jacobs, J. (1984). Cities and the Wealth of Nations:  Principles of Economic Life (New  York: Random House). Katz, L.F. and Murphy, K.M. (1992). ‘Changes in relative wages, 1963–​1987: supply and demand factors’. The Quarterly Journal of Economics 107: 35–​78. Kolko, J. and Neumark, D. (2007). Business Location Decisions and Employment. Dynamics in California (San Francisco, CA: Public Policy Institute of California). Kollmeyer, C. (2009). ‘Explaining deindustrialization:  how affluence, productivity growth, and globalization diminish manufacturing employment’. American Journal of Sociology 114: 1644–​1674. Lopez, H. and Servén, L. ‘Too poor to grow’. World Bank Policy Research Working Paper Series, October 1 http://​ssrn.com/​abstract=1447205 (last accessed 25 April 2017). Marano, C. and Tarr, K. (2004). ‘The Workforce Intermediary: Profiling the Field of Practice and Its Challenges’ in R. Giloth (ed.) Workforce Intermediaries for the Twenty-​first Century, pp. 93–​123 (Philadelphia, PA: Temple University Press). Mishel, L., Bernstein, J., and Allegretto, S. (2006). The State of Working America 2006/​07 (Ithaca, NY: Cornell University Press). Mitra, S. (2014). ‘Anchoring Transnational Flows:  Hypermodern Spaces in the Global South’ in F. Miraftab and N. Kudva (eds) Cities of the Global South Reader, pp. 106–​114 (Abingdon: Routledge). Molina, F. (1998). Making Connections: A Study of Employment Linkage Programs (Washington, DC: Center for Community Change). Moretti, E. (2012). The New Geography of Jobs (Boston, MA: Houghton Mifflin Harcourt). Neumark, D., Wall, B., and Zhang, J. (2011). ‘Do small businesses create more jobs? New evidence for the United States from the National Establishment Time Series’. Review of Economics and Statistics 94: 16–​29. North, D.C. (1955). ‘Location theory and regional economic growth’. The Journal of Political Economy 63: 243–​258. Oliver, M. and Shapiro, T.M. (1995). Black Wealth/​White Wealth a New Perspective on Racial Inequality (New York: Routledge). Ong, P. and Loukaitou-​Sideris, A. (eds) (2006). Jobs and Economic Development in Minority Communities (Philadelphia, PA: Temple University Press). Osterman, P. (1999). Securing Prosperity: The American Labor Market: How it has Changed and What to do About it (Princeton, NJ: Princeton University Press). Pastor, M., Jr., Benner, C., and Matsuoka, M. (2009). This Could be the Start of Something Big: How Social Movements for Regional Equity are Reshaping Metropolitan America (Ithaca, NY: Cornell University Press). Pastor, Jr., M., Dreier, P., Grigsby III, J.E., and Lopez-​Garza, M. (2000). Regions That Work: How Cities and Suburbs Can Grow Together (1st edition) (Minneapolis, MN:  University of Minnesota Press).

808   Chapple Pede, V.O. (2013). ‘Diversity and regional economic growth:  evidence from US counties’. Journal of Economic Development 38: 111–​127. Peters, A. and Fisher, P. (2004). ‘The failures of economic development incentives’. Journal of the American Planning Association 70: 27–​38. Piore, M.J., and Sabel, C.F. (1984). The Second Industrial Divide:  Possibilities for Prosperity (New York: Basic Books). Porter, M.E. (1995). ‘The competitive advantage of the inner city’. Harvard Business Review May/​June: 55–​7 1. Porter, M.E. (1998). ‘Clusters and the new economics of competition’. Harvard Business Review 76: 77–​90. Porter, M.E. (2003). ‘The economic performance of regions’. Regional Studies 37: 545–​546. Psacharopoulos, G. and Patrinos, H.A. (2004). ‘Returns to investment in education: a further update’. Education Economics 12.2: 111–​134. Quadrini, V. (2000). ‘Entrepreneurship, saving, and social mobility’. Review of Economic Dynamics 3: 1–​40. Salkin, P.E. ‘Community benefits agreements: opportunities and traps for developers, municipalities, and community organizations’. Planning and Environmental Law 59: 3–​8. Sassen, S. (1991). The Global City:  London, New  York, Tokyo (Princeton, NJ:  Princeton University Press). Saxenian, A. (1994). Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA: Harvard University Press). Sen, A. (1999). Development as Freedom (New York: Anchor). Stern, D.I. (2004). ‘The rise and fall of the environmental Kuznets curve’. World Development 32: 1419–​1439. Stiglitz, J.E. (1996). ‘Some lessons from the East Asian miracle’. The World Bank Research Observer 11: 151–​177. Stiglitz, J. (2012). The Price of Inequality (London: Penguin). Storper, M. (2013). Keys to the City: How Economics, Institutions, Social Interaction, and Politics Shape Development (Princeton, NJ: Princeton University Press). Storper, M. and Walker. R. (1989). The Capitalist Imperative (Oxford: Blackwell). Swanstrom, T. and Banks, B. (2009). ‘Going regional: community-​based regionalism, transportation, and local hiring agreements’. Journal of Planning Education and Research 28: 355–​367. Teitz, M.B. (1994). ‘Changes in economic development theory and practice’. International Regional Science Review 16: 101–​106. The Economist (2013). ‘Gini out of the bottle’. 26 January. Tiebout, C.M. (1956). ‘Exports and regional economic growth: rejoinder’. Journal of Political Economy 64: 69. Treuhaft, S. and Rubin, V. (2013). Economic Inclusion:  Advancing an Equity-​Driven Growth Model (Oakland, CA: PolicyLink). U.S. Government Accountability Office (2006). ‘Empowerment zone and enterprise community program: improvements occurred in communities, but the effect of the program is unclear’. Report GAO-​06–​727 (Washington, DC: Government Accountability Office). U.S. Small Business Administration (2010). Small Business Procurement Data Shows Significant Progress Toward 23 Percent Federal Contracting Goal (San Francisco, CA: U.S. Small Business Administration Press Office).

Just Growth   809 Weber, R.N. and Smith, J.L. (2003). ‘Assets and neighborhoods: the role of individual assets in neighborhood revitalization’. Housing Policy Debate 14: 169–​192. Weir, M. and Rongerude, J. (2007). ‘Multi-​level power and progressive regionalism. building resilient regions’. UC-​Berkeley Institute of Urban & Regional Development Working Paper 2007-​15 http://​www.iurd.berkeley.edu/​publications/​workingpapers.shtml (last accessed 25 April 2017). Yin, J.S. (1998). ‘The community development industry system: a case study of politics and institutions in Cleveland, 1967–​1997’. Journal of Urban Affairs 20: 137–​157.

Chapter 43

P olicy Th rou g h Practice:  L o c a l C om munit i e s , Se l f-​ Organiz ation , a nd P ol i c y Jennifer Clark Introduction Economic geography, as a disciplinary practice, fixes the lens of analysis on both the scale of economic action and the processes that determine how economic resources are distributed and concentrated across places. In other words, economic geographers study where the governance and coordination of economic activities occur (national, regional, or local), and how resources are allocated within and across cities, regions, and nation states (Clark et al., 2000; Christopherson and Clark, 2007). Consequently, economic geographers track, document, and analyse the dynamic and multiscalar shifts of policy responsibilities and authorities between scales of governance, as well as the processes of economic accumulation and disinvestment that happen in places and over time. Those processes are heavily influenced by policy choices at the sub-​national level (Martin, 2001; Christopherson and Clark, 2007; Clark, 2013). Economic geographers thus view institutions—​private and public—​as actors and as sites of action in policy processes that shape variation in the self-​organization practices of local and regional communities (Boschma and Frenken, 2009; MacKinnon et al., 2009). The first empirical challenge in understanding how these local and regional institutions operate is defining them. Public policy tends to categorize institutional intermediaries by legal status: private, public, or non-​profit. From an economic geography perspective, how institutional intermediaries operate within and across markets matters more than their legal status. And, increasingly, institutional intermediaries operating in the space between established institutions vary significantly in terms of form across cities and regions. These ‘third-​sector intermediaries’ are not exclusively non-​profits, but can also be for-​profits, philanthropies, quasi-​public (public–​private partnerships), benefit corporations, or public

Policy Through Practice    811 sector entities. They exist both as single-​site community-​based organizations and as multiscalar networks (vertical, horizontal, or both). This chapter focuses on these institutional intermediaries and how they contribute to the evolving practices of self-​organizing within local communities. The rapid development of local ‘third-​sector intermediaries’ is directly linked to the combined trends of devolution and privatization by neo-​liberal national governments in recent decades. In the USA, the UK, and Canada, local governments have faced difficult choices over the last three decades: (i) pull back from meeting the expanded scope and coverage of service obligations devolved to the local level, or (ii) meet expanded and new obligations leveraging alternative revenue strategies. The cumulative consequences of thirty years of these choices are now reflected in significant regional variation in the organization and implementation of social and economic policies at the local scale. This variation affects the capacities of places to provide not only everyday services to citizens, but also longer-​term infrastructure investments that influence locational decisions by firms, the crux of regional economic competitiveness in a global economy. In recent years, this has extended beyond supports for regional innovation systems to the hard information and communication infrastructure that enables innovation ecosystems—​fibre, wireless, cellular networks. Increasingly, it has become economically important for cities to produce an urban form that facilitates the transition to flexible work systems: dense, mixed-​use development with multi-​modal transit access. It is worth highlighting that third-​sector strategies bypass the metropolitan regionalism discourse of the last forty years and focus again on the city as the prioritized space for political and economic engagement. The examples identified here are characteristic of this approach of distributed networks organized as multiscalar coalitions promoting innovative urban governance and capacity building through policy innovation diffusion models. Common among them is the use of non-​governmental networked institutions to influence and seed city-​scale adoption of policies characterized through real-​time engagement in multiscalar and cross-​regional practice. What follows is a discussion of the developing ‘third-​sector intermediaries’ in cities and regions across the USA that illustrate the ways in which institutional intermediaries are operating in local and regional economies both through city governments and by developing new ways to design, implement, and diffuse policy models through new models of implementation. The chapter concludes with a discussion of the economic policy implications of these new modes of programme design, delivery, and decision-​making for regional economies and uneven development.

Localizing Policy Beginning with the significant national economic policy changes of the 1980s, policy responsibilities and authorities shifted significantly between national governments and their subnational counterparts. National-​scale policy reforms since the Reagan and Thatcher periods prioritized the devolution of policy responsibilities to the local level. Further, these policy initiatives encouraged privatization of economic activities, particularly public-​sector services, or services once perceived as more efficiently or equitably provided by the public sector (Bolan, 1973; Barnes and Ledebur, 1998).

812   Clark Although much of the critique of privatization has focused on the erosion of social services and the dismantling of worker protections in the USA in particular, the privatization trend covered a vast amount of economic terrain in terms of the affected market domains. For example, privatization affected housing markets, labour markets, financial markets, business services, and social services (from social welfare to automobile insurance). The subsequent patchwork of geographically uneven and partial regulatory regimes has encouraged ‘jurisdiction shopping’ among firms as they make decisions about the expansion of existing operations and the location of new operations. This patchwork has contributed to greater differentiation in the market institutions across regional economies—​in their coverage, services, specializations, and sophistication (Malecki, 2004; Feldman and Martin, 2005; Christopherson and Clark, 2007). In the USA and Canada, the variation between states and provinces in the regulation of some markets (particularly labour markets) can exceed the variation seen across nation states (Clark, 1989; Finkin, 2001; Peck, 2002a; Stone, 2004; Leigh and Clark, 2010). In addition to the uneven privatization of formerly public services, the models of policy programming by the public sector vary from locality to locality. These programme models diverge in terms of mode of implementation and level and quality of coverage. Devolution has enabled this flexibility in policy design and implementation and emphasizes the value of ‘tailoring’ policy for local conditions. Within some nation states there is significant variation to accommodate. In the USA and Canada, jurisdictions differ in the size of state and local governments in terms of geography, population, and markets. Further, the underlying asset and factor conditions between places differ substantially in ways that affect politics and economies, including natural resource endowments (e.g. energy), industrial legacies, rural and urban population distributions, cultural conditions, and the orientation and character of internationalization. As a consequence, the density of institutional and governance frameworks vary at the local scale. Jurisdictions with larger populations and larger economies often have more capacity to develop and deliver policy programming at the local and regional scale. In many ways, subnational implementation flexibility and tailoring is necessary, inevitable, and desirable. This argument has been made by many economic geographers in the context of regional innovation policies (Gertler and Wolfe, 2002; Clark, 2010). In fact, there has been a near policy consensus around the rationale for flexibility in regional innovation and development policies aimed at building competitive local and regional economies (Clark and Christopherson, 2009). The question of coverage and quality of social services, however, has remained more divisive and ideological. This chapter focuses on the production of local and regional policy variation within local communities. The chapter distinguishes between two different ways in which local communities have pursued self-​organizing in response to the increase in autonomy facilitated through devolution. Firstly, through existing democratic processes for the purposes of expanding public services or increasing the role of the state in the regulation and/​or coord­ ination of private market actors. And, secondly, through the expansion of ‘third-​sector institutions’ (local and regional non-​governmental organizations) operating within private market spaces to expand the provisioning of public services and collective goods or enhance the efficient delivery of new or existing services (often using novel technology) beyond the boundaries of local government.

Policy Through Practice    813

Filling the Gaps: What Neo-liberalism Leaves Behind Economic geographers have documented the negative consequences of neo-​liberal trends, and we have noted significant declines in the quality and scope of public services and the economic disadvantages of the privatization of public goods (Eisenschitz and Gough, 1996; Peck, 1996; Peck, 2002b; Hackworth, 2007; Birch and Mykhnenko, 2009). Labour markets have garnered particular attention in the analysis. As a result, labour markets provide a familiar and compelling illustration of the ways in which local communities have experienced and then used evolving approaches to local organizing through third-​sector intermediaries to tailor policy to localized priorities, conditions, and capacities. The key question then becomes: ‘When the state steps away from mediating market failures and provisioning public services, what emerges in its place?’ The decline of labour unions, first in the private sector and later in the public sector, coupled with the rise of for-​profit labour market intermediaries (including various forms of temporary employment firms) demonstrated the ability of neo-​liberalism to erode effective and seemingly embedded institutions (Clark, 1989; Harrison, 1994; Martin et al., 1996; Doussard, 2013). In other words, the ideological transition from a Keynesian to a neo-​liberal approach to governmental action did not simply change the scope of public sector activity in markets, but also affected the scope and coverage of third-​sector intermediaries. In redefining the scope and function of labour-​market intermediaries like labour unions, the transition to neo-​liberalism decoupled labour market services from labour market advocacy. The shift created a new and expanded market for private for-​profit labour market intermediaries while eroding the importance of the advocacy and bargaining services provided by labour unions (Benner, 2003; Van Jaarsveld, 2004; Carnoy et al., 1997; Peck and Theodore, 2002). It is important to recall that the national-​scale neo-​liberal project championed devolution and deregulation ostensibly to ‘empower’ localities and promote flexible forms of regional governance. Conversely, it was largely assumed that the emergent forms of flexible regional governance would align with the neo-​liberal national model and would lean towards a diminished role of the state rather than expanded role. In other words, policymakers expected that local communities would seek industry partnerships and private market solutions to intractable issues of infrastructure management and social service delivery in line with national political priorities. Shrinking budgets would subsequently force cost savings in labour markets and promote a leaner, more efficiency-​based model for the public sector at all scales. Thus, devolution in the context of neo-​liberalism is operationally distinct from arguments about local and regional policy flexibilities for the purposes of accommodating differences in regional economic conditions. As a political process, devolution is often more about the replication of a model across scales, not the tailoring of programming to accommodate regional variation. That said, it has been the case that the rise of neo-​liberalism at the national scale has led to waves of privatization of public services in the USA, the UK, and Canada and the devolution of those services to state, provincial, and local actors (Peck and Theodore, 2015). In response to greater responsibilities without a corresponding increase in capacities or resources, local

814   Clark governments ‘contracted out’ to private market actors everything from social services to transportation systems (Warner and Gerbasi, 2004). This accompanying and necessarily deregulated environment created less rigorous standards for evaluation, performance, and efficiency simply because of the lack of capacity to measure, analyse, and ensure results (Clark, 2013). Devolution shifted the cost of service provisioning to local governments that were thus faced with either increasing revenues (through taxes, fees, fines, or borrowing) or cutting the services provided. In the USA, the increase in the use of fines and fees by local governments for revenue generation is a case in point. Often, increased economic autonomy, including the authority to generate revenue, did not accompany the greater responsibility for public services. Services subsequently declined or failed to expand as technologies changed, leaving a gap unevenly addressed by private market actors (Ashton et al., 2016). In response to this policy environment a variety of divergent trends emerged. There has not been a simple replication of national neo-​liberal policies at the local and regional scale. In part, this is a function of the demand for new and different public services motivated by new technologies and expanding markets. For example, the rapid development and diffusion of information and communication technologies have put pressure on local governments to consider how to design and deliver an entirely new category of services. Not surprisingly, when the public sector is slow to provide a public service, other institutional intermediaries emerge. These intermediaries show a great deal of regional variation in coverage, scope, and form. They often develop from the bottom up rather than the top down, thus making it difficult to differentiate them based on a standardized typology or compare them neatly across communities. It is possible, however, to categorize them by function. These intermediaries are stepping into the spaces as new facilitators for economic activities ranging from labour to innovation to infrastructure to networked systems. In no place is their presence more evident than in cities (Clark, 2012, 2014). They are negotiating the porous boundaries between firms, governments, and formal institutions. And, although they have evolved in the shadow of neoliberal privatization, they vary significantly in their character—​bottom-​up, community-​based, and solution-​driven (Pastor et al., 2009; Benner and Pastor, 2012). In effect, a technology-​savvy generation has become a new cohort, re-​discovering a ‘public’ or ‘civic’ orientation and developing new ways to expand the capacity to act in the spaces that neoliberalism left behind.

Local and Regional Responses: Policy Innovation as Design Practice As noted in the previous section, there are several ways in which local communities have pursued the push towards self-​organization that devolution required. The empirical emphasis in this chapter is on the expanding third-​sector strategies and the institutional intermediaries that implement them operating outside of existing democratic processes to pursue policy innovations—​both in policy design and implementation. In order to discuss this evolving category, it is necessary to distinguish it from the policy practices operating within existing democratic processes—​city ordinances, referenda, bond issues, state laws, and advocacy or

Policy Through Practice    815 lobbying coordinated with political parties. Third-​sector strategies are effectively agnostic about the role of the state in mediating market conditions. These institutional intermediaries and the strategies they pursue simultaneously accept the role of the state as given and as dynamic. They emphasize flexibility in policy design and in the structure of implementing organizations. In this way, their actions parallel political efforts to pursue an increased role of the state, as well as those corresponding efforts to pursue a decreased role of the state through increased local and regional autonomy. The functionality of third-​sector intermediaries is not contingent on political circumstances.

Progressive Local Government in the Public Sector It is important to emphasize the point about regional variation in response to devolution and disinvestment in the public sector. The ways in which public or collective goods and services are provisioned from place to place as the state steps back is not fixed. The question of who steps in to fill these policy gaps—​local government, the private sector, traditional non-​ profits, or no one at all—​does remain contingent on place, as well as time. In some cases, where the public sector pulls back, the work just no longer happens. In some cases the work is reconfigured to reduce coverage or scope for the goods or services provided. And, in some cases, the work is picked up by intermediaries who find a way to allocate public services or collective goods using market mechanisms. A significant number of local governments have not implemented neo-​liberal policy priorities in alignment with the shift in national policy towards privatization. Some localities have adapted to the agency they inherited from the national scale by developing innovative tools for implementing policies affecting regional competitiveness from localized healthcare systems to energy-​efficiency programmes through progressive local governments. These policy responses are reactions to the inaction of national governments on topics ranging from climate change to demographic transition. For example, over 1,000 US city mayors have signed the US Conference of Mayor’s Climate Protection Agreement, effectively ratifying the terms of the Kyoto Protocol at the local level.1 These actions are generally affected through traditional policy mechanisms, such as the passage of ordinances or laws through city councils. They also often result from political advocacy aligned with party affiliations. These efforts challenge a ‘taken-​for-​granted’ political assumption that devolution would produce the local replication of national (or even state) neo-​liberal policy priorities. The assumption was that the devolution project would breed neo-​liberal market intermediaries to occupy the space and take up the work of provisioning public and collective goods and services from the public to the private sector. The reality is more complex and nuanced. Instead, policy flexibility has resulted in diversity of local implementation of policy priorities producing increased variation across policy domains. The global recession motivated policy innovation at the local level to expand from the relatively limited world of environmental advocacy and discussions about sustainability to an array of economic policy interventions that affect firm strategies, labour markets, and innovation systems. The result is the development of policy innovations designed to expand markets that push back on the practices of policy diffusion that originally formed as top-​down processes (Peck and Theodore, 2015). These diffusion models are increasingly bottom-​up and horizontally networked.

816   Clark Several decades after the neo-​liberal project began, localities have begun to modify economic policy models and governance regimes, and also to expand the network of actors engaged in policy-​making and systems governance at the local-​scale (Clavel and Wiewel, 1991; Clavel, 2010). Handed the reins of regional autonomy, local governments have begun to engage in intentional acts of localized agency in keeping with the free-​market, deregulatory, and privatization models articulated at the national scale. Increasingly, however, local actors have found spaces to form distinctively different models and institutions under that same umbrella of national policy prescriptions. Frequently those local actors exist in parallel to the public sector stymied by both deregulation and disinvestment.

Third-​sector Strategies: New Forms of Local Policy Innovation and Diffusion The discussion of traditional uses of local governments to create and maintain policy variation at the sub-​national scale highlights what is distinct about the expanding role of third-​ sector strategies and the evolution of the institutional intermediaries that promote them. Firstly, these institutions and the strategies they promote are less transparent than strategies initiated by local governments. Secondly, they are intertwined with market-​based approaches to technology diffusion, regional innovation, and economic competitiveness. There is significant emphasis on the use of technology to expand markets, increase access, and, ultimately, to build out a flexible and networked infrastructure of service provisioning that enables many forms of entrepreneurship (social, public, and traditional) within existing institutions and in parallel to them. The disinvestment in the public sector has not just affected the provisioning of public and collective goods and services, but it has also limited the development of policy thinking about both the processes of policy design and implementation. New technologies have produced a range of new policy questions, many of which require decisions about the design and adoption of new models. The absence of clarity about best practices leaves policymakers searching for adoptable models, often without the tools to assess the varied and evolving implications from a public-​sector perspective. This issue was recently highlighted by Jamie Peck and Nik Theodore: The modern policymaking process may still be focused on centres of political authority, but networks of policy advocacy and activism now exhibit a precociously transnational reach; policy decisions made on one jurisdiction increasingly echo and influence those made elsewhere; and global policy ‘models’ often exert normative power across significant distances (2015, p. 3).

The following subsections describes three models of empirical examples from the design and diffusion of urban innovations in the US context.2 These models illustrate how non-​governmental actors assumed programming responsibilities abandoned by the public sector, as well as the new spaces that emerged in the new economy. These third-​sector intermediaries underscore significant regional differences in how policy value is developed through the process of ‘collective’ action and ‘public’ design of policy (as opposed to ‘public policy’). Notably, just as technology plays a critical role in expanding new services and capacities to

Policy Through Practice    817 offer old services more efficiently within and across local communities, technology also enables innovative approaches to policy design and diffusion. These institutional intermediaries also act in a variety of policy arenas or program delivery areas. Principally, there are four types of policy spaces. All four touch on the expanded ability to deliver services at the local (almost always urban) scale. Firstly, they work on provisioning new services through the diffusion of new technologies (municipal wireless, fibre, small cells). Secondly, they expand on how to provide old services through the diffusion of new models (social and human services; labour market reproduction). Thirdly, they organize access to services to new or emerging populations (ageing, immigrant, accessibility-​ challenged, and younger generations). Finally, they develop new forms and mechanisms of ‘democratic engagement’ that, in many ways, substitute formal governmental mechanisms of decision-​making with civic innovation. The actors involved are using new approaches to urban governance by adopting a technology diffusion model rather than traditional policy strategies. The approach to policy exchange is flexible and loosely coordinated. It privileges cities and skips states. This strategy contrasts with traditional approaches to sub-​national policy-​making that focuses on statehouse lobbying and developing coalitions among established interest groups (Lowi, 1967). Each model uses a different strategic approach. The first approach involves seeding self-​ organization at the local-​scale through direct investments in city governments by external, national, or international non-​governmental organizations. The second strategic approach involves bottom-​up organizing using a distributed network. This second approach often starts as a single-​site, community-​based pilot of programme delivery that later connects/​ coordinates with a loosely distributed network of similar organizations. The third approach begins as an intentional national network deployed in multiple sites through organizations operating outside the public sector. A key difference between these three strategies is in their relationship to city governments. In the first case the strategy operates through the public sector and in the second and third case new institutional forms emerge to operate in parallel to it. The examples of how these three strategic approaches operate illustrate the key differences.

Model 1: Strategic Capacity Seeding in Cities Strategic seeding of self-​organizational capacity in cities largely occurs through philanthropic funding intended to seed new capacity through technical assistance and professional staff placed in city governments. These investments are explicitly urban, made directly to cities, and are deeply connected to technology-​led and market-​driven approaches to achieving shared civic-​oriented goals, such as urban resilience, sustainability, and efficiency. One case of this strategic approach is the Innovation Delivery teams (I-​teams) seeded in Atlanta, Chicago, Louisville, Memphis, and New Orleans by the Bloomberg Philanthropies in 2011 and expanded to twelve more cities in 2014.3 An additional iteration of the approach is the City Energy Project seeded in ten US cities in 2014 (Atlanta, Boston, Chicago, Denver, Houston, Kansas City, Los Angeles, Orlando, Philadelphia, and Salt Lake City) by the Bloomberg Philanthropies, the Kresge Foundation, and the Doris Duke Foundation, and supported by technical staff in two national organizations, the National Resource Defense Council and the Institute for Market Transformation.4 The City Energy Project also funded

818   Clark staff positions to administer the three-​year initiative within selected city governments. The Rockefeller Foundation’s ‘100 Resilient Cities’ project is another example that involves placing ‘resilience officers’ in city governments and facilitating a network of policy exchange. In this model, the philanthropy is seeding self-​organization by providing technical assistance and professionalized staff directly to the recipient city inline with the philanthropies strategic priorities for urban innovation and sustainability.

Model 2: Bottom-​up Best Practices The second strategic approach follows more closely a traditional community-​based advocacy approach. Self-​organized and mission-​driven institutions emerge in order to address policy gaps or market failures. Increasingly these institutional intermediaries are forming loose networks for best practice policy exchange. Key examples of this form revolve around small-​firm advocacy and scale up, as well as labour market advocacy and incumbent worker training. Information exchange about market conditions and strategic approaches to accessing suppliers, materials, and other resources are especially important to Maker Movement actors, such as Maker’s Row, ADX Portland, SF Made, and the national network formed by some of these intermediaries, the Urban Manufacturing Alliance.5 A labour market example of a similar model is the New York City-​based Freelancers Union, which supports a network of ‘New Mutualism’ sites.6 Similarly, the National Domestic Workers Alliance runs advocacy campaigns in multiple cities while simultaneously using information technology to create national portals to establish certifications, define market norms, set industry standards, and start up its own domestic worker job-​matching services.7 This model operates outside of city governments. And, because the institutions are not dependent on local political conditions, they can operate in places without majority political support for their work.

Model 3: Capacity-​building Franchise Arrangements The third strategic approach to localized policy diffusion, the capacity-​building franchise approach, operates through the private sector rather than city governments, as is the case in Model 2. However, unlike in Model 2, these institutional intermediaries are market driven rather than mission driven. Typically, the service or programme provided is delivered through a purpose-​designed new institutional intermediary. The ‘franchise approach’ has evolved with multiple local programmes deployed in tandem across communities to form a networked model from the beginning rather than evolving it over time. In this way, this model overlaps with Model 1’s commitment to a multiscalar distributed network intended to facilitate loose collaborations motivated by best-​practice exchange from the beginning—​ with the explicit goal of establishing a network of programme implementation. Because these intermediaries are market driven, they are more often formed as for-​ profit rather than non-​profit organizations and look more like the occupational, sectoral, and technology-​specific intermediaries that have emerged in policy arenas across regional economies (Clark, 2014; Bryson et  al., 2015).8 Key examples of this model include Tech Shop, which operates membership-​based maker spaces in cities across the USA in order to provide small businesses and entrepreneurs with access to flexible production spaces and

Policy Through Practice    819 equipment. Another example is General Assembly, which operates a flexible technology training model in the USA and the UK to provide technology workers with access to skill upgrading and networking opportunities. A growing number of co-​working spaces use a for-​profit membership model to provide flexible work spaces and shared services to contract contingent workers. This model ventures further into private-​market spaces than the other two and competes both with more established and larger business and training services firms (including for-​profit universities) with mission-​driven non-​profits in Model 2. The boundaries between these models are relatively porous. Some third-​sector intermediaries begin life as for-​profits with a social entrepreneurship orientation and later reassess their organizational structure to become non-​profit organizations or re-​file as B-​corps (benefit corporations) in places where that legal status is available. Similarly, although some of these intermediaries originate as a set of networked projects, some evolve from singles sites into a distributed network. The dynamic nature of these institutional intermediaries is one reason to keep them categorized together as third-​sector intermediaries based on their function in mediating markets and filling policy gaps rather than their location or legal status.

Policy Implications The role of institutional intermediaries in shaping regional competiveness is not a new subject of inquiry in economic geography. Most recent theoretical turns in the discipline, including evolutionary economic geography, emphasized the role of institutions (MacKinnon et al., 2009, Boschma and Frenken, 2009; Gertler, 2010). However, the theoretical literature has yet to develop a body of empirical work with tangible examples illustrating policy implications to a broader audience. What is clear is that intermediaries defy easy policy categorization and are explicitly, and increasingly, tied to innovative governance in city regions rather than nation states. In the wake of both economic and political transitions—​ranging from devolution to the Great Recession—​local communities have adopted varied responses to governance and self-​organization, reorienting their regional economies and the institutions that bind them together. These responses are uneven, incomplete, and regionally differentiated. Recent trends towards devolution in the USA, the UK, and Canada have added a new intensity to discussions about regional autonomy, self-​governance, and multiscalar policy frameworks. There is an active policy debate about how to balance the role of the nation state with the autonomy of regions and localities in the context of a global economy. These discourses are simultaneously political and economic, and weave together popular movements for increased political autonomy and increased self-​determination in sub-​national regions with more explicitly regional economic discussions around privatization, internationalization, and the role of the state (as well as the appropriate scale of that state) in economic policy decision making, implementation, and regulation. The debate about scale is about growth, governance, and the functional alignment of the two. Ultimately, these are questions shaped by economic geographies. Peaceful and democratic independence referendums in Scotland, Quebec, and Catalonia underscore the ongoing debates about scale of governance and point to the varied localized

820   Clark responses to neo-​liberal policies adopted by nation states.9 In turn, questions about the engagement and disengagement from the European Union by nation states like the UK and Greece highlight the dynamic and multiscalar character of these discussions about the role of the nation state in shaping policies for regional communities who perceive a distinction between the national policy regimes under which they are governed and the values of their communities. For example, following the 2014 Scottish independence referendum and the national election in the UK in 2015, the First Minister and head of the Scottish National Party, Nicola Sturgeon, explained the crucial point about fiscal autonomy and scale of action to a US audience at the Council of Foreign Relations in Washington DC in June 2015: But we (Scotland) want maximum fiscal powers within the United Kingdom, short of Scotland being an independent country. Why do we want that? Not for its own sake, but because the more powers we have, the more fiscal responsibility we have, the more ability we have to shape things like our system of social security, the more able we will be to grow our economy, to make sure we’re doing the things and pursuing the policies that help us to attract investment and create jobs and grow our economy faster and more sustainably. So its powers and responsibilities for a purpose, and we will make those arguments that in the Westminster parliament as the debate about for the autonomy for Scotland continues over the—​the weeks and months to come.10

And so the distinction between the expansion and contraction of the role of the state through existing governance mechanisms remains central to the ability of local communities to self-​ organize, even as the development of and investment in third-​sector regional intermediaries captures attention and resources. Institutional intermediaries operating outside of the public sector—​those operating in the third sector—​have the ability to shake the status quo by reconfiguring essential power asymmetries between scales of governance, notably access to information. However, formal local autonomy is required to develop models of fiscal independence capable of supporting a broader array of public services and to manage the territorial boundaries of the highly differentiated spaces that emerge on the other side. In other words, institutional intermediaries can pursue some expanded services and provide some collective goods through the third sector, but, as Nicola Sturgeon notes, they may not be able to finance those actions over time. Technological change enables new forms of local organizing in response to shifting market conditions. Portals and social media platforms have proven effective for rapid information exchange and knowledge diffusion and. minimize many of the transaction costs associated with building coalitions and coordinating constituencies. Effective portals like Maker’s Row reduce supply-​chain search costs, thus mitigating the costs to individual small producers and entrepreneurs of finding both suppliers and customers. For places covered and served by such intermediaries, flexible production systems are more economically viable. Similarly, effective labour market matching portals like the Freelancers Union (on the more high-​tech side of the labour market) or the National Domestic Workers Alliance (on the more low-​tech side) increase the availability of labour market information available both to the workers and employers. In effect, these intermediaries provide the required structure to make flexible labour and flexible production work. Consequently, places with a density of these institutional intermediaries are more competitive in the emerging knowledge economy, regardless of the services provided by

Policy Through Practice    821 the public sector. Increasingly, it is local organizing through third-​sector strategies that produces variations in regional capacity to absorb technology and adapt to new forms of work and production. These third-​sector strategies highlight innovative approaches to localized self-​organizing. The solutions, however, are highly contingent, and they tend to be territorially bounded in ways that limit the diffusion of best practices. For example, third-​sector institutional intermediaries often attempt to organize access to services to new or emerging populations but find that real resource constraints—​primarily urban infrastructure investments—​limit access. For example, Internet portals may contain extensive labour market information, but for workers without reliable Internet access, the information is still inaccessible. Crowdsourcing strategies that aggregate information on the best bicycle routes across the city can provide useful information to citizens but they cannot build new bicycle lanes. Both policy and market failures produce dynamic spaces for institutions to develop innovative models for mediating these gaps. Thus, neo-​liberalism produces a range of new ‘value propositions’. These are new spaces for market interventions and revenue generation to fund collective actions. Social impact bonds or ‘pay for success’ arrangements are just one innovative financial instrument operating in this space. Pioneered in the UK to use public monies that would be used to incarcerate reoffending former prisoners, the mechanism uses a portion of those funds to pay a third-​sector intermediary to provide services to avoid reincarnation in the first place. The social impact bond arrangement, and other such arrangements, allows governments to use financing approaches to leverage anticipated costs to finance early interventions—​particularly in public safety, education, and other social interventions. What is clear from these third-​sector strategies is that private-​sector policy solutions are not necessarily adversarial or extractive. In some cases, they enable the expansion of public services through direct delivery or by expanding the revenue opportunities for city governments. They do operate, however, outside of traditional governance mechanisms (the third sector) and are often less transparent to citizens and policymakers (complex financial arrangements or unfamiliar or novel technologies). Priorities are set whereby revenue can be generated not by what is selected directly by voters or city councils. It is well understood among economic geographers that neo-​liberalisms often erode the base of established and effective institutions that mediate power asymmetries between cap­ ital and labour in clear and important ways. However, the core elements of neo-​liberalism—​ devolution, deregulation, and disinvestment—​do not necessarily lead to the same set of shared, replicable ideologically neo-​liberal outcomes across places. Instead, actors, institutions, and coalitions in local communities pursue a wide variety of strategies to shape places that reflect local priorities. Through third-​sector strategies, local communities have found a way to work around the disinvestment of the public sector. The conclusion, then, is that devolution and deregulation lead to new forms of organizing and new sites of action—​not just new markets. Local communities, and specifically cities, are where that action plays out. For economic geography, this produces an evolving and complex landscape of actors and processes that provision collective goods and once ‘public’ services in local and regional economies. The configuration of these processes and the characteristics of these third-​sector actors are locally tailored, territorially bounded, and contingent across time and space. In other words, the economic landscape produced is increasingly varied, uneven, and, in profound ways, unequally equipped and prepared for the challenges of the knowledge economy

822   Clark and its new forms of work and production. For economic geographers, this dynamic envir­ onment underscores the importance of the core disciplinary project of tracking, documenting, and analysing the dynamic and multiscalar shifts of policy responsibilities and authorities between scales of governance, as well as the processes of economic accumulation and disinvestment that happen in places and over time.

Notes 1. See US Conference of Mayors Climate Protection Agreement: http://​www.mayors.org/​climateprotection/​agreement.htm (last accessed 27 April 2017). 2. This is not to say that these processes and the evolving institutional forms that enable them are unique to the USA. There are also examples in the UK and Canada, as well as other countries. See Peck and Theodore (2015) for examples of policy model diffusion across Latin America. 3. See ‘Bloomberg philanthropies expands innovation teams program to 12 new American cities’ http://​www.bloomberg.org/​press/​releases/​bloomberg-​philanthropies-​expands-​ innovation-​team-​program-​12-​new-​american-​cities/​ (last accessed 27 April 2017). 4. See City Energy: http://​www.cityenergyproject.org/​about/​ (last accessed 27 April 2017). 5. See SF Made at http://​www.sfmade.org; Makers Row at http://​makersrow.com/​how-​it-​ works, and also The Urban Manufacturing Alliance at http://​urbanmfg.org/​about (all last accessed 27 April 2017). 6. See the Freelancers Union at https://​www.freelancersunion.org and ‘What is new mutualism’ at: https://​www.freelancersunion.org/​blog/​dispatches/​2013/​11/​05/​what-​new-​mutualism/​(both last accessed 27 April 2017). 7. See the National Domestic Workers Alliance at http://​www.domesticworkers.org/​who-​ we-​are (last accessed 27 April 2017). 8. See Anderson (2013), Miller Center (2014), and Hatch (2013). 9. See Toro (2012). See also Bennhold and Erlanger (2015) and Reuters (2015). 10. See Council on Foreign Relations (2015).

References Anderson, C. (2013). ‘Maker movement’ Wired, 16 April https://​www.wired.com/​2013/​04/​ makermovement/​(last accessed 27 April 2017). Ashton, P., Doussard, M., and Weber, R. (2016). ‘Reconstituting the state: city powers and exposures in Chicago’s infrastructure leases’. Urban Studies 53: 1384–​1400. Barnes, W.R. and Ledebur, L.C. (1998). The New Regional Economies: The U.S. Common Market and the Global Economy (Thousand Oaks, CA: SAGE). Benner, C. (2003). ‘Labor flexibility and regional development: the role of labour market intermediaries’. Regional Studies 37: 621. Benner, C. and Pastor, M. (2012). Just Growth: Inclusion and Prosperity in America’s Metropolitan Regions (London: Routledge). Bennhold, K. and Erlanger, S. (2015). ‘Nicola Sturgeon, star of Scottish politics, vows to secure more power’. The New York Times, 8 June https://​www.nytimes.com/​2015/​06/​09/​world/​europe/​ nicola-​sturgeon-​star-​of-​scottish-​politics-​vows-​to-​secure-​more-​power.html (last accessed 27 April 2017).

Policy Through Practice    823 Birch, K. and Mykhnenko, V. (2009). ‘Varieties of neoliberalism? Restructuring in large industrially dependent regions across Western and Eastern Europe’. Journal Of Economic Geography, 9: 355–​380. Bolan, R.S. (1973). ‘Planning and the new federalism’.. Journal of the American Institute of Planners 39: 226. Boschma, R. and Frenken, K. (2009). ‘Some notes on institutions in evolutionary economic geography’. Economic Geography 85: 151–​158. Bryson, J.R., Clark, J., and Vanchan, V. (eds) (2015). Handbook of Manufacturing Industries in the World Economy (Northampton, MA: Edward Elgar). Carnoy, M., Castells, M., and Benner, C. (1997). ‘Labour markets and employment practices in the age of flexibility: a case study of Silicon Valley’. International Labour Review 136: 27–​48. Christopherson, S. (2002). ‘Why do national labor market practices continue to diverge in the global economy? The “missing link” of investment rule’. Economic Geography 78: 1–​20. Christopherson, S. and Clark, J. (2007). Remaking Regional Economies: Power, Labor, and Firm Strategies in the Knowledge Economy (New York: Routledge). Clark, G.L. (1989). Unions and Communities Under Siege: American Communities and the Crisis of Organized Labor (Cambridge and New York: Cambridge University Press). Clark, G.L., Feldman, M.P., and Gertler, M.S. (2000). The Oxford Handbook of Economic Geography (Oxford and New York: Oxford University Press). Clark, J. (2010). ‘Coordinating a conscious geography: the role of research centers in multi-​ scalar innovation policy and economic development in the US and Canada’. Journal Of Technology Transfer 35: 460–​474. Clark, J. (2012). ‘Is there a progressive approach to innovation policy?’ Progressive Planning Winter: 17–​21. Clark, J. (2013). Working Regions: Reconnecting Innovation and Production in the Knowledge Economy (London: Routledge). Clark, J. (2014). ‘Manufacturing by design:  the rise of regional intermediaries and the re-​ emergence of collective action’. Cambridge Journal of Regions, Economy and Society 7: 433–​448. Clark, J. and Christopherson, S. (2009). ‘Integrating investment and equity: a critical regionalist agenda for a progressive regionalism’. Journal of Planning Education and Research 28: 341. Clavel, P. (2010). Activists in City Hall: The Progressive Response to the Reagan Era in Boston and Chicago (Ithaca, NY: Cornell University Press). Clavel, P. and Wiewel, W. (1991). Harold Washington and the Neighborhoods: Progressive City Government in Chicago 1983–​1987 (New Brunswick, NJ: Rutgers University Press). Council on Foreign Relations (2015). ‘Assessing Scotland’s future. A conversation with Nicola Sturgeon’ http://​www.cfr.org/​united-​kingdom/​assessing-​scotlands-​future/​p36589 (last accessed 27 April 2017). Doussard, M. (2013). Degraded Work: The Struggle at the Bottom of the Labour Market (Minneapolis, MN: University of Minnesota Press). Eisenschitz, A. and Gough, J. (1996). ‘The construction of mainstream local economic initiatives: mobility, socialization, and class relations’. Economic Geography 72: 178–​195. Feldman, M. and Martin, R. (2005). ‘Constructing jurisdictional advantage’. Research Policy 34: 1235–​1249. Finkin, M. (2001). ‘International governance and domestic convergence in labour law as seen from the American Midwest’. Indiana Law Journal 76: 143–​172.

824   Clark Gertler, M. (2010). ‘Rules of the game: the place of institutions in regional economic change’. Regional Studies 44: 1–​15. Gertler, M.S. and Wolfe, D.A. (2002). Innovation And Social Learning: Institutional Adaptation in an Era of Technological Change (New York: Palgrave Macmillan). Hackworth, J.R. (2007). The Neoliberal City:  Governance, Ideology, and Development in American Urbanism (Ithaca, NY: Cornell University Press). Harrison, B. (1994). Lean and Mean: The Changing Landscape of Corporate Power in the Age of Flexibility (New York: Basic Books). Hatch, M. (2013). The Maker Movement Manifesto: Rules for Innovation in the New World of Crafters, Hackers, and Tinkerers (New York: McGraw-​Hill). Leigh, N.G. and Clark, J. (2010). ‘North American Perspectives on Local and Regional Development’ in A. Pike and J. Tomaney (eds) Handbook Of Local And Regional Development, pp. 515–​526 (London: Routledge). Lowi, T. (1967). ‘The public philosophy: interest group liberalism’. The Americal Political Science Review 61: 5–​24. MacKinnon, D., Cumbers, A., Pike, A., Birch, K. and McMaster, R. (2009). ‘Evolution in economic geography:  institutions, political economy, and adaptation’. Economic Geography 85: 129–​150. Malecki, E.J. (2004). ‘Jockeying for position:  what it means and why it matters to regional development policy when places compete’. Regional Studies 38: 1101–​1120. Martin, R.L. (2001). ‘Geography and public policy: the case of the missing agenda’. Progress In Human Geography 25: 5–​17. Martin, R.L., Sunley, P., and Wills, J. (1996). Union Retreat and the Regions:  The Shrinking Landscape of Organized Labour (London:  J. Kingsley Publishers; Regional Studies Association). Miller Center (2014). ‘Building a nation of makers: six ideas to accelerate the innovative capacity of America’s manufacturing SMEs’ https://​www.nist.gov/​sites/​default/​files/​documents/​mep/​data/​MilsteinReport-​Manufacturing.pdf (last accessed 27 April 2017). Pastor, M., Benner, C., and Matsuoka, M. (2009). This Could be the Start of Something Big: How Social Movements for Regional Equity are Reshaping Metropolitan America (Ithaca, NY: Cornell University Press). Peck, J. (1996). Work-​Place:  The Social Regulation of Labour Markets (New  York:  Guilford Press). Peck, J. (2002a). ‘Labor, zapped/​growth, restored? Three moments of neoliberal restructuring in the American Labor market’. Journal of Economic Geography 2: 179–​220. Peck, J. (2002b). ‘Political economies of scale: fast policy, interscalar relations, and neoliberal workfare’. Economic Geography 78: 331. Peck, J.A. and Theodore, N. (2002). ‘Temped out? Industry rhetoric, labor regulation and economic restructuring in the temporary staffing business’. Economic and Industrial Democracy 23: 143–​175. Peck, J. and Theodore, N. (2015). Fast Policy:  Experimental Statecraft at the Thresholds of Neoliberalism (Minneapolis, MN: University Of Minnesota Press). Reuters (2015). ‘Political infighting in Catalonia clouds independence drive.’ http://​www.reuters. com/​article/​us-​spain-​catalonia-​idUSKBN0OX2DB20150617 (last accessed 27 April 2017). Stone, K. (2004). From Widgets to Digits: Employment Regulation for the Changing Workplace (New York: Cambridge University Press).

Policy Through Practice    825 Toro, F. (2012). ‘Quebec’s near win.’ The New York Times, 5 September. Van Jaarsveld, D.D. (2004). ‘Collective representation among high-​tech workers at microsoft and beyond: lessons from Wash Tech/​CWA’. Industrial Relations 43: 364–​385. Warner, M. and Gerbasi, J. (2004). ‘Rescaling and reforming the state under NAFTA: implications for subnational authority’. International Journal of Urban and Regional Research 28: 858–​873.

Chapter 44

In novation Hi g h ways a nd the Geo g ra ph y of Inclusiv e  G row t h Anita M. McGahan and Janice Gross Stein Introduction ‘Innovation for inclusive growth’ (IIG) has emerged as an important concept in the fields of economic development and health. At the core of IIG is the idea that the objective of growth policy must be to advance the prosperity of all, with special attention to the implications of growth for the poor (George et al., 2012). The reasoning for the emphasis on remediating poverty is that the consequences of inequality are widespread: when some citizens are not engaged, then society leaves fallow important resources, incurs welfare costs, cultivates political and social disaffection among the marginalized, and suffers ethically. An import­ ant mechanism for increasing inclusiveness is the encouragement of entrepreneurialism in resource-​limited settings (Radjou et al., 2011). A related idea is that the maximum return on investment from such entrepreneurialism arises when low-​cost products and services invented in settings of poverty are exported to relatively wealthier settings (Govindarajan and Trimble, 2011; Kumar and Puranam, 2012). Export may contribute to local comparative advantage and, eventually, the compounding economic benefits that arise from trade (Amsden and Chu, 2003). Despite the popularity of these ideas, the geography of IIG has not been fully considered. A separate literature on the economic geography of innovation (GOI) deals primarily with the challenge of improving local prosperity by encouraging the development of local industry—​usually in a city or region or nation (henceforth we shall refer primarily to ‘communities’ for simplicity, but we mean the range of geographical units). The important sub-​ streams of this literature emphasize different mechanisms for achieving prosperity. The first focuses on industrial clusters, which include groups of firms engaging in similar productive activities, as well as their suppliers, distributors, financiers, and customers. The second deals with national systems of innovation (NSIs), which arise from the activities of universities, patent systems, governmental technology investments, and other institutions that foster

Innovation Highways and the Geography of Inclusive Growth    827 entrepreneurialism and the commercialization of technology in established firms. While initially dealing with competitiveness at the level of nations, the NSI framework has also been applied to the level of cities and regions. In this chapter, we seek to link these literatures. We advance the idea of innovation highways as a new construct that links GOI theories on the mechanisms of prosperity with IIG theories on the mechanisms of inclusion. We argue that governance over innovation highways determines their inclusiveness, while the location of and speed on innovation highways shape the ways they advance prosperity. We conclude that further study will open up areas of inquiry regarding Innovation Highway location, speed, and governance to identify opportunities for improving prosperity and inclusiveness simultaneously, thus avoiding trade-​offs between prosperity and inclusiveness.

Brief Review of Core Constructs on the Economic Geography of Innovation The literature on GOI is extensive and has been described elsewhere in this Handbook. Here, we identify only a few of the major constructs for the purpose of later elucidating the gains from integration between the GOI and IIS literatures. The unifying idea that we seek to develop is that the GOI literature, with important exceptions, has focused primarily on mechanisms for raising prosperity in a particular community, neighbourhood, city, region, or nation. The basic logic is that proximity leads to knowledge spillovers, agglomer­ ation economics, dynamic insights, and specialization (Feldman, 1994; Aoyama et al., 2011). Collectively, all of these benefits of proximity enable commercializable innovation. Two primary mechanisms of this enablement have been identified and studied extensively. Scholars including Gertler (2003, 2010), Krugman (1991), and Porter (1990, 2000) have demonstrated that the co-​location of firms and their trading partners in clusters stimulates competitive innovation, promotes knowledge spillovers, increases agglomeration economies (of both scale and scope), and allows for specialization (Feldman, 1994; Malecki, 1997). Talented and creative personnel seek employment in clusters because of the opportunities that accompany specialization and scale. As firms in clusters benefit from local innovation capacity and become known internationally, their advantages compound and become reinforced (Spencer et al., 2010). This literature echoes many of themes in the literature on the microstructure of cities (Jacobs, 1969). The second mechanism is the NSI. The core argument in the literature on NSIs is that the institutional environment in a city, region, or nation is critical for enabling innovation (Malerba, 2004, 2005; Malerba and Nelson, 2011; Malerba and Adams, 2014; Nelson and Sampat, 2001). Taxation policy, investment incentives, trade policy, universities, education systems, and policies regarding the private ownership of intellectual property all influence private incentives for basic investments in technology development and the environment for commercialization of technologies once they are embedded in business models. Broadly, the institutional environment must be sufficient both to support commercial success and meet the needs of the local community (Rodrik, 2008; Murmann, 2013).

828   McGahan and Stein In both traditions, a principal conceptualization of performance is economic prosperity, often measured by growth, and the unit of analysis is a particular place. Comparisons are often drawn by assessing over time and between places the changes that arise from the implementation of particular programmes or policies designed to improve prosperity (i.e. such as through differences-​in-​differences comparisons of places with differential implementation of a policy (Banerjee and Duflo, 2011)). The goal is to identify the conditions under which agglomeration and the institutional environment give rise to the economic growth of the community.

Brief Review of the Literature on Inclusive, ‘Reverse’ Innovation While studies on the effectiveness of economic development policies have long been central to the fields of economics and political science, the literature on IIG has developed largely independently of research on the GOI. As a consequence, it differs in several major ways. One major difference between the GOI and IIG literatures is in the IIG’s focus on innov­ ation rather than growth as the primary target of policy and measure of performance (Lee, 2013). The theoretical conceptualization of ‘innovation’ employed is expansive and includes product innovation, process innovation, and even knowledge creation generally (Fagerberg et al., 2012). Growth and innovation may or may not be related in the short run (innovation may be costly or may not generate appropriable benefits), although they are related empiri­ cally in the long run (Adner and Kapoor, 2010; Gates, 2011). Many of the innovations celebrated in the IIG tradition involve simplifying existing commercialized products, processes, and technologies for lower cost (Govindarajan and Trimble, 2011). In some instances, the simplification that lowers costs also improves quality, thus pushing forward the productivity frontier (Porter, 1985). A second distinguishing feature of research on IIG is the emphasis on the distributional consequences of innovation: IIG studies take the remediation of inequality as its primary objective, and thus stand in contrast with GOI literature’s focus on growth (Cohen and Klepper, 1991, 1992). Like GOI research, IIG research takes a particular place as the primary unit of analysis, but, unlike GOI research, studies in the tradition of IIG point to mechanisms of openness as the conduit through which resources flow into and out of a particular place. The IIG literature stipulates that places in the Global South may be more innovative than those in the Global North both by necessity and because of the absence of institutional constraints on creativity (Govindarajan and Trimble, 2011; George et al., 2012). These claims must, of course, be interrogated to investigate their robustness and veracity. Yet the examples that have emerged in this literature suggest that, under at least a range of important circumstances, the innovation capacity of entrepreneurs and inventors in resource-​limited settings is important for remediating local inequality, fostering local growth, and improving prosperity in remote, globally established communities primarily by lowering the costs of established products and services. In the IIG tradition, three different strands of thinking about mechanisms have developed. The first, in the critical policy domain, focuses on the empirical phenomenon of

Innovation Highways and the Geography of Inclusive Growth    829 increasing inequality and persistent impoverishment in resource-​limited settings despite decades of foreign aid and policy intervention. Critical policy studies (i.e. Easterly, 2001, 2006) demonstrate how foreign aid to low-​income countries has led to social dislocation, political instability, and even adverse economic outcomes in low-​income countries (Moyo, 2010). At the heart of the challenge is the exportation of unaffordable systems even by the standards of high-​income countries into settings where infrastructure, norms of interaction, and complementary assets are not available (McGahan, 2012). Well-​intentioned interventions often have the unintended consequences of centralizing wealth and exacerbating inequality by displacing local entrepreneurs in poor communities rather than promoting inclusion. This literature calls for policies and interventions designed primarily to advance inclusiveness rather than aggregate growth (McGahan et al., 2013). A second stream of research in sociological and management studies demonstrates how ‘bricolage’—​making do with what’s at hand—​in impoverished environments has led to important technological breakthroughs. Bricolage is the hallmark of ingenuity, and is rife in resource-​limited settings, although insufficient to remediate poverty except under the most unusual of circumstances (Bhatti, 2012; Bradley et al., 2012). Yet several important breakthroughs based on bricolage have been documented (Lee and Lim, 2001; Ansari et al., 2012). These have occurred when entrepreneurs have specialized knowledge, fortuitous access to relevant resources, and awareness of the opportunity for invention. Much more research is needed to identify when bricolage remediates poverty rather than arises as an artefact of inequality (Hall et al., 2012). A third stream emphasizes ‘reverse innovation’ and was inspired by the procedural and product breakthroughs of institutions such as at the Aravind Eye Hospital in India (Govindarajan and Ramamurti, 2011; Govindarajan and Trimble, 2011; Stein, 2013). Originally motivated to offer cataract surgery to anyone who needed it, the hospital developed procedures that radically lowered the cost of care without compromising quality relative to standards established for resource-​rich settings. As the hospital continued operations, the volume of its services led to improvements in quality even beyond what is commonly achieved in some of the world’s most prestigious hospitals. As cost reductions compounded, the hospital and its approaches became ‘disruptive’ (Bower and Christensen, 1995)  to health systems designed under less stringent circumstances (Topol, 2011). Reverse innov­ ation occurs when breakthroughs in resource-​limited settings have the capacity to improve prosperity in resource-​rich countries either by lowering costs or improving quality or both (Govindarajan and Ramamurti, 2011; Govindarajan and Trimble, 2011; Markides, 2012; Tharoor, 2012). Noting the successes of such organizations, General Electric chief executive officer Jeffrey Immelt announced that the company would concentrate innovation activities in resource-​limited environments rather than invest incrementally on technologies developed in high-​income countries, where design improvements tend to accumulate in unaffordably costly ways (Immelt et al., 2009). These three streams all emphasize the importance of inclusiveness, and yet are not fully integrated either theoretically or normatively (see Shane, 1993; Simanis and Hart, 2008, 2009). A reason for this lack of integration is each stream’s reliance on phenomenological and empirical observation to point to critical gaps in standing theory. The emphasis in critical studies and, to a lesser extent, in the reverse-​innovation stream is on what’s missing rather than on how to develop more effective interventions. The emphasis in bricolage and in

830   McGahan and Stein the reverse-​innovation streams is the under-​theorizing of established practices in resource-​ limited settings. What is needed is to confront these gaps fully. Integrating insights from these streams is the central agenda of the emerging theory of IIG (George et al., 2012). As scholars identify the core constructs that characterize and describe the unique and unifying characteristics of inclusive entrepreneurship in resource-​limited settings, hypotheses emerge about how the poor are enfranchised and ultimately become successfully integrated into the framework for growth in their communities. Robust theories of inclusive innovation must account for the ex ante distribution of the risks and costs of innovation, as well as of the benefits (Schumacher, 1974; Thorsteinsdóttier et al., 2004; McGahan et al., 2013). When the benefits of commercialization are pursued without consideration for the vulnerability of the poor to the risks and costs of failed innovation (e.g. see Prahalad, 2004), commercialization itself loses legitimacy. The literature on IIG cannot develop theory until it considers the ex ante criteria that describe effective and fair policy rather than the ex post distributional consequences (McGahan, 2012; McGahan et al., 2013; Stein, 2013). A central problem in the practice of reverse innovation is the exportation of risks of development into resource-​limited settings, such as when pharmaceutical trials are conducted on impoverished patients. When those patients that bear the risks of failed innovation are not enfranchised in the benefits of successful innovation, then the distribution is not truly inclusive (McGahan, 2012). In other words, inclusive innovation must enfranchise the poor into opportunity for success, as well as of failure. Such enfranchisement is robust only when the poor have governance rights. The literature on democracy suggests that inclusive innovation requires enfranchisement of the poor in governance, as only the enfranchised are capable of asserting and claiming their rights to the benefits of inclusivity (Pierre and Peters, 2002; Kooiman, 2003; Klein et al., 2011; McGahan, 2012). Enfranchisement occurs with engagement in the processes by which innovation is governed. Governance in this instance includes decision rights, property rights, and managerial rights associated with the selection of relevant economic opportunities, the deployment of resources in pursuit of the opportunities, the distribution of the risks from failure, and the distribution of the gains from success (Rhodes, 1996, 1997; Skelcher, 2005; Klein et al., 2011). Recent studies suggest that the governance adaptation that characterizes inclusive innovation is successful only under certain conditions. Firstly, governance must clearly specify the constituencies included in the governance process while allowing for engagement of specialized experts and consultation with non-​governing stakeholders (Ostrom, 1990; Ansari et  al., 2012; Rezaie et  al., 2012). Governance for inclusive growth fails when institutional structures are so strict that outside experts are effectively barred from offering advice (consider the risks of building roads without engineering expertise); similarly, it fails when institutional structures are ill specified to the point that consultation of stakeholders becomes so extensive that governance decisions are bogged down in bureaucracy; it also fails when the marginalized have no voice from the beginning over the distribution of potential risks and benefits. Success in achieving inclusiveness depends, in large part, on clarity about boundaries (Libecap, 1989; Ostrom, 1990). Secondly, the results of innovation must be distinguished from the process of governance. Innovation is risky and may fail, and yet those enfranchised in the governance of innovation must not be excluded because of the failure of the innovation (Libecap, 1989; Ostrom, 1990; Klein et al., 2011). Consider the pharmaceutical industry, where the success rate on new drug

Innovation Highways and the Geography of Inclusive Growth    831 development is as low as one in 10,000. Robust IIG requires that an outcome of failed drug discovery does not constitute grounds for judging the process of governance as failed, or for disenfranchising the poor (or any other constituency) from governance processes. Effective governance over innovation must specify rules ex ante for distributing the risks of failure and allocating the fruits of success (Libecap, 1989; Ostrom, 1990; Rezaie and Singer, 2010; Klein et al., 2011). The tension between democratic governance and the unilateral leadership authority required to manage innovation processes effectively must be calibrated. Adaptation in governance is critical as the types of decisions, allocation processes, and risk-​management issues change over time as technology advances (Piore and Sabel, 1984; Libecap, 1989; Klein et al., 2011). Robust governance institutions are designed to resist change and to insure against failure, yet improvements in the technology of decision-​making, resource allocation, and risk management must be accommodated for governance structures to remain relevant. A critical tension arises as leaders charged with responsibilities for implementing well-​governed initiatives confront obstacles in the governance process designed for enfranchisement. In such situations, inclusiveness demands not only continued engagement of the poor, but also a process for negotiating the transition to updated governance approaches (Klein et al., 2011). This can be a challenging and demanding process that can consume time and resources, but is fundamental to a fair distribution of risks and benefits. In the next section, we suggest that integration of ideas, constructs, and relationships from the GOI and IIG traditions requires considering governance of both communities and of technological advance. The GOI and IIG streams each assess performance by making claims that explicitly or implicitly rest on the boundaries of the governable unit. • The GOI’s cluster tradition emphasizes inbound and outbound trade from the cluster as a primary indicator of economic prosperity. The performance of a cluster is often assessed relative to that of other clusters on the volume, value, and terms of trade. • The GOI’s NSI tradition points to the global competitiveness of a nation, region, or cluster as a central objective of public policy. The performance of the geographical unit is evaluated through assessment of its integration into the global financial institutions and global supply chains. • The IIS’s critical tradition suggests that, when a portion of the population is disenfranchised from economic opportunity, then the prosperity of the community is blunted relative to its potential. The community underperforms. This tradition points to trade-​ offs between local economic activity and inequality that can be broken through relationships between the geographical unit and external trading partners. • The IIS’s bricolage tradition points to underutilized fixed resources as the building blocks of invention. This literature emphasizes human ingenuity and problem solving—​informed by imported insights regarding technological opportunity—​as the keys for unlocking invention. In this line of research, the templates for invention reflect globally available knowledge, frameworks, and prototypes. • The IIS’s reverse-​innovation literature suggests that low-​cost innovations in resource-​ limited settings represent commercializable inventions that may be globally competitive. Through outbound trade, inventors may prosper even in local settings. Recent research in this stream suggests that the realization of the benefits of such trade depends on the availability of receptor facilities in remote markets for locally invented goods.

832   McGahan and Stein Each of the traditions contains elements of logic that describe how prosperity—​either through contemporaneous economic performance or innovation—​depends on the relationships between activity in a community and outside actors (Prahalad and Mashelkar, 2010). Yet relatively little is known about the structure of these connections, how they are created, and how they are governed.

Innovation Highways: Enabling Prosperity In this section, we introduce the metaphorical construct of the innovation highway as central to the relationships between the GOI and IIG literatures, and more generally to the study of economic geography. Innovation highways are conceptualized as connecting a community to other communities and to global systems. A community’s highways—​the pathways by which knowledge, transactions, and other economic assets flow across the community’s borders—​contribute to resource creation, market opportunity, and, perhaps, most importantly, collaboration opportunities. We argue that how a highway is governed will affect the distributive consequences of innovation. Examples of innovation highways include: • Trade agreements that enable transactions between economic actors in different communities. The World Trade Organization (WTO) was formed in 1994 as the result of the Uruguay-​round of negotiations under the General Agreement on Tariffs and Trade. The rules of the WTO were negotiated by representatives of member nations, each charged with governance responsibility in their jurisdictions. The WTO’s policies on intellectual property arrangements were immediately criticized for arising from governance processes that did not fully consider the interests of the poor in member countries. The rules governing trade in intellectual property were subsequently amended after more extensive consultation with constituencies that represented the interests of the poor. These amendments have been fundamental in enabling reductions in the price of critical medications that are essential to the treatment of diseases of the poor. Here the link between governance and the distribution of benefits from innovation is clear. • Economic relationships between emigrants in a destination country and family members in the home country. In Canada, new rules are emerging to govern the gifting of money from emigrants in Canada to their family members in the home country. The new rules consider that emigrants are often sponsored by their families because of the economic opportunities available in Canada. When emigrants start businesses, they often return payments to their families in originating countries; in other words, the returns to entrepreneurship are shared across jurisdictional boundaries. The new rules are designed to govern such transactions. • Contractual relationships between companies in a particular geography and those in remote areas. Over the last decade, the outsourcing of important innovation activities by corporations across international jurisdictional boundaries has grown significantly. Pharmaceutical companies, semiconductor manufacturers, software developers, and a range of other high-​technology businesses frequently write contracts that stipulate

Innovation Highways and the Geography of Inclusive Growth    833 desired innovative outcomes, as well as the terms on which such outcomes are paid. Governance over these contracts is often subject to informal processes rather than enforceable arrangements. As a result, subcontractors may either not be paid or may be underpaid for the innovations that they generate. • Student exchanges that occur regularly across universities located in different communities. Students from Harvard University in Cambridge, Massachusetts, regularly travel to the Great Lakes University in Kisumu, Kenya, in an educational exchange. The transfer of knowledge and the cultivation of relationships between the Cambridge and Kisumu communities enable innovation that could not otherwise occur: students collaborate to form non-​governmental organizations and start businesses, which must operate by the jurisdictional rules of the communities in which they are formed. Incorporation in Cambridge eases the burden on donors from the USA who seek documentation regarding non-​profit status to support the tax deductibility of contributions. Incorporation in Kisumu enables the employment of workers on site. To begin our analysis, we take the innovation highway as dyadic relationship between communities. A  community’s network of innovation highways includes its system of dyadic relationships. We conjecture that the dyadic highway, as well as a community’s network of highways, is important to economic prosperity. The metaphor of a highway suggests a number of important corollaries. The capacity for exchange in each direction on the highway represents the amount of innovation that occurs through collaboration by the partners. The balance of bilateral volume represents the character of the knowledge that may be exchanged (e.g. students from Cambridge bring relatively greater formal disciplinary training to Kisumu; students from Kisumu relatively greater understanding of the institutional environment for development to Cambridge). The speed of travel reflects the readiness for translation of critical innovation inputs across the communities. Tolls represent the transactional costs. Entry and exit points suggest that innovation exchange between two communities may be augmented or impeded en route. The infrastructure in each community for accepting and distributing the transferred knowledge is akin to distribution centres for trafficked goods and services. The allocation of responsibilities for repairing and maintaining the highways reflects the robustness of community commitment to the relationship and the nascent governance system. Although much more development of this idea is needed, we stipulate that the concept of the innovation highway between communities is important for understanding the performance of the communities on criteria that are central to both the GOI and IIG literatures. GOI research describes the factors that drive the prosperity of communities, but has tended to refer primarily to internal characteristics of the community rather than to the relationships between the community and outsiders and how these relationships are governed. IIG research deals primarily with outside relationships and internal equity, but also does not pay explicit attention to governance. The innovation highway accommodates both perspectives and offers an approach for integrating insights from the two streams through an analysis of governance mechanisms. We consider that innovation highways are the institutions of exchange between communities, and that to function they must be governed.

834   McGahan and Stein

The Governance of Innovation Highways: Enabling Inclusiveness Inclusive governance demands that the poor have a direct and immediate voice in decision-​ making about the pursuit of innovation opportunities, the allocation of scarce resources, the distribution of the risks and costs of innovation, and the distribution of the rewards to innovation. This voice must be represented not only internally, such as during elections, but also in governance over external arrangements with other communities. Enfranchisement also means that the poor participate in the opportunities created by the construction of innovation highways, and in the resolution of disputes between communities over how innovation between them will be governed. Representatives of the poor, in other words, have a direct voice in rule-​making about the innovation highway and the risks and benefits that innovation may bring into their communities. A conversation about governance of this kind is currently ongoing in Canada, where companies propose to build new pipelines to move oil to foreign markets, and aboriginal communities who claim rights to the land along the route insist on a voice in the governance of the risks that new pipelines may bring to their communities, as well as an agreement ex ante on the share of benefits they will receive once the pipeline is built. These conversations have been long and difficult, in part because no meta-​governance arrangements are in place to regulate these conversations and the parties will consequently turn to slow judicial processes to resolve differences. Here a governance deficit is blocking progress. Inclusiveness demands that all members of society are enfranchised in the processes for determining which innovation highways will be constructed, and whether and how such highways will be opened to participation across the community. IIG stipulates inclusiveness over the terms of trade, investment, and other inter-​community transactions. Crucially, the poor must be enfranchised in consideration over how risk is governed and socialized so that the poor benefit directly (not through second order redistributive mechanisms) from innovation. This conceptualization challenges those who argue that economies will grow through innovation, and the poor will benefit indirectly from greater prosperity. At the same time, the GOI literature on agglomeration demands that innovation highways be constructed to contribute to local advantages that accumulate comparatively across the involved communities. The communities on both ends of the innovation highway must benefit comparatively from the exchange in the sense of classical theories of comparative advantage. Innovation highways must be constructed to make the partnering communities globally competitive. As the interlocked communities become integrated, then their mutual performance reflects their joint competitiveness in the global economy. As innovation highways develop, many of the most important opportunities for innov­ ation involve the creation of public goods. These public goods must be governed in ways that account for their public character. The risk from poor governance is underinvestment, unfair allocation of operational risk, and the channelling of profits in ways that deplete mutual performance. Harmonization of governance rules across the interlocked parties becomes central to the mutual performance of the communities. For example, harmonization of drug-​ approval processes, institutional review boards, and so on, across jurisdictional boundaries requires enfranchisement of poor nations in drug-​approval processes over pharmaceuticals

Innovation Highways and the Geography of Inclusive Growth    835 that may be sold primarily in poor communities. As the connections between communities strengthen—​that is, as the innovation highway between them becomes central to their performance—​then governance over the innovation highway requires integration, negotiation, and, ultimately, harmonization of rules across the communities.

Conclusion The GOI and IIG literature, developed independently, each point to the importance of relationships between communities, as well as to the internal characteristics of communities. The GOI points to the importance of internal agglomeration economies and supporting institutions, while the IIG emphasizes internal equity and external relationships. In this essay, we suggest that both internal and external facets of communities are central to their prosperity and inclusiveness, and propose the construct of the ‘innovation highway’ as analogous to the cluster for understanding inter-​jurisdictional relationships. The governance of the innovation highway, itself an outcome of negotiation between participating communities, may be crucial to the effective performance of each participating community on both growth and inclusiveness dimensions. The system of innovation highways created by a community may put significant demands on the community’s governance capabilities, and yet may also offer unprecedented opportunities for prosperity. Our analysis raises a number of questions about the robustness of governance over innovation highways for future research. Inclusiveness in the process of governance over innovation highways carries the promise of removing economic, political, and ethical barriers that have plagued community exchange. Design of appropriate governance systems for innovation highways is central not only to their immediate contributions to the performance of linked communities, but also to the eventual integration of the communities for mutual benefit. Future research will need to examine alternative governance mechanisms and the impact of these mechanisms and processes on inclusive innovation and growth.

References Adner, R. and Kapoor, R. (2010). ‘Value creation in innovation ecosystem’. Strategic Management Journal 3: 306–​333. Amsden, A. and Chu, W. (2003). Beyond Late Development (Cambridge, MA: MIT Press). Ansari, S., Munir, K., and Gregg, T. (2012). ‘Impact at the “bottom of the pyramid”: the role of social capital in capability development and community empowerment’. Journal of Management Studies 49: 813–​842. Aoyama, Y., Murphy, J.T., and Hanson, S. (2011). Key Concepts in Economic Geography (London: SAGE). Banerjee, A.V. and Duflo, E. (2011). Poor Economics: A Radical Rethinking of the Way to Fight Global Povert (New York: Public Affairs). Bhatti, Y.A. (2012). ‘What is frugal, what is innovation? Towards a theory of frugal innovation’ http://​papers.ssrn.com/​sol3/​papers.cfm?abstract_​id=2005910 (last accessed 28 April 2017).

836   McGahan and Stein Bower, J.L. and Christensen, C.M. (1995). ‘Disruptive technologies: catching the wave’. Harvard Business Review https://​hbr.org/​1995/​01/​disruptive-​technologies-​catching-​the-​wave (last accessed 28 April 2017). Bradley, S.W., McMullen, J.S., Artz, K., and Simiyu, E. M. (2012). ‘Capital is not enough: innovation in developing economies’. Journal of Management Studies 49: 684–​7 18. Cohen, WM. and Klepper, S. (1991). ‘Firm Size Versus Diversity in the Achievement of Technological Advance’ in Z.J. Acs and D.B. Audretsch (eds) Innovation and Technological Change: An International Comparison, pp. 183–​203 (Ann Arbor, MI: University of Michigan Press). Cohen, W.M. and Klepper, S. (1992). ‘The tradeoff between firm size and diversity in the pursuit of technological progress’. Small Business Economics 4: 1–​14. Easterly, W. (2001). The Elusive Quest for Growth: Adventures and Misadventures in the Tropics (Cambridge, MA: MIT Press). Easterly, W. (2006). The White Man’s Burden: Why the West’s Efforts to Aid the Rest Have Done So Much Ill and So Little Good (New York: Penguin Press). Fagerberg, J., Mowery, D.C. and Nightingale, P. (2012). ‘The heterogeneity of innovation: evidence from the Community Innovation Surveys’. Industrial and Corporate Change 5: 1175–​1180. Feldman, M.P. (1994). The Geography of Innovation (Boston, MA:  Kluwer Academic Publishers). Gates, B. (2011). ‘Innovating with impact: financing 21st century development’. Report to the G20 leaders, Cannes Summit (November). George, G., McGahan, A.M., and Prabhu, J. (2012). ‘Innovation for inclusive growth: towards a theoretical framework and a research agenda’. Journal of Management Studies 49: 661–​683. Gertler, M.S. (2003). ‘Tacit knowledge and the economic geography of context, or the undefinable tacitness of being (there)’. Journal of Economic Geography 3: 75–​99. Gertler, M.S. (2010). ‘Rules of the game: the place of institutions in regional economic change’. Regional Studies 44: 1–​15. Govindarajan, V. and Ramamurti, R. (2011). ‘Reverse innovation, emerging markets, and global strategy’. Global Strategy Journal 1: 191–​205. Govindarajan, V. and Trimble, C. (2011). Reverse Innovation:  Create Far From Home, Win Anywhere (Boston, MA: Harvard Business School Publishing). Hall, J., Matos, Sheehan, L., and Silvestre, B. (2012). ‘Entrepreneurship and innovation at the base of the pyramid:  a recipe for inclusive growth or social exclusion’? Journal of Management Studies 49: 785–​812. Immelt, J. R., Govindarajan, V., and Trimble, C. (2009). ‘How GE is disrupting itself ’. Harvard Business Review 87: 56–​65. Jacobs, J. (1969). The Economy of Cities (New York: Random House). Kooiman, J. (2003). Governing as Governance (London: SAGE). Klein, P., Mahoney, J., McGahan A.M., and Pitelis, C. (2011). ‘Strategy and the libecap paradox: efficiency and co-​adaptation of organizations and institutions’. University of Toronto working paper (February). Krugman, P. (1991). Geography and Trade (Cambridge, MA: MIT Press). Kumar, N. and Puranam, P. (2012). India Inside: The Emerging Innovation Challenge to the West (Boston, MA: Harvard Business School Publishing). Lee, K. (2013). Schumpeterian Analyses of Economic Catch Up (Cambridge:  Cambridge University Press).

Innovation Highways and the Geography of Inclusive Growth    837 Lee, K. and Lim, C. (2001). ‘Technological regimes, catching-​up and leapfrogging: the findings from Korean industries’. Research Policy 30: 459–​483. Libecap, G.D. (1989). Contracting for Property Rights (Cambridge:  Cambridge University Press). McGahan, A.M. (2012). ‘Paradoxes of innovation in health and their resolution in embedded innovation’. Munk Monitor 2: 14–​17. McGahan, A.M., Rezaie. R., and Cole, D.C. (2013). ‘Embedded Innovation in Health’, in D. Soman, J. Stein, and J. Wong (eds) Innovating for the Global South: Towards an Inclusive Innovation Agenda (Toronto: University of Toronto Press). Malecki, E. (1997). Technology and Economic Development: The Dynamics of Local, Regional and National Competitiveness (2nd edition) (London: Addison Wesley Longman). Malerba, F. (2004). Sectoral Systems of Innovation (Cambridge: Cambridge University Press). Malerba, F. (2005). ‘Sectoral Systems of Innovation: How and Why Innovation Differs Across Sectors’ in J. Fagerberg, D. Mowery, and R. Nelson (eds) The Oxford Handbook of Innovation, pp. 380–​406 (Oxford: Oxford University Press). Malerba, F. and Adams, P. (2014). ‘Sectoral Systems of Innovation’ in M. Dodgson, D. Gann, and N. Phillips (eds) The Oxford Handbook of Innovation Management, pp. 183–​203 (Oxford: Oxford University Press). Malerba, F. and Nelson, R. (2011). ‘Learning and catching up in different sectoral systems: evidence from six industries’. Industrial and Corporate Change 20: 1645–​1675. Markides, C.C. (2012). ‘How disruptive will innovations from emerging markets be?’ Sloan Management Review 54: 1. Moyo, D. (2010). Dead Aid: Why Aid is Not Working and How There is a Better Way for Africa (New York: Farrar, Straus and Giroux). Murmann, P.J. (2013). ‘The coevolution of industries and important features of their environment’. Organization Science 24: 58–​78. Nelson, R. and Sampat, B. (2001). ‘Making sense of institutions as a factor shaping economic performance’. Journal of Economic Behaviour and Organization 44: 31–​54. Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action (New York: Cambridge University Press). Pierre, J. and Peters, G. (2002). Governance, Politics, and the State (Basingstoke: Macmillan Press). Piore, M.J. and Sabel, C.F. (1984). The Second Industrial Divide:  Possibilities for Prosperity (New York: Basic Books). Porter, M.E. (1985). Competitive Advantage (New York: Free Press). Porter, M.E. (1990). The Comparative Advantage of Nation (New York: Free Press). Porter, M.E. (2000). ‘Locations, Clusters, and Company Strategy’, in G.L. Clark, M.P. Feldman, and M.S. Gertler (eds) The Oxford Handbook of Economic Geography, pp. 253–​274 (Oxford: Oxford University Press). Prahalad, C.K. (2004). The Fortune at the Bottom of the Pyramid (Philadelphia, PA: Wharton School Publishing). Prahalad, C.K. and Mashelkar, R.A. (2010). ‘Innovation’s Holy Grail’. Harvard Business Review https://​hbr.org/​2010/​07/​innovations-​holy-​grail (last accessed 28 April 2017). Radjou, N., Prahbu. J., and Ajuja, S. (2011). Jugaad Innovation (San Francisco, CA: Jossey-​Bass). Rezaie, R., McGahan, A.M., Daar, A.S., and Singer, P.A. (2012). ‘Innovative drugs and vaccines in China, India and Brazil’. Nature Biotechnology 30: 923–​926.

838   McGahan and Stein Rezaie, R. and Singer P.A. (2010). ‘Global health or global wealth?’ Nature Biotechnology 28: 907–​909. Rhodes, R.A.W. (1996). ‘The new governance: governing without government’. Political Studies 44: 652–​667. Rhodes, R.A.W. (1997). Understanding Governance. Policy Networks, Governance, Reflexivity, and Accountability (Buckingham: Open University Press). Rodrik, D. (2008). ‘Second-​best institutions’. American Economic Review 982: 100–​104. Schumacher, E.F. (1974). Small is Beautiful (London: Sphere Books). Shane, S. (1993). ‘Cultural influences on national differences in rates of innovation’. Journal of Business Venturing 8: 59–​74. Simanis, E. and Hart, S. (2008). Beyond Selling to the Poor: Building Business Intimacy through Embedded Innovation (New York: Cornell University Press). Simanis, E. and Hart, S. (2009). ‘Innovation from the inside out’. MIT Sloan Management Review http://​sloanreview.mit.edu/​article/​innovation-​from-​the-​inside-​out/​ (last accessed 28 April 2017). Skelcher, C. (2005). ‘Jurisdictional integrity, polycentrism, and the design of democratic governance’. Governance 18: 81–​110. Spencer, G., Vinodrai, T., Gertler, M.S., and Wolfe, D.A. (2010). ‘Do clusters make a difference? Defining and assessing their economic performance’. Regional Studies 44: 697–​7 15. Stein, J.G. (2013). ‘Frugal Innovation and Development Assistance’ in D. Soman, J. Wong, and J. Stein (eds) Innovating for the Global South, pp. 151–​167 (Toronto: University of Toronto Press). Tharoor, S. (2012). ‘India’s hope lies in frugal innovation’. New Straits Times 23 July http://​ www.nst.com.my/​opinion/​columnist/​india-​s-​hope-​lies-​in-​frugal-​innovation-​1.111326 (last accessed 28 April 2017). Thorsteinsdóttir, H., Saénz, T.W., Quach, U., Daar, A., and Singer, P.A. (2004). ‘Cuba—​ innovation through synergy’. Nature Biotechnology 22 (Supplement): DC19–​DC24. Topol, E.J. (2011). ‘Medicine needs frugal innovation’. MIT Technology Review https://​www. technologyreview.com/​s/​426336/​medicine-​needs-​frugal-​innovation/​ (last accessed 28 April 2017).

Chapter 45

Sho cking Aspe c ts of Regiona l Devel opment :  Towa rd s an Ec onom ic G e o g ra ph y of Resili e nc e Ron Martin The Rise of ‘Resilience Thinking’ Every now and again a concept emerges within the academic literature that, for some reason or other, suddenly spreads out from the specific field(s) in which it originated into a whole variety of disciplines and applications. Typically, such concepts have a certain ‘generic’ character, in the sense that they capture a feature or process that is thought to be applicable to, or can be adapted so as to be rendered applicable to, quite diverse contexts. But as they travel across disciplinary boundaries such concepts inevitably require some degree of re-​ specification or re-​interpretation relevant to the particular context in question, and this itself can both enrich the concept and raise significant theoretical and methodological issues concerning its meaning and relevance (Wimmer and Kössler, 2006). One notion that has undergone this sort of intellectual journey is that of resilience. The use of this concept can be found in early-​1970s’ ecology, as part of the study of the stability and persistence of ecological systems in response to natural (environmental) and human-​ induced disturbances and perturbations (see the pioneering study by Holling, 1973). It also emerged at about the same time in psychopathology and developmental psychology studies, to help understand how individuals (especially children) cope under adversity (Garmezy, 1971, is one of the first such studies). Over the last two decades or so, both of these fields have directed renewed attention to the notion, elaborating it considerably in the process (e.g. Luthar and Becker, 2000; Luthar, 2003; Folke, 2006; Walker and Salt, 2006; Masten, 2014). But at the same time, the idea of resilience has also attracted growing interest from several other disciplines, including management studies, organizational science, urban planning,

840   Martin comparative politics, development studies and policy, and environmental science, particularly studies of the impact of climate change (Adger, 2000; Sheffi, 2007, 2015; Seville, 2009; Pelling, 2011; Lee et al., 2013; Evans and Reid, 2014; Caniglia et al., 2016). Even a new journal, Resilience, has been established. And according to two key contributions (Zolli and Healy, 2012; Rodin, 2015), resilience is precisely the sort of analytical tool we need in order to understand and confront what they argue is an increasingly uncertain and risk-​prone world. Given this disciplinary diffusion of the concept, it is not too surprising that it should find its way into economic geography and regional studies, and over the last decade the literature concerned with the resilience of city and regional economies has grown apace (see e.g. Bristow, 2010; Cambridge Journal of Regions, Economy and Society, 2010; Hassink, 2010; Hudson, 2010; Pike et al, 2010; Simmie and Martin, 2010; Fingelton et al., 2012, 2015; Martin, 2012; Boschma, 2015; Martin and Sunley, 2015, ). Of course, the increasing popularity of a notion is no guarantee of its profundity, and the use of the concept of resilience in economic geography inevitably raises as many questions as answers. Does the notion aid our understanding of (uneven) regional development under conditions of instability, uncertainty, and risk? Does the process of (uneven) regional development aid our understanding of resilience? Some geographers query whether the notion has any value-​added, and is just another term for ‘competitiveness’ or ‘sustainability’ (see e.g. Bristow, 2010; Hudson, 2010).1 But as Scott (2013) points out, the value of the concept of resilience relative to the notions of competitiveness and sustainability is its explicit emphasis on shocks, disruptions, and unknowable perturbations (Tompkins and Adger, 2004), and how such disruptions interact with processes of gradual and incremental change across temporal and spatial scales (Folke, 2006). As I have argued elsewhere (Martin, 2012; Martin and Sunley, 2015), while the notion of resilience certainly attracts debate, and is characterized by several as-​yet-​unresolved issues—​not least its precise meaning and how it should be theorized—​it nevertheless helps to stimulate new ways of thinking about change in the economic landscape. My purpose in this chapter is to set out what this new way of thinking involves, what is distinctive about an economic geography of resilience, and what it might contribute to our understanding of uneven regional development.2

Regional Development as a Shock-​prone Process The first defining feature of a ‘resilience perspective’ on the economic landscape is a recognition that uneven geographical development itself is not some smooth or slowly changing phenomenon, but an inherently shock-​prone process, subject to all sorts of disruption, perturbation, and interruption. To be sure, many of the main forces shaping the economic landscape are slow and incremental, or perhaps are phase-​like in evolution. But historical experience—​and, indeed, recent events (not least the global financial crisis of 2008–​09)—​ indicate that sudden and unexpected events and dramas frequently disturb and punctuate the momentum and direction of economic growth and development (Ormerod, 2010). It is in this context that the notion of resilience takes on potential relevance. The idea of resilience is intended to capture and describe the extent to which and ways in which an entity or system

Shocking Aspects of Regional Development    841 is able to withstand or recover positively from a shock or disturbance that has disrupted the ‘normal’ position, state, behaviour, or functioning of that entity or system.3 Applying this idea to the study of regional development immediately raises the question of what we mean by ‘normal’ and ‘shock’. To some extent the answer to this question depends on the spatial and temporal scales of our analysis. Despite what some spatial economists might assume, the economic landscape is never in equilibrium but in constant flux: old firms, products, jobs, and technologies are continually disappearing, just as new firms, products, jobs, and technologies are continually emerging. According to Joseph Schumpeter it is this ongoing process of what he called ‘creative destruction’ that drives the evolution of the economy, ‘incessantly destroying the old one, incessantly creating a new one’ (Schumpeter, 1942, p. 83). At an aggregate scale, such as the level of a national economy, the net result of this ongoing process of ‘mutation’ (as Schumpeter called it) may appear as a relatively slow and steady rate of structural–​ technological change. But at the micro-​level, it involves a multitude of ‘local disruptions’, as individual firms in particular places close or contract and workers are laid off. For the individual workers concerned, such local events may well constitute adverse personal shocks. Whether such events disturb a locality’s employment or growth path as a whole will depend on the scale of the firm closure or contraction and whether other offsetting local employment opportunities are available or created. There is no guarantee that the positive side of ‘creative destruction’, the opening of new firms and creation of new jobs, will occur precisely in those places where firms and jobs are destroyed. It is when local firm closures or contractions assume a scale such that there is an identifiable net negative effect on that locality’s employment or growth path, that it becomes possible to talk of a local economic ‘shock’. Thus, for example, the loss of a major or dominant employer locally, perhaps because of a move to another location, or because of closure following a post-​takeover ‘rationalization’ by a new non-​local parent company, can be highly disruptive, leading directly, and via various negative multiplier effects on other firms and workers, to a rise in local unemployment, a reduction in local gross domestic product (GDP) per capita, reduced household incomes, and greater dependence on welfare. While such ‘idiosyncratic’ locally specific shocks may be hardly noticeable and appear as mere ‘noise’ at the level of the national economy as a whole, and are part of the ‘normal’ workings of capitalism, they can, nevertheless, pose significant challenges for the individual localities and communities concerned. Sometimes, such disruptions take on a multi-​locational character, affecting several locations more or less simultaneously. For example, an entire national industry may undergo intense contraction, or even disappear totally, perhaps because of the rise of major competitors elsewhere, or the closure of several spatially distributed branches of a major company, or the rationalization of a public-​sector industry, with the result that adverse ‘destructive’ shocks are imposed on several places more or less at the same time. A graphic illustration is the coal mine privatization and closure programme in the UK in the 1980s and 1990s. In 1983 there were 170 coal mines across the UK. A relentless process of closure and abandonment then ensued. By 2000 almost 150 collieries had been closed down, some 25 in 1985 alone, literally decimating the industry. In many of the localities concerned, mining was a major or significant employer. The large number of jobs typically lost from a local pit clos­ ure itself, together with the knock-​on effects on other local trades and firms, represented a serious and negative shock to the local economies concerned. It proved difficult for the unemployed coal miners to find new employment locally, and most redundant miners faced

842   Martin significant geographical and occupational immobilities. How does a local economy affected in this way respond and recover? How long does it take for other types of activity to create jobs to compensate for those lost by the closure? The recovery of employment in the pit closure areas has varied considerably, with those localities most dependent on coal for their job base (e.g. Easington, Bolsover, and Newark), experiencing the slowest subsequent growth of employment, while others much less dependent (e.g. Selby, Rotherham, and Doncaster) faired much better. In the worst affected areas, local employment had still not returned to pre-​closure levels two decades after the shutdown (Ormerod, 2008). The similar rationalization of the UK’s steel plants, also in the 1980s and 1990s, involving a number of major plant closures, is another example of an industry-​wide, multi-​location shock. And then, at an aggregate level, there are major ‘macro-​level shocks’, such as recessions, financial crises, political crises, technological upheavals, wars, and so on, that disrupt whole national economies, or even much of the global economy (Briguglio, 2004; Briguglio et al., 2009). These may occur relatively infrequently, but tend to be widespread and severe in their effects, with few, if any, localities escaping their consequences. However, the precise impact is likely to vary from place to place. As an example, consider Figure 45.1, which shows how the Great Recession of 2007–​09 in the USA affected per capita GDP in the cities of New York, Chicago, Atlanta, and Phoenix. Three distinctive features are immediately apparent. Firstly, the impact of the recession was clearly much greater proportionately in the cities of Phoenix and Atlanta than in Chicago and New York: the cities differed in their resistance to the downturn, as measured in terms of the fall in per capita GDP. Secondly, the cities also varied in the speed and extent of their recoverability. In New York, per capita GDP had recovered to its 2007 pre-​shock level by 2011. In Phoenix, by contrast, even by 2014 per capita GDP was still 13 per cent below its 2007 level. In fact, thirdly, the data for these four cities suggest that the more severe a city’s downturn in per capita GDP in the recession—​the lower a city’s

Onset of Great Recession

GDP per Capita (2007 = 100)

110 105 100 95 90 85

New York

Chicago

Atlanta

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

80

Phoenix

Figure  45.1  Differing City Responses to the Great Recession in the USA, 2007–​09: New York, Chicago, Atlanta, and Phoenix Compared. Note: GDP, gross domestic product. Source: https://​research.stlouisfed.org/​

Shocking Aspects of Regional Development    843 resistance—​the slower its subsequent recovery has been. Indeed, in the case of Phoenix and Atlanta the effect of the recession may well turn out to have been not merely transient in nature, but involving a permanent downturn shift in the growth path of per capita GDP. A basic question is thus what determines such differential impacts and why some places recover more rapidly than others.4 With such events, the ‘memory’ or ‘remanence’ effects in some localities and places can be long lasting, lingering on long after the shock itself (Blanchard and Katz, 1992; Simmie and Martin, 2010; Fingelton et  al., 2012; Doran and Fingleton, 2013; Evans and Karecha, 2014; Cowell, 2013; Cellini and Torrisi, 2014; Fingleton et al., 2015). Research has found that national economies that are more prone to frequent or major shocks tend to have lower overall long-​run rates of growth (Cerra and Saxena, 2008; Cerra et al., 2009); that is, shocks can have permanent effects. An immediate issue is whether and to what extent this outcome also applies at the subnational scale. Do localities, cities, and regions that are much more susceptible to idiosyncratic shocks, or much more severely affected by nationwide and general disruptions, tend also to have lower long-​run growth paths as a consequence? And does slow growth itself increase a locality’s, city’s, or region’s vulnerability or susceptibility to shocks (Martin, 2012; Augustine et al., 2013; Hans and Goetz, 2013)? In other words, is there a two-​way recursive interaction between shocks and underlying long-​run growth? If so, the notion of resilience becomes highly relevant for understanding uneven regional and city development. Typically, in most applications of resilience across various disciplines, shocks are seen as sudden, often unexpected, events, and usually as negative in nature and impact; hence, the interest in how well the system of interest copes with and recovers from the shock. In principle, however, a ‘shock’ may be of a positive kind. (Winning the lottery might possibly be so described, certainly for the lucky individual concerned, and it could well lead to a major adjustment in that person’s outlook and lifestyle!) In an economic–​geographical context, positive ‘shocks’ would include a major increase in the demand for a locality’s, city’s, or region’s products or services, the opening of a major new employer, a substantial new infrastructural investment, or a major technological advance, for example one that boosts the competitiveness of an area’s industries or leads to the emergence of new and novel activities. Events of this kind certainly require adjustments in the local economy in question, and may lead to pressures on existing infrastructures and resources, such as housing, public transport, and local skills. And unless these infrastructures and resources are upgraded and extended, local supply constraints could emerge that dampen and even undermine the positive nature of, and opportunities afforded by, the ‘shock’. But whether we include adjustments to ‘positive shocks’ under the rubric of resilience is perhaps open to debate. Responding to a major local developmental opportunity seems rather different from recovering from a negative disruption that undermines or destabilizes a local economy. What is relevant is that a positive shock in one locality or region, because of, say, a local technological breakthrough, may have distinctly negative consequences for the industries and jobs based on an old technology in other localities and regions. An example is the widespread introduction of highly efficient automatic loom technology in the US textile industry in the first two decades of the twentieth century. This presented the erstwhile world-​leading textile firms of Lancashire, UK, with a major technological–​competitive shock, to which they were extraordinarily slow to respond, resulting in the failure and closure of literally thousands of factories across the towns and cities in the region (Aldcroft, 1968).5 What such occurrences suggest, again, is that the geographies of shocks and of local reactions to them are integral to the process of

844   Martin uneven regional development: what is a major positive economic boost or development in one region or locality (possibly at a significant geographical remove) may well give rise to a negative shock in another. In addition to these various types of shock, some writers have queried whether the notion of resilience should be restricted to the analysis of the impact of sudden unpredictable adverse disturbances, or whether it is also applicable in relation to how firms, workers, and institutions react to and cope with ‘slow burn’, that is, incrementally increasing, processes, pressures, and changes. This is not a straightforward issue. Many of the processes that drive economic growth and development are ‘slow burn’ in the sense that they occur continually, are ongoing, and even quasi-​predictable in nature. Firms continuously face pressures arising from constantly shifting competition, technology, and the like, and have to respond, adjust, and adapt accordingly if they are to survive and prosper. One could argue that this need to remain ‘dynamically competitive’, to adapt continually to and anticipate constantly changing market and technology conditions is a pre-​requisite of resilience, and that whether a firm has successfully built resilience is only revealed at those junctures when shocks occur. Unless the notion is restricted to actually occurring sudden shocks, the concept would have little distinctive or incisive meaning. In effect, the term could be used in relation to almost any form of change, and hence as a consequence would lose its explanatory value. Having said this, however, slow-​burn processes need not be unrelated to shocks. For example, certain pressures, although slow and incremental, may build up cumulatively, and eventually reach a ‘tipping point’ or threshold at which they trigger a distinctive ‘shock’ event. One example might be the emergence and rise of more productive, cheaper, or technologically advanced competitors that slowly but progressively capture existing firms’ market share (the Lancashire textile firms referred to earlier fit this situation). The original firms may ignore or decide not to adjust (e.g. by re-​equipping to increase efficiency) in response to this growing pressure. But this may eventually lead to situation where the original firms can no longer operate profitably, and are forced to close down, with possibly damaging effects on their specific localities. Sudden shocks, in other words, can be ‘a long time in the making’. But all this is not to suggest that ‘resilience’ is simply a property of an economy (or any type of system, for that matter) that is only present or ‘happens’ when a sudden shock occurs. It may become particularly evident—​become ‘revealed’—​at such junctures, but it is a feature that is continuously produced and reproduced over time as part of, and latent in, that economy’s development and functioning. Further, and this is a crucial point, resilience is not some unchanging characteristic: rather, it may evolve—​increasing or decreasing—​depending on the nature and direction of a region’s developmental path. It is the changing ability of a regional economy—​its firms, workers, and institutions—​to adapt over time that embues that economy with the resilience to minimize its vulnerability to and ability to recover successfully from shocks. Other things being equal, failure to adapt will gradually erode resilience, so that the economy becomes increasingly vulnerable to shocks and less likely to recover from them when they occur. Yet again, this highlights the need for a ‘developmental perspective’ on regional resilience, with implications for how we theorize the notion. In considering that task, however, we need to consider what we actually mean by resilience when applied in a regional or local context, and how we measure it.

Shocking Aspects of Regional Development    845

Responding to Shocks: The Notion of Regional Economic Resilience Despite the long-​established use of the notion of resilience in ecology and psychopathology, even in those fields there is still debate over how to define and operationalize the concept, and over how it differs from other seemingly similar ideas, such as stability, homeostasis, persistence, dynamic robustness, elasticity, and the like (Grimm and Vissel, 1997; Luthar and Becker 2000; Brand and Jax, 2007). According to Brand and Jax, as the concept of resilience has spread not just across ecology, but also other related disciplines, so it has acquired multifarious meanings and confusions, perhaps to the point at which its ‘conceptual clarity and practical relevance are critically in danger’ (2007, p. 1). In their view, the ‘success of the concept in stimulating research across disciplines on the one side, and the dilution of the descriptive core on the other, raises the fundamental question what conceptual structure we want resilience to have’ (Brand and Jax, 2007, p. 2). This question applies no less in economic geography than in other fields. It was perhaps inevitable that economic geographers should turn to ecology for definitions of resilience: after all, the concept has attracted an extensive literature in that discipline (see Gunderson and Pritchard, 2002), and human interaction with ecosystems has itself been a recurring theme in geographical enquiry. Early on, the eminent ecologist Holling distinguished between ‘engineering’ and ‘ecological’ resilience (Holling, 1996). The former, which is the closest to the etymology of the term resilience, from the Latin resilire, ‘to regain form and position elastically’ (following a disturbance), emphasizes the self-​restorative bounce back of a system or entity to its pre-​shock state or position following a disruption (Table 45.1). The focus of attention is not so much on the scale or nature of the shock or disruption, but on the speed with which the system or entity returns to its ‘normal’ (typically assumed to be its ‘equilibrium’) state or trajectory. Shocks, in other words, are assumed be transitory in their impacts, with no permanent or remanent effects. Most ecologists, however, prefer to work with the second definition of ‘ecological resilience’, which is interpreted as ‘the magnitude or scale of disturbance that can be absorbed before the system changes in structure by changes in variables and processes that control its behaviour … Resilience is this context is a measure of robustness and buffering capacity of the system to changing conditions’ (Berkes and Folke, 1998, p. 12). The focus of attention here, then, is on the absorptive capacity of a system, on its dynamics when pushed ‘far from’ its ‘normal’ (again typically interpreted as ‘equilibrium’) state or position; or, more precisely, how far it can be pushed from that state by a shock before it is unable to return to it. But this version of resilience is not unproblematic. For example, a system may be pushed beyond its ‘absorptive capacity’ by a shock and end up permanently in a different (new ‘equilibrium’) state as a consequence. But what if that new state is more favourable that the old one, measured on some performance or sustainability criterion, for example? The issue here is not one of absorbing the shock and stability of structure, but one of shock-​induced positive adaptation to a superior structure and state. This is surely as much (if not more) an indication of resilience than if the system had managed to absorb the shock and resumed its former (but actually less favourable) position.

846   Martin Table 45.1 Different Conceptions of Regional Economic Resilience: From Bounce Back to Adaptive Development Conception

Interpretation and features

Resilience as ‘self-​ restorative ‘bounce back’ from shocks

Shock produces self-​correcting and autopoietic processes that restore the economy back to its pre-​shock state or path: focus is on speed and extent of ‘bounce back’; assumes shocks are merely transient events, with no permanent or remanent effects.

Resilience as ‘ability to absorb’ shocks

The size of a shock that an economy can absorb or tolerate without undergoing any significant change in structure or identity. Focuses on stability of structure and functionality. If the shock exceeds the economy’s ‘absorptive capacity or ‘threshold’ it may not be able to return to its pre-​shock state or path, and may move to an alternative, typically less favourable, state or path.

Resilience as ‘adaptive development’ in response to, or anticipation of, shocks

Capacity of an economy to restructure, reorientate, and transform its structure, function, and identity in a positive direction so as to emerge from the shock on a favourable path. This may be its pre-​shock path or involve ‘bounce forward’ to an alternative superior path.

Now both of these two definitions or concepts of resilience can be given economic–​ geographical interpretations. Thus, we find Hill et al., for example, defining regional economic resilience as ‘the ability of a region to recover successfully from shocks to its economy that either throw it off its growth path or have the potential to throw it off its growth path, but do not actually do so’ (2008, pp. 4–​5). This is essentially the self-​restorative bounce-​back interpretation of resilience. This idea is shown schematically in Figure 45.1, where a stylized regional growth path a–​b is disrupted by a shock at point b, experiences contraction b–​c, but then recovers to point d, which is where it would have been in the absence of the shock, and then resumes its pre-​shock path (d–​e). Pendall et al. suggest that this interpretation can also be given an equilibrium connotation as ‘regional growth in output and population, or rates of unemployment, poverty or labour force participation, poverty, can be considered at least partly equilibrium phenomena’ (2010, p. 73). This interpretation is open to question, however. Not only does it imply that each regional economy is always and everywhere tending towards an equilibrium growth path, but also that there are automatic self-​correcting forces and mechanisms that are activated by shocks and which serve to restore that equilibrium path. This is worringly close to subscribing to a neoclassical view of the socio-​economy, with its assumptions of free market forces, free mobility of factors of production, rational behaviour on the part of agents, and self-​correcting dynamics. As argued earlier, regional and local economies are rarely in equilibrium. In fact, the assumption of equilibrium is not necessary. The ‘canonical’ bounce-​back model of resilience could still be invoked under the less stringent assumption that a regional economy has some sort of long-​run (or ‘normal’) underlying growth path (set by, for example, the region’s workforce base, its productivity growth rate, and the like) of the sort traced out by the line a–​b–​c–​d–​e in Figure 45.1, so that while in the short run the rate of recovery from a shock can exceed that long-​run rate (c–​d), when the regional economy approaches

Shocking Aspects of Regional Development    847 the full employment of its resources, growth is once again limited by the long-​run (‘normal’) rate. This is essentially the assumption made by Friedman (1988), who interprets business fluctuations as movements downwards from, and recoveries as movements back to, a long-​ run upward-​sloping ‘maximum feasible growth ceiling’. Likewise, the ‘ecological’ model of resilience can be given an economic counterpart. Most economists, even of a mainstream persuasion, now assume that an economy might have multiple equilibrium states or paths, and that if a shock is sufficiently severe such an economy may be unable to return to its original equilibrium state or path, and instead is moved to an alternative equilibrium position, in which case ‘hysteresis’ is said to have occurred. Thus, Romer defines hysteresis as a situation ‘where a one-​time disturbance permanently affects the path of an economy’ (Romer, p. 471), the implication being that the economy is shifted from one equilibrium to another. Again, however, the assumption of multiple equilibria is not crucial to this idea of alternative economic time paths, only that there are certain forces that make for the reproduction or reinforcement of a particular path which when given a particularly severe shock might give rise to or be replaced by a new pattern that then becomes similarly reproduced and reinforced, until such time that another disruptive occurs. Consider again Figure 45.2. Instead of bouncing back to point d, and thence resuming its pre-​shock growth path, a regional economy may be unable to absorb the shock, such that the disruption leaves it with a weakened or depleted productive structure or workforce, and the economy emerges on a less favourable growth path—​either one where the pre-​shock growth rate is resumed but with a smaller productive base, c–​f, or even one where the rate of growth is also lower, c–​g. Such a region would be deemed to lack resilience. However, the outcome of a shock need not be negative in nature. A deep shock may, in fact, leave a region’s basic economic structure more or less intact, but stimulate major productivity-​enhancing investments and practices within its firms. In this case the region’s economy might undergo

Employment or output

Positive hysteretic response: bounce forward to raised growth path and growth rate j Positive hysteretic response: h bounce forward to raised growth path and resumption of pre-shock growth rate e

Bounce back to pre-shock growth path andgrowth rate

f

Negative hysteretic response: lowered growth path and resumption of pre-shock growth rate Negative hysteretic response: lowered growth path and slower growth

i Shock d g

b a

c Time

Figure 45.2   Stylized Reactions of a Regional Economy to a Shock.

848   Martin a one-​off upward shift in its growth path, say to i–​h, and then resume its pre-​shock growth rate, or, if the improvement in productivity can be maintained, it might move to a much improved path, such as i–​j. In each case, the region’s economy can be said to have experienced a shock-​induced ‘bounce forward’ to a new path, even though its structure is largely unchanged, and thus to display resilience. However, once we abandon any necessary commitment to equilibrium the way is open for conceptions of resilience that may be much more dynamic, and realistic. Economies satisfy many of the properties that characterize and define complex adaptive systems (Krugman, 1994, 1996; Beinhocker, 2006; Martin and Sunley, 2007; Bristow and Healy, 2015). Such systems need never be in equilibrium, are self-​organizing, highly open, with spatially distributed components and dynamics, and agents (firms and workers) that display adaptive behaviour. Further, in recent formulations of the theory of complex systems the notion of ‘robustness’ is used not to signify the stability or stasis of system structure in the face of shocks, but to denote the ability (or even the necessity) of a system to undergo changes in structure and identity precisely in order to preserve or regain certain core functions and performances (Jen, 2003; Kitano, 2004; see also Janssen and Anderies, 2007). Robustness is therefore a form of adaptive resilience, as the system undergoes positive (and possibly purposive) adaptation in anticipation of or in response to perturbations or disruptions. In an economic setting, core performances and functions would include such obvious features as full employment (or low unemployment), a favourable economic growth rate, rising real incomes, and the like. In fact, a certain degree of structural change and adaptation may well be required for a region simply to ‘bounce back’ to its pre-​shock growth path, and will almost certainly be involved in those cases of ‘bounce-​forward’ resilience (Dissart, 2003; Lang, 2012). One of the interesting trends in resilience research in behavioural psychopathology is that of linking resilience to individual development (O’Dougherty Wright et al., 2013; Masten, 2014). Resilience in this work refers to the process of, capacity for, and outcomes of successful adaptation. An individual’s capacity to adapt in the face of adversity is to be understood in terms of that individual’s development, and by recognizing that the latter is itself shaped by that capacity. The focus is on an understanding of the processes leading to resilience in development. This approach draws on developmental systems theory, and rather than asking the question why an individual is resilient, the emphasis is on the developmental pathways and trajectories, the multiple-​level interactions and relational networks, and the wider resources that influence that individual’s resilience as it changes and develops over time. Context and history therefore assume importance, and how these influence risk, vulnerability, and resistance to shocks, on the one hand, and adaptive recovery, on the other. And all this is used to help inform the issue of intervention, of fostering and building an individual’s adaptive resilience. These ideas would seem to resonate closely with the issue of regional economic resilience. The resilience of a regional or local economy is also about successful adaptation. Likewise, resilience is also inextricably interwoven with the process of regional development, and the particular pathways that that development takes, with the local and extra-​local resources available to firms and workers, and with local context and history.6 Resilience is thus not some singular static attribute of a regional or local economy, but a multifaceted developmental process (see Figure 45.3) comprising four distinct but interrelated dimensions or stages: risk or vulnerability, resistance, reorientation or reorganization, and recoverability. In broad terms, the vulnerability of a regional or local economy refers to those conditions, features, and attributes of that economy which influence its risk of being

Shocking Aspects of Regional Development    849 RISK

RESISTANCE

REORIENTATION

RECOVERABILITY

Scale, Nature and Duration of Shock

Pre-Shock Regional Developmental Pathway

Vulnerability and Exposure to Shocks

Depth of Reaction to Shock

Extent and Nature of Adjustment to Shock

Post-Shock Regional Developmental Pathway

Regional Economic Structures, Resources, Capabilities Competences Local (and National) Institutions Nature and Extent of Supportive policies and Measures

Figure 45.3   Regional Economic Resilience as Process. Source: Martin and Sunley (2015).

destabilized or disrupted by adverse events or forces arising without or within that region. It refers to the region’s susceptibility to or degree of exposure to shocks. Resistance refers to the scale of impact of a shock on a region’s economy. We know, for example, that different regions react differently to a national recession or crisis (Martin, 2012; Martin and Sunley, 2015; Fingleton et  al., 2015; Martin et  al., 2016 (see also Figure 45.1)), and this resistance will again be shaped by a region’s economic developmental or evolutionary pathway: some aspects of that pathway may make for resistance, others may weaken the region’s resilience. The key point is that the inherited structural features and functions of a region’s economy, moulded by its developmental trajectory up to the time of a shock will influence and condition its vulnerability and resistance to that shock. But, at the same time, the impact of the shock itself may result in or set in motion various changes to those features and functions. What matters is how far and in what ways a region’s inherited structures and functions are adaptable, and to what extent such adaptation and reorientation is needed for the region to emerge on a favourable recovery path. Once we think about resilience as a process of the sort depicted in Figure 45.3, two implications would seem to follow. Firstly, that resilience is both a consequence and a possible determinant of the ongoing evolution of a region’s developmental pathway. Secondly, what this, in turn, suggests is a need to embed the idea of regional resilience in a theory of (uneven) regional economic development, not simply as some sort of ‘add-​on’ to that theory, but as integral to it.

Embedding Resilience in Explanations of (Uneven) Regional Development The immediate challenge here, of course, is what sort of theory of (uneven) regional development? The issue is that we do not have a single theory that commands general

850   Martin agreement or acceptance. Rather, what we have is a range of explanatory frameworks, concepts, and perspectives. In fact, if anything, over the last two or three decades economic geographers and regional studies scholars have shied away from the idea of constructing an encompassing or general theory of the development and evolution of economic landscape. Instead, much of the intellectual effort in these disciplines has been directed at adding to an ever-​expanding plethora of concepts, partial theories, and accounts, resulting in what Barnes and Sheppard (2010) have called ‘intellectual solitudes’, which emphasize this or that aspect or determinant of regional development, or which focus on this or that type of regional economy. However, notwithstanding this current array of often only loosely connected (and sometimes competing) ideas and approaches, it is nevertheless possible to distinguish between various broad perspectives, even if these are more ‘ideal–​typical’ characterizations than coherent paradigms. Of interest is not only how each of these might see the role of shocks, but also what interpretation is put on the idea of resilience (Table 45.2). While many economic geographers may not subscribe to the so-​called ‘new economic geography’ (NEG) or ‘new spatial economics’ that has developed over the last three decades, and notwithstanding its several limitations (see e.g. Garretsen and Martin, 2010; Martin, 2010a), this body of theory has been highly influential in policy circles, and thus is worth noting here. While NEG does not expressly refer to the idea of resilience, it does include discussions of how hypothetical (and highly simplified) economic landscapes might respond to shocks. In NEG models, resilience would be interpreted as the stability of an equilibrium spatial pattern of economic activity in the face of major shifts in transport costs, in trade barriers, policy interventions, and the like. In such models, a shock above a critical threshold (defined by the parameters of the model in question) induces a shift to a new equilibrium spatial configuration. The shift is assumed to happen instantaneously, and to exhibit path dependence (that is to say, is dependent on the initial, pre-​shock, spatial equilibrium). Essentially, then, NEG theory can be used to investigate—​in the abstract if not empirically—​ both straightforward ‘bounce back’ and ‘absorptive’ types of resilience, with the latter admitting the possibility of shock-​induced hysteretic shifts to alternative equilibrium spatial configurations. Arguably, the attraction of NEG models is that they can be used to explore ‘what if ’-​type questions, and this might be useful in resilience studies, and even possibly in analysing alternative policy interventions in response to shocks. But their usefulness is limited. One problem in these NEG models is that the multiple equilibria they ‘allow’ are assumed to be somehow ‘latent’ in the system of interest; indeed, the different alternative spatial equilibria are built into the structure of these models ab initio. Furthermore, and, critically, these models provide highly incomplete explanations of regional development. In fact, they are not really about development at all, but rather about comparative statics—​comparing (hypothetical) economic landscapes before and after shocks: there is no real history. In complete contrast as an approach to the study of uneven regional development, geographical political economy types of perspective do claim to take history seriously, in the sense of interpreting the long-​run evolution of economic landscapes as inextricably bound up with the instabilities and contradictions that characterize the dynamics of capital accumulation. Shocks, and particularly major periodic over-​accumulation crises, are seen as inherent to the capitalist system. The specific origins of such crises can vary, and they can occur at various spatial scales (from the local to the global), but they almost invariably lead

Shocking Aspects of Regional Development    851 Table 45.2 Embedding Resilience in Regional Development Theory? Some Possibilities Theory

Role of shock/​implied interpretation of resilience?

New economic geography theory

‘Resilience’ as stability of an equilibrium spatial pattern of economic activity in the face of shocks. A shock above critical threshold induces shifts to new spatial equilibrium pattern.

Geographical political economy

Shocks as major over-​accumulation or competitive crises that set in train the search by capital for radically new ‘technological and spatial fixes’. ‘Resilience’ of regions depends on balance between local in situ technological fix and flight of capital to more buoyant or cheaper locations. Shocks exacerbate tensions between capital and labour.

Evolutionary-​ Schumpeterian theory

Shocks as ‘gales’ of creative destruction and ‘competitive selection’. ‘Resilience’ as regional economic ‘fitness’, and ‘positive re-​orientation’ of a region’s industrial–​technological system. Innovation and sectoral diversity assumed to improve regional resilience.

Path-​dependence theories

In standard path-​dependence theory over time technological and structural lock-​in reduces regional resilience; externally originating shocks serve to ‘de-​lock’ established regional development paths. In path-​dependent positive adaptation theories, regional resilience can be continually renewed/​maintained; shocks may stimulate the emergence of or branching into new regional development paths.

Institutionalist approaches

Institutions (both formal and informal, local, and extra-​local) and governance arrangements and priorities can shape regional resistance to and recoverability from shocks. Stickiness of outmoded institutions can reduce regional resilience, while supportive and proactive institutional interventions can assist recoverability.

to disinvestment of capital and job losses in certain sectors and locations, and possible shifts of production and jobs to others, as firms search for some sort of ‘fix’ by which to recover profitability and market share (Harvey, 2006). Under this perspective, the issue of the resilience of a local economy would depend on the nature and extent of this ‘fix’. Firms may respond by in situ restructuring and reorganization, or by pursuing a strategy of technological upgrading, or even by relocating spatially. Conceptually and empirically, then, the question of local resilience would thus be situated both within a wider geographical context (which can be global), and recognized as involving potential conflicts between capital and labour. Indeed, this perspective would approach the whole issue of resilience in terms of ‘resilience for whom?’ (Hudson, 2010). What may assist the resilience of a firm (e.g. reorganizing the labour process, slimming down the workforce, or in extremis abandoning one location for another, more profitable one), can be at the cost of severe disruption of local jobs and incomes. A geographical political economy perspective on regional resilience would thus necessarily emphasize the social and distributional aspects and consequences of different resilience strategies on the part of capital. Evolutionary economic geography also seeks to take history explicitly into account, albeit from a rather different point of view. Although here, too, resilience itself is not a key

852   Martin concept or consideration, an evolutionary perspective can give valuable insight into the idea. The key interest in evolutionary economic geography is on how firms, industries, and local economies evolve and adapt through time. This links with those definitions of resilience that define it in terms of positive adaptation. For evolutionary economic geographers, as for evolutionary economists, the Schumpeterian idea of ‘creative destruction’ mentioned earlier is never far below the intellectual surface. Shocks and disruptions to the economy—​whether of a general or locally specific nature, and arising from a variety of sources—​would be seen as leading to local or more generalized bouts of creative destruction involving processes of ‘competitive selection’ (Metcalfe, 1998). What matters during such disruptions is the (competitive) ‘fitness’ of individual firms, as that shapes their ability or otherwise to resist and recover from the disruption. Given that technological change plays a key role in economic evolution, innovation takes on a particular significance as a determinant of a firm’s, industry’s, or regional economy’s ability to resist and recover from shocks (Clark et al., 2010; Boschma, 2015). Thus, this perspective would emphasize how the mechanisms that are believed to drive economic evolution—​such as the creation of new novel forms of economic activity and growth (new technologies, products, firms), and the competitive market selection from these new forms—​in turn shape the resilience of firms, industries, and regions. So an evolutionary economic geographical approach could well be used to throw light on the recursive nature of uneven regional resilience and development referred to above. One particular version or aspect of evolutionary economic geographical thinking that possibly merits separate mention as being of relevance to regional resilience is that of path dependence (Martin and Sunley, 2006). Despite its frequent (and often uncritical) invocation, this concept is far from straightforward. Essentially, it refers to the ways in which and extent to which, at any point in time, the structural, technological, organizational, and institutional forms of an economy inherited from its past development shape, constrain, or enable its future development (Martin and Sunley, 2006). Most applications of the path-​dependence idea use its narrowest interpretation, or expression, namely that of ‘lockin’ to a particular developmental path, attributed to the self-​reinforcing effects of various increasing returns mechanisms (see Martin and Sunley, 2006). So construed, path dependence would equate with the progressive rigidity of spatial economic structures and configurations. Now, whether this rigidity or lock-​in makes for resilience—​as a high degree of ‘lock-​in’ would imply a high degree of stability in the face of shocks—​or whether, on the contrary, it renders the ‘locked-​in’ form and pattern of regional development increasingly vulner­ able to shocks because of its inflexibility and lack of adaptability is an open empirical question. Much may well depend on the particular type of regional development path involved. A region that becomes ‘locked’ into an economic structure centred around heavy, capital-​ intensive industries, such as steel production, is likely over time to become increasingly vulnerable to shocks, and less resilient to them. Indeed, in this type of path dependence it takes a shock to ‘unlock’ a regional economy from its existing and inherited path. There is a parallel here with the so-​called ‘adaptive cycle model’ used in ecology (Simmie and Martin, 2010; Martin and Sunley, 2011). Here, in the early stages of development of an ecosystem its resilience increases, but as the system approaches maturity and a steady state, and resources become locked into a particular stable pattern and use, so rigidity increases and resilience declines (Gunderson and Holling, 2002). If an external shock then disturbs the ecosystem, it

Shocking Aspects of Regional Development    853 may not be able to absorb the disruption, resources are destroyed and released, and recovery to the original type ecosystem may not be possible, leading to a shift in its form and function. A not dissimilar scenario can be envisaged for certain types of local and regional economies (for an application, see Simmie and Martin, 2010). In contrast, a region ‘locked’ into a mode of development based on information and communication technology-​type activities, which by their very nature are more flexible and allow greater scope for branching into related and complementary fields, may exhibit high and sustained resilience. In fact, this latter possibility links to a more expansive, evolutionary form of path dependence which permits ongoing, yet still path dependence-​conditioned economic adaptation (see Martin, 2010b). Under this conception, a highly innovative economic base will encourage more or less continuous branching, adaptation, and mutation, a high new firm formation rate, and hence a constant renewal of variety, all factors that are likely to make for high resilience to shocks. A path-​dependence perspective thus allows for different resilience outcomes and dynamics. A further possible framework, or vantage point, for theorizing and studying regional resilience is that of institutional economic geography. In recent years, increasing attention and emphasis has been given to the role of social, legal, political, and other institutions, both formal and informal, in mediating, moulding, and regulating patterns of regional economic growth and development. An institutional perspective on regional economic resilience would focus on both how institutions can serve to foster or hinder economic adaptability, and on how resilient and adaptable institutions are themselves. Both vantage points can be found in the literature dealing with the role of institutions in the resilience of socio-​ ecological systems, for example (Handmer and Dover, 1996; Harrison, 2003; Herrfahrdt-​ Pähle and Pahl-​Wostle, 2012). On the whole, institutions tend to change more slowly than the economies in which they are embedded and which they tend to support (Setterfield, 1997). This can be a feature that makes for regional economic resilience, since as such, stable institutions—​both formal, such as local and central government, and informal, for example social capital (local community conventions and networks of cooperation and mutual support)—​reduce uncertainty, promote trust, and allow confidence in longer-​term decision-​ making. In this way, institutions can help to stabilize a local or regional (or indeed national) economy, and provide resources and structures that contribute to that economy’s resilience to shocks. In fact, some built-​in formal institutional arrangements are activated automatically in response to certain types of economic disruption. A national unemployment benefit system, for example, will operate to channel income support to communities severely hit by a recession or by locally specific economic disruptions, such as a company closure, while at the same time the tax burden on such areas will fall. There is a degree of automatic spatial redistribution and stabilization—​in effect ‘risk sharing’—​at such times, albeit limited in scale. Or central states may intervene with targeted measures and support in the case of cities or regions hit by adverse local events, for example in the form of investment support to local firms, local infrastructural projects and the like. Financial institutions can also play a formative role. How they choose to respond to the indebtedness of local firms and households at times of economic recession or crisis can either mitigate or intensify the impact of such events. At the other end of the scale, local specific institutions, such as local regional governments, development agencies, business associations, employment agencies, and community organizations, may help local firms, workers, and households weather shocks.

854   Martin However, institutions can themselves become ‘locked-​in’, and increasingly rigid and even dysfunctional, and to that extent may undermine a region’s economic resilience, and may even be the source of disruptive shocks. Institutions are inherently conservative, frequently bureaucratic and self-​serving. They can, over time, become increasingly mal-​aligned with the needs of the economy, and be slow or unable to respond at times of economic disruption. How institutions themselves change, and how they can become captured by particular dominant power groups, elites, and their associated political and ideological dispositions can be instrumental in both generating and ameliorating shocks. The general movement within many nation states over the last thirty years towards neo-​liberal policies of deregulation, privatization, and, most recently, fiscal austerity, has arguably intensified the inherent instability of global capitalism and exposed many regions and cities to increased risk and vulnerability. The dismantling in many countries of much of the regulatory and supervisory frameworks that controlled financial systems is a striking example. The deregulation of banking and financial markets permitted an historically unprecedented level of speculative financial activity and indebtedness, and contributed in no small measure to the consequential banking crisis that broke in 2007, which triggered in its wake the deepest recession for more than eighty years. How different types of institutions, local, national, and even global, impinge on the evolution of the space economy is an ongoing area of research. But what is clear is that institutional arrangements can exert an important influence on local and regional resilience, both positive and negative.

‘Building’ Regional Economic Resilience Even if the idea of resilience is a relatively recent one in the social, management, and related sciences, it has quickly spiralled off into the policy realm. The literature already abounds with policy discussions and suggestions of how to ‘build’ resilience in organizations, businesses, cities, and local communities, and numerous bodies, both local and national, have sprung up concerned with this task (see e.g. Tompkins, 2007; Denhardt and Denhardt, 2010; World Bank, 2012; Biggs et al., 2015; 100resilientcities.org). How a regional or local economy’s resilience can be improved and supported is obviously a valid task for policymakers to address. But as the forgoing brief discussion suggests, while our understanding of what makes for such resilience is beginning to take shape, we are far from having a coherent and universally agreed conceptualization and explanation of the notion, of exactly how it relates to economic development, or what its precise determinants are. While there is something in the adage that ‘too much analysis can lead to policy paralysis’, the lack of such a commonly agreed understanding means that policy programmes and strategies aimed at ‘building’ local economic resilience may rush ahead of a thorough appreciation of what, precisely, it is that they are supposed to build, and how this aim can best be achieved. Some international policy and consultancy bodies have identified what they believe to be the key ‘characteristics’ or ‘qualities’ of resilient cities and regions that by implication should therefore be the foci of measures and actions aimed at ‘resilience building’ (see Table 45.3). But such ‘characteristics’, while logical enough, tend to be rather vague and generic. In order to become specific policy objectives they would need to be translated into what they mean in terms of the determinants of economic development. Put another way, the determinants of

Shocking Aspects of Regional Development    855 Table 45.3 ‘Characteristics’ of Resilience: The Key Qualities Identified by the Rockefeller Foundation and the United Nations Rockefeller/​Arup

United Nations

Reflectiveness

Situational awareness

Resourcefulness Robustness

Self-​regulating

Diversity Flexibility Integration

Adaptability

Integrated Redundancy Inclusiveness Sources: da Silva (2014); Rodin (2015); www.cityresilienceindex. org.

resilience are themselves the potential policy levers by which resilience can be strengthened, and those determinants are closely linked to those that make for successful regional growth and development more generally. However, as we have seen, there are different theories of regional development, each emphasizing different types of key factors, forces, and processes. It is not straightforward, therefore, to distil clear policy messages from across these—​and indeed other—​theories and perspectives. Nevertheless, some key foci can be suggested. From a policy perspective, responding with new policies and measures following a major disruption is less effective than seeking to reduce the vulnerability of a locality or region to shocks in advance of any such disruptions (Robb, 2000; Hamel and Välikangas, 2003; Marcos and Macaulay, 2008; MacKinnon and Derickson, 2013). Policy, in other words, should focus on building ‘anticipatory’ adaptive resilience. This means that, on the one hand, local development and policy organizations need to identify the likely future challenges and shocks that could impact and disrupt their economies, firms, workers, infrastructures, and other key assets, and, where possible, encourage or promote a form of development that reduces the local economy’s vulnerability or exposure to such possible shocks. On the other hand, it also means encouraging and promoting those features, structures, and conditions that make for local economic adaptability, so that the locality can successfully re-​orientate its activities and structures as the wider economic, technological, and competitive environment changes. As mentioned earlier, these twin aims seem to underpin much of the policy debate and discussions around building resilience against climate change. But similar imperatives apply in thinking about improving an area’s economic resilience. At a very broad level, at least three such sets of local features and factors can be identified as potential arenas for policy intervention (Figure 45.4). Of key importance is a regional economy’s ‘dynamic competitiveness’, the ability of its firms to respond to and compete

856   Martin successfully in constantly shifting and changing markets (Martin, 2006). This, is turn, will depend on a range of structural characteristics that define the region’s developmental path and its associated externalities. Diversity of economic activities and export markets, the flexibility of the supply chains used by local firms, a high propensity of those firms to innovate and invest in new products and technologies, the production of a highly educated, skilled, and creative workforce (which, in turn, depends on a high-​quality local educational system), a modern local infrastructure, and a ready access to investment finance—​all these are likely to increase the resistance of a regional economy to shocks and to speed up its recovery from them. Supporting innovation, investment, and a vibrant local industrial ecosystem (so as to ensure the creation of economic diversity) would seem a particularly important element of any policies aimed at increasing and maintaining local economic resilience. A second set of characteristics that bear upon regional resilience, and which are associated with the region’s economic performance, are what might be grouped under the term ‘business confidence’, this, in turn, having to do with the local business climate, the existence of a vibrant entrepreneurial culture (and thence high new firm formation rate), a positive ‘can-​ do’ outward-​looking but locally committed attitude within the business community, and a willingness to network, and share knowledge and awareness. A confident, entrepreneurial business community drives successful economic growth and development, and the latter, in turn, reinforces that confidence. Under these conditions, even if disrupted by a major shock, Structural diversity Switchable markets Supply-chain flexibility Innovation and creativity Skilled/educated workforce Modern infrastructures Access to finance Structure

Dynamic Competitiveness

Resistance RESILIENCE Recoverability Business Confidence

Institutional Support

Culture

Governance

Locally committed business culture Entrepreneurship Positive attitude and outlook Situation awareness Effective networking

Supportive institutions Leadership Proactive attitude Coordinated policy-making Developmental stance Long-term vision Resource allocation

Figure 45.4   Some Potential Policy Foci for ‘Building’ Regional Adaptive Resilience.

Shocking Aspects of Regional Development    857 a buoyant and confident regional economy is much more likely to recover quickly and successfully than a region in which firms lack competitiveness, growth is slow, and business confidence is correspondingly weak: a confidence in the underlying strength of a region’s economy encourages investment and innovation, which reinforces that strength and further boosts confidence. The converse circular process is likely to operate in a slow-​growing, weakly performing region. We are beginning to understand how local economic conditions can affect the psychology of individuals, their outlook, attitudes, and behaviour (Rentfrow et al., 2008; Rentfrow, 2010, 2013), and how once formed such local traits tend to persist and may feed back to reproduce those same local economic conditions. The influence of local differences in psychological factors on local economic resilience is thus a topic that warrants further research.. The extent to which and ways in which policymakers might help influence and shape these two sets of interrelated factors is itself bound up with the nature of economic governance, both locally and more widely (Lebel et al., 2006). While governance is far from easy to measure, it has become increasingly recognized as a formative driver of local economic development. Just as national ‘varieties of capitalism’ can be distinguished by their different models of economic governance and management (Hall and Soskice, 2001), and these different forms have been shown to influence national growth and performance (Schneider and Paunescu, 2012), so both national-​level and local-​level forms of economic governance and institutional structure may well influence the geographies of resilience to shocks. The national-​level system of economic (and political) governance can exert considerable influence over the ability of individual localities, cities, and regions to recover from major shocks. For example, how a national state decides to react to a major recession may help or hinder particular regions. National fiscal or monetary policies do not necessarily affect every region or city equally, or even in the same direction. Particular forms of stimulus may benefit certain areas and assist their recovery while doing nothing or very little to help others. The reaction of states to the recent financial crisis, and especially the imposition of fiscal austerity programmes, illustrates this point. In the UK, for example, while many predicted the banking crisis of 2007–​08 would impact most heavily on London, the nation’s financial centre, the scale of the UK government’s support for the financial sector (reckoned to be as much as £1 trillion in direct bailouts of banks, capital guarantees, and quantitative easing) helped London to escape the worst of the recession and has fuelled its strong subsequent recovery. Meanwhile, the same government’s fiscal austerity cuts to welfare programmes and grants to local governments have fallen disproportionately on the less prosperous northern cities and regions struggling to recover from the crisis (Beatty and Fothergill, 2014). While the national-​level system of economic governance is thus important, local forms of governance are also critical. A supportive and synergistic nexus of local institutions (e.g. from local government to business and trade associations to labour organizations to local finance bodies) that are willing to cooperate and collaborate, can provide leadership, coordinated and proactive policy-​making, and a common commitment to an agreed long-​term vision for the regional economy. As such, potentially they can play a formative role in shaping a region’s or city’s economic resilience. Local institutions—​especially, of course, local or city governments, through their tax-​raising and public-​spending powers, and their control over utilities, planning, and infrastructures—​bear both directly and indirectly on the competitive success of the local economy, and hence on local economic adaptability. A key issue here is the territorial organization of government and political power. Is a highly centralized

858   Martin structure (of the sort found in the UK, for instance) more or less conducive to local economic stability and resilience than a decentralized and devolved system (of the sort found, say, in Germany or Canada)? This is especially important in relation to fiscal powers. On the one hand, the more that local public spending on utilities, infrastructure, education, skills training, business support, and the like depend on funding from central government (and the degree of such dependence varies significantly from country to country—​see e.g. Slack, 2017), the more restrictive the room for independent local strategic manoeuvre is likely to be. On the other hand, the greater is local fiscal autonomy, the more potentially vulnerable is a local economy to disruptions to its tax base and other local income streams, possibly arising from a major negative shock to its economy. This issue of the degree of devolution of fiscal and related powers to local states and authorities has become a key topic of political and academic debate in recent years, the more so given the general context of fiscal austerity (see e.g. Cambridge Journal of Regions, Economy and Society, 2010). Not surprisingly, the more successful is a city or regional economy, the more likely it is to desire greater fiscal and spending autonomy, so as to determine it own priorities and its own destiny. Less prosperous cities and regions, with less favourable local tax bases are likely to be less enthusiastic. The implications of local fiscal and political devolution for the issue of local economic resilience is a subject in need of detailed investigation. The sets of local economic factors around which policies for ‘building’ resilience might be focused that are shown in Figure 45.4 are by no means exhaustive: far from it. Rather they merely indicate the complexities faced by any policymaker with this aim in mind. Three points are worth emphasizing. Firstly, ‘building resilience’ should not be viewed as an independent policy activity: it has to be seen as inseparable from the more general task of promoting stable—​and sustainable—​local economic growth and development. And this takes us back to the fundamental challenge of how we explain (uneven) regional development. Secondly, ‘building resilience’—​like promoting successful local economic development more generally—​is ultimately about agency: about mobilizing business talent, the skills and creativity of workers, and the commitment and support of key institutions, all around shared goals and values. And, thirdly, ‘building resilience’ it is not a ‘one-​off ’ activity: resilience, like development, has to be continually renewed and ‘rebuilt’.

Regional Resilience: Fad or Fundamental Feature? An obvious question is whether the notion of ‘resilience’ is just another faddish addition to the extensive collection of analytical constructs and concepts that now populates economic geography, or whether it captures a fundamental dimension of regional development and merits serious attention as such. There is no doubt that the concept has swept into all sorts of academic enquiry in recent years, and has even become part of everyday journalistic and popular discourse, often without any clear or precise definition attached. And it may be that this widespread invocation of the concept has been stimulated by the belief that socio-​economic life has become more uncertain, risky, and unpredictable. If ‘resilience thinking’ is merely a fad, then like all fashions, it will, in time, become unfashionable. But the fact is that

Shocking Aspects of Regional Development    859 ‘shocks’ have always been a feature of capitalism, an inescapable part of its very nature and dynamics. Regional development is an inherently shock-​prone process. The idea of ‘resilience’ provides a potentially valuable way of thinking about the impacts and implications of this instability, of the role of shocks in economic development, and, correspondingly, acts as a stimulus to thinking about ways to allay their worse effects—​social, economic, and environmental. This is not to argue however, that the concept of resilience is unproblematic. Several issues remain to be resolved. One such issue is how we measure resilience. Scanning the economic geography literature reveals a range of approaches, from descriptive case studies, to simple indices and indicators, to statistical time series methods, to causal explanatory models (see Martin and Sunley, 2015). Common to all these different approaches, however, is the underlying problem of comparing the actual outcome of a shock to a regional or local economy with what the developmental or growth path of the region would have been in the absence of the shock—​in other words, the problem of the ‘counterfactual’. A further issue is the need to theorize the notion, not as an independent feature or attribute of the economic landscape, but as part of the very uneven development of that landscape. The question of why resilience varies from one locality or region to another is inseparable from the question of why economic development itself varies across geographical space. And this need to increase our understanding—​both theoretical and empirical—​of the causal processes and factors that promote or hinder the resilience of local and regional economies is all the more pressing given the increasing reference to the notion in policy discourses. There is nothing worse than policy based on fads rather than fundamentals.

Notes 1. For more general critiques see e.g. Hanley (1998), Kaplan (1999), and Davoudi and Porter (2012). 2. A much longer exposition of the notion of regional economic resilience, on which this chapter draws can be found in the paper by Martin and Sunley (2015). 3. The qualifier ‘positively’ is important, as it emphasizes that the key interest is in favourable outcomes and responses, and not in what might be called ‘perverse’ resilience, that is the resistance to change of an adverse or dysfunctional state of affairs (e.g. an oppressive political or military regime). 4. For studies of the spatial differences in the impact of the Great Recession in the USA, see Doran and Fingleton (2013), Balland et al. (2015), and Doran and Fingleton, (2015). Likewise, the spatial impact of the crisis in Europe has also varied substantially across regions and cities: see Fingleton et al. (2015) and Brakman et al. (2015). 5. As such, this collapse is another example of the industry-​wide, multi-​locational shock referred to earlier. The reasons why the Lancashire firms were slow to both anticipate and respond to the new competition from US firms have been much debated. Both Frankel (1955) and Tyson (1958) suggest that technological ‘interrelatedness’ and ‘lock-​in’ to an obsolete nineteenth-​century technology within the Lancashire firms were responsible for their slow switch to automatic looms. This lock-​in turn is attributed to the high degree of horizontal specialization (and hence interrelatedness) across the Lancashire cotton firms, which meant that the introduction of new weaving technology required complementary

860   Martin changes in spinning and other stages of production, creating substantial problems of coordinating technological change across numerous independent firms. 6. It is in this context that the frequent use of the term ‘resilience’ takes on meaning in the debate and discussions surrounding the impact of climate change and global warming. Responding to climate change is often stated as requiring two kinds of (interrelated) action: mitigating climate change by reconfiguring our forms and processes of economic growth so as to reduce the emission of greenhouse gases; and adapting to climate change, that is, altering the organization of physical, infrastructural, and spatial forms of the economy so as to be able to operate under a different climate from that of today. Mitigation can be seen as action intended to reduce the build up of the likely eventual shock, and adaptation as action intended to reduce its impact.

References Adger, W.N. (2000). ‘Social and ecological resilience: are they related?’ Progress in Human Geography 24: 347–​364. Aldcroft, D.H. (1968). The Development of British Industry and Foreign Competition 1875–​1914 (London: Allen and Unwin). Augustine, N., Wolman, H., Wial, H., and McMillen, M. (2013). ‘Regional economic capacity, economic shocks and economic resilience’. MacArthur Foundation Network on Building Resilient Regions, Working Paper, May. Balland, P.‐A., Rigby, D., and Boschma, R. (2015). ‘The technological resilience of US cities’. Cambridge Journal of Regions, Economy and Society 8: 167–​184. Barnes, T. and Sheppard, E. (2010). ‘Nothing includes everything: towards engaged pluralism in anglophone economic geography’. Progress in Human Geography 34: 193–​214. Beatty, C. and Fothergill, S. (2014). ‘The local and regional impact of the UK’s welfare reforms’. Cambridge Journal of Regions, Economy and Society 7: 63–​79. Beinhocker, E.D. (2006). The Origins of Wealth:  Evolution, Complexity and the Radical Remaking of Economics (New York: Random House). Berkes, F. and C. Folke (eds) (1998). Linking Social and Ecological Systems:  Management Practices and Social Mechanisms for Building Resilience (New York: Cambridge University Press). Biggs, R., Schluter, M., and Schoon, M.L. (eds) (2015). Applying Resilience Thinking:  Seven Principles for Building Resilience in Social–​Ecological Systems (Cambridge:  Cambridge University Press). Blanchard, O. and Katz, L. (1992). ‘Regional evolutions’. Brookings Papers on Economic Activity 92: 1–​75. Boschma, R. (2015). ‘Towards an evolutionary perspective on regional resilience’. Regional Studies 49: 733–​751. Brakman, S., Garretsen H., and Marrewijk, C. (2015). ‘Regional resilience across Europe: on urbanisation and the initial impact of the Great Recession’. Cambridge Journal of Regions, Economy and Society 8: 225–​240. Brand, F. and Jax. K. (2007). ‘Focusing the meaning(s) of resilience: resilience as a descriptive concept and a boundary object’. Ecology and Society 12: 1–​23. Briguglio, L. (2004). ‘Economic Vulnerability and Resilience: Concepts and Measurements’ in L. Briguglio and R.E.J. Kisanga (eds) Economic Vulnerability and Resilience of Small States (Malta: Islands and Small States Institute).

Shocking Aspects of Regional Development    861 Briguglio, L., Cordina, G, Farrugia, N., and Vella, S. (2009). ‘Conceptualising and measuring economic resilience’. Oxford Development Studies 37: 229–​247. Bristow, G. (2010). ‘Resilient regions: re-​“place”ing regional competitiveness’. Cambridge Journal of Regions, Economy and Society 3: 153–​167. Bristow, G. and Healy, A. (2015). ‘Crisis response, choice and resilience: insights from complexity thinking’. Cambridge Journal of Regions, Economy and Society 8: 241–​256. Cambridge Journal of Regions, Economy and Society (2010). ‘The Resilient Region’. 3: 1–​167. Caniglia, B.S., Vallee, M., and Frank, B. (eds) (2016). Resilience, Environmental Justice and the City (London: Routledge). Cellini, R. and Torrisi, G. (2014). ‘Regional resilience in Italy: a long run analysis.’ Regional Studies 48: 1779–​1796. Cerra, V. and Saxena, S., (2008). ‘Growth dynamics: the myth of recovery’. American Economic Review 98: 439–​457. Cerra, V., Panizza, U., and Saxena, S. (2009). ‘International evidence on recovery from recessions’. IMF Working Paper 09/​183. Clark, J., Huang, H.-​I., and Walsh, J.P. (2010). ‘A typology of industrial districts: what is means for regional resilience’. Cambridge Journal of Regions, Economy and Society 3: 121–​138. Cowell, M. (2013). ‘Bounce back or move on: regional resilience and economic development planning’. Cities 30: 212–​222. Da Silva, J. (2014). City Resilience Index:  Understanding and Measuring City Resilience (London: Rockefeller Foundation/​Arup International Development). Davoudi, S. and Porter, L. (2012). ‘Resilience:  a bridging concept or a dead end?’ Planning Theory and Practice 13: 2. Denhardt, J. and Denhardt, R. (2010). ‘Building Organisational Resilience and Adaptive Management’ in J. Reich, W. Reich, A.J. Zautra, and J.S. Hall (eds) The Handbook of Adult Resilience, pp. 333–​349 (New York: The Guilford Press). Dissart, J.C. (2003). ‘Regional economic diversity and regional economic stability: research results and agenda’. International Regional Science Review 26: 193–​204. Doran, J. and Fingelton, B. (2013). ‘US metropolitan area resilience: insights from dynamic spatial panel estimation’. Paper presented at the Annual Conference of the Regional Science Association International (British and Irish Section), University of Cambridge, 22 August. Doran, J. and Fingelton, B. (2015). ‘Resilience from the Micro Perspective’. Cambridge Journal of Regions, Economy and Society 8: 205–​224. Evans, R. and Karecha, J. (2014). ‘Staying on top: why is Munich so resilient and successful?’ European Planning Studies 22: 1259–​1279. Evans, B. and Reid, J. (2014). Resilient Life: The Art of Living Dangerously (New York: Polity Books). Fingleton, B., Garretsen, H., and Martin, R. (2012). ‘Recessionary shocks and regional employment’. Journal of Regional Science 52: 109–​133. Fingleton, B., Garretsen, H., and Martin, R. (2015). ‘Shocking aspects of monetary union: the vulnerability of regions in Euroland’. Journal of Economic Geography 15: 907–​934. Folke, C. (2006). ‘Resilience: the emergence of a perspective for social—​ecological systems analysis’. Global Environmental Change 15: 253–​267. Frankel, M. (1955). ‘Obsolescence and technological change in a maturing economy’. American Economic Review 45: 296–​319. Friedman, M. (1988). ‘The “plucking model” of business fluctuations revisited’. Working Paper in Economics E88–​48, Hoover Institution, Stanford University.

862   Martin Garmezy, N. (1971). ‘Vulnerability research and the issue of primary prevention’. American Journal of Orthopsychiatry 41: 101–​116. Garretsen, J.H. and Martin, R.L. (2010). ‘Rethinking new economic geography models: taking geography and history more seriously’. Spatial Economic Analysis 5: 127–​160. Grimm, V. and Vissel, C. (1997). ‘Babel, or the ecological stability discussions: an inventory of terminology and a guide for avoiding confusion’. Oecologica 109: 323–​334. Gunderson, L. and Holling, B. (eds) (2002). Panarchy:  Understanding Transformations in Human and Natural Systems (Washington, DC: Island Press). Gunderson, L. and Pritchard, L. (eds) (2002). Resilience and the Behaviour of Large Scale Systems (Washington, DC: Island Press). Hall, P. and Soskice, D. (2002). Varieties of Capitalism:  The Institutional Foundations of Comparative Advantage (Oxford: Oxford University Press). Hamel, G. and Välikangas, L. (2003). ‘The quest for resilience’. Harvard Business Review September: 1–​15. Handmer, J.W. and Dovers, S.R. (1996). ‘A typology of resilience: rethinking institutions for sustainable development’. Industrial and Environmental Crisis Quarterly 9: 482–​511. Hanley, N. (1998). ‘Resilience in social and economic systems: a concept that fails the cost-​ benefit test?’ Environment and Development Economics 2: 221–​262. Hans, Y. and Goetz, S.J. (2013). ‘Predicting the economic resilience of US counties from industry input–​output accounts’. Paper presented at the Southern Regional Science Association Annual Meeting, Washington, DC, USA, 5 April 2013. Harrison, N. (2003). ‘Good governance:  complexity, institutions, and resilience’. Paper Presented to the Global Environmental Change Research community, Montreal Canada, 16–​18 October. Harvey, D. (2006). Spaces of Global Capitalism (London: Verso). Hassink, R. (2010). ‘Regional resilience: a promising concept to explain differences in regional economic adaptability?’ Cambridge Journal of Regions, Economy and Society 3: 45–​58. Herrfahrtdt-​Pähle, E. and Pahl-​Wostle, C. (2012). ‘Continuity and change in social-​ecological systems: the role of institutional resilience’. Ecology and Society 17: Article 8. Hill, E., Wial, H., and Wolman, H. (2008). ‘Exploring regional economic resilience’. Working Paper 2008, 4, Institute Urban and Regional Development, UC Berkeley. Holling, C.S. (1973). ‘Resilience and stability of ecological systems’. Annual Review of Ecology and Systematics 4: 1–​23. Holling, C.S. (1996). ‘Engineering Versus Ecological Resilience’ in P. Schulze (ed.) Engineering Within Ecological Constraints, pp. 31–​44 (Washington, DC: National Academy Press). Hudson, R. (2010). ‘Resilient regions in an uncertain world: wishful thinking or a practical necessity?’ Cambridge Journal of Regions, Economy and Society 3: 11–​26. Janssen, M. and Anderies, J.M. (2007). ‘Robustness of socio-​ecological systems to spatial and temporal variability’. Society and Natural Resources 20: 1–​6. Jen, E. (2003). ‘Stable or robust? What is the difference?’ Complexity 8: 12–​18. Kaplan, H.B. (1999). ‘Toward an Understanding of Resilience: A Critical Review of Definitions and Models’ in M.D. Glantz and J.R. Johnson (eds) Resilience and Development: Positive Life Adaptations, pp. 17–​83 (New York: Plenum). Kitano, H. (2004). ‘Biological robustness’. Nature Reviews Genetics 5: 826–​837. Krugman, P. (1994). ‘Complex landscapes in economic geography’. American Economic Review 84: 412–​416. Krugman, P. (1996). The Self-​Organising Economy (Oxford: Oxford University Press).

Shocking Aspects of Regional Development    863 Lang, T. (2012). ‘How do cities and regions adapt to socio-​economic crisis?’ Raumforsch Raumordnung 70: 285–​291. Lebel, L., Anderies, J.M., Campbell, B., Folke, C., and Hatfield-​Dodds, S. (2006). ‘Governance and the capacity to manage resilience in regional social-​ecological systems’. Ecology and Society 11: 9. Lee, A.V., Vargo, J., and Seville, E. (2013). ‘Developing a tool to measure and compare organisations’ resilience’. Natural Hazards Review 14: 29–​41. Luthar, S.S. (ed.) (2003). Resilience and Vulnerability: Adaptation in the Context of Childhood Adversities (Cambridge: Cambridge University Press). Luthar, S. and Becker, B. (2000). ‘The construct of resilience: a critical evaluation and guidelines for future work’. Child Development 7: 543–​562. MacKinnon, D. and Derickson, K. (2013). ‘From resilience to resourcefulness: a critique of resilience policy and activism’. Progress in Human Geography 37: 253–​270. Marcos, J. and Macaulay, S. (2008). ‘Organisational resilience: the key to anticipation, adaptation and recovery’. Paper, Cranfield School of Management, Cranfield University. Martin, R.L. (2006). ‘Economic Geography and the New Discourse of Regional Competitiveness’ in S. Bagshi-​Sen and H. Lawton-​Smith (eds) Economic Geography: Past, Present and Future, pp. 159–​172 (London: Routledge). Martin, R.L. (2010a). ‘The “New Economic Geography”: Credible Models of the Economic Landscape?’ in A. Leyshon, R. Lee, L. McDowell, and P. Sunley (eds) The SAGE Handbook of Economic Geography, pp. 53–​72 (London: SAGE). Martin, R.L. (2010b). ‘Roepke lecture in economic geography—​rethinking regional path dependence: beyond lock-​in to evolution’. Economic Geography 86: 1–​27. Martin, R.L. (2012). ‘Regional economic resilience, hysteresis and recessionary shocks’. Journal of Economic Geography 12: 1–​32. Martin, R.L. and Sunley, P. (2006). ‘Path dependence and regional economic evolution’. Journal of Economic Geography 6: 395–​437. Martin, R.L and Sunley, P. (2007). ‘Complexity thinking and evolutionary economic geography’. Journal of Economic Geography 7: 573–​601. Martin, R.L. and Sunley, P. (2011). ‘Conceptualizing cluster evolution: beyond the life cycle model’. Regional Studies 7: 1299–​1318. Martin, R.L. and Sunley, P. (2015). ‘On the notion of regional economic resilience: conceptualisation and explanation’. Cambridge Journal of Economic Geography 14: 1–​42. Martin. R.L, Sunley, P.J., Tyler, P., and Gardiner, B. (2016). ‘How regions react to recessions: resilience and the role of economic structure’. Regional Studies 50: 561–​585. Masten, A.S. (2014). Ordinary Magic: Resilience in Development (New York: Guilford Press). Metcalfe, J.S. (1998). Evolutionary Economics and Creative Destruction (London: Routledge). O’Dougherty Wright, M., Masten, A.S., and Narayan, A.J. (2013). ‘Resilience Processes in Development: Four Waves of Research on Positive Adaptation on the Context of Adversity’ in S. Goldstein and R.B. Brooks (eds) Handbook of Resilience in Children, pp. 15–​37 (New York: Springer). Ormerod, P. (2008). ‘Resilience after local economic shocks’. Applied Economic Letters 17: 503–​507. Ormerod, P. (2010). ‘Risk, recession and the resilience of the capitalist economies’. Risk Management 12: 83–​99. Pelling, M. (2011). Adaptation to Climate Change:  From Resilence to Transformation (London: Routledge).

864   Martin Pendall, R., Foster, K., and Cowell, M. (2010). ‘Resilience and regions: building understanding of the metaphor’. Cambridge Journal of Regions, Economy and Society 3: 71–​84. Pike, A., Dawley, S., and Tomaney, J. (2010). ‘Resilience, adaptation and adaptability’. Cambridge Journal of Regions, Economy and Society 3: 59–​70. Rentfrow, P.J. (2010). ‘State differences in personality: towards a psychological geography of the United States’. American Psychologist 65: 548–​558. Rentfrow, P.J. (ed.) (2013). Geographical Psychology: Exploring the Interaction of Environment and Behaviour (Washington, DC: American Psychological Association). Rentfrow, P.J., Gosling, S.D., and Potter, J. (2008). ‘A theory of the emergence, persistence, and expression of regional variation in basic traits’. Perspectives on Psychological Science 3: 339–​369. Robb, D. (2000). ‘Building resilient organizations’. OD Practitioner 32: 27–​32. Rodin, J. (2015). The Resilience Dividend: Managing Disruption, Avoiding Disaster, and Growing Stronger in an Unpredictable World (London: Profile Books). Romer, P. (2001). Advanced Macroeconomics (New York: McGraw Hill). Schneider, M.R. and Paunescu, M. (2012). ‘Changing varieties of capitalism and revealed comparative advantage from 1990 to 2005: a test of the hall and Soskice claims’. Socio-​Economic Review 10: 731–​753. Schumpeter, J. (1942). Capitalism, Socialism and Democracy (London: Routledge). Scott, M. (2013). ‘Resilience: a concept for rural studies?’ Geography Compass 7/​9: 597–​610. Setterfield, M. (1997). Rapid Growth and Relative Decline (Basingstoke: Macmillan). Seville, E. (2009). ‘Resilience: great concept … but what does it mean for organisations?’ https://​ir.canterbury.ac.nz/​handle/​10092/​2966 (last accessed 30 April 2017). Sheffi, Y. (2007). The Resilient Enterprise: Overcoming Vulnerability for Competitive Advantage (Cambridge, MA: MIT Press). Sheffi, Y. (2015). The Power of Resilience: How the Best Companies Manage the Unexpected (Cambridge, MA: MIT Press). Siegel, P.B., Johnson, T.G., and Alwang, J. (1995). ‘Regional economic diversity and diversification’. Growth and Change 26: 261–​284. Simmie, J. and Martin, R. (2010). ‘The economic resilience of regions: towards an evolutionary approach’. Cambridge Journal of Regions, Economy and Society 3: 27–​43. Slack, E. (2017). ‘The Geography of Local Public Finance’ in R.L. Martin and J. Pollard (eds) Handbook of the Geographies of Money and Finance, pp. 253–​278 (Chichester: Edward Elgar). Tompkins, J.A. (2007). ‘Four steps to business resilience’. Industrial Management 49: 14–​18. Tompkins, E.L. and Adger, W.N. (2004). ‘Does adaptive management of natural resources enhance resilience to climate change?’ Ecology and Society 9: 10. Tyson, R.E. (1968). ‘The Cotton Industry’ in D.H. Alcroft (ed.) The Development of British Industry and Foreign Competition, 1875–​1914 (London: Allen and Unwin). Walker, B. and Salt, D. (eds) (2006). Resilience Thinking: Sustaining Ecosystems and People in a Changing World (Washington, DC: Island Press). Wimmer, A. and Kössler, R. (eds) (2006). Understanding Change: Models, Methodologies and Metaphors (Basingstoke: Palgrave). World Bank (2012). Building Urban Resilience:  Principles, Tools and Practice (Washington, DC: World Bank). Zolli, A. and Healy, A.M. (2012). Resilience:  Why Things Bounce Back (London:  Headline Publishing Group).

Author Index

Aalbers, M. B., 9, 540–​1, 543–​4, 546, 551, 614, 618 Aarland, K., 272, 274 Abernathy, F. H., 433 Acemoglu, D., 67, 99, 230, 233, 521, 759 Acs, Z. J., 262 Adams, J. D., 255, 335–​6 Adams, P., 827 Addo, K. A., 671 Adger, W. N., 667, 672, 840 Adler, P. S., 531 Adner, R., 828 Adorno, T., 305 Agarwal, R., 272, 287, 297, 300 Aghion, P., 150, 250, 759, 762, 771 Aglietta, M., 115, 471 Agnew, J. A., 169 Agrawal, A. K., 275, 279, 333, 359 Aguiar, L., 277 Ahlbrandt Jr, R. S., 541 Ahmed, S., 667 Ai, C., 654 Ainslie, G., 202 Aitken, R., 616, 619, 623 Ajibade, I., 672 Akerlof, G. A., 657, 658 Akudugu, J. A., 783, 784 Alacer, J., 332, 334–​5, 337–​9, 348 Albouy, D., 509 Aldcroft, D. H., 843 Aldrich, H. E., 215 Alexander, N., 428, 432, 435 Alexander, P., 83 Alfaro, L., 348 Alford, M., 456 Ali, I., 777–​81 Ali-​Yrkkö, J., 332 Allen, F., 658 Allen, J., 183, 185, 328, 454, 647–​8

Allen, M., 757, 765 Allsopp, C., 657–​8 Almenberg, J., 196 Alvaredo, F., 49, 794 Ambachtsheer, K., 597 Amin, A., 172, 186, 230, 291, 296, 312, 471 Amsden, A., 25, 29, 826 Anderies, J. M., 848 Andersen, K. V., 318 Anderson, B., 277, 488 Anderson, C., 5, 276, 822 Andersson, Å. E., 501 Ang, A., 593, 601 Ang, S., 416 Anheier, H. K., 305, 800 Ansari, S., 829, 830 Anthony, S., 579 Antràs, P., 399 Aoyama, Y., 11, 576, 827 Appadurai, A., 307 Appelbaum, E., 528–​30, 794, 801 Applebaum, H. A., 522–​4, 529 Appleyard, L., 543, 620 Appleyard, M. M., 255 Appold, S. J., 217 Apte, U., 409 Arce, A., 454 Armington, M. M., 803 Armitage, S., 640 Armstrong, R. C., 713 Arndt, H. W., 771 Arndt, S. W., 399 Arnott, R., 603 Arora, A., 272, 337, 414 Arrighi, G., 566 Arrow, K. J., 150, 202, 220, 251, 252 Arthurs, H. W., 252, 486, 491 Arzaghi, M., 347, 356 Ascani A., 376, 782, 783

866   author Index Asheim, B. T., 218, 222, 235, 392 Ashton, P., 546, 814 Asmussen, C. G., 412 Athey, S., 279 Athreye, S., 408–​9, 420 Atkinson, A. B., 40–​1, 772 Attewell, P., 521 Au, C.-​C., 85–​7, 92 Audia, P. G., 215, 325, 334 Audretsch, D. B., 259, 325–​6, 331–​2, 337, 348–​9, 368, 399 Auerswald, P. E., xxxi, 245, 250–​1, 255, 261 August, B., 314 Augustine, N., 843 Authers, J., 8 Autor, D. H., 67, 69, 273, 278, 511, 795 Auty, R. M., 170, 719–​20, 734 Aversano, N., 727 Avnimelech, G., 148 Aydos, P., 761 Azoulay, P., 336 Bachher, J. S., 597, 605–​6 Bacolod, M., 506, 511 Baffes, J., 652–​4, 657, 723 Bai, C., 86 Bailey, I., 686, 688 Bailey, R. G., 247, 253 Bailey, T., 529 Bair, J., 383–​4, 389, 394, 396, 449–​50, 452, 454 Baker, J. L., 774 Baker, W. E., 510 Bakija, J., 54 Bakker, G., 310, 311 Bakker, K., 170, 683 Balasubramanian, S., 276 Baldwin, C., 332, 338, 369, 419 Baldwin, R., 375, 399 Balland, P. A., 218–​19, 859 Ballas, D., 50 Banerjee, A. V., 104, 803, 828 Banerjee, M., 97 Banks, B., 801 Banks, J., 286 Banks-​Leite, C., 254 Bannon, I., 721 Bansal, P., 685–​6, 699

Baptist, S., 759, 765 Barca, F., 782, 786 Bardi, U., 752 Barley, S. R., 522 Barnard, H., 307 Barnes, T. J., 11, 161, 163–​5, 167, 169, 171–​2, 234, 239, 396, 470–​1, 672, 850 Barnes, W. R., 811 Barnett, A., 667, 671 Barnett, C., 616 Baron, J., 200, 205 Barrett, S., 761, 762 Barrientos, S., 453 Barrios, S., 671 Barro, R. J., 145, 499, 503 Barrowclough, D., 306 Barry, A., 658 Bartik, T. J., 155, 798–​9 Bartlett, C. A., 186 Barton, D., 592 Barua, A., 672 Basak, S., 654 Basco, S., 278 Bassens, D., 618 Bassett, T., 668 Bates, T., 798, 803 Bathelt, H., xxxi, 9, 179–​80, 182–​7, 189–​90, 205, 208, 218, 238, 299, 309, 312, 328, 338, 383, 604 Batt, R., 529 Battilana, J., 221, 419 Baum-​Snow, N., 511 Baye, M., 276 Bayus, B. L., 419 Beatty, C., 857 Beaverstock, J. V., 187, 376 Bebbington, A., 719, 726 Bebchuk, L. A., 598 Becattini, G., 325 Bechky, B. A., 294, 296 Beck, T., 545 Becker, G. S., 196, 208–​9, 499, 501, 509–​10 Becker, J., 115, 120, 131, 133, 845 Behrens, K., 347, 353 Beinhocker, E. D., 848 Belk, R. W., 456 Bell, A. M., 337 Bell, D., 276, 501, 575

author Index   867 Bell, M., 217, 277, 338 Bell-​Masterson, J., 254 Belussi, F., 216 Benartzi, S., 204 Benfratello, L., 372 Benner, C., 488, 793, 796, 801, 813–​14 Bennhold, K., 822 Beramendi, P., 64 Berchman, M., 523 Berg, A., 31, 775, 796 Bergene, A. C., 475 Berger, S., 332 Bergmann, L., 165 Berkes, F., 845 Berman, E., 68 Berndt, C., 167, 454 Berndt, E., 765 Bernhardt, A., 528–​9, 801 Bernstein, P. L., 566 Berry, B. J. L., 3 Berry, C. R., 503, 511 Berry, M., xxxi, 19, 24–​5 Bertram, R., 597 Bertrand, O., 412 Berube, A., 73, 74 Besky, S., 452, 454 Besley, T., 99, 104, 544 Beugelsdijk, S., 338 Bezemer, D., 99 Bhattacharjea, A., 552 Bhattacharya, B., 102 Bhatti, Y. A., 829 Bianchi, C., 437–​8 Bicchetti, D., 654–​5 Bier, Susanne, 314 Biggs, R., 854 Billo, E., 451 Binz, C., 220–​2 Birch, K., 813 Birdsall, N., 796 Birkenholtz, T., 668 Birkinshaw, J., 375 Birtchnell, T., 53 Bishop, P., 220 Black, D., 88, 509 Blackburn, R., 611 Blackler, F., 291

Blakely, E. J., 799 Blanchard, O., 843 Bleakley, H., 362 Blinder, A. S., 278, 407, 409 Block, F., 151 Blomley, N., 169 Blömstrom, M., 372 Bloom, N., 250, 263, 273, 334 Blowfield, M., 448–​9, 451–​2 Bluestone, B., 68, 466–​7, 470, 794 Blum, B., 277, 506 Boddy, M. J., 468, 541 Boden, M. A., 305 Bodie, Z., 647, 649 Boeckler, M., 167, 454 Boero, R., 670 Bogers, M., 286 Bohle, H., 668 Böhm, S., 683 Bok, R., 437 Bolan, R. S., 811 Bolt, J., 99 Bolton, P., 771 Bor, D., 209 Borghini, S., 187 Borjas, G., 521 Bos, J. W. B., 646, 655, 660 Boschma, R. A., xxxi, 2, 9, 12, 116, 165, 188, 213–​18, 220–​3, 230, 238, 260, 331, 733, 810, 819, 840, 852 Bosello, F., 669 Bosker, M., 90 Boston, T. D., 803 Bostrom, N., 756 Bosworth, T., 101 Botts, H. A., 604 Boudreau, K., 330 Bourdieu, P., 487 Bourguignon, F., 770 Bouvard, V., 605 Bouwer, L., 667–​8 Bowen, A., 749–​50, 755, 758–​9, 765 Bower, J. L., 829 Bowman, A., 478 Boyer, R., 115, 234, 472 Boykoff, M. T., 665 Bradford, C., 541, 546

868   author Index Bradley, S. W., 829 Bradshaw, M. J., 665, 732, 735 Bradshaw, T. K., 799 Brady, D., 794 Brady, M., 669, 675 Brady, T., 288 Brailly, J., 188 Braithwaite, J., 566, 638 Brakman, S., 172, 859 Brand, F., 845 Brando, C., 451 Brandth, B., 488 Branscomb, L. M., 255 Brass, D. J., 188 Brassett, J., 620 Braun, B., 170 Braverman, H., 522 Breau, S., 54 Breiger, R. L., 187 Breman, J., 101 Brenner, N., 169, 792 Brenner, T., 216 Breschi, S., 188, 218 Bresnahan, T. F., 271, 273, 330, 335 Bridge, G., 170, 665, 683, 732, 734, 738 Briguglio, L., 667, 842 Brillo, B. B., 437 Brinks, V., 287, 289, 291, 297–​8, 300 Bristow, G., 840, 848 Brito Henriques, E., 313 Brittan, S., 544 Brock, W. A., 765 Broekel, T., 218 Broome, J., 758 Brown, C., 209 Brown, J., 219 Brown, J. R., 271 Brown, T., 291–​2 Brown, W., 216, 477 Brownell, A., 360 Brückner, M., 775 Brunner, S., 762 Brunsson, N., 420 Brusco, S., 235 Brusoni, S., 420 Bryan, D., 648 Bryceson, D. F., 775

Bryden, R., 339 Brynjolfsson, E., 275–​7 Bryson, J. R., 818 Buchanan, J. M., 633 Bucholz, D., 71 Buckley, P., 368 Budish, E., 578 Buenstorf, G., 216 Bulkeley, H., 683 Bulloch, G., 419 Bulte, E. H., 721 Bumpus, A. G., 170, 665, 686 Bunyaratavej, K., 412 Burawoy, M., 522, 531 Burgess, R., 104 Burnham, M., 672 Burroni, L., 235 Burt, R. S., 188 Burt, S., 439 Burton, G. M., 603 Burton, R., 667 Busch, L., 454 Butchart, S. H. M., 754 Butler, J., 486 Büyükşahin, B., 654 Byatt, D., ix Bygrave, W. D., 601 Caerlewy-​Smith, E., 208 Cai, F., 81, 83, 90–​2 Cairncross, F., 270 Caletro, J., 53 Çaliskan, K., 167 Callon, M., 167, 287, 298, 454, 684, 686 Calvert, K., 733 Calvo, G. A., 545 Campbell, C. J., 705 Campbell, J. C., 207 Campbell, J. Y., 597 Caniglia, B. S., 840 Cantner, U., 219 Cantwell, J., 186, 312, 367, 370, 375–​6 Capone, G., 222 Card, D., 67, 801 Cardoso, President E., 796 Carey, D., 738 Carlin, B. I., 595

author Index   869 Carlino, G., 339, 348, 510 Carmel, E., 416 Carmody, P., 396 Carnoy, M., 813 Carpenter, J. N., 598 Carr, D. L., 373 Cartier, A., 79 Casale, M., 667 Cashin, P., 654 Casper, S., 155 Cassis, Y., 566 Castaldi, C., 220 Castellani, D., 370 Castells, M., 576, 580, 794 Castranova, E., 580 Castree, N., 2, 170, 466, 474–​5 Castronova, E., 580, 583, 585 Cattaneo, O., 382, 399 Caves, R. E., 305–​6, 308, 310–​11, 313, 372–​4, 509 Cavoli, T., 27–​8 Cellini, R., 843 Cerra, V., 843 Chacar, A. S., 336, 338 Chamberlin, E., 629, 635 Chaminade, C., 313 Chan, A., 83 Chan, C. K-​C., 456 Chan, K. W., xxxii, 78, 80–​4, 86–​93 Chandler, A., 336 Chandra, K., 109 Chang, H. J., 29 Chang, S. D., 92 Chant, S., 97 Chapple, K., xxxii, 792, 800 Charlot, S., 271 Chatterjee, M., 672 Chatterji, A., 360 Chatty, R., 168 Chautard, A. ix Chen, B., 91 Chen, M., 348 Chen, M. A., 779 Chesbrough, H., 419 Cheshire, P., 152 Chetty, R., 795 Chinitz, B., 333, 353, 360, 797 Choi, J., 276

Chor, D., 399 Christaller, W., 499–​500 Christensen, C. M., 829 Christophers, B., 168, 614, 648 Christopherson, S. , viii, xxxii, xxxiii, 9, 234, 309, 313, 452, 486–​7, 499, 501, 577, 647, 732, 735–​8, 810, 812 Chu, W., 826 Chung, P.-​P., 92, 334 Chuprinin, O., 596 Churchill Semple, E., 718 Churchill, W., 631 Cingano, F., 771–​2, 775 Cistulli, V., 772 Clapp, J., 650 Clark, G. L., ix, xxi, xxxii, 1, 3, 8, 58, 129, 159, 168, 180, 182, 190, 196–​7, 200–​8, 238, 338, 391, 396, 449, 466, 468, 542–​3, 546, 569, 572, 577, 596–​8, 603–​6, 614, 616, 623, 634, 649, 658, 684, 699, 733, 810, 813 Clark, J., xxxiii, 735, 810, 812, 814, 818, 852 Clark, J. R. A., 577, 593, 683 Clark, P. B., 596, 601 Clark, T. N., 509 Clavel, P., 816 Cliff, D., 579 Clower, T. L., 736 Coase, R. H., 250, 629, 634, 683, 751 Cobble, D. S., 522 Cochrane, A., 468 Coe, N. M., xxxiii, 1, 10–​11, 181, 307–​8, 312–​13, 383–​5, 387, 389, 391–​6, 399, 427–​8, 432, 434, 436, 452–​3, 466, 474–​5, 478, 486, 488, 491, 561, 612, 621, 623, 734–​5 Coenen, L., 223, 732 Cohen, D., 503 Cohen, W. M., 335–​6, 828 Cohendet, P., 186–​7 Colby, C. C., 718 Cole, A., 54 Coles, A.-​M., 253 Coll, S., 735 Collier, P., 146, 762 Collier, S., ix, 721 Collins, S., 58, 101 Combes, P.-​P., 91, 172, 347–​9, 359 Cook, A. C. G., 551

870   author Index Cook, I., 454 Cooke, P. N., 116, 214, 222, 235, 306, 383 Coomes, O. T., 670 Cooper, B. S., 802 Cooper, K., 57 Coppock, S., 618, 620 Corbridge, S., xxxiii, 102, 107–​8, 551 Corden, W. M., 719 Corfee-​Morlot, J., 675 Cornell, B., 606 Corpataux, J., 543, 551 Cortes, P., 521 Cortright, J., 339 Costello, C., 754 Couto, V., 408–​9, 411 Cova, B., 286 Coval, J. D., 596 Cowell, M., 843 Cowen, D., 163, 172 Cowen, T., 40 Cox, K., 116 Coyle, D. M., 799 Crandall, R., 274 Cranford, C. J., 487 Cray, A., 800 Crescenzi, R., 152, 216, 371–​2, 377 Crespo, N., 372 Crevoisier, O., 181 Cribb, J., 42 Crotty, J., 726 Crouch, C., 189, 236 Crutzen, P. J., 755 Cruz, A., 671 Cuddington, J. T., 722 Cunningham, S., 305 Currah, A. D., 313, 434–​5 Currid-​Halkett, E., 42, 312, 505 Curry, L., 166 Cusmano, L., 222 Cutler, A. C., 637–​8 Cutler, T., 616, 618 Cutter, S., 668, 670, 672, 676 Czelusta, J., 721 da Rocha, A., 434 da Silva, J., 855 da Silva, L. I. L., 796

da Vinci, L., 255 Dabla-​Norris, E., 55 Daft, R. L., 280 Dahl, M. S., 215, 325, 337–​8 Dakora, E. A. N., 438 Dales, J. H., 683 Daly, H. E., 318, 756 Dani, D., 245 Dani, L., xxxiii Dannenberg, A., 761–​2 Darwin, C., 247 Das, R., 102 Dasgupta, P., 145, 757, 759 Datu, K., 377 Davenport, T., 408–​9, 417, 421 Davenzati, G. F., 230 Davey, E., 750 Davidson, J., 582 Davidson, L., 73 Davidsson, P., 599 Davies, A., 288 Davies, R., 649 Davies, S., 116 Davoudi, S., 859 Dawley, S., 220 Dawson, J., 436 De Cian, E., 669 de Graaff, N., 735 De Groot, H. L. F., 219 De La Cadena, M., 725 de la Fuente, A., 503 de Landa, M., 170 de Lemos, E., 761 de Ruyter, K., 289–​90, 292 de Sherbinin, A., 669 De Vaus, D., 189 de Vries, J., 563 Dean, M., 453 Defever, F., 376 Delgado, M., xxxiii, 181, 324–​6, 330–​5, 337–​9, 348, 353 Dell, M., 665, 667 DeLong, B., 102 Demirbag, M., 412 Deng Xiaoping, 24 Denhardt, J., 854 Denhardt, R., 854

author Index   871 Denike, K., 166 Denning, L., 740 Dépelteau, F., 187 Derickson, K., 855 Derrida, J., 169 Dertouzos, M. L., 332 Desilver, D., 73 Desrochers, P., 220 DeVerteuil, G., 41 Devi, R., 108 Deville, J., 614, 618, 622 Dewar, M., 799 Dholakia, N., 298 Diamond, D. W., 545 Diamond, R., 511, 512 Dib, L. A., 434 Dicken, P., 12, 68, 181, 183, 185–​6, 312, 382–​3, 385, 391, 393, 399, 558, 561, 734, 735 Dietz, S., 760 DiNardo, J. E., 67 Ding, W. W., 25, 275 Dissart, J. C., 798, 848 Dixon, A. D., 35–​6, 168, 391, 543, 550, 572, 597–​8, 606, 634 Djellal, F., 184 Dobbs, R., 774 Dodge, M., 581, 585 Doeringer, P. B., 522 Doh, J., 412 Doherty, B., 456 Doherty, M. E., 205, 428 Doig, J. W., 634 Dolan, C. S., 449, 451–​4 Dollar, D., 105 Domanski, D., 651–​2 Dombos, T., 456 Domenech, R., 503 Domm, P., 579 Donaghey, J., 420 Donahue, S. M., 310 Donald, B., 486 Donaldson, J., 106 Donnelley II, E., 605 Dorado, S., 419 Doran, J., 843, 859 Dore, R., 611, 618 Doreian, P., 188

Dorling, D., xxxiv, 39–​41, 46, 51, 53–​5, 57, 59 Dorn, D., 278 Dornbusch, R., 655 Dörry, S., 392 Dosi, G., 213 Dossani, R., 407, 409, 411 Douglas, I., 672 Doussard, M., 466, 528, 813 Dovers, S. R., 853 Dow, K., 668 Downes, T., 272 Drache, D., 718 Drahos, P., 566, 638 Dranove, D., 273 Dreber, A., 196 Dreger, C., 68 Dreze, J., 106 Driffield, N., 372 Drucker, P., 501 Du Bry, T., 527 du Gay, P., 612, 623 Dube, A., 801 Dubey, A., 774 Duflo, E., 828 Dumais, G., 337, 347 Duménil, G., 165, 612, 648 Dunford, M., 115, 770 Dunne, T., 335 Dunning, J. H., 186, 335, 367–​8, 370, 372, 377 Dupuis, M., 617 Duranton, G., xxxiv, 11, 153, 172, 271, 274, 332, 337, 339, 347–​9, 352–​4, 359–​60, 376, 503, 775 Dybvig, P. H., 545 Dyer, S., 486 Dymski, G. A., xxxiv, 168, 539, 542, 550–​2, 612, 614, 620 Dynarski, S., 57 Eakin, H., 667, 671–​2 Easterlow, D., 203 Easterly, W., 99, 101, 145, 197, 829 Eastwood, R., 104 Eaton, J., 509, 545 Eberts, R., 793, 796 Ebner, A., 235 Eckstein, Z., 509 Economy, E. C., 710

872   author Index Edgcomb, E., 803 Edlund, L., 509 Ehrenfeld, D., 752 Ehrlich, A. H., 713 Ehrlich, P. R., 707, 713, 751 Einav, L., 277 Einstein, A., 764 Eisenschitz, A., 813 Eisinger, P. B., 799 Elberse, A., 310–​11 Elbing, S., 236, 239 Elden, S., 736 Eliashberg, J., 310–​11 Elizabeth II, Queen, 544, 551 Ellis, C. D., 598, 605 Ellison, G., 276–​7, 324–​5, 331, 354, 360 Ellison, S. F., 276–​7 Elms, D. K., 382, 399 Elton, E. J., 592 Emirbayer, M., 187 Emmitt, S., 522 Endo, G., 436, 438, 444 Engelen, E., 546, 550, 611, 623 Engels, A., 687 Engels, B., xxxiv, 19 Engels, F., 504 England, K. V. L., 476 Enright, M., 335 Eppinger, S. D., 332, 338 Epstein, G. A., 539, 647 Ericson, R., 250 Erlanger, S., 822 Erlich, M., 524–​5 Ernst, D., 383 Erramilli, M., 416 Erten, B., 722, 723 Erturk, I., 61–​18, 620 Essletzbichler, J., 165, 214, 216, 220, 261, 732 Ethiraj, S., 40–​10, 414 Ettlinger, N., 185–​6 Evans, R., 840, 843 Evenson, R. E., 263 Everitt, B. S., 326 Ewers, R. M., 254 Fagerberg, J., 255, 262, 828 Faggio, G., 360

Fairris, D., 801 Falck, O., 509 Fallick, B., 351 Fama, E. F., 544, 599 Fan, S., 655, 772 Fang, L. H., 596, 606 Farla, J., 732 Farmer, J. D., 759, 763 Farole, T., 233, 734 Farrigan, T., 75 Fattouh, B., 653, 655, 657–​8, 660 Faulconbridge, J., 181, 185, 189, 287, 295, 297, 299–​300, 546, 603–​4, 623 Fawcett, E., 148 Fay, M., 750, 775 Fearon, J., 97 Featherstone, D., 169, 490 Feenstra, R. C., 399 Fehr, E., 208 Feldman, M. P., ix, xxi, xxxiv, 1, 116, 143–​4, 147–​9, 151, 153–​5, 221, 259, 262, 324–​6, 331–​2, 334, 337–​8, 348–​9, 368, 812, 827 Ferguson, N., 567 Fernández, V. R., 398 Fernández-​Macías, E., 47 Ferrari, P., ix Feser, E. J., 325–​6, 506, 798 Fiaschetti, M., 209 Fielding, D., 99 Fik, T. K., 167 Fine, B., 159 Fine, J. R., 529 Fingleton, B., 260, 843, 849, 859 Fink, L., 570 Finkel, G., 523 Finkel, M. L., 737 Finkin, M., 812 Finlayson, A., 616–​18 Fiol, C. M., 215 Firebaugh, G., 64 Fischer, M. M., 503 Fisher, G. M., 64 Fisher, P., 799 Fitzgerald, J., 802 Fitzgerald, W., 740 Fjeldstad, Ø. D., 183, 419 Flamm, K., 272

author Index   873 Fleming, L., 219 Flew, T., 305 Flodman Becker. K., 779 Flohr, S., 300 Florida, R., xxxv, 64, 73, 309, 413, 499–​502, 504–​6, 508–​11, 795 Flyer, F., 334 Fogelman, C., 668 Folbre, N., 476 Foley, D. K., 758 Folke, C., 839, 840, 845 Foray, D., 223 Forbes, B. C., 14 Ford, J., 675 Ford, President G., 524 Forde, C., 486 Forman, C., xxxv, 269, 270–​3, 275–​7 Fornahl, D., 216, 220 Fors, G., 370 Fort, T. C., 278, 332, 338 Foster, D. P., 598 Foster, L., 250, 263 Fothergill, S., 857 Foucault, M., 169, 616 Fourcade, M., 187 Fowler, C. S., 172 Foxon, T. J., 732 Francis, A., 418 Franke, N., 295 Frankel, J. A., 653 Frankel, M., 859 Franks, D., 671 Franz, M., 453 Franzen, D., 208 Frazier, T., 669–​70, 675 Frederiksen, L., 308–​9 Freedland, M., 493 Freeman, C., 222, 261 Freeman, R., 409 Freidberg, S. E., 454 French, K. R., 599 French, S., 547, 550, 603, 611, 615–​17, 619, 622, 646 Frenken, K., xxxv, 9, 213–​21, 230, 261, 325, 339, 733, 810, 819 Freund, C., 277 Fried, J. M., 598

Friedman, M., 448, 585, 847 Friedman, T. L., 270, 277, 413, 508, 719 Friedmann, J., 376 Froebel, F., 501 Frohlich, T. C., 72 Froud, J., 611, 612 Frydman, C., 54 Frynas, J. G., 448–​9 Fuchs, E., 332 Fudge, C., 468 Fudge, J., 486–​7, 490 Fujita, M., 146, 717, 775 Fuller, A. T., 666 Funder, D. C., 200 Furman, J., 63 Füssel, H.-​M., 672 Gabe, T. M., 505–​6 Galbraith, J. K., 31–​2, 34, 36, 548, 568 Gambardella, A., 330, 335 Gandhi, I., 101 Gandhi, M., 107 Gandhi, R., 101 Gans, J., 338 Garcia Haro, M. A., 286–​7 Garcia-​Perpet, M.-​F., 167 Garmezy, N., 839 Garretsen, H., 172, 850 Garsten, C., 688 Garud, R., 220, 732 Garzik, J., 583 Gaspar, J., 271, 274 Gates, B., 505, 828 Gates, G., 510 Gavrilets, J., 245, 251, 256, 264 Geels, F. W., 253 Gehman, J., 732 Geif, A., 199 Geisst, C. R., 566 Genus, A., 253 George, G., 826, 828, 830 George, L., 276 Georgescu-​Roegen, N., 165 Gerbasi, J., 814 Gereffi, G., 186, 375, 383–​4, 393, 396, 399, 451, 717 Gerrans, P., 209

874   author Index Gerschenkron, A., 551 Gertler, M. S., ix, xxi, 1, 116, 182, 222, 230, 232–​5, 238–​9, 309, 338, 466, 471–​2, 568, 576–​7, 604, 812, 819, 827 Getz, C., 453 Ghandnoosh, N., 529 Ghani, E., 360 Ghauri, P. N., 368 Ghemawat, E., 413 Ghose, A., 277 Ghoshal, S., 186 Gibb, M., 784 Gibbon, P., 383, 396, 452–​3 Gibbons, M., 299 Gibson, R., 187, 189 Gibson-​Graham, J. K., 1, 171, 474 Giddens, A., 162, 305 Gigerenzer, G., 200–​1 Gilad, S., 638 Gilbert, E., 618 Gilbert, W. S., 634 Gillett, S., 274 Gillingham, K., 759 Giloth, R., 802 Ginsburgh, V. A., 305 Girard, M., 288 Giroud, X., 336 Gitman, L., 740 Giuliani, E., 187, 217–​18, 413 Glaeser, E., 11, 146, 152, 219, 259, 271, 274, 309, 324–​6, 331, 333, 335, 347, 352–​3, 360, 500, 502–​3, 506, 509, 511, 775, 798 Glaister, K., 412 Glasmeier, A. K., xxxv, 41, 63, 67, 216, 671 Glassman, J., 169, 398 Gleeson, B., 171 Glückler, J, xxxvi., 9, 179–​80, 182–​6, 188–​90, 205, 208, 219, 604 Glyn, A., 568 Gobillon, L., 348 Godley, W., 549 Goerzen, A., 376 Goetz, S. J., 843 Goger, A., 449 Goldfarb, A., xxxvi, 269–​70, 272, 275, 277 Goldmanis, M., 277 Goldring, L., 488

Goldsmith, J., 586 Golfetto, F., 187 Golledge, R., 197 Golosov, M., 758 Gomberg-​Muñoz, R., 527 Goodman, L., 71 Goodwin, J., 187 Goolsbee, A., 271–​2, 277 Gordon, I. R., 182, 309 Gordon, R. B., 752, 765 Gordon, R. J., 5 Gornick, J., 49–​50, 52 Gorse, C., 522 Gorton, G., 650 Gospel, H., 409 Gottlieb, P. D., 509 Gottschalg, O., 598 Gough, J., 813 Govindarajan, V., 826, 828–​9 Grabelsky, J., 524–​5 Grabher, G., xxxvi, 116, 182, 188, 205, 216, 218, 235, 286–​7, 289–​96, 299–​300, 308–​9, 383 Graeber, D., 41 Graf, H., 219 Graham, J. R., 596 Graham, M., 576 Graham, S., 603 Granovetter, M., 154, 205 Graves, B., 523 Graves, P. E., 509 Gravner, J., 251 Green, A. E., 488 Green, M., 466 Greenbaum, R. T., 800 Greenhut, M. L., 167 Greenstein, S., xxxvi, 269, 271–​3 Greenwood, B. N., 279 Greenwood, R., 599 Greer, R. J., 649 Gregson, N., 392 Griffith, D. A., 412 Grigg, D., 170 Grillitsch, M., 235, 239 Grimm, V., 845 Grimshaw, D., 530 Grindle, M., 102 Gripaios, P., 220

author Index   875 Grönroos, C., 286 Gross, J., 800 Grosse, M., 776 Grossman, G., 147, 374, 399 Grote, M., 577, 578 Grove, K., 668 Gruber, D. A., 307–​8 Gruber, M. J., 592 Grubesic, T. H., 272 Guarin, A., 456 Guenther, C., 220 Gulbrandson, I. T., 294 Gunderson, L., 253, 257, 845, 852 Gurley, J. G., 551 Gustafsson, B., 774 Guthman, J., 448, 451, 593–​4, 601, 617 Guy, F., 369 Guzman, J., 337–​8 Gyourko, J., 511 Haalboom, B., 451 Haberly, D., 563, 567 Habito, C. F., 777 Hacker, P. M. S., 208 Hackworth, J. R., 813 Haefliger, S., 287, 300 Hagan, J., 519 Hagerman, L. A., 601 Hagerstrand, T., 197 Haigh, N., 419 Haldane, A., 207, 649 Hall, B., 356 Hall, J., 829 Hall, P. A., 188–​9, 222, 234, 237, 239, 699, 857 Hall, S., 543, 547, 611, 622–​3 Hallegatte, S., 666, 670, 750, 762 Halpin, S., 672 Haltiwanger, J., 798 Hamel, G., 855 Hamilton, G. G., 427 Hamilton, K., 751 Hamlin, K., 91 Hammer, I., 183 Hammermesh, D. S., 275 Hampwaye, G., 784 Handmer, J. W., 853

Haniotis, T., 652–​4, 657 Hanley, N., 859 Hans, Y., 843 Hanson, S., 466, 472, 476 Haraway, D., 170 Harcourt, G. C., 163 Harden, C. P., 675 Hardie, I., 620 Hardoy, J., 672 Hargadon, A. B., 294, 296 Harker, C., 486 Harris, R. S., 596 Harrison, B., 68, 466–​7, 470, 472, 794, 813 Harrison, M., 454 Harrison, N., 853 Harrison, P., 671 Harrison, R. T., 543 Harriss-​White, B., 101 Hart, J., 667 Hart, S., 829 Hartley, J., 305 Hartog, M., 220 Harvey, D., 41, 64, 115–​16, 162, 168, 170–​2, 469, 473, 475, 541, 544, 550, 568–​9, 578, 647, 648, 690, 699, 742, 851 Haskel, J. E., 372 Hassink, R., 116, 216, 223, 260, 840 Hatch, M., 822 Hatfield, I., 491 Hau, H., 577 Hausmann, R., 100 Hawley, J. P., 603 Hay, C., 619 Hay, I., 41 Hayek, F. A., 585 Hayes, T. J., 495, 683, 687 Hayter, R., 181, 223, 716, 719, 732 Haythornthwaite, C., 287 Headey, D., 655 Heal, G., 750, 757–​9 Health, A., 651–​2 Healy, A., 840, 848 Hebb, T., 449 Hecker, D., 337 Heim, B. T., 54 Heimans, J., 763 Helleiner, E., 650

876   author Index Helm, D. xxxvii, 637, 640, 703, 705–​6, 713, 761–​2 Helper, S., 333–​4 Helpman, E., 147, 399 Helsley, R., 349, 360 Henderson, B. J., 654 Henderson, G., 165 Henderson, J., 383, 385, 387, 391–​3, 435, 561 Henderson, J. V., 11, 85–​8, 92, 273, 334, 336, 347, 349, 356 Hendrikse, R. P., 547, 618 Hendry, C., 219 Henn, S., 185, 299 Hennessey, P., 544 Hennig, B. D., 41 Henning, M., 252, 260–​1, 325 Henrich, J., 208 Henry, N., 477 Hepburn, C., xxxvii, 749–​51, 755, 758, 759–​62, 765 Herman, A., 454 Hernández Romero, M. A., 527 Herod, A., 169, 392, 466, 473–​5 Herrfahrdt-​Pähle, E., 853 Hertel, T., 667 Herzog, L., 688 Hesmondhalgh, D., 305, 308 Hess, M., 383, 387, 392, 394, 396 Hidalgo, C. A., 220, 252, 259–​60 Hilferding, R., 541 Hill, E., 846 Hindman, D. B., 272 Hirsch, B. T., 771 Hirsch, P. M., 305, 307–​8 Hirshleifer, J., 598 Hobday, M., 420 Hochberg, Y. V., 596 Hockerts, K., 419 Hodbod, J., 675 Hodgson, G., 205 Hoffman, A., 419 Hogarth, R. M., 205 Holland, T., 55 Holling, B., 852 Holling, C. S., 245, 253, 257, 260–​1, 264, 839, 845 Holman, N., 546 Holmes, J., 475

Holmes, N., 73–​4 Hong, H., 202 Hood, N., 375 Hoover, E. M., 500 Hope, C., 758 Hopenhayn, H. A., 250 Hopwood, A. G., 687 Horkheimer, M., 305 Horner, R., 389–​90, 394 Horowitt, G., 153 Hortacsu, A., 277 Hoskins, C., 309, 311 Hotelling, H., 500, 756–​7, 760 Hotz-​Hart, B., 399 Hough, P. A., 454 Howells, J., 287 Howitt, P., 250, 759 Howkins, J., 502 Hsieh, C.-​T., 263, 352 Huang, Y., 79, 91 Hubbert, M. K., 705–​7, 712 Huberman, G., 202 Hudson, R., 181, 312, 393, 840, 851 Hughes, A., xxxvii, 383, 448–​50, 452 Hughes, T. P., 631 Humphrey, J., 383, 396, 415, 436, 438, 452 Humphreys, D., 654, 658 Humphreys, M., 738 Humphries, J., 476 Hurley, S., 197 Hutton Ferris, D., 190 Hutzschenreuter, T., 412 Hwang, E. L., 290, 298 Hwang, V. W., 153 Hymer, S., 368–​9, 377 Iammarino, S., xxxvii, 147, 216, 220, 366, 369–​70, 374–​6, 413 Ianchovichina, E., 778, 786 Ibañez, J., 671 Ibert, O., xxxviii, 286–​7, 289, 291–​5, 297, 300 Ibragimov, R., 596 Ietto-​Gillies, G., 368, 371, 375 Immelt, J. R., 829 Inglehart, R., 510 Inkpen, A., 416 Innis, H. A., 716, 718, 720, 723

author Index   877 Irwin, S. H., 653, 655–​6 Isaksen, A., 218, 220, 222 Isar, Y. R., 305 Isard, W., 163, 473 Isenberg, D. J., 245 Iskander, N., xxxviii, 519, 522, 525–​7 Ivković, Z., 577, 596 Iyer, B., 417 Jacks, D. S., 655–​6 Jackson, M., 101 Jackson, P., 491 Jackson, S. L., 719, 723–​4 Jackson, T., 750 Jacobs, J., 151, 219–​20, 259, 261, 297, 309, 360, 499–​500, 509, 797–​8, 827 Jacobs, M., 758 Jacobsson, K., 688 Jacobus, R., 800 Jaffe, A., 275, 334, 336, 348 Jäger, J., 133 James, A., 488 Jameson, H., 493 Jank, W., 277 Janssen, M., 848 Jansson, J., 295 Jarecki, H. G., 657 Javorcik, B. S., 372 Jax, K., 845 Jeannerat, H., 286, 298 Jeffcut, P., 306 Jeffers, J., 667 Jen, E., 848 Jenkins, R., 102, 108, 450 Jennings, J., 306 Jennings, W. W., 596 Jensen, J. B., 278 Jensen, P., 409, 412, 420 Jeppesen, L. B., 293–​4 Jerrett, D., 722 Jessop, B., 115, 742 Jevons, W. S., 703, 712 Jian, T., 772 Johannisson, B., 183–​4 Johns, J., 313 Johns, R., 450 Johnsen, S., 252–​3

Johnson, B., 151, 183–​4 Johnson, K., 671 Johnson, L., 673, 683 Johnson, N., 579 Johnson, P., 55, 99 Jonas, A. E. G., 472 Jones, A., 10, 53, 181, 185, 542, 672 Jones, B. F., 275 Jones, C., 305 Jones, M., 201 Jordhus-​Lier, D. C., 392, 466, 474, 478 Jorgenson, D., 273 Jovanovic, B., 250 Juravich, T., 522 Jurek, J. W., 599 Jurgenson, N., 287 Just, S. N., 294 Kahneman, D., 2, 197, 199–​200, 202, 205, 207 Kain, S., ix Kaiser, A., ix, 58, 313 Kakwani, N., 772, 778 Kale, S., 107 Kalleberg, A. L., 486, 528–​30 Kalnins, A., 336 Kanbur, R., 36, 770, 772, 785–​6 Kane, E. J., 545 Kannan, P. K., 277 Kannothra, C. G., xxxviii, 407, 418 Kant, I., 160 Kaplan, H. B., 859 Kapoor, R., 828 Karch, H., 208 Karecha, J., 843 Kash, D. E., 255 Katz, B., 291, 292 Katz, C., 476, 477 Katz, L. F., 149, 273, 794, 843 Katznelson, I., 541 Kauffman, S. A., 251–​3, 263 Kawamura, Y., 297 Kay, J., 649, 657–​9 Kearns, J., 263 Kebede, A. S., 669 Keeble, D., 183 Keeley, B., 490

878   author Index Keen, S., 549 Keith, A., 570 Keller, M. R., 151 Kelly, P. F., 393, 472, 475 Kemeny, T., 152–​3, 358 Kemp, D., 726 Kemp, J., 653 Kennedy, J. F., 14 Kenney, M., 407, 409, 411, 415, 501 Kerr, E. T., 347 Kerr, W. R., xxxviii, 325–​6, 331, 333, 335, 339, 348, 353–​6, 359–​60 Kessides, C., 779 Ketels, C., 330, 338–​9 Ketokivi, M., 332 Keynes, J. M., 197, 201, 203, 545, 548–​9, 585, 658, 690, 712 Khaire, M., 307 Khaled, M., 765 Khan, J., 713 Khan, M., 596 Kiatpongsan, S., 65 Kierzkowski, H., 399 Kim, L., 383 Kim, W. B., 92 Kim, Y., 488 Kindleberger, C. P., 566, 659 King, D., 763 King, R., 120 Kinnel, R., 597 Kirchain, R., 332 Kirilenko A., 579 Kirshen, P., 669 Kislev, Y., 263 Kitano, H., 848 Kitchin, R., 581, 585 Klagge, B., 558 Klasen, S., 777–​8, 785–​6 Klein, G., 198 Klein, P., 830–​1 Klenow, P. J., 263, 272, 352 Klepper, S., 215–​16, 220, 347, 828 Kletzer, L. G., 278 Klier, T. H., 374 Kline, P., 153 Klooster, D. J., 683, 687 Kneale, J., 616–​17

Knight, E., 684, 686 Knight, E. R. W., 604 Knight, Eric, 634 Knight, F. H., 25, 197, 201 Knight, J., 772 Knorr Cetina, K., 291, 296, 298, 684 Knorringa, P., 455–​6 Knox, D., 209 Knox-​Hayes, J., xxxviii, 208, 543, 665, 683–​7, 690, 692, 699 Kochhar, R., 525 Kogler, D. F., 213, 220, 348 Kohli, A., 101 Kok, I., ix, 734, 737 Kokko, A., 372 Kolk, A., 734 Kolko, J., 271, 274, 799 Kollmeyer, C., 794 Kominers, S., 339, 348, 354–​6, 359 König, J., 299 Koo, J., 325 Kooiman, J., 830 Koopman, R., 399 Kortum, S., 601 Korzeniewicz, M., 186, 375, 396 Kössler, R., 839 Kossoy, A., 685 Kountouris, N., 493 Kozul-​Wright, Z., 306 Kraay, A., 105 Krackhardt, D., 188 Krausmann, E., 671 Krautkraemer, J., 757 Kremer, M., 351 Krippner, G. R., 611, 647–​8, 659 Krueger, A. B., 801 Krueger, J. I., 200 Krugman, P., 11, 54–​5, 70, 146, 159, 166, 172, 253, 324, 545, 716–​17, 827, 848 Kruten, T., 49–​50, 52 Kumar, K., 411 Kumar, L., 669 Kumar, Nirmalya, 826 Kumar, Nitish, 109 Kumhof, M., 552 Kuznets, S., 35, 722, 771, 794 Kvande, E., 488

author Index   879 Labban, M., 646–​8 Laborie, L., 632 Labrianidis, L., 129 Lacity, M., 418, 419 Laeven, L., 539 Laffont, J. J., 148 Lafond, F., 763 Lafontaine, F., 336 Lagendijk, A., 181 Lagendijk, V., 632 Lai, K. P. Y., xxxix, 611, 615, 617, 620, 622–​3 Laibson, D., 202 Laitin, D., 97 Lakhani, K. R., 295 Lal, P., 671–​2 Lal, R., 271 Lall, S., 372 Lam, T. C., 92 Lambooy, J. G., 213, 221 Lampert, B., 722 Landers, J., 800 Landry, C., 501 Lane, C., 35, 396 Lang, T., 848 Langdale, J. V., 306 Lange, M., 99 Langley, P., 611, 613–​14, 616–​18, 620–​1 Langner, B., 292 Lansing, D. M., 687 Lardy, N., 81, 82 Larner, W., 612, 615–​16 LaRochelle-​Côté, S., 491 Larsen, M. M., xxxix, 407, 411–​12, 420 Lash, S., 551 Laube, W., 783–​4 Laughlin, G., 579 Laursen, K., 293–​4 Lave, J., 522–​3 Lawrence, F., 784 Lawson, T., 159, 230 Lawson, V., 466, 477 Lazear, E. P., 360, 603 Lazega, E., 188 Lazzeretti, L., 306 Lazzeretti, R., 306, 309 Le Billon, P., 719, 721, 724 Leamer, E. E., 280, 576

Lebel, L., 857 Leber, C., 579 Leborgne, D., 115 Ledebur, L. C., 811 Lee, A. V., 840 Lee, B., 648 Lee, C., 41, 83, 164, 166, 434 Lee, D. J., 522 Lee, K., 828–​9 Lee, R., 277, 391, 474, 546–​7, 594, 647 Lee, S.-​W., 310 Lee, Y. S., 434 Lefebvre, H., 161, 690 Lehdonvirta, V., 580 Leiblein, M. J., 597 Leichenko, R., xxxix, 75, 665, 667, 669–​73, 675 Leigh, N. G., 812 Leijonhufvud, A., 548–​9 Leonhardt, D., 49–​50, 52 Leontief, W., 163 Lepawsky, J., 392 Leppälä, S., 220 Lerner, J., 601 Leslie, D. A., 383 Lessard, D., 334, 338 Lessig, L., 585 Leuthold, R. M., 656 Levi-​Faur, D., 638 Levin, S., 251 Levina, N., 411 Levine, R., 99, 101 Levinthal, D. A., 252, 335–​6 Levitt, R. E., 597, 605 Lévy, D., 165, 612, 648 Levy, David L., 686, 734 Levy, F., 150, 530 Levy, S., 584 Lewin, A., 408–​10, 412 Lewis, A., 92 Lewis, H, 486 Lewis, M., 606 Leyshon, A., 11, 168, 540–​3, 547, 550–​1, 559, 581, 613–​14, 616–​17, 619, 622, 650 Li, F., 724 Li, P. F., 182, 187 Li, Q., 83 Li, S., 86

880   author Index Li, W., 168, 620 Li, X., 80, 85–​6, 88 Li, Y., 41, 71, 88, 92 Liang, Z., 82 Libecap, G. D., 830–​1 Lieber, E., 270 Lieberman, M., 336, 338 Liecke, M., 313 Lievens, A., 287, 294 Lilley, S., 617 Lillie, N., 452 Lim, C., 829 Lim, K. F., 82 Lin, C.-​Y., 756 Lin, J., 362 Lin, J. Y., 83 Lin, S., 31, 79 Lin, Y., 91 Lincoln, W., 353 Linnane, C., 740 Linneman, P. D., 509 Lipietz, A., 115, 120, 133, 471 Lipton, M., 104 LiPuma, E., 648 Lissoni, F., 188, 218 Listokin, D., 546 Littler, J., 450 Liu, J., 79, 86 Livanos, I., 488 Liverman, D., 170, 665, 667–​8, 671, 675 Lloyd, S., 448, 450 Lo, A., 597 Lo, V., 577 Lobo, J., 252 Lockwood, J. W., 579 Loconto, A., 453, 454 Lohmann, L., 684–​7 Long, J., 419 Longino, H., 160, 171 Lopez, J. H., 474–​5, 796 Lorenzen, M., xxxix, 187, 305–​9, 312–​13, 318, 414–​15 Lösch, A., 166, 473, 500 Lotay, J. S., 686, 695 Loukaitou-​Sideris, A., 798 Love, M. S., 201 Lovell, H., 687

Lovering, J., 467–​9, 471 Low, N., 171 Low, P., 382, 399 Lowe, M., 433, 434 Lowe, N., xl, 151, 153, 155, 519, 522, 525–​7 Lowi, T., 817 Lu, M., 91 Lubart, T., 504 Lucas, R., 147, 207, 263, 500, 509 Luers, A., 667 Luetchford, P., 454 Lundstrom, S., 778, 786 Lund-​Thomsen, P., 453 Lundvall, B. Å., 151, 183–​4, 189, 234 Luo, X., 277, 399 Luo, Y., 420 Luthar, S. S., 839, 845 Lüthje, C., 297 Lyautey, M., 14 Lyon, T. P., 449 Ma, L. J., 79 Mabro, R., 655, 657, 660 Macaulay, S., 855 McCabe, M., 280 McCann, P., xl, 147, 182, 216, 223, 309, 366–​7, 369, 374–​6, 413 McCartney, M., 104 McCoy, C., 209 McCubbin, S., 667 McDermott, C. J., 654 McDowell, J., 196 McDowell, L., 168–​9, 466, 472, 475–​8, 486–​7 McElroy, C. A., xli, 715 McFall, L., 613, 622 McGahan, A. M., xli, 826, 829–​30 McGill, S., ix, xli, 209, 645 McGillivray, M., 145 McGinnis, J., 690 McGranahan, D., 505 McGranahan, G., 669, 774–​5, 781 McGrath, C., 188 McGrath-​Champ, S., 475 McIlwaine, C., 97 McKinley, T., 785–​6 McKinney, K., 279 McNerney, J., 252

author Index   881 McNichol, E., 72 McSweeney, K., 670 MacDonald-​Korth, D., 570–​1 Mace, G., 710 Macher, J., 275 Machlup, F., 501 MacIntosh, J., 597 MacKenzie, D., 167, 579, 684, 686–​7 MacKinnon, D., 214, 221, 223, 237, 389–​91, 394, 453, 733, 735, 810, 819, 855 Maddison, A., 21, 25, 143, 147 Maggioni, M. A., 215 Magkilat, B., 408 Mahalanobis, P. C., 104 Mahoney, J., 99 Mahr, D., 287, 294 Maillat, D., 181 Maintz, J., 290 Majkgård, A., 417 Maldonado, J., 672 Malecki, E. J., 576, 812, 827 Malerba, F., 827 Malhotra, A., 492 Malmberg, A., 181–​2, 186, 221, 230, 309, 328, 561 Maloney, M., 669 Malthus, T. R., 707, 712–​13 Mamdani, M., 99 Mankiw, N. G., 499, 503 Mann, G., 164, 169, 648 Manning, S., xl, 407–​18, 420–​1 Manso, G., 595 Mao Zedong, 79, 82 Marano, C., 802 Marcos, J., 855 Maré, D. C., 500, 509 Mariani, M., 334, 336, 339 Marino, E., 675 Markard, J., 732, 742 Markides, C. C., 829 Markoff, J., 417 Markusen, A., 169, 325, 373, 576, 719 Marlet, G., 505 Marrocu, E., 505 Marron, D., 613–​14 Marshal, J. C., 208 Marshall, A., 146, 215, 220, 245, 262–​3, 309, 324, 330–​2, 335, 347, 349, 499–​500, 688

Marshall, J., 546 Martin, F., 231 Martin, R., xl, 9, 12, 64, 115–​16, 165, 169, 172, 213–​14, 216, 220–​1, 223, 232, 237–​9, 260, 551, 558, 611–​12, 614, 616, 618, 620–​1, 665, 733, 741, 810, 812–​13, 839–​40, 843, 848–​50, 852–​3, 856, 859 Martin, X., 416 Martinich, J., 672 Maru, Y., 676 Marvin, S., 603 Marx, K., 144, 162–​5, 169, 172, 501, 504, 541, 688, 690 Mashelkar, R. A., 832 Maskell, P., 181–​2, 186–​7, 221, 230, 299, 309, 328, 412 Maslow, A. H., 510 Mason, C. F., 686–​7 Mason, C. M., 543 Mason, R., 409 Massey, D., 2, 115, 133, 466–​8, 470, 473, 475, 477, 501, 569 Massini, S., 407 Masten, A. S., 839, 848 Masters, M. W., 653 Mateos-​Garcia, J., 292, 296 Mathur, V. K., 503 Maugeri, L., 653, 752 Maurer, B., 603, 618 May, R., 207 Mayer, C., 712 Mayer, J., 652 Mayes, R., 451 Maystre, N., 654–​5 Mazzucato, M., 148, 151, 230 Meadows, D. H., 706–​7, 751 Mechler, R., 667–​8 Meegan, R., 115, 467–​8 Meisenzahl, R. R., 150 Mellander, C., xli, 64, 73, 499, 505, 510–​11 Memedovic, O., 338 Mendel, G., 247–​8 Meng, X., 91 Menzel, M. P., 216 Merton, R. C., 647 Merton, R. K., 190 Mestieri, M., 278

882   author Index Metcalfe, J. S., 167, 852 Metters, R., 409–​10 Meyer-​Stammer, J., 784 Michaelowa, A., 685 Michaelowa, K., 685 Michelacci, C., 353 Micheletti, M., 449 Michielsen, T. O., 683 Mikkelson, G., 55 Milanovic, B., 49–​50, 52, 168, 770 Milberg, W., 396, 399 Milkman, R., 528–​9 Miller, M. H., 646, 648, 656 Millner, A., 758 Mills, B. F., 272, 333 Mincer, J., 503, 509 Minsky, H. P., 13, 540, 547–​51, 659 Miozzo, M., 407 Mirowski, P., 9, 159, 170 Mirus, R., 309 Mische, A., 187 Mishel, L., 795 Mishkin, F. S., 558 Mitchell, D., 167, 169 Mitchell, K., 476 Miteva, D. A., 761 Mithas, S., 278, 409, 413, 420 Mitra, S., 803 Mittal family, 46 Mobius, M., 275 Modi, N., 102 Moenaert, R. K., 336 Mohan, G., 722 Mokyr, J., 150 Mol, M., 412 Molina, F., 800 Molloy, R. E., 54 Molotch, H. S., 541 Molyneux, M., 476 Monaghan, A., 495 Monk, A. H. B., xlii, 543, 572, 577, 591, 596–​8, 601, 603, 606, 623, 634, 649–​50 Montgomerie, J., 619 Montgomery, A. W., 449 Moore, D., 586 Morawetz, N. J., 307 Moreno, E. L., 774, 775

Moreno, K., 40 Moretti, E., 153, 334, 795 Morgan, K., 116, 230, 309, 383 Morishima, M., 160, 164–​5 Morrison, A., 222, 563 Morrison, C., 671 Morrison, P., 64, 218 Moser, S., 667 Moskowitz, T. J., 596 Moss, M. L., 271, 280 Mossig, I., 313 Moudud, J. K., 166 Moulaert, F., 115, 184 Mowery, D., 275 Moyo, D., 829 Mudambi, R., 186–​7, 307, 312, 318, 368, 375–​6, 414–​15 Muellerleile, C. M., 646, 659 Mukim, M., 352 Müller, Felix C., 291, 300 Mulligan, G. F., 167 Muneepeerakul, R., 252 Muniesa, F., 298 Murcia, C., 254 Murmann, J. P., 221, 827 Murnane, R., 150 Murphy, D., 605 Murphy, J. T., 10, 181, 185, 392, 396, 542 Murphy, K. M., 599, 794 Murray, F., 330 Murray, J. Y., 412 Musson, D., ix Mutabazi, K., 672 Mutch, A., 188 Mutebi, A. M., 437 Mutersbaugh, T., 453 Muth, J. F., 251 Muzio, D., 603 Myers, N., 709 Mykhnenko, V., 813 Nabeshima, K., 25–​7 Nachum, L., 183–​4 Nadvi, K., 450, 452–​3, 455–​6 Nagar, R., 477 Nageswaran, V. A., 654 Nagurney, A., 167

author Index   883 Naidu, C., 102 Nakamoto, S., 584 Nardinelli, C., 503 Narula, R., 367, 372 Nash, J. C., 719 Nathan, D., 396 Naughton, B., 80 Nayyar, D., 19, 22–​5, 29, 31, 34–​5 Neary, J. P., 373–​4, 719 Nebiolo, M., 75 Neff, G., 291, 298 Neffke, F., 216, 220, 252, 260–​1, 325–​6 Nehru, J., 104, 107 Neil, D., 597 Neilson, J., 383, 396, 398, 452–​4 Nel, E., 784 Nelson, R. R., 151, 213–​14, 221, 234, 250, 262–​3, 827 Neumann, J., 669 Neumark, D., 798, 799 Newberry, D. M. G., 656 Newman, K. S., 528, 546 Nguyen, H. T. H., 437 Nicholls, R. J., 669 Nicolaus, M., 501 Niederman, F., 411 Nieto, M., 412 Nightingale, P., 295 Niosi, J., 420 Noah, T., 70 Nolan, P., 571 Noland, M., 510 Nordås, H. K., 428 Nordhaus, W. D., 499, 757–​8, 761–​2 Norris, P., 510 North, D. C., 151, 155, 208, 232, 797 Northrop, L., 579 Northrup, D. O., 524 Northrup, H. R., 524 Norton, M. I., 65 Novy-​Marx, R., 606 Nowatzki, N. R., 41 Nudds, M., 197 O’Brien, K., 667, 670, 676 O’Brien, R., 168, 541–​2, 570, 576 O’Connor, J., 170, 648

O’Connor, K., 577, 604 O’Dougherty Wright, M., 848 O’Hern, M., 286 O’Neill, P., xlii, 603–​4, 628, 632–​3, 639, 647 O’Sullivan, D., 648 Obama, B., 55, 63, 72 Obeng-​Odoom, F., 774 Ocampo, J. A., 722–​3 Ofer, G., 83 Offe, C., 633 Oh, J., 309, 311 Ohlin, B., 500 Okishio, N., 165 Olds, K., 170, 391 Oliver, M. L., 802 Ong, P., 798 Öniş, Z., 75 Ono, Y., 272, 336 Opal, C., 775 Ormerod, P., 840, 842 Orshansky, M., 64 Ortega-​Argiles, R., 223 Ostale, E., 437–​8 Östensson, O., 651, 653–​4, 656 Oster, S. M., 275 Osterman, P., 529–​30, 794 Ostrom, E., 760, 830–​1 Ostry, J. D., 31, 775 Ottaviano, G. I. P., 166, 510 Ottoviano, G. I. P., 717 Ouma, S., 392, 396, 454, 457, 683 Overby, E., 276–​7 Overholt, W., 726 Overman, H., 152, 339, 347, 354 Owen, J. R., 726 Owen-​Smith, J., 182–​3, 188 Ozcan, K., 286–​7, 289, 294 Pacella, A., 230 Pachucki, M. A., 187 Paci, R., 505 Page, S. E., 510 Pahl-​Wostle, C., 853 Pai, M., 416 Pakes, A., 250 Palan, R., 561, 567 Palavicini-​Corona, E. I., 783

884   author Index Palladino, G., 523, 524 Palma, J. G., 495 Palmer, D., 586 Palpacuer, F., 449–​50, 452 Panagariya, A., 101 Pandiella, G., 672 Pandit, K., 509 Pani, E., 546 Panitz, R., 188 Paolisso, M., 672 Papadopoulos, D., 617 Papandrea, F., 311 Pardoe, J., 673 Pariser, E., 585 Parkinson, M., 775 Parr, J., 166, 374 Parrilli, M. D., 383 Parsons, J. E., 649, 655, 657 Pasqualetti, M. J., 742 Pasquali, P., 92 Passmore, J., 170 Pastor, M., 793, 796, 801, 814 Pastor, M. Jr, 793, 796, 801, 814 Patchell, J., 223, 732 Pathak, P., 620 Patibandla, M., 409–​10, 414 Patrinos, H. A., 801 Pattberg, P., 673–​4 Patterson, J. G., 604 Paunescu, M., 857 Pavan, R., 511 Pavlova, A., 654 Payne, B. C., 596 Peck, J., xlii, 11, 35, 64, 169–​70, 172, 188, 235–​7, 383, 391, 465–​7, 470–​6, 478, 487–​91, 528, 568, 699, 812–​13, 815–​16, 822 Pede, V. O., 798 Peeters, C., 408–​9, 412 Pelling, M., 672, 840 Pendall, R., 846 Penning-​Rowsell, E., 673 Pennings, J. M. E., 656 Penrose, E., 370 Peoples, S., 667 Peri, G., 510 Perkins, D. H., 92 Pernia, E. M., 772, 778

Perramond, E. P., 684 Persson, H., 372 Persson, T., 99 Peters, A., 799 Peters, G., 830 Petersen, B., 410, 414 Petersen, S., 723 Petrovic, M., 427 Pettit, P., 196 Pfeiffer, A., xlii, 749 Phalippou, L., 598 Phelps, E. S., 149–​50 Phelps, N. A., 186 Phillipson, N., 144 Philo, C., 41 Pickering, C., 671 Pickett, K. E., 41–​2, 55 Pickles, J., 389, 396 Pierre, J., 830 Pigou, A. C., 501, 751 Pike, A., 235, 237, 239, 260, 391, 603, 611–​12, 649, 656, 659, 782–​3, 786, 840 Piketty, T., 5–​6, 9, 31, 41, 54, 104, 149, 168, 495, 511, 772 Piller, F., 288 Pincetl, S., 666 Pinch, S., 182 Pinch, T., 297 Pindyck, R. S., 654, 656, 758 Piore, M. J., 246, 456, 499, 501, 522, 529, 798, 831 Pires, R., 456 Pisano, G. P., 332–​3, 336 Pischke, J.-​S., 521 Piscitello, L., 370 Pittel, K., 759 Plambeck, E. L., 686 Plantinga, A. J., 686–​7 Plehwe, D., 170 Pliske, R., 198 Plummer, P., 162, 166 Polanyi, K., 6, 7, 231, 465 Polanyi, M., 280, 576 Pollard, J., 168, 172, 542, 547–​8, 603, 611–​12, 622, 649, 656, 659, 673 Pollert, A., 493 Ponte, S., 384, 396, 452–​3 Ponzetto, G. A. M., 271, 274

author Index   885 Porter, L., 859 Porter, M. B., 246 Porter, M. E., 181, 216, 324–​6, 330–​6, 339, 349, 399, 499, 508, 797–​9, 827–​8 Posner, E., 601 Posthuma, A., 396 Potiowsky, T., 328 Potter, A., 215 Potter, J., 783 Potter, R., 717, 719 Potts, J., 286, 305 Pouder, R., 216 Powell, C., 187 Powell, W. W., 182–​3, 188, 217, 368 Power, D., 169, 295, 306, 309, 580 Prahalad, C. K., 286, 298, 830, 832 Pratt, A. C., 306–​9 Pratt, G., 466, 472, 476–​7 Prebish, R., 722 Pred, A., 8, 72, 575 Preda, A., 684 Preston, B., 668–​70 Price, M., 72 Prieger, J. E., 272 Pritchard, B., 396, 452–​3, 454 Pritchard, L., 845 Prno, J., 724 Probert, J., 396 Protsiv, S., 330 Pryce, G., 666 Pryke, M., 623, 647–​8, 673 Przeworski, A., 97 Psacharopoulos, G., 801 Puga, D., 153, 274, 332, 337, 347–​9, 353, 360, 376 Pulkkinen, R., 221 Puranam, P., 411, 826 Pykett, J., 204 Quadrini, V., 798 Quah, D., 759 Quatraro, F., 220 Quealy, K., 49–​50, 52 Quesnay, F., 162 Quheng, D., 774 Quiggin, J., 31 Quigley, J. M., 775

Raco, M., 633 Radetzki, M., 649, 654–​8, 722 Radjou, N., 826 Rafferty, M., 648 Rahm, D., 736 Rainnie, A., 392 Ramamurti, R., 829 Ramaswamy, V., 286–​7, 289, 294, 298 Ramos, R. A., 770, 776–​7, 780, 786 Ramsey, F. P., 756–​7 Rani, P., 36 Ranieri, R., 770, 776–​7, 780, 786 Rank, M. R., 69 Rankin, W. J., 628, 630–​1 Rantisi, N. M., 232 Rao, C., 416 Rauch, J. E., 503 Rauh, J. D., 596, 606 Rauniyar, G., 785–​6 Raven, R., 732 Ravid, G., 312 Ray, D., 145 Raynolds, L. T., 452 Reagan, R., 148, 467, 524, 811 Reardon, T., 434, 436, 438 Redding, S., 352 Reddy, P., 409 Rees, J. A., 170 Rees, M., 756 Reid, J., 840 Reimer, J., 522–​3 Reimer, S., 383 Reinecke, J., 420 Rekers, J. V., 235, 239 Ren, Y., 296 Rentfrow, P. J., 857 Resseger, M. G., 506 Restino, C., 725 Rezaie, R., 830–​1 Rhiney, K., 667, 669 Rhodes, R. A. W., 830 Ribot, J., 668, 675 Ricardo, D., 147, 163–​4 Rid, T., 586 Rifkin, J., 492 Rigby, D. L., 165, 214, 216, 220, 568 Rightor, N., 736–​8

886   author Index Riisgaard, L., 451 Riles, A., 299 Rinallo, D., 187 Rindfleisch, A., 286 Ritsatos, T., 727 Ritzer, G., 287 Rivkin, J. W., 332 Roback, J., 509 Robb, D., 855 Robert-​Nicoud, F., 347 Roberts, B. H., 775 Roberts, J., 291, 296 Roberts, K. H., 296 Roberts, S., 11 Robertson, M. M., 170 Robinson, D. T., 598 Robinson, J., 99 Rockström, J., 750, 755, 761 Rodgers, G., 487 Rodin, J., 840, 855 Rodríguez, A., 412 Rodríguez-​Pose, A., xlii, 230, 770, 781, 783, 785–​6 Rodrik, D., 29, 33, 98–​9, 150, 161, 208, 230, 233, 827 Rogerson, C. M., 783–​4 Rogerson, J. M., 783 Romer, P., 147, 152, 220, 263, 501, 759, 847 Rongerude, J., 801 Rook, D., 209 Rosales, J., 721 Rosansky, V. I., 649 Rose, A. K., 260, 653 Rose, N., 617 Rosen, S., 509 Rosenblat, T., 275 Rosenthal, S. S., 245, 334, 339, 347–​8, 354, 358, 360 Ross, A., 477 Ross, M., 720 Ross, N., 41 Ross, R. J. S., 83 Rosser, A., 720 Rosser, J. B. Jr, 162 Rossi-​Hansberg, E., 399 Rostow, W. W., 630, 633, 640 Rotemberg, J. J., 654

Round, J., 488 Rouwenhurst, K. G., 650 Roy, A., 169 Rozelle, S., 772 Rubenstein, J. M., 374 Rubery, J., 476 Rubin, V., 802 Ruiz, N., 353 Runciman, D., 41 Rutherford, T., 466, 474, 480 Ruwanpura, K., 455 Rycroft, R., 255 Sabel, C. F., 246, 499, 501, 585, 798, 831 Sachs, J. D., 97–​100, 109, 720 Sadler, D., 448, 450 Sælen, H., 758 Saez, E., 54, 70–​1, 149, 495 Safford, S., 529 Sagoe-​Addy, K., 671 Sai, D., 774 St John, C., 216 Saiz, A., 500, 503, 510 Sako, M., 407–​9, 413, 417 Sala-​i-​Martin, X., 770 Salais, R., 155 Salamon, L. M., 800 Salkin, P. E., 800 Salt, D., 839 Salzinger, L., 527, 531–​2 Samers, M., 168, 476 Sampat, B. N., 827 Sampsa, S., 601 Sampson, R. J., 512 Samuelson, P. A., 499, 579, 633 Samuelson, W. A., 203 Sanders, D. R., 653, 655–​6 Sandor, R., 654–​5, 657 Sandrea, I., 739 Sandve, A., 451 Sankhe, S., 774–​5 Sarvary, M., 271 Sassen, S., 41, 565, 569, 614, 795 Sato, Y., 596 Sauer, C. O., 718 Sautter, B., 216 Savage, L., 466, 474

author Index   887 Saxena, S., 843 Saxenian, A., 235, 259, 312, 325, 330, 334, 337, 348–​9, 351, 415, 499, 501, 576, 798 Sayer, A., 41, 170, 180, 190, 469 Scaramozzino, P., 655 Schafran, A., 7 Scharfstein, D., 599 Scheibelhut, T., 597 Schindler, S., 392 Schipper, F., 631, 640 Schmidt, K. M., 208 Schmitz, H., 383, 396, 415 Schneider, M. R., 857 Schoenberger, E., 396 Schot, J., 631, 640 Schrank, A., 151, 154, 456 Schröder, M., 236, 239 Schroeder, S. K., 550 Schumacher, E. F., 830 Schumpeter, J. A., 150, 219, 252, 256, 261, 300, 722, 841 Schwarz, D., 576 Scitovsky, T., 149 Scott, A. J., 116, 164, 167, 169, 182, 306, 309, 313, 391, 396, 466, 470–​1, 499, 501, 506, 568, 580, 684, 718 Scott, D., 671 Scott, M., 840 Scribner, S., 522 Seamans, R., 279 Sedita, S. R., 216 Seidel, V. P., 292 Seigworth, G. J., 618 Sellers, P., 47 Selwyn, B., 392–​3 Selwyn, P., 120 Sembenelli, A., 372 Sen, A., 106, 148–​9, 800 Sensoy, B. A., 598 Serafinelli, M., 351 Servén, L., 795 Seto, K., 667 Setterfield, M., 853 Seville, E., 840 Seyfang, G., 594 Shabani, M., 551 Shackleton, R., 435

Shafir, E., 200, 203 Shah, A., 109 Shah, S. K., 287, 295, 297, 300 Shale, J., 45 Shane, S., 829 Shannon, R., 439 Shapiro, J. M., 509, 511 Shapiro, T. M., 802 Sharkey, P., 512 Sharma, D., 417 Sharma, P., 100 Sharma, R., xliii, 19, 33, 408, 591, 601, 604, 634, 671 Sharma, S., 97 Sharman, J. C., 567 Sharpe, W. F., 200, 206, 597 Shaver, J. M., 334 Shaw, E. S., 551 Shaxson, N., 567 Sheffer, D. A., 597 Sheffi, Y., 840 Shelton, T., 577, 580 Shen, B., 687 Sheppard, E., xliii, 160–​7, 169–​72, 470–​1, 850 Shih, W. C., 332–​3 Shiller, R., 40, 657–​8 Shreck, A., 453 Shu, Q., 86 Shultz, G. P., 713 Sidaway, J. D., 547, 618 Sider, A., 740 Siemiatycki, M., 637 Siggelkow, N., 252 Silva, J., 667, 670, 672 Silva, O., 353 Silvennoinen, A., 726 Silver, M. L., 522–​4 Simanis, E., 829 Simmie, J., 223, 840, 843, 852–​3 Simon, C. J., 503 Simon, H. A., 2, 197, 199, 204–​5, 207, 751 Simon, J. L., 707, 713, 752 Simons, P., 209 Simonton, D. K., 509 Sin, H. L., 451 Sinai, T., 276, 277 Sinclair, D., 605

888   author Index Singer, H. W., 722 Singer, P. A., 831 Singh, M., 101 Sinha, A., 102 Sinha, K., 409 Sinn, H., 649 Sjoberg, O., 83 Skelcher, C., 830 Skoufias, E., 666, 672 Slack, E., 858 Slater, D., 658 Slater, G., 486 Smart, S., 740 Smith, Adam, 143–​4, 162, 164, 167, 499, 509, 629–​30, 688–​9 Smith, Adrian, 383, 392 Smith, B. E., 466, 476–​7 Smith, F., 488 Smith, J. L., 656, 658, 803 Smith, N., 116, 161, 168, 393, 469, 686, 716 Smith, S., 41, 203, 453 Smulders, S., 749, 757 Snyder, C. M., 280 Soete, L., 261 Soja, E., 161 Sokol, M., 550 Solari, S., 620 Solecki, W., 670 Solinger, D., 83, 90 Solow, R., 102, 147, 757, 759 Sölvell, Ö., 330 Somanathan, T. V., 654 Sommeiller, C., 72 Son, H. H., 777–​8 Sonderegger, P., 413 Sorenson, O., 215, 218, 325, 334, 337–​8, 601 Soskice, D., 188, 222, 234, 699, 857 Sotarauta, M., 221 Soto, M., 503 Soyez, D., 724 Spatareanu, M., 372 Spence, M., 97, 598 Spencer, G., 827 Spencer, H., 246–​7, 255, 262 Sraffa, P., 164, 167 Srikanth, K., 411 Stabell, C., 183

Staber, U., 216 Stafford, E., 599 Stam, E., 215 Stambaugh, R. F., 201 Standing, G., 495, 528 Stangler, D., 254 Star, S. L., 159 Stark, D., 188, 288, 291, 298 Starks, L. T., 598 Starosta, G., 393, 398 Starrett, D., 166 Stathakis, G., 120 Steedman, I., 163, 167 Steiger, T. L., 522–​3 Stein, J. G., xliii, 826, 829–​30 Steinmueller, W. E., 292, 296 Stephan, P., 147 Stern, D. I., 757, 794 Stern, J., 637 Stern, N., 6, 750, 757–​8; see also Stern Review Stern, P., 671 Stern, S., 279, 337–​9 Sternberg, R. J., 305, 504 Sternberg, T., 723, 725 Stewart, K., 57 Stiglitz, J. E., 29, 31, 55, 198, 656, 752, 756–​7, 759, 794–​6, 800 Stiles, T., 39 Stiroh, K. J., 273 Stokey, N. L., 759, 765 Stolarick, K., 503, 505 Stone, K. V. W., 486, 491, 812 Storey, K., 726 Storper, M., xliii, 116, 143, 152–​5, 168, 182, 197, 202, 214, 230, 232, 280, 309, 313, 325, 337, 370, 385–​6, 391, 399, 411, 466, 468, 471, 499, 501, 552, 576, 604, 674, 734, 793, 795, 804 Stotesbury, N., 58 Strambach, S., 221 Strange, S., 567, 569, 649 Strange, W. C., 245, 334, 339, 347–​9, 354, 358, 360 Strauss, K., xliv, 197, 208, 472, 480, 485–​7, 491, 616 Streeck, W., 237–​8, 611 Stripple, J., 683 Strumsky, D., 252, 255

author Index   889 Stuart, T., 215 Stupnytska, A., 20 Sturgeon, N., 820 Sturgeon, T. J., 383–​4 Subianto, P., 34 Subramanian, A., 103 Sugden, F., 673 Suharto, President, 34 Suire, R., 216, 217 Sukarno, President, 34 Sundararajan, A., 492 Sunley, P., 9, 115–​16, 165, 213–​14, 216, 220, 223, 232, 237–​9, 260, 393, 665, 733, 741, 840, 848–​9, 852, 859 Sunstein, C., 198, 205, 207 Sutton, J., 318 Swann, P., 330 Swanstrom, T., 801 Sweeney, B., 475 Swyngedouw, E., 115, 383, 648, 668, 683 Sydow, J., 116 Syverson, C., 263, 270 Taeube, F., 413 Tambe, P., 279 Tamoschus, D., 300 Tan, C. H., 620, 622–​3 Tang, A. M., 81–​2 Tang, K., 650, 654, 659 Tanner, A. N., 220, 223 Tanner, T., 672 Tansley, A., 246–​7, 256 Tapscott, D., 294–​5, 297 Tarr, K., 802 Tate, C., 669 Täube, F. A., 312, 318 Taylor, G., 3 Taylor, M. J., 392 Taylor, M. S., 765 Taylor, P. J., 551, 561 Taylor, R. S., 370 Taylor, S., 669 Tecu, I., 348 Teitz, M. B., 799 Temin, P., 530 Ter Wal, A. L. J., 188, 218 Terazono, E., 658

Teubal, M., 148 Teytelboym, A., xliv, 749 Thaksin Shinawatra, Colonel, 34 Thaler, R., 198, 205, 207 Tharoor, S., 829 Thatcher, C., 670 Thatcher, M., 148, 468, 470, 811 Thelen, K., 189, 237, 238, 239, 611 Theodore, N., 64, 169, 188, 235, 237, 472–​3, 480, 488–​91, 528, 699, 813, 815–​16, 822 Thetford, T., 803 Thiel, J., 313 Thisse, J.-​F., 146, 166, 717, 775 Thomas, A., 669, 673 Thomas, B., 54, 151 Thompson, P., 504 Thompson, W. R., 500, 503–​4 Thorat, S., 774 Thornton, J. R., 771 Thorp, S., 726 Thorsteinsdóttir, H., 830 Thrift, N., 168, 170, 172, 230, 290, 295, 298, 312, 383, 391, 471, 542, 551, 559, 614, 620, 650 Throsby, D., 305 Tian, F., 88 Tickell, A., 169, 172, 383, 471, 568 Tiebout, C. M., 797 Tien, H. Y., 81 Tijmstra, S., 783, 785 Timmons, J., 601 Timms, H., 763 Tindall, T., 637, 640 Tirole, J., 148 Tokatli, N., 396 Tol, R. S. J., 758 Tomaney, J., 230 Tomei, J., 675 Tomlinson, J. B., 307 Tompkins, E. L., 840 Tompkins, J. A., 854 Topol, E. J., 829 Topolova, P., 104 Toro, F., 822 Torrance, M., 604, 634–​5 Torre, A., 218 Torres, R., 491, 493 Torres, S., 99

890   author Index Torrisi, G., 843 Toulson, D., 578 Townsend, A. M., 271, 576 Towse, R., 305 Tracey, P., 180, 182 Treanor, J., 418 Treguer, D., 750 Treuhaft, S., 802 Trigilia, C., 235 Trimble, C., 826, 828–​9 Tripathi, S., 774 Tripathy, A., 332, 338 Trippl, M., 220 Tripsas, M., 287, 297, 300 Truffer, B., 223 Trump, D., 7, 33 Tsampra, M., xliv, 113 Tschakert, P., 668 Tschang, F., 420 Tseng, M. M., 288 Tsui, K. Y., 80 Tucker, C., 277 Tufekci, Z., 417 Tufts, S., 466, 474 Tuomi, T., 307 Turi, P., 299 Turner, B., 676 Turner, M., 352, 672 Turnovsky, S. J., 656 Turok, I., 313, 774, 775, 781 Tversky, A., 197, 199–​200, 202, 205, 207 Tversky, B., 203 Tyson, R. E., 859 Ullman, E. L., 503, 508 Urahn, S., 57 Urkidi, L., 725 Urry, J., 299, 551 Urwin, R., 606 Uzzi, B., 183, 188 Vaast, E., 411 Vacas–​Soriano, C., 47 Valencia, F., 539 Valentine, G., 169 Valenzuela, A. Jr, 529 Valiante, D., 650, 651

Välikangas, L., 855 Vallee, G., 491 Van Alstyne, M., 275, 492 Van de Ven, A., 261, 409 van den Berge, M., 220, 223 Van den Bulte, C., 336 van der Molen, M., 646, 655, 660 Van der Ploeg, F., 758, 762 van der Woude, A., 563 van der Zwan, N., 618, 623 Van Gaasbeck, K., 274 Van Jaarsveld, D. D., 813 Van Laak, D., 631 Van Maanen, J., 522 Van Reenen, J., 250, 263 van Woerkens, C., 505 van Zeebroeck, N., 275 Vancura, P., 669, 671 Vang, J., 313 Varoufakis, Y., 34 Varshney, A., 104 Vaughan-​Williams, N., 620 Vázquez-​Barquero, A., 783 Veblen, T. B., 230, 232, 262 Vedres, B., 188 Veitch, J. M., 542 Venables, A. J., 146, 182, 197, 202, 280, 309, 373, 399, 411, 604, 717, 762, 770, 772 Venkatasubramanian, K., 675 Verma, R., 409, 410 Vernon, R., 332, 368–​70, 377, 501 Viceira, L. M., 597 Vicente, J., 216–​17 Vilhuber, L., 279 Viner, J., 250 Vinodrai, T., 238–​9 Vinterberg, T., 314 Vissel, C., 845 Vitali, S., 562 Vlaar, P., 411 Voelzkow, H., 236, 239 Vogel, H. L., 310–​11 Vogt-​Schilb, A., 761 von Braun, J., 667 von Hippel, E., 182, 271, 287, 289, 293, 295, 332, 338, 419 von Mises, L., 690

author Index   891 von Thünen, J. H., 499–​500 von Trier, L., 314 Vorley, T., 306 Vosko, L. F., 486–​7 Vural, L., 450 Wachsmuth, D., 792 Wade, R., 29 Wadhwa, V., 510 Wagner, G., 756 Waine, B., 616, 618 Wainwright, T., 170, 546, 551, 594 Waite, L., 486 Waldfogel, J., 276–​7 Waldinger, R., 527 Walker, B., 839 Walker, R., 115, 168, 214, 466, 468, 795 Wallace, J., 88 Wallerstein, I., 716 Wallsten, S. J., 245 Walsh, J., 474 Wan, G., 91 Wang, F., 92 Wang, X., 88 Wang, Y., 80 Wanvik, T. I., 454 Warde, A., 469 Warner, A. M., 720 Warner, M., 814 Warren, A., 486 Warren, G., 597 Wasko, J., 307 Waterman, D., 310 Watson, A., 486 Wattal, S., 279 Watts, H. D., 215 Watts, M. J., 668, 720 Webber, M. J., 165, 197, 568 Webber, S., 673, 674 Weber, A., 499, 500 Weber, R. N., 803 Weber, S., 287 Wei, Y. H. D., 41 Weick, K. E., 296 Weil, D., 524, 528, 529 Weinhold, D., 277 Weinstein, B. L., 736

Weintraub, E. R., 207 Weir, M., 801 Weisbenner, S., 577, 596 Weissenbacher, R., 115, 120, 131 Weitzman, M. L., 263, 758 Wellman, B., 272 Wells, M., 527 Wells, O., ix Wenger, E., 523 Wenting, R., 215–​16, 221 Werner, M., 389, 394, 454 Weszkalnys, G., 715, 722 Weterings, A., 220, 223 Weyl, E. G., 601 Whatmore, S., 170 Wheeler, T., 667 Whitacre, B. E., 272 Whitaker, J., 278, 409, 413, 420 White, H., 145 Whiteman, J., 197 Whitford, J., 151, 154 Whiting, S., 80 Whitley, R., 234, 308, 386 Whitson, R., 488 Whittington, K. B., 188 Whoriskey, P., 54 Wick, K., 721 Widodo, J., 34 Wiertz, C., 289–​90, 292 Wiewel, W., 816 Wildman, S. S., 311 Wilhelm, W., 563 Wilkie, C., xliv, 770, 786 Wilkins, B., 40 Wilkinson, R. G., 41–​2, 54–​6 Williams, A. D., 294–​5, 297 Williams, A. T., 603 Williams, C. C., 488 Williams, G., 572 Williams, K., 611 Williams, P., 541 Williams, R., 546 Wills, J., 391, 466, 474, 477, 486 Wilson, D., 20 Wilson, E. O., 709 Wilson, G. A., 688 Wilson, M., 410

892   author Index Wimmer, A., 839 Winders, J., 466, 476–​7 Winkler, A. E., 275 Winkler, D., 396, 399 Winter, S. G., 151, 213–​14, 250, 262–​3 Winters, A., 97 Wiseman, M., 592 Wisner, B., 668 Withagen, C., 758, 762 Witsoe, J., 107 Wojan, T., 505 Wójcik, D., ix, xxi, 1, 8, 168, 238, 338, 542–​3, 551, 557, 560, 562–​4, 567, 569–​72, 577, 594, 603–​5, 660, 699 Wolfe, D. A., 235, 812 Wolpert, J., 197 Wong, K., 529 Wong, K.-​Y., 167 Wong, S., 79 Woo, M. Y-​en, 28 Wood, A., 11 Wood, G., 35 Wood, P. A., 569 Wood, S., 434, 439, 442 Woon, C. Y., 486 Working, H., 656, 657 Worthen, H., 523 Wright, C., 687 Wright, G., 721 Wright, S., 247–​9, 251–​2 Wright, T. P., 251 Wrigley, N., xliv, 11, 427–​8, 432, 434–​6, 440, 455, 603, 733 Wu, J., 339 Wu, T., 277, 586 Wyly, E. K., 41, 546, 614 Xiong, W., 650, 654, 659 Yadav, L., 107–​9 Yamane, A., 668

Yang, D., 81, 491, 772 Yang, Z., 758 Yaqub, O., 295 Yeaple, S. R., 374 Yeung, G., 92 Yeung, H. W.-​c., xliv, 11, 35, 162, 179, 181, 185, 382–​5, 387–​97, 399, 416–​17, 623, 734–​5 Yew, C. P., 88 Yin Yang, 634 Yin, J. S., 802 Young, H. P., 598 Yousuf, H., 55 Yudhuyono, President, 34 Yusuf, S., 25–​7, 97 Zademach, H.-​M., 313 Zaheer, A., 188 Zaheer, S., 412–​13 Zalik, A., 657 Zanfei, A., 370 Zeckhauser, R., 203 Zelizer, V. A., 613, 616, 618 Zentner, A., 277 Zeuli, K., 331 Zhang, J., 82, 235–​6 Zhang, X., 772 Zhao, M., 339 Zhou, K., 86 Zhu, F., 279 Zhuang, J., 774, 777, 779–​81 Zlolniski, C., 527 Zoller, T. D., 262 Zolli, A., 840 Zook, M. A., xlv, 270, 272, 279, 543, 575–​8, 580, 585–​6 Zucman, G., 70, 71 Zukauskaite, E., 230 Zvan, B., 597 Zwick, D., 298 Zyontz, S., 339 Zysman, J., 542, 551

Subject Index

Aberdeen, 735 aboriginal communities, 834 Accenture, 408, 415–​16 accession: of EEC, 120, 130–​2; of EU, 113; of World Trade Organization (WTO), 437 acquisitions, 259, 312, 336–​7, 369, 374, 561, 566, 599 adaptability, 260–​1, 783, 852; economic, 853, 855, 857; versus adaptation, 261 adaptation, 116, 154, 189, 198, 214, 216, 222, 246, 249, 260–​1, 271, 670, 673–​5, 845, 848–​9, 851–​2, 860; economic, 853; governance of, 830; successful, 848; versus adaptability, 261 adivasis, 107 ADX Portland, 818 Aegean, 124 aerial differentiation, 3–​4, 8 Africa, 4, 98–​9, 408, 415, 419, 433, 565, 719, 722, 773, 779; African, 26, 777, 784; African, countries, 28 African Development Bank, 749, 770, 785 agglomeration(s), 5, 13, 28, 78, 83, 86–​8, 91, 114, 116, 146–​7, 152–​3, 155, 180–​1, 214, 253, 274, 324, 330–​2, 336–​9, 347–​9, 352, 354–​6, 359, 361, 370, 374, 568, 717–​18, 735, 797–​9, 827–​8, 834; advantages, 184; benefits, 270, 331, 333–​4, 353, 564; cluster, 334; diseconomies, 7; economies, 86, 88, 90, 308–​9, 330–​1, 333, 335, 337, 347–​8, 352, 354, 361, 385, 577, 827, 835; external, 334–​8; geographical, 413; internal, 335–​8; models of, 349; perspective, 352; regional, 154; spatial, 87, 91, 202, 309; urban, 500, 792, 795 agrarian reform, 107 agriculture/​agricultural, 22, 23, 26, 30, 33, 34, 75, 81, 82, 83, 84, 92, 105, 120, 122, 123, 124, 125, 126, 130, 132, 161, 427, 428, 453, 527, 667, 670, 671, 672, 683, 687, 710, 723, 724, 736;

decollectivization of, 82; dryland, 669; economy, 738; geographers, 170; labour, 78 Ahold, 429–​30, 438–​9 Airbnb, 492 AJR model, 99 Alabama, 72 Alaska, 669, 724 Alberta, 637, 706 Aldi, 429–​30 Alibaba, 441, 570; Alibaba Group, 441 Amazon, 276, 429–​31, 440–​1, 443, 492–​3, 570, 709 amenities, 500, 506, 509–​12; absent, 350; cultural, 630; neighbourhood, 800; role of, 509 American Airlines, 410 American dream, 66, 69 Amsterdam, 563, 565–​6, 571 Andes, 724 Andhra Pradesh, 102, 106 Angola, 5, 7 Antarctic, 707 Anthropocene, 13, 161, 665–​6, 674–​5, 755, 765 Apple, 7, 78, 441, 721 apprenticeship: programmes, 525–​6; system, 523–​4 Aravind Eye Hospital, 829 arbitrage, 389, 417, 543, 571, 578, 586 Arctic, 705, 707, 715, 727; Arctic Circle, 705; Arctic Ocean, 723 Argentina, 434, 456, 565, 724 artificial intelligence, 255, 411, 417 Asia, 4, 19–​20, 24, 26–​32, 34–​6, 97, 408, 410, 433, 441–​2, 467, 565, 604, 719, 723–​4, 792; East, 27–​8, 30, 34, 36, 104, 389, 395, 399, 428, 434, 544–​6, 552, 794; North, 28; South, 28, 104, 399, 436, 453; South East, 28, 30, 399, 428, 436–​7, 439, 793

894   Subject Index Asian, 12, 20, 24, 33–​5, 134, 390, 408, 434, 571, 777; countries, 20, 22, 31; development, 24, 27; economies, 20–​1, 26, 32, 389; industrialization, 25; industrializers, 67; region, 19–​21, 27, 32, 35; tigers, 25, 27, 29, 35 Asian Development Bank (ADB), 21–​4, 31, 34, 36, 399, 749, 770–​1, 777, 785 Asian Infrastructure Investment Bank, 36 Assam, 106 Association of Southeast Asian Nations (ASEAN), 27–​8, 374, 399 Athens, 12, 129, 134 Atlanta, 565, 817, 842–​3 Atlantic, 467, 472, 669, 710; North Atlantic, 73, 172 Attiki, 123 Auchan, 429, 430 austerity, 6, 12, 28, 55, 113–​14, 119, 122, 126, 130, 133–​6, 149, 455, 490, 550, 857; anti-​austerity, 494; fiscal, 113–​14, 120, 122–​3, 127, 129–​30, 133–​5, 854, 857–​8; policies, 131 Austin, 328, 333 Austin–​Round Rock, 328 Australia, 3, 23, 27–​8, 33, 49–​51, 99, 209, 234, 317, 408, 456, 565, 604, 685, 720, 761, 774 Austria, 49, 50, 117 automation, 68, 417–​18, 467 Avasant Group, 419 B2C, 441 Baden-​Württemberg, 235, 386 Baltic, 117, 120 Baltimore, 565 Bandra Kurla, 107 Bangalore, 409, 413, 415, 418, 453, 565, 570 Bangkok, 307, 390, 565, 772 Bangladesh, 20–​3, 26, 30, 32, 89, 455, 773 Bank of Japan, 654 bankers, 53, 620; pioneer, 595 banking, 13, 26, 32, 397, 541, 545, 563, 566, 569–​7 1, 577, 583–​5, 615–​16, 619–​21, 658; crisis, 854, 857; deregulation of, 854; investment banks, 570; mobile, 570; retail, 619; rural, 107; see also investment bankruptcy, 740; business, 129; public, 119, 133 Baraboo, 333 Barbados, 410 Barcelona, 631

Barclays Bank, 53, 571 Barings, 566 Barnes and Noble, 276 Barrick Gold, 724 Bavaria, 235 Bayes’ theorem, 201, 203; Bayesian theorists, 201 behaviour: competitive, 180, 199; human, 196, 198, 202; individual, 199, 205, 207–​8, 732; legitimate, 199; observed, 196–​7, 199, 204, 208–​9, 292; behaviouralism, 197, 199 behavioural, 548, 558; approaches, 3, 202; biases, 649; concept, 371; finance literature, 202; geographies, 616; health, 57; perspective, 12; psychology, 200, 203–​5; psychopathology, 848; responses, 260; revolution, 12, 199, 202, 204, 207; scientists, 203; turn in economics, 8, 197 Beijing, 11, 24, 28, 32, 307, 413, 565 Belgian Congo, 99 Belgium, 49–​50, 131, 430, 490, 563 benchmark cluster definitions (BCD), 325–​6, 328, 330–​1, 339 benefits, 44, 47, 56, 59, 75, 81, 83, 90, 151, 202, 273, 332, 488, 505, 601, 630, 692, 710–​11, 760, 777, 792, 831; distribution of, 144; economic, 215; environmental, 683; housing, 56; social, 84; welfare, 612 Berkeley, 8, 54 Berlin, 307, 313 Bermuda, 562, 567 Bharatiya Janata Party (BJP), 101 Bhojpur, 107 BHP Billiton, 721 Bhumihar caste, 107 Bihar, 102, 106–​9 biodiversity, 55, 683, 697, 699, 710, 727, 751, 754–​6, 761–​2, 764; loss, 725; prices, 761 biopharmaceuticals, 325, 332 biopolitics, 617 biotechnology, 153, 795 Bitcoin, 279, 581–​7 BITNET, 275 Blackrock, 570 blockchain technologies, 581–​2, 585–​7 Bloomberg Philanthropies, 817 Boca Raton, 565 Boise City, 333

Subject Index   895 Bolivian, 1 Bollywood, 306–​7, 309, 312–​13, 315–​18 Bolsa Familia, 796 Bolsover, 842 bonuses, 527, 563, 570, 615 boom–​bust cycles, 648, 655, 659, 735 Borders (store), 276 Boston, 202, 259, 325, 328, 359, 565, 570, 817 Botswana, 100, 773 Brahmin caste, 107 Brazil, 4, 20, 43, 75, 86, 89, 235, 307, 310–​11, 352, 434, 441, 455–​6, 562, 565, 650, 685, 761, 773, 796; Brazilian, 456, 544 Bretton Woods, 34, 67, 374, 542, 794 Brexit, 571 BRIC/​BRICs/​BRICS, 4, 20, 75, 562 bricolage, 829, 831 Bridgeport, 73, 333 Britain, 56, 146, 494, 620, 631, 703; Victorian, 613; see also England; Scotland; United Kingdom;  Wales British, 40, 99, 315, 470, 494, 551, 616, 619, 705, 711, 750; colonies, 567; government, 620; workers, 490 British Columbia, 637 British Petroleum (BP), 735, 741 British Virgin Islands, 567 Brno, 416 broadband, 271–​2, 274, 279 Brookings Institution, 73 Brundtland Commission, 704, 713, 781 Brunei, 28 Bucharest, 413 Budapest, 565 Buenos Aires, 565 Bulgaria, 120, 131 Bureau of Labour Statistics, 73 Burlington, 333 business process outsourcing (BPO), 408, 417 business-​to consumer (B2C) industries, 333, 334, 441 Cadbury, 449 Calgary, 565 California, 54, 335, 351 California, University of, 548 Cambodia, 28 Cambridge, Mass, 359, 833

Canada, 23, 48–​52, 54, 99, 234, 339, 431, 488, 491, 565, 718, 720, 774, 794, 811, 812, 813, 819, 822, 832, 834, 858; Canadian, 54, 134, 232, 495, 509, 718 Canary Wharf, 485 Capability Maturity Model Integration (CMMI), 420 Cape Town, 784 capital: accumulation, 81, 102, 107, 151, 251, 477, 629, 633–​4, 742, 850; globalized, 115; gains, 64, 71, 75, 621; investment, 64, 66, 71, 75, 543, 580, 725; mobility, 64, 577, 801 capitalism, 4, 8, 12, 29, 35–​6, 115, 131, 144, 146–​7, 159–​61, 163–​5, 167–​8, 171, 450, 466, 472, 495, 541, 544, 549, 558, 580, 592, 596–​7, 599, 603–​5, 614, 629–​33, 640, 648, 841, 859; agrarian, 104; Anglo-​American, 238; casino, 649; Chinese, 235; climate, 684; contemporary, 384; crisis of, 6, 35; crony, 149; finance, 99, 612; finance-​led, 603, 648; financial, 596–​7, 621; German, 238; global, 19, 35–​6, 544, 854; globalized, 171, 477; globalizing, 163, 171–​2; knowledge-​based, 499–​500, 508, 512; modern, 699; national, 115; organization of, 469; patrimonial, 511; regulatory, 638; shareholder, 649; transformations of, 236; variegated, 35; varieties of, 188–​9, 222, 234–​6, 465, 543, 550, 857 capitalist, 79, 597; accumulation, 115, 541; change, 623; cities, 629; countries, 467; crisis, 542, 550; democracies, 208; development, 115; dynamics, 395, 548; economy/​economies, 34, 144, 146, 159–​66, 168–​7 1, 188, 386, 391, 544, 548; firms, 384; formations, 550; free-​market, 723; growth, 580; industrialization, 795; labour process, 469, 479; modes of production, 621; non-​ capitalist, 604; objectives, 448; production, 622; reproduction, 541; restructuring, 479; social relations, 115; societies, 612; system, 144, 468, 540, 591–​4, 604, 850; take-​off, 630; trade, 717; transformation, 471 capitalization, 577, 582 carbon: accounting, 684, 686–​7; 685, 694, 697; emissions, 5, 683–​4, 686–​7, 708, 710, 761, 765; markets, 13, 161, 170, 683–​9, 694–​5, 697–​8; pricing, 692, 757, 761; sinks/​sinking, 689, 697; tax, 675, 758

896   Subject Index carbon dioxide, 5, 685, 710–​11, 754, 765; see also CO2 Carbon Disclosure Project, 674, 686 Caribbean, 410, 561, 567, 794 Caribbean Data Services, 410 Carnegie Mellon, 199 Carrefour, 429–​30, 438–​40 Casino, 430, 440–​1 caste system, 109 Catalonia, 7, 819 Cayman Islands, 560, 567, 571 CB Richard Ellis (CBRE), 431–​2, 442 Central, 124 Central America, 4, 416, 434 Central Attiki, 124 Central Bank of Russia, 586 Central Europe, 134, 389, 428, 560 Changzhou, 28 Channel Islands, 560 Chennai, 100, 413, 665 Chesapeake Energy, 739 Chevron, 721 Chhattisgarh, 102, 107, 109 Chicago, 565, 578, 795, 817, 842 Chicago Board of Trade, 655 Chile, 434, 437–​8, 495, 685, 724; Chilean, 217 China, 4–​7, 12, 19–​36, 41, 66–​7, 75, 78–​93, 97, 99–​100, 104–​5, 168, 235, 263, 310–​11, 352, 383, 389–​90, 408–​9, 417, 434, 441, 453, 455–​6, 488, 547, 561–​2, 565, 572, 645, 650, 653, 685, 687, 705, 710, 715, 721–​3, 726–​7, 735, 772–​3, 794; Chinese, 4, 8, 11, 19, 24–​8, 30–​4, 79–​80, 82, 84–​8, 90–​1, 235, 456–​7, 571, 614, 722–​3 China Sea, 34 Choices (cooperative community), 531–​2 Chongqing, 235 Cisco Systems, 7, 235, 409 Cities Alliance, 786 Citigroup, 563 City Energy Project, 817 city size, 78, 85, 86, 87, 88, 349, 351 clean energy, 696–​8, 757, 760; clean technology, 7, 764 Cleveland, 274 climate: change, 6–​8, 13, 30, 591, 601, 665–​76, 683–​9, 691–​2, 697–​9, 705, 715–​16, 719, 721,

725, 727, 750, 757–​8, 760–​1, 763–​5, 793, 803, 815, 840, 855, 860; see also global warming; economics, 665, 671; shocks, 666–​8, 670, 672, 675; stability, 757 Club of Rome, 706–​7, 709, 712, 752 cluster(s), 12, 182, 184, 214–​19, 246, 256, 307, 315, 318, 324–​6, 330–​2, 334–​7, 347–​8, 351–​5, 358–​61, 385, 389, 391, 399, 410, 413–​14, 501, 577, 797, 799, 827, 831; agglomerative, 348; analysis, 326, 330, 506; approaches, 181, 216, 328; benchmark, 325, 328, 330; business, 793; connections between, 312–​13; cluster(s), creative industry, 306–​9, 311–​12; cultural industry, 309–​10; definition(s), 325–​6, 328; development of, 352; entertainment, 313–​17; film, 311; geographical, 420, 509; high-​tech, 353, 413–​14; industrial, 213, 236, 376, 499, 501, 826; IT, 418; liberal, 236; localized, 456; manufacturing, 181–​2, 184; multifunctional, 471; multi-​linkage, 328; multiple, 338, 355; professional services, 184; regional, 182–​3, 330, 333; region-​specific, 330; relationships, 181, 182; research, 181, 183, 186, 325; role of, 331; specialized, 307, 353; strong, 338–​9; structure of, 356; study of, 353, 360; subregional, 28; talent, 504; temporary, 187, 299; theory, 184, 330; urban, 183; US, 332, 333 clustering, 146, 181, 215, 222, 306, 308–​9, 326, 328, 349, 354, 360, 508, 511–​12, 795, 799; industry, 214; of creative production, 308; of talent, 499–​500, 502, 509, 511; spatial, 215, 217; urban, 184 CO2, 30, 685–​8, 694, 696, 754–​5, 757–​8, 761, 765; emissions, 30, 687, 758–​9; see also carbon dioxide; greenhouse gas(es) emissions coal, 27, 80, 359, 469–​70, 668–​9, 675, 703–​4, 708–​9, 712–​13, 715, 723, 725, 751–​2, 754, 762–​3, 765; mine closures, 841–​2; miners’ strike, 469 co-​creation, 287–​91, 294, 298; definition, 286, 299; geographies of, 300; logic of, 289; mapping, 290; practices, 287, 298–​300; situating, 289; spatial logistics of, 299; temporary, 299; types of, 291; typology of, 290; value, 286, 287 cognition, 2, 197, 199, 204–​5, 207–​8 cognitive legitimacy, 215 CoinMarketCap, 582

Subject Index   897 Cold War, 25, 632, 718–​19, 722 collective bargaining, 493, 520 co-​location, 12–​13, 185, 197, 202, 215–​17, 296–​7, 325–​6, 332–​3, 335–​6, 409, 416–​17, 564, 827; benefits, 349; cross-​border, 376 Cologne, 11, 236 Colombia, 773 colonialism, 99, 454, 719; postcolonial, 172, 449, 454, 457, 718, 720 commerce, 145, 441, 629, 693–​4; electronic, 276–​7, 440–​3; net based, 471; patterns of, 3 commercialization, 255, 261–​2, 827, 830; of Internet, 272 commodification, 170–​1, 673, 687; of labour, 476 commodities super-​cycle, 716, 721–​2, 727 commoditization, 410, 413–​14, 419, 421 Commodity Futures Modernization Act, 650 Commodity Futures Trading Commission (CFTC), 656–​7 communication(s), 7, 64, 154, 161, 163–​4, 172, 183, 185, 218, 269–​7 1, 279–​80, 293, 299, 312, 371, 374, 390, 394, 338, 411–​12, 478, 501, 575, 579, 584, 742, 761, 811; channels, 412; costs, 269–​7 1, 273–​5, 277–​9; digital, 275, 279; electronic, 270, 278; fibre-​optic networks, 560; global, 449, 632; Internet-​enabled, 274; patterns, 274; technology, 316; technologies, 185, 375, 763; transfer, 411 communism, 6; rise of, 144; communist, 81–​2, 102, 630, 723 Communist Party of India, 102 community benefits agreements (CBAs), 800–​1 competition, 8–​9, 33, 67, 101–​2, 131–​2, 148, 164, 166–​7, 215–​16, 252, 259, 297, 311, 314, 349, 390, 415, 432, 438, 470, 478, 542, 569, 621, 637, 669, 720, 797, 844, 851–​2, 859; forces of, 635; imperfect, 558; import, 332; intra-​firm, 375; local, 418, 434; oligopolistic, 794; perfect, 166, 167; spatial, 97, 166; status, 293 competitiveness, 84, 114, 127, 133, 180–​1, 219, 330, 391, 512, 568, 811, 816, 827, 834, 840, 843, 857; cultural, 311; dynamic, 855; firm and regional, 324; global, 831; industrial, 542, 544; inner-​city, 799; local, 512; national, 398; pro-​growth, 784; regional, 792, 793, 815

complexity theory, 214 Congo, 709 Conlumino, 442 Connecticut, 72–​3 conservation, 256–​8, 683, 687, 697 consolidation, 113, 257, 410, 435–​7, 439, 566, 719; industry, 571; political, 796 construction, 29, 32, 73, 120, 122–​7, 129, 131–​2, 135, 356, 434, 457, 468, 475, 491–​2, 522–​7, 529, 614, 634, 761, 834; carbon market, 686; industry, 522, 526; jobs, 530, 793, 800; pipeline, 735; tools, 526; well, 740; workers, 520, 522 consumerism, 24, 580 consumption, 1, 3, 5, 24, 27, 33, 55, 59, 83, 91, 102, 113, 131–​3, 135, 149, 161, 168, 172, 287, 298, 306, 313, 316, 339, 385, 427, 435, 443–​4, 452, 454, 457, 501, 545, 549, 576, 580, 593–​4, 601, 611, 613, 616, 619, 628, 690–​4, 706, 712–​13, 726, 753, 756–​9, 795; consumer behaviour, 277, 279; creative, 308–​9; cultural, 580; decline, 129; excessive, 754–​5; geography of, 287; goods, 349; growth, 104; household, 119; local, 797; mass, 47, 632; meat, 716; natural resources, 591; of financial products, 612; passive, 294; patterns, 182, 398, 797; personal, 613; post-​consumption, 397; private, 123, 135, 633; resource, 720 convenience stores, 429, 435–​6, 439 Copenhagen, 306, 313–​15 Corbridge, S., 97 core–​periphery, 113, 115, 117, 120, 126, 130, 133–​5, 253, 360, 368–​9, 376–​7, 717, 727; contradictions, 113–​14, 131; divergence, 114; dynamics, 722, 727; polarization, 120; political economy, 115; research, 719; theory, 716–​18, 720; see also periphery Cornwall, 475 corporate citizenship, 448; culture, 397, 435, 801; responsibility, 448–​51, 455 corporate social responsibility (CSR), 13, 411, 419, 449–​55, 457, 726 Corporation for Enterprise Development, 802–​3 corruption, 26, 32–​4, 97, 100, 154, 727; government, 726, 739 Cosco, 134

898   Subject Index cost curve, 349, 351, 629 Costa Rica, 416, 453, 687; Costa Rican, 454 cost–​benefit analysis, 630, 760–​1, 798 Council of Foreign Relations, 820 Craigslist, 279 creative: class, 500, 502, 504–​6, 510–​11; class dynamic, 795; destruction, 256, 258, 722, 841, 851–​2; industries, 8, 13, 215, 305–​11, 313, 317, 486, 509 credit: risk, 614; scores, 614; scoring, 612–​14, 616 Credit Suisse, 40, 65 Crete, 124 Croatia, 118, 120, 122 cronyism, 26, 154 crowdsourcing, 419–​20 cultural hegemony, 307, 311; Western, 307 customer, 180, 287–​91, 408; base, 613; care, 451; pressures, 385; role of, 286; Schumpeterian, 300; site, 293 customer–​supplier connections, 354; customer–​supplier interactions, 349 customization, 289, 306; mass, 288 cyber security, 7 Cyprus, 117, 119, 133, 135, 567 Czech Republic, 120, 416, 434, 440 dalit community, 107, 109 Danish Film Act, 314 Danish Film Institute, 314, 318; Danish, films, 315 Dar es Salaam, 669 Darwinian, 252; Darwinism, 214; Darwinian, theory, 247; see also Provincial Darwinism model Dayton, 333 DCMS, 305 debt crisis, 113, 131–​2, 134, 544; Latin American, 545 decision-​making: individual, 197, 199–​200, 202, 204–​6 decolonization, 720 deforestation, 30 deindustrialization; see under industrialization Delaware, 561 Delhaize, 430, 439

Delhi, 24, 89, 100, 107; New Delhi, 101–​2 Dell’s Idea Storm Forum, 292; Dell strategy, 288 Deloitte, 431, 433, 438, 440–​1 Denmark, 48–​51, 120, 306, 314, 489; Danish 313–​15 Denver, 817 Department for International Development, 781, 786 dependency theory, 35 Depression, 65, 374 deregulation, 8, 437, 443, 449–​50, 471, 479, 486, 542, 546, 611, 618–​19, 813, 816, 821, 854; financial, 650; see also under financial; global financial crisis (GFC); regulation; reregulation derivatives, 559–​60, 563, 566, 646–​50, 652, 656, 659–​60, 691–​4, 696; commodity, 646; trading, 647; weather, 673 desertification, 30 deskilling, 418, 467; workplace, 468 Detroit, 215, 274, 331 deunionization, 69; see also labour/​ trade unions Deutsche Bank, 547, 563 Deutsche Börse, 579 developing countries, 5, 63, 66–​8, 83, 92, 97, 408–​9, 417–​18, 436, 720, 722, 750, 772–​5, 783, 792, 796, 800, 803 development, 12–​13, 20, 23–​4, 26, 66, 75, 86, 107–​8, 115, 135, 144–​5, 147–​50, 154, 179, 183, 185, 220, 222, 230, 235, 407, 418, 479, 670, 673, 705, 711, 715–​17, 719–​20, 722–​3, 725–​7, 734, 736–​7, 741, 760, 763, 770, 775, 782–​4, 794, 797, 815, 820, 831, 833, 840, 843–​4, 849–​53; adaptive, 846; agencies, 853; agreements, 800; benefits, 398; capacity, 801; capitalist, 115, 188, 592; cluster, 181, 184, 218, 352; commercial, 134; community, 800, 802; dimensions of, 786; disparities in, 146; early stages of, 66; East Asian, 388; economic, 2, 8, 12, 20, 22, 29, 35, 79, 97, 144–​5, 148–​55, 188, 190, 199, 221, 233, 300, 305, 307, 352, 366, 376, 383, 395, 398, 418, 427, 433, 473, 499–​500, 509, 512, 601, 619, 630, 666, 674, 716–​17, 720, 722, 783, 785, 826, 854, 856, 859; endogenous, 799; energy, 734; equality, 728; financial, 623; foundations

Subject Index   899 of, 145; from within, 799; geographical, 144, 168, 236, 317; geography of, 146; global, 722; industrial, 30; industry-​led, 716; inequitable, 772; infrastructure, 635, 638, 640; initiatives, 419; level of, 67; local, 394, 784, 801, 855; local and regional, 389, 391; local economic, 857; long-​term, 260; market, 397, 622; model, 69; national, 31, 80, 622; objectives, 724; of Asia, 27; of China, 27, 90; of countries, 260; of privatization, 133; orderly, 155; participatory, 787; paths/​ pathways, 261, 794; planning, 667; policies, 147, 155, 398, 672, 721, 828; political, 720; pressure of, 715; process(es), 288, 290, 292–​3, 633, 721, 783; product, 187; programmes, 793; projects, 293; prospects, 26; real-​estate, 800; regional, 14, 114, 181, 213, 219–​22, 246, 371, 387, 389, 399, 499, 503–​4, 506, 561, 623, 699, 735, 738–​9, 796, 812, 841, 850, 852, 855, 858–​9; regional and local, 793; resource, 724, 727; resource-​based, 716; resource-​led, 725; retail, 444; societal, 66; socio-​economic, 775, 776; spatial, 19, 25; strategies, 92, 782, 784, 785, 796; studies, 383; sustainable, 8, 704, 716, 719, 749, 858; technology, 827; territorialized, 388; theorists, 719; trajectory, 261; uneven, 2, 8, 10, 14, 19, 26, 29, 35, 115, 214, 394, 542, 667, 716–​18, 795, 803, 811, 840, 844, 849, 859; uneven regional, 850; urban, 71, 222, 541, 722; urban and regional, 391; user-​driven, 291; ventures, 591; workforce, 801–​3 devolution, 79, 389, 800–​1, 811–​15, 819, 821, 858 diaspora(s), 312, 316; communities, 415; diasporic connections, 313; Indian, 316; waves, 415 dichloroacetic acid (DCA) forum, 297 differentiation, 2, 7, 169, 180, 189, 246, 550, 684, 779, 812; axes of, 486; definition, 246 digital spaces, 580; new, 575, 580–​1, 585–​7 digitization, 309–​10, 409, 413 dirty energy, 759, 760 disinvestment, 435, 469, 810, 815–​16, 821–​2, 851; see also investment Disney, 312 diversification, 23, 152, 202, 214, 22–​1, 331, 373, 594, 596, 603; international, 374; livelihood,

672; over-​diversification, 595; portfolio, 596, 650; regional, 222, 260 division of labour, 143, 469, 499, 509, 629; international, 114, 133, 367; new international, 25, 35; social, 185; spatial, 13, 67–​9, 115, 120, 130–​1, 183, 307–​8, 311, 317, 466, 469, 475, 480, 501 Dogecoin, 581 Doncaster, 842 Dongguan, 82, 92 Doris Duke Foundation, 817 Dow Jones Index, 650 dual system (eryanzhi), 78, 80, 82–​3 Dubai, 317, 565 Dublin, 565, 571 Durban, 784 Dutch East India Company, 563, 566, 594 dynamic stochastic general equilibrium (DSGE) model, 545, 548, 552 Easington, 842 East India Company, 716 Eastern Gangetic Plains, 673 economic crisis, 70, 133, 337, 440; Asian, 434; global, 6–​7, 100, 108, 428, 442, 444, 549 economic development: community, 801; equitable, 786; regional, 797, 799 economic growth, 3, 5, 12, 21–​2, 25, 31, 33, 55, 63, 66–​7, 78, 80, 83, 92, 97–​106, 108–​9, 144–​5, 147–​8, 150, 152, 199, 273, 305, 331, 479, 499, 501–​3, 505, 509–​12, 539, 544, 550, 557, 580, 620, 629, 649, 666–​7, 670–​1, 674, 687, 703–​4, 706, 711, 713, 722, 726, 749–​50, 755–​8, 764, 770–​7, 779–​84, 786, 794, 796, 828, 840, 844, 848, 856, 860; benefits of, 772, 778; continued, 757; equitable, 14, 770–​1, 776–​86; in China, 772; inclusive, 777–​80, 785; indefinite, 754; inequitable, 775; models, 764; national, 722; rapid, 710; regional, 793, 796, 798, 853; stages of, 630; sustainable, 712, 780, 858; urban, 774, 781, 783 economic relations, 4, 7, 181; global, 715 economies of overview, 184 economies of scale, 5, 28, 85, 146, 183, 185, 249, 373, 501, 567, 603, 613, 717; location-​specific, 374 economism, 161

900   Subject Index Ecorys, 126 ecosystem(s), 13, 153, 245–​7, 250, 253–​5, 259–​62, 628, 669, 674–​5, 683, 695–​6, 699, 709–​10, 712–​13, 751, 755, 811, 845, 852–​3, 856; biological, 247, 252, 255; boundaries, 254; concept, 262; definition, 246–​7; development of, 256; disturbed, 258; diversified, 261; economic, 246–​8, 251–​7, 261–​3, 510; entrepreneurial, 264; eframework, 251; interdependencies, 258; mapping, 253; national entrepreneurial, 262; outcomes, 261; resilience, 260; social, 257; structures, 258; succession, 255, 256; view, 250 education, 25–​6, 31, 68, 74–​5, 92, 98–​9, 104, 122–​6, 145, 147, 151, 206, 272, 316, 478–​9, 502–​6, 509, 519, 521, 614, 628, 630, 749, 772, 774, 779–​80, 795–​6, 798, 801–​4, 821, 827, 858; access to, 419; consumer, 261; financial, 595, 618; formal, 309; higher, 495; system, 630; universal, 794 educational attainment, 500, 502–​6, 801 Egypt, 419 electrification, 104; rural, 107 electronic trading, 570 email spam, 586 emigrants, 832; see also immigrant(s); immigration; migrants; migration Emilia-​Romagna, 235, 798 emissions markets, 685, 686 employee turnover, 234, 561 employment, 22, 64, 67–​9, 72–​3, 82–​3, 85–​6, 90, 92, 104, 107, 109, 114, 116–​18, 120, 122–​3, 125–​7, 129, 132, 134–​5, 150, 154, 232, 234, 259, 273–​4, 326, 330–​3, 353, 407, 415, 418–​19, 465–​8, 472–​3, 475, 477–​8, 486–​7, 489, 494–​5, 503, 505–​6, 509, 523, 528, 545, 564, 570–​1, 615, 669, 671, 777–​80, 785, 795, 801, 804, 813, 827, 833, 841, 847–​8; agencies, 491, 492, 853; benefits, 490; collapse, 133; decent, 779; fair, 479; flexible, 529; full, 568; global, 418; growth, 219, 273–​4, 331, 353; informal, 779; insecure, 527; insecurity, 487; intermediated, 488; international, 312; markets, 107; non-​agricultural, 779; non-​standard, 492; norms, 472; part-​time, 490; precarious, 487, 490–​1, 495; private, 126;

productive, 780; protective, 528; public-​ sector, 796; regulation, 493; relations, 469; restructuring, 802; rights, 467, 49–​4; security, 472; self-​employment, 114, 127, 130–​2, 135, 486, 491–​2, 798; specialization, 328; standards, 493; temporary, 491; triangular, 493; women’s, 476 endowments, 230, 261, 373, 386, 398, 465, 591, 596, 602, 645, 652; natural resource, 812 energy transitions, 14, 675, 732–​4, 742; see also thermal energy England, 143, 488; see also Britain; United Kingdom Enron, 279 entrepreneurial, 13, 24, 31, 102, 113, 116, 126, 130, 135, 149, 153, 184, 223, 245–​6, 261–​2, 264, 287, 312, 353, 415, 493, 509, 617, 621, 799, 803, 856; activity, 257; networks, 798; opportunity, 258; potential, 257 entrepreneurialism, 150–​1, 826–​7 entrepreneurs, 25–​6, 32, 98, 153, 155, 185, 215, 220, 246, 250, 252, 259, 261, 272, 300, 338, 353, 504, 506, 577, 601, 793, 798, 818, 820, 828–​9; civic, 239; returnee, 415; user, 300 entrepreneurship, 100, 114, 120, 122, 127, 130–​2, 134–​5, 144, 154, 164, 186, 248, 254, 256, 261–​2, 326, 330–​3, 338, 348, 353, 360, 398, 415, 470, 504, 792, 797–​9, 802–​3, 816, 819, 832; inclusive, 830; social, 419 environment, 1–​3, 5, 11, 21, 40, 80, 91–​2, 107, 170, 181, 187, 196, 198–​9, 203, 205, 209, 214, 219, 221, 236, 249–​50, 252, 257, 289, 296, 299, 324, 334, 370–​1, 392, 448, 450, 453, 557, 594–​5, 602, 637, 658, 674–​6, 689, 703, 715, 718, 721, 723–​4, 735, 749, 751, 756, 822, 833, 855; built, 633; deregulated, 814; financial, 616; global, 619; institutional, 793, 827, 828; natural, 687, 689, 694, 697, 698; policy, 814; regulatory, 741; zero-​growth, 764 environmental, 2–​4, 8, 14, 19, 24, 30–​1, 33, 55, 80, 98, 152, 161, 172, 199, 254, 349, 392, 398, 420, 437, 448–​9, 580, 632–​4, 675, 687–​8, 692, 694, 697, 716, 718, 720–​1, 725, 737, 739–​40, 750, 755–​6, 759, 765, 775, 781, 786, 839–​40, 859; advocacy, 815; auditing, 451; challenges, 764; change, 715, 719, 721, 723; conditions, 474; constraints, 716, 727, 750,

Subject Index   901 760, 764; cycles, 686; damage, 724, 726, 757; degradation, 30, 349–​50, 683, 765; determinism, 3–​4, 8, 172, 199, 674, 718, 723; diversity, 501; externalities, 683; finance, 685, 694–​5; hazards, 668, 670, 737; impact assessment, 725; issues, 749; justice, 666; limits, 723; management, 683; outcomes, 639; policy/​policies, 757, 762; pressures, 750; problems, 754, 755, 761; protection, 449, 721, 728, 741, 750; resistance, 724; resources, 684, 693; responsibility, 450; socio-​ environmental, 690–​1, 694; sustainability, 557, 594; systems, 687; vulnerability, 725 Epirus, 124 equitable growth, 785, 793, 798–​9 equity capital, 636, 640 Erkner, 300 essentialism, 204, 207 Estonia, 118, 131, 491 Ethical Trading Initiative, 453 ethical: arguments, 66; barriers, 835; codes, 452, 454; consumers and producers, 456; consumption, 452; issues, 456; labour codes, 453–​4; performance, 449; responsibility, 450; standards, 449, 452, 455; trade, 452, 455 ethnicity, 2, 474, 510, 780 EU24, 129; EU27, 125–​6; EU28, 117, 120, 127 Eurobank Properties, 134 Europe, 6, 9, 21, 26–​7, 29, 33–​4, 47, 59, 67, 113–​14, 117, 120, 126, 129–​32, 134, 143, 154, 311, 313–​14, 408, 410, 433, 441, 456, 491, 493, 519, 565, 570, 604, 620, 631, 685, 687, 708, 718–​19, 726, 793–​4, 859; East, 389; Eastern, 428, 434, 560; medieval, 558, 594; Western, 311, 315, 467, 471, 488 European, 4, 7, 34, 44–​5, 48, 53, 55–​7, 92, 113–​14, 119–​20, 122, 126, 130–​1, 133, 135, 154, 234, 305, 311, 314–​16, 330, 339, 361, 374, 376, 408, 416, 428, 431, 486, 495, 503, 542, 631–​2, 770, 783; countries, 489; Eastern, 567; integration, 113, 133, 632; retailers, 443; Western, 306, 310–​11, 408, 429 European Bank for Reconstruction and Development (EBRD), 749 European Banking Authority (EBA), 53 European Central Bank (ECB), 113, 115, 119, 135, 581–​3

European Cluster Observatory, 330, 339 European Commission (EC), 113, 115, 119, 126–​7, 131, 133, 135, 305, 656, 770, 783, 785; see also troika (IMF/​EC/​ECB) European Economic Community (EEC) accession, 120, 130 European Monetary Union (EMU), 113, 122, 127, 131–​2, 135 European Single Market, 374 European Union (EU), 34, 44–​6, 48, 53, 56, 92, 113–​14, 117–​18, 120, 122–​3, 125–​7, 129–​35, 154, 374, 376, 488–​9, 491, 505, 571, 576, 582, 631, 685, 726, 792, 820; referendum, 7, 571 European Union Emissions Trading System (EU ETS), 685, 687 Eurostat, 129–​30, 396, 398, 488–​9, 726 Eurozone, 6, 113–​14, 119–​20, 127, 129, 131–​5, 539; crisis, 12, 119, 122; southern, 130 EU-​SILC, 44–​6, 48, 56, 120 evolution, 9, 116, 131, 170, 197, 214, 216, 218–​19, 234, 237–​9, 245–​9, 252–​3, 255–​6, 262, 331, 338, 366, 540, 566, 605, 699, 709, 849–​50; city, 353; economic, 733, 852; of interest communities, 297 evolutionary: approach, 215, 260, 264, 397; biology, 238, 246–​7, 251, 262, 264; evolutionary economic geography (EEG), 9, 14, 213–​23, 237–​8, 733, 741, 819, 852; perspective, 213, 223, 261, 852; theory, 198, 250 exchange value, 689–​92, 694–​8 exchange-​traded products (ETPs), 651 exhaustible resources, 751, 754, 760 Expedia, 582–​3 exploitation, 30, 83, 85, 183, 256–​8, 721, 803; of knowledge, 375; sustainable, 783 exports, 23–​7, 34, 78, 134, 278, 307, 310–​11, 314–​15, 373, 398, 719, 723, 740, 792, 797, 799 externality pricing, 688, 694–​5 Exxon, 721, 735, 741 Facebook, 7, 335 factory system, 143 Fairfax Financial Holding Ltd, 134 Fairtrade International, 450, 452–​4, 456;, standards, 452–​3 Far East, 316 fascism, 144

902   Subject Index Federal Reserve Board, 54, 71 Federal Reserve System, 54, 71, 164, 572, 650 feminism, 478 feminist, 159, 170, 474; geography, 675; theory, 673 feminization, 486–​7; feminized, 476–​7 fictitious capital, 647–​8; thesis, 656 finance capital, 541; environmental, 693; see also organic finance financial: advisors, 614–​15; advisors and business services (FABS), 558, 561–​3, 565–​72; citizenship, 542, 619–​22; crisis/​ crises, 4, 28, 32, 36, 122, 127, 129, 134, 207, 392, 455, 486, 494, 539–​40, 544–​8, 550–​1, 618, 620, 622, 645, 694, 722, 726, 842, 857; deregulation, 854; instability, 13, 539–​40, 548–​51; markets, 6, 9, 168, 197, 201, 433, 439–​40, 539–​40, 545–​6, 551, 568–​70, 579, 597, 611–​12, 614, 616, 618, 621, 646–​9, 653, 655, 658–​9, 735, 738, 740, 812; planning, 611, 615–​16, 618; services, 53, 325, 331, 542, 544, 558, 561, 563–​4, 570, 594–​9, 603, 613, 619–​21, 699, 802; financial technology (fintech), 7; see also under deregulation; global financial crisis (GFC) financialization, 6, 13, 35, 120, 168, 387, 397, 472, 539, 550, 562, 603, 611–​12, 614–​15, 617–​23, 636, 646–​9, 652–​4, 655–​60, 666–​7, 673, 718, 725, 727; biofinancialization, 617; carbon, 684; definition, 539; of everyday life, 613, 616; thesis, 647, 657–​8 Finland, 48–​52, 58, 120 fintech, 570 fish stocks, 30, 754 fisheries, 671, 751, 756, 764 fitness landscapes, 246, 248–​9, 251–​3, 256 flash crash, 579 flexibility, 115, 132, 183, 236, 288, 420, 471, 479, 488, 492, 561, 812, 815, 856; policy, 815 flooding, 665, 668, 670, 672 Florida, 74 Flower Power movement, 8 food industry, 592–​5, 598, 601; food security, 28, 594, 774 Forbes (list of the world’s biggest companies), 721 Ford Motor Company, 102, 469

Fordism, 471, 486, 567–​9; Keynesian, 632; peripheral, 115, 120, 131; post-​, 391, 471–​2, 568 Fordist, 113–​15, 120, 127, 391, 469, 471–​2, 477, 479, 568; development, 115; era, 794; post-​, 466, 472, 477, 501; semi-​Fordist structures, 130 foreign, 389; aid, 98, 829; bonds, 740; capital, 436–​7; competition, 470; country, 368; domestic worker, 485; exchange, 559, 569, 571; exchange crisis, 544; firms, 386, 410, 414; influence, 6; intervention, 724; investment, 101, 445, 667; laws, 561; markets, 432, 444, 834; operations, 433; ownership, 232; policy, 26; retailers, 438; rivals, 416; sales, 440; sourcing, 278; suppliers, 386; units, 411 Foreign Business Act (1999), 434 foreign direct investment (FDI), 25–​8, 65, 68, 80, 85, 92, 132, 134–​5, 186, 235, 317, 369, 371–​4, 386, 410, 414, 434, 436–​7, 439, 562–​3, 567, 720; horizontal (HFDI), 373; outward, 374; vertical (VFDI), 373 forestry, 671, 683, 697, 699, 715; carbon, 687 forests, 30, 247, 628, 672, 692, 696, 698, 751, 765 formalization, 435–​7, 757, 779 fossil fuel(s), 30, 668, 675, 688, 690, 694–​5, 697, 705, 708–​9, 732–​4, 736, 738–​9, 751, 754, 757–​9; imports, 750 Foucauldian, 453 Foxconn, 236 fracking, 161, 705–​6 fracturing, 479, 669, 733, 739; hydraulic, 740 France, 44, 47–​51, 53, 56–​8, 86, 99, 115, 310, 430, 433, 441, 489–​90, 542, 564–​5, 740; French, 14, 47, 99, 311, 452, 631, 640 Frankfurt, 565, 571 Fraport, 134 Freelancers Union, 818, 820, 822 futures, 33, 559, 566, 579, 616, 619, 648, 650–​3, 655–​8, 660; commodity, 655; contracts, 647; exchanges, 645; markets, 653; oil, 648; shale reserve, 739 G7, 24, 34 Gallup–​Knight survey, 510 Gamaliel Foundation, 801

Subject Index   903 gambling, 586, 648 gaming companies, 581, 587 Ganges, River, 30 Gangetic plains, 109 Gansu province, 5 Gartner Inc., 580 gas extraction, 14, 734, 738; see also natural gas;  shale gender, 8, 162, 197, 466, 472, 474, 478–​9, 487, 546, 622, 672, 780; norms, 472; relations, 465, 469; roles, 673 gene mutation, 248–​9 General Agreement on Tariffs and Trade, 630, 832 General Assembly, 819 General Electric (GE), 410, 829 geography of innovation (GOI), 826–​8, 831–​5 Germany, 25, 42, 44, 47–​53, 56–​8, 86, 99, 115, 117–​18, 120, 131, 234–​5, 238, 315, 383, 415, 430–​1, 433, 441, 489, 491, 542, 564–​5, 571, 740, 858; German, 232–​3, 236, 238, 300, 311, 410, 414, 509, 542–​3, 547, 614 Ghana, 419, 454, 784; Ghanaian, 454, 457 Gini coefficient, 2, 39, 48, 86, 773, 786; Gini index, 41; Gini measure of inequality, 51, 490 Glasgow, 565 Glass–​Steagall Act, 650 Glencore, 727 global cities, 75, 369, 376, 377 global city network (GCN) literature, 376–​7 Global Commission on the Economy and Climate, 749 Global Conference on Economic Geography, 11, 339 global delivery models (GDMs), 416 global finance, 168, 186, 542, 546–​7, 550, 558, 572, 575 global financial crisis (GFC), 6, 8, 20, 23, 25–​8, 34–​5, 40, 83, 99, 117, 130–​1, 134, 440, 455, 551, 612, 636, 645, 649–​51, 654, 658, 716, 721–​2, 725–​6, 840 global financial networks (GFNs), 562–​3, 566, 570–​2, 621, 623; evolution of, 567 Global Green Growth Institute, 749 Global Minotaur, 34

Global North, 443, 450, 452, 454–​5, 475, 669, 828 global production network(s) (GPN/​GPNs), 181, 186, 308, 312–​13, 317, 382–​96, 397–​9, 415, 443, 449, 452–​4, 457, 561–​2, 567, 571–​2; approach, 10, 561 global sourcing, 27, 407–​13, 417–​20, 428, 433 Global South, 4, 11, 14, 71, 75, 450, 454–​6, 479, 667, 669, 672, 720, 770–​6, 779–​85, 803, 828 global value chain(s) (GVC/​GVCs), 13, 186, 369, 375, 382–​5, 393–​4, 396, 398, 449, 451–​4, 457; see also value chains global warming, 803, 860; see also climate change globalization, vii, xxi, xxxix, xliii, xliv, xlv, 3, 5–​7, 10, 13, 19, 31–​2, 35–​6, 64, 66–​7, 69, 115, 131, 168, 171, 186, 222, 270, 277, 366, 368, 373–​4, 391–​2, 427, 429, 431–​5, 444, 450, 477, 508, 543, 562, 666–​7, 718, 799, 801, 803; economic, 387, 389, 427, 443; expanding, 539; financial, 541, 543; financial de-​globalization, 571; modern, 369, 378; of corporations, 449; finance, 547; processes, 428; research, 186; retail, 13, 427–​9, 432, 435, 437, 439–​40, 443–​5; ruthless, 383; strategies, 435 Golden Quadrilateral, 100 Goldman Sachs, 20, 563, 567, 570, 650 Google, 7, 359, 570 Google Books Ngram Viewer, 10 GPN 1.0, 391–​5, 399; GPN 2.0, 383, 395–​6, 398–​9 Gramm–​Leach–​Bliley Act, 650 Great Lakes University, 833 Great Recession, 11, 51, 331, 528, 819, 842, 859 Greece, 12, 47–​50, 52, 113–​14, 117–​20, 122–​7, 129–​6, 489–​90, 820; Greek, 119, 122–​4, 126–​7, 129, 130–​6, 489–​90, 820; Greek, economy, 113–​14, 120, 122, 126–​7, 129–​31, 133–​5 greed, 40, 55 green growth, 14, 749–​5 1, 754, 756–​62, 764–​5; definition, 751; strong, 758; weak, 750, 758 Green Growth Summit, 750 green paradox, 758, 762 Green Revolution, 104 green technologies, 223, 601

904   Subject Index greenhouse gas(es), 6, 591, 665, 668, 683–​9, 694, 698, 754–​5, 757, 860; emissions, 6, 591, 665, 668, 683, 689, 698, 750, 765, 803; see also CO2 Greenland, 710, 723 Greenville, 333 Guangdong, 31, 92, 235 Guangzhou, 5, 28 Guatemala, 773 guerrilla warfare, 109 Gujarat, 102, 106, 108 Gurgaon, 107, 570 H&M, 429, 430–​1 Halliburton, 739, 741 Hanoi, 565 Harris County, 735 Harvard University, 99, 104, 478, 505, 833 Haryana, 106 health, 5, 24, 30, 57, 99, 104, 115, 126, 147, 273, 594, 611, 617, 628, 630, 672, 710, 749, 774, 779–​80, 826; and safety, 450, 453; global financial, 722; healthcare, 57, 451, 492, 504, 510, 671, 780, 799, 801, 815; human, 593; insurance, 410, 617; of ecosystems, 246; products, 429, 440; public, 145; systems, 630, 829 hedge funds, 560, 563, 596, 602, 645, 652, 653, 655 Hellenic Post, 134 Hellenic Republic Asset Development Fund (HRADF), 133 Hellenikon, 134 Hewlett Packard, 410 HFC-23, 687 hidden costs, 411–​12, 599 high-​frequency trading (HFT), 578–​80, 587 Himanshu, 104 Hindi, 315–​16; Hindu, 101, 108 holey adaptive landscapes, 256, 259 Hollywood, 307–​9, 311, 313–​17, 349 Holocene, 755 home bias, 12, 202, 596 Home Depot, 431, 440–​1 homelessness, 7 homogenization, 307, 543, 593 Honduras, 670, 687, 773

Hong Kong, 5, 25, 28–​9, 47, 67, 235, 307, 443, 485, 561–​2, 564–​5, 567, 570–​1 Hope & Co., 566–​7 Hotelling model, 756, 757 House of Commons, 631 housing, 8, 24, 510, 512, 541, 611, 614, 674, 736, 800, 843; affordable, 672; benefits, 56; boom, 619; costs, 56, 59, 71, 511, 798, 802, 803; density, 272; finance, 541, 546; market(s), 127, 129, 152, 203–​4, 352, 666, 812; price bubbles, 542, 606; prices, 71, 669, 736, 795; regulations, 152; values, 71, 511 Houston, 565, 817 HSBC, 485, 564 hukou system, 78, 80–​84, 87–​8, 90–​2, 352 human capital, 13, 25, 65–​6, 73, 98, 109, 144, 230, 465, 499–​503, 505–​6, 508–​9, 511–​12, 520–​2, 525, 530, 704, 707, 711, 797; local, 375; model, 530 Human Development Index, 106, 145 human resources, 451, 494 human rights, 449 Hungary, 120, 122, 434, 565 hunger, 713, 723, 774, 796 Hurricane Mitch, 670 Hurricane Sandy, 665, 673 IBM, 153; IBM Global Services, 408 Iceland, 47 IIS, 827, 831 IKEA, 294–​5, 429–​31, 443, 452 immigrant(s), 316, 353, 510, 519–​20, 525–​7, 530, 563; unskilled, 519; workers, 13, 519–​20, 525–​9; see also emigrants, immigration; Latino immigrants; migrants; migration immigration, 25, 220, 316, 353, 506, 510, 519–​20; anti-​immigrant rhetoric, 525, 528; policy, 490, 520; scholarship, 519 impact sourcing (IS), 411, 419–​20 imperialism, 35 income, 1–​2, 5, 12–​13, 21–​2, 31, 39–​59, 63–​5, 67–​75, 97–​8, 104, 119, 122, 126, 133, 146–​7, 149, 154, 197, 272–​3, 487, 495, 503, 506, 541, 545, 670, 672, 724, 776–​8, 786, 795, 797, 801, 829, 853; access to, 419; aggregate, 144; contraction, 134; convergence, 152; convergence or divergence of, 273;

Subject Index   905 definition, 69; differences, 279; disparities, 144, 146, 670; disposable, 120; distribution, 31, 39, 41–​5, 47, 55, 72–​3, 104, 115, 149, 487, 780, 798; groups, 438; growth, 796; inequality, 39–​40, 42–​3, 46, 48, 55, 57–​8, 63, 72–​3, 75, 362, 490, 495, 511, 528, 530, 773, 792, 794–​5, 798, 802; levels, 667; losses, 113; low, 529, 803; national, 712; per capita, 143, 145, 147, 150; polarization, 495; poverty, 103–​4, 106, 108, 802; redistribution, 786; tax, 46, 75, 115 indebtedness, 74, 620, 853–​4; external, 101; over-​, 614 India(n), 4, 7, 12, 20–​32, 34, 75, 86, 89, 97, 99, 100–​9, 235, 263, 306–​7, 310–​13, 315–​18, 352, 408–​10, 415–​17, 419, 436, 453, 455–​6, 493, 562, 565, 570, 650, 665, 673, 829; economy, 100; purchase power, 315; Inditex, 430, 431 Indochina, 28 Indonesia, 20–​5, 27–​31, 34, 86, 89, 436–​9, 565, 685, 773 Industrial Areas Foundation, 801 industrial location, 467, 501 industrial relations, 232, 392, 470, 475; German, 233 Industrial Revolution, 20, 144, 147, 150, 501; Fourth, 417 industrialization, 22, 24–​5, 30, 79–​83, 88, 90, 92, 104, 113, 116, 120, 132, 653, 707, 717, 773; capitalist, 795; deindustrialization, 67, 113, 115, 129–​30, 467, 470, 528, 542, 718, 79–​4, 801; flexible, 115, 127; re-​industrialization, 467 industries: local, 326, 797; traded, 325–​6 inequality, 2, 4–​9, 12, 14, 31, 36, 39–​46, 48, 51, 53–​4, 57, 64–​7, 69, 71–​5, 91, 103, 131, 150, 168, 171, 467, 479, 495, 500, 511, 541, 557, 572, 749, 770–​2, 774–​7, 780, 786, 792–​3, 795, 800–​1, 826, 828–9, 831; deepening, 591; economic, 31, 33, 39–​41, 43–​4, 51, 53, 64, 71, 495, 511, 771–​2, 775; effect on growth, 795–​6; geographical, 512; global, 47, 392, 394; growth of, 803; income, 12, 39–​44, 46–​8, 52–​6, 58, 63–​75, 197, 362, 487, 528, 530, 774, 794–​5; increasing, 795, 829; interpersonal, 773; interpersonal and territorial, 771, 772; of busts, 723; of well-​being, 512; reduction, 796, 800–​1; regional wage, 273; rise of, 793; rising, 511,

792; social, 472; spatial, 5, 105, 108, 168, 511, 620, 670, 674; territorial, 773; urban, 31; i wage, 73, 511; wealth, 55–​6, 63–​6, 69–​72, 74, 802 infant mortality, 106–​7 inflation, 55, 203, 544, 568, 585, 619, 650, 655 informal economy, 1, 779 informalization, 467, 479 information communications technology (ICT), 25–​6, 234, 338, 397, 409, 417, 419, 433, 561, 814, 853: see also information technology (IT) information flows, 575–​80, 587; global, 581 information technology (IT), 32, 155, 271–​5, 277–​8, 325, 328, 333, 408–​10, 416–​18, 435, 558, 561, 570, 586, 613, 671, 795, 818; enterprise, 273; health, 273; investment, 278; see also information communications technology (ICT) Infosys, 408, 410, 415–​17 infrastructural Europeanism, 631 infrastructure, 13, 25, 29, 34, 71, 80, 98, 130, 132, 135, 145, 147, 151, 154, 204, 259, 261, 269–​70, 316, 352, 359, 361, 371, 390, 408–​10, 412, 414, 417–​18, 444, 525, 560, 564, 570, 591, 601, 604, 628–​40, 650, 665, 668–​70, 672, 698, 722, 726, 734–​5, 737–​8, 742, 763, 795, 797, 804, 811, 813–​14, 816, 829, 833, 856, 858; governance, 734; investment, 759; politics of, 628, 639; programmes, 761; projects, 801; soft, 630; urban, 630, 821 ING Group, 571 innovation, 5, 7, 11–​14, 29, 75, 98, 116, 127, 144, 147–​55, 184, 186, 188–​9, 217, 219–​20, 222, 233–​5, 248, 258, 261, 263, 270, 275, 280, 286–​90, 299–​300, 325–​6, 330–​3, 337–​8, 351, 360, 367, 370–​2, 375, 389–​90, 395, 407, 409, 412, 420, 435, 470, 504, 510–​12, 521, 531, 557, 559, 566, 568, 599, 605, 612, 632, 684, 687, 719, 750, 754, 760, 763–​4, 792, 795, 799, 814, 816–​17, 822, 826–​8, 830–​4, 851–​2, 856–​7; approach, 759; business model, 417, 419; civic, 817; commercial, 275; cultural product, 307; diffusion of, 202; energy, 591; financial, 560, 646; for inclusive growth (IIG), 826–​8, 830–​5; geography of, 147; green, 759; highway(s), 14, 827, 832–​5;

906   Subject Index innovation (cont.) hubs, 795; inclusive, 830; localization of, 275; localized systems of, 245; networks, 399; policy, 811, 815; producer-​driven, 294; product, 593, 594; recombinant, 219; regional, 214, 223, 235, 500, 811, 812; regional innovation system (RIS), 222; resource and infrastructure, 591; reverse, 829–​31; role of, 751; service, 411; socio-​technical, 733; spatil, 287; spatiality of, 298; structure of, 255; studies, 312; technological, 147, 398, 733; urban, 818; user-​driven, 300 Institute for Fiscal Studies, 55, 59 Institute for Market Transformation, 817 Institute for New Economic Thinking (INET), 9 Institute for Public Policy Research, 492 Institute of Developing Economies Japan External Trade Organization (IDE-​JETRO), 382, 399 institutional intermediaries, 810–​11, 814–​21 institutionalism, 208, 393, 478 institutions, viii, 35, 98–​101, 103, 116, 145, 150–​4, 188, 205, 208, 214, 217, 221–​2, 230–​9, 246, 259, 275, 318, 324, 328, 330, 335–​6, 352, 371, 386, 390, 413, 450, 452, 454–​5, 471, 473, 480, 526, 532, 547, 557, 618–​20, 623, 634, 654, 760, 762, 779, 793–​4, 797–​8, 813, 816, 819, 821, 826, 829, 833, 835, 844, 854, 858; anchor, 802; banking, 26, 577, 584; capital market, 234; China’s, 78, 80–​1, 91, 235; complementarity of, 236; definition of, 232, 238; deliberative, 188; democratic, 67; economic and social, 238; evolution of, 239; financial, 29, 202, 539, 558, 581, 584, 586, 611, 613–​14, 618–​19, 622, 647, 650, 654, 686, 802, 831, 853; formal, 530, 814; formal and informal, 2, 97, 231–​3, 561, 851, 853; functions of, 232, 238; governance, 831; government, 436, 450; India’s, 100; international, 749, 785; knowledge-​based, 500, 511; labour, 529; labour market, 66, 68, 75, 521, 528; local, 145, 214, 216, 222, 236, 309, 386–​7, 389, 394, 804, 810, 857; market, 812; mission-​driven, 818; national, 231, 236–​7, 452; national vs regional, 236, 239, 686; nature of, 231, 854 non-​firm, 385, 387; non-​governmental, 811; policy-​oriented,

785; political, 33, 471, 546; private, 151, 810; protective, 520; proto-​, 527, 530; public, 151, 155, 629, 810; regional, 576, 785, 810; regulatory, 741; role of, 151, 213–​14, 22–​3, 230, 234, 279, 393, 520, 542, 794, 819, 821, 853; social, 387; specialized, 325; state, 230, 392, 456, 478; structure of, 155; sub-​national constellations of, 235; supranational, 115, 117, 131; third-​sector, 812; transnational, 638 integration, horizontal, 373–​4; vertical, 367, 370, 372–​4, 384, 452 Intel, 7 intellectual property (IP), 295, 308–​9, 317, 580, 586, 827, 832; rights, 311 Intergovernmental Panel on Climate Change, 5, 666, 670, 672, 675, 803 intermediation, 395, 487, 490, 492, 592, 597, 613–​14; digital, 492; financial, 599, 604, 605, 606 internalization, 370, 372, 377, 634, 694; advantages, 367; knowledge, 376; of externalities, 694 International Energy Agency, 713 International Labor Organization (ILO), 398, 451, 453, 487, 779, 787 International Monetary Fund (IMF), 4, 24, 28, 47, 55, 63, 66, 74, 113, 115, 119, 131, 133, 135, 434, 545, 552, 601, 630, 796; see also troika (IMF/​EC/​ECB) international production, 367–​8, 370, 394 internationalization, 127, 130, 186, 370, 386, 415, 429, 440, 444, 470–​1, 812, 819; competitive, 135 Internet, 13, 153, 172, 269–​9, 287, 290, 296–​7, 299, 353, 374, 417, 419, 428, 440, 485, 493, 543, 575, 584, 586, 821; access, 274; adoption, 270, 272, 275–​6, 279; age, 759; applications, 279; behaviour, 277; diffusion of, 275, 278; investment, 273; market platforms, 420; technology, 271, 274 Internet of things, 172, 417 internet service providers (ISPs), 272; dial-​up, 272 Interstellar Kredit, 581 investment(s), 4–​6, 8, 13, 24–​5, 27, 29, 33, 36, 44, 64, 68–​9, 71, 75, 83, 85, 98, 100, 102, 104, 107–​9, 115, 123, 127, 129, 133–​5, 148, 151, 155, 201, 232–​3, 269, 273, 310–​11, 314–​15, 328, 338,

Subject Index   907 352, 369, 372–​4, 394, 396, 398, 409, 411, 421, 428, 433–​4, 437, 439, 466, 468, 501, 520–​1, 543, 549, 558–​60, 562–​6, 572, 579, 592, 594–​9, 601, 604–​5, 612, 614–​22, 630–​1, 634–​7, 648–​51, 653–​4, 659–​60, 694, 710–​11, 719, 722–​3, 726, 735–​9, 741, 750, 757, 779, 786, 795–​7, 803, 820, 826–​7, 834, 843, 847, 857; banks, 351, 564, 565, 567, 650, 652; capital, 597; climate, 758; community, 449; compensating, 712; development path, 372; direct, 382, 386, 817; equity, 389; finance, 856; financial, 612, 615, 621, 646–​7, 652–​4; flag-​planting, 439; foreign, 368–​9; foreign inward, 101–​2; government, 65, 149; greenfield, 369; industry, 596, 602; infrastructural, 88; infrastructure, 25, 631, 637, 811, 821; institutional, 601, 604–​6; internal, 595; international, 720; Internet, 275, 277; inward, 386, 735; IT, 273, 275, 277–​8; large-​scale, 135, 151; liberalization of, 450; long-​term, 601, 696, 722, 742; mineral reserves, 734; mutual fund, 202; neglect, 107; opportunities, 592; optimism, 722; overseas, 470; passive, 652; patterns, 398, 667; professionals, 201; public-​sector, 470; real-​estate, 127; retail, 434; returns, 734, 738, 740, 826; risks, 594; short-​term, 741; social, 630, 801; speculative, 648, 738, 740; speculatory, 135; state, 796; support, 853; urban, 817; workforce, 467; see also disinvestment invisible hand, 29, 167, 298, 688; of the market, 452 Ionian Islands, 124 Ireland, 48–​52, 58, 119–​20, 129, 133, 410, 489, 562, 565, 571; Northern Ireland, 488 Israel, 5, 48–​52, 488, 565 Istanbul, 307, 565 Italy, 44, 47–​53, 56, 58, 113, 118, 120, 130–​1, 235, 386, 487, 489, 494, 565–​6, 594 Jacobs’ externalities, 151, 219–​20, 261 Jakarta, 89, 565 Jamshedpur, 107 Japan, 19–​23, 25–​8, 30, 32, 34, 42, 49–​50, 53, 58, 67, 86, 89, 99, 234, 307, 310, 389, 430, 433, 441, 443, 542, 565, 685, 794; Japanese, 25, 27, 36, 50, 311, 432, 542, 571; Japanese, factories, 501 Jersey City, 73

Jharkhand, 102, 107, 109 job loss(es), 67, 123, 125, 129, 407, 467–​8, 519, 851 Johannesburg, 565, 784 joint ventures, 312, 442 JP Morgan Stanley, 567 just growth, paradigm, 14, 754, 793 Kansas City, 817 Karnataka, 106 Kenya, 419, 456, 833 Kerala, 106 Keynesian, 115, 470–​1, 477, 540, 545, 548–​9, 558, 568, 629, 632, 758, 813; Keynesianism, 149, 170, 629; Keynesians, 548; non-​Keynesian, 552; post-​Keynesian, 230, 549; see Keynes, John Maynard Kimberly-​Clark, 292 Kingfisher, 431 Kiribati, 674 Kisumu, 833 Knight Capital Group, 578 Knowledge Capital Model (KCM), 373–​4 knowledge: economy, 144, 187, 488, 492, 500–​2, 511, 820–​1; flows, 182–​3, 185–​6, 316, 336, 347, 351, 359, 368, 371; networks, 185, 214, 217–​19, 253, 375–​6, 577; production, 222, 287, 290–​1, 293, 295–​6, 299, 413; services, 183, 410, 413–​14; services clusters (KSCs), 413–​14, 415; work, 409, 413; workers, 501–​2 Koeri caste, 107 Kolkata, 100, 107 Korea, 21; Korean, 27, 50; North Korea, 34; South Korea, 20–​8, 30, 32, 34, 49–​50, 58, 67, 100, 335, 389–​90, 434, 436, 440, 443, 565, 685 Kraft Foods, 292 Kresge Foundation, 817 Kurmi caste, 107 Kuwait, 47 Kuznets curve, 479, 765, 794 Kyoto, 683–​4 Kyoto Protocol, 683–​5, 815 labour: migrant, 82–​3, 90, 92; relations, 69, 119, 134, 389, 465; rights, 456, 491–​2; rural, 82, 104 labour market(s), 13, 25, 66, 68–​9, 108–​9, 130, 168, 182, 185, 234, 324, 351, 408, 465–​7, 471–​2, 476–​9, 486–​90, 492–​3, 495, 500, 511,

908   Subject Index labour market(s) (cont.) 519–​21, 525–​8, 530–​2, 564, 666, 602, 793–​4, 801–​2, 812–​13, 815, 818, 820–​1; female, 488; feminization of, 472, 478; flexible, 493; free, 164; internal, 490; local, 309; national, 82, 491; participation, 802; regional, 674; regulation, 64, 65, 68, 486, 490; reproduction, 817; rural, 104; secondary, 352; specialized, 185; UK, 488; urban, 90, 197; US, 67–​8 labour movement(s), 466–​7, 474–​5, 480, 492, 494, 531, 801 labour power, 164, 491, 719; definition, 465 Labour Party, 469 labour/​trade union(s), 68, 168, 233, 385, 392, 450, 453, 465, 474, 487, 490–​1, 493–​5, 520, 522–​6, 529, 568, 800–​1, 813; see also deunionization;  union Labuan, 565 L-​advantages, 370, 372 Lagos, 307, 672 Lahore, 316 Lancashire, 843–​4, 859 Laos, 28 Latin America, 4, 97, 408, 428, 433, 437, 467, 719, 773, 785, 792–​4, 822; Latin American, 29, 544–​6 Latino: construction workers, 525; immigrants, 520, 525–​7; middle class, 796 Latvia, 120, 122 Leeds, 551 Lego, 294, 295 Leibniz Institute, 300 Leninist, 79 Lewis turning point, 83, 92 liberalism, 64; see also neoliberalism liberalization, 101, 103, 147, 449, 450; banking, 621; market access, 434 Liechtenstein, 562 Lima, 565 Linux (operating system), 287, 295, 296 Litecoin, 581 literacy, 99, 106–​7; financial, 206, 595, 616, 618 Lithuania, 120, 122 local economic development (LED), 783–​4; definition, 783, 787 local enterprise partnerships (LEPs), 237

localization, 146, 217, 245, 308, 330, 334, 390–​1; de-​localization, 592, 594; economies, 215, 217, 334, 391, 797–​8; of innovation, 275; strategic, 435 location theory, 160, 377, 499 lock-​in, 116, 123, 182, 214, 216–​17, 238, 670, 686, 734, 759, 762, 851–​2, 854, 859; definition, 734 logical positivism, 170 London, 9, 29, 53, 236, 307, 309, 317, 468, 485–​6, 543, 547, 551, 560, 564–​7, 570–​1, 669, 857; City of, 492, 546, 551, 620; property market, 203 London School of Economics, 544 long-​term investors (LTIs), 591–​2, 594–​9, 601–​5 Los Angeles, 155, 548, 565, 817 loss aversion, 2, 203 Louisiana, 73, 739 Louisville, 817 Lund, 11 Luxembourg, 47, 131, 492, 562–​3, 567, 571 Luxembourg Income Study, 48 LVMH, 430–​1 Maastricht Treaty, 631 Macao, 5, 29 Macedonia, 124 McKinsey & Company, 123, 591 McKinsey Global Institute, 374 Madhya Pradesh, 102, 106 Madison, 333 Madrid, 565 Maharashtra, 102, 106 Maker Movement, 818 Maker’s Row, 818, 820, 822 malaria, 98, 295 Malaysia, 20–​3, 25, 27, 29–​30, 32–​4, 389–​90, 434, 437, 565 Malaysian Institute for Economic Research, 400 Malta, 118 Malthusian, 147, 707 Manchester school of economic geographers, 383, 389, 393 Manga (cartoons), 313 Manhattan, 356, 564, 570 Manila, 89, 565

Subject Index   909 manufacturing, 5, 22–​3, 25, 27–​8, 32–​3, 69, 74, 82, 88, 120, 122–​7, 129, 131–​2, 181, 183–​4, 234, 253, 261, 263, 274, 325–​6, 331–​7, 339, 353, 361, 369, 382, 384–​5, 413, 416, 427, 429, 443, 453, 467, 470–​1, 475, 511, 576, 674, 695, 717, 719, 726, 794–​5, 822; capital, 99; clusters, 182–​4; firms, 292; process, 708; technologies, 232; textiles, 420 Maoists, 98, 107, 109 Marcellus shale play, 738 marginalized, 474, 477, 777–​8, 780, 792, 801, 826, 830 market capitalization, 7, 582 market economies, 79, 145, 146, 477, 633; coordinated, 222, 234–​5; emerging, 456, 645; foreign, 373; German, 236; liberal, 222, 234, 235 market failure, 26, 148, 520, 592, 689, 757, 762, 813, 818, 821 marketization, 6, 7, 167, 171, 449, 667, 673, 683 Marshallian, 308; agglomeration(s), 324; externalities, 151, 215–​17, 219, 273; thinking, 215; view, 214, 218 Marwaris, 102 Marxian, 161–​2, 164, 166, 172, 468, 473, 541, 544, 558; crisis theory, 540, 544; Marxism, 162, 393, 478, 541; Marxist, 41, 102, 115, 159, 418, 467, 473, 487, 630, 640, 648, 693, 699; Marxist, tradition, 647; neo-​Marxist, 487 Massachusetts Institute of Technology (MIT), 54, 67, 359, 706 Mediterranean, 120 megacities, 4, 7, 89, 669 Mekong, 709 Melbourne, 565 Memphis, 817 Menlo Park, 335, 356, 565 mentorship, 523, 526 mergers, 259, 278, 369, 374, 561, 566, 599, 727 Metro, 72–​3, 429–​30 Metro Cash & Carry, 453 metropolitan, 12, 73, 134, 510–​11; areas, 64, 73–​4, 86, 89, 184, 254, 275, 508, 510–​11, 795, 798; councils, 469; regions, 24, 28, 73, 376, 503, 506, 795, 811; statistical area (MSA), 254–​5, 272, 326, 564

Mexico, 86, 89, 254, 339, 352, 565, 685; Gulf of Mexico, 706; Mexican crisis, 544; Mexico City, 307, 565 Miami, 565, 606 micro-​entrepreneurship, 129, 135, 803 Microsoft, 409–​10 Middle East, 316 migrants, 82–​3, 85, 90–​1; in-​migrants, 675, 799; migrant workers, 82, 90–​1, 486; rights, 473, 490 migration, 24, 69, 72, 78–​9, 81–​3, 91, 152, 185, 349–​50, 352, 486, 511, 798; economic, 488; in-​, 512; inefficient, 352; internal, 352; non-​economic, 797; racialized, 476; restrictions, 86; rural–​urban, 83, 105; see also outmigration Milan, 307, 565 milieu school, 181 Miller Center, 822 Minera Yanacocha, 724 mineral depletion, 754, 765; mineral rights, 737, 739–​40 mining, 161, 470, 671, 683, 715, 721, 723–​4, 726–​7, 841; projects, 724; region, 715 Mississippi, 72 modernization, 108, 135, 428, 470, 631, 640, 688, 773; ecological, 686; period, 470; modernization, retail, 438 monetarism, 470; monetary policy, 203–​4, 611; money laundering, 567, 569, 576, 586 Mongolia, 715, 723–​4, 725–​6; Mongolian, 723, 725 monopolies, 75, 149, 630; monopolization, 635 Montreal, 565 Moody’s Investors Service, 740 Morgan Stanley, 563, 565–​7 Morocco, 419 Moscow, 89, 413, 565 Motorola, 409 Mozambique, 670 multinational(s), 131, 216, 220, 369, 415; corporations (MNCs), 122, 132, 415, 674, 720, 722, 724–​5; enterprise(s) (MNE/​MNEs), 307, 312, 366–​8, 414; firms, 25, 348, 361, 366, 501 multi-​sectorality, 163, 167 Mumbai, 24, 100, 107, 306–​7, 313, 315–​17, 565

910   Subject Index Munich, 570 Musahar caste, 109 Muslims, 107 mutation, 214, 247–​9, 252, 841, 853 Myanmar, 8, 28 N–​( 8 + 3), 20, 24, 26, 30–​3 Namibia, 773 Nampa, 333 Nanjing, 28 nanotechnology, 153 Nantong, 28 Nasdaq Index, 7 Nash equilibrium, 166 NASSCOM, 408 National Bank of Greece, 127, 129 National Domestic Workers Alliance, 818, 820 National Resource Defense Council, 817 National Rural Employment scheme, 101 national systems of innovation (NSIs), 826–​7, 831 nationalism, 6, 9; resource, 724–​5; rise of, 144 nationalization, 720 natural capital, 14, 696, 704, 709–​13, 749–​51, 754–​7, 760–​1, 764; definition, 751; depletion, 720; exhaustible, 757; global, 761, 762; renewable, 754 natural gas, 33, 669, 675, 705, 738, 765; see also gas extraction;  shale natural resources, 1, 22, 28, 501, 591, 646, 667, 685, 696–​9, 703, 706, 717–​18, 752, 756, 758; depletion of, 797; renewable, 710, 754 natural selection, 198, 248 Nayak, P., 101 neoclassical, 231, 392, 473, 846; economics, 172, 213, 689; economists, 530; models, 544; production functions, 163; theory, 114, 146, 477, 544, 647; thought, 230; tradition, 230 Neo-​Europes, 99 neoliberal, 75, 103, 148, 168, 171, 398, 468, 470–​1, 477, 549, 612, 61–​18, 621, 667, 811, 813–​16, 820; era, 794; governments, 611; ideology, 793; neoliberalism, 34–​5, 101, 170, 477, 800–​1, 813–​14, 821; hyper-​neoliberalism, 550; neoliberal policies, 616, 854; neoliberalization, 6, 168, 448, 451–​3, 457, 490, 618, 666;

neoliberalized, 470, 472, 474, 619; see also liberalism Netherlands, 5, 42, 46, 48–​52, 58, 120, 430, 489–​90, 505, 543, 546, 560, 563, 565, 567, 571, 594, 622 network: dynamics, 218–​19, 387–​8, 438; embeddedness, 386–​7; theory, 188, 393, 454 networking, 127, 216, 222, 355–​6, 360, 819; infrastructure, 269 Nevada, 335 New Deal, 467, 629 new economic geography (NEG), 19, 35, 147, 214, 368, 374, 552, 717, 850 new growth, 115, 804; markets, 616; paths, 219, 220, 223; platforms, 107; strategies, 612; theories, 152, 250; trajectories, 131 New Jersey, 570, 634, 673 New Orleans, 817 New York, 29, 48–​50, 52, 72–​3, 89, 274, 307, 313, 317, 325, 333, 349, 547, 551, 560, 564–​7, 571, 578, 592, 619, 634, 842; New York City, 797, 818 New Zealand, 49–​50, 99, 456, 685 Newark, 73, 333, 842 Newcastle, 546 Newtonian, 162 Nigeria, 86, 419, 773; Nigerian, 485; Nigerian, oil production, 720 Nikonian, 295 NK framework, 253; NK model, 251 Nollywood, 307 non-​governmental organizations (NGOs), 222, 450, 455, 724–​5, 812, 817, 833 non-​renewables, 703–​4, 709, 711–​13; depletion of, 711 Nordisk Film, 314 North America, 19, 21, 26, 238, 316, 317, 433, 456, 467, 471, 488, 493, 706, 719, 740, 792 North American, 56, 232, 328, 487 North American Free Trade Agreement (NAFTA), 374 North Carolina, 525 North Sea, 711 Northern Rock, 546 Norwalk, 73 Norway, 49–​51, 58, 120, 315, 491, 720; Norwegian, 547 nuclear, 675, 705, 708

Subject Index   911 Oakland, 333, 356 obesity, 605; epidemic, 593–​4 offshore, 13, 311, 410, 416; financial centres, 576; jurisdictions, 558, 560–​3, 565, 567, 571; resources, 706; sourcing, 410 offshoring, 129, 131–​2, 307, 312, 375, 399, 407, 409–​10, 415 oil, 22, 671, 675, 690, 703, 705, 707–​8, 711–​12, 715, 720–​1, 723, 734–​7, 739–​40, 751–​2, 754, 765; crises, 69, 757; deposits, 708; extraction, 736, 738; fields, 740–​1; firms, 738; foreign, 33; futures, 648; independence, 706; industry, 736, 741; markets, 646; pipelines, 834; price volatility, 645; prices, 646, 723, 737, 740; production, 738, 741, 742; regions, 735; reserves, 707; resources, 705; shock, 707; supply, 705–​6; transport of, 737; wells, 705 Olympia, 333 Oman, 100 online shopping, 440, 442 Ontario, 637 organic finance, 13, 592–​3, 596, 599, 600–​3, 605; see also finance capital; organic foods, 592–​5, 601; organic investors, 592–​3, 595, 601 Organisation for Economic Co-​operation and Development (OECD), 50, 55, 65, 74, 305, 374, 382, 384, 396, 398–​9, 428, 442, 488, 490, 494, 567, 633, 635, 749, 751, 770–​3, 775, 777, 785–​6 Organization of the Petroleum Exporting Countries (OPEC), 707 O-​Ring model, 351 Orissa, 106, 107, 109 Orlando, 817 outmigration, 120, 129, 131, 673–​4; see also migration outsourcing, 270, 338, 375, 384, 386, 390, 394, 399, 408, 410–​12, 417–​19, 490, 602, 832; firms, 272; global, 415, 418–​20; global service, 419; micro, 492 ownership (O) advantages, 370–​1, 373; ownership–​location–​internalization (OLI) paradigm, 367, 369–​72, 376–​7 Oxfam, 41, 65, 571 Oxford, 11, 634; Oxford University, 58, 339 Oyu Tolgoi mine, 725

Pacific, 30, 104, 433, 565; Western, 34 Pacific Economic Cooperation Council, 399 Pakistan, 20–​3, 89, 316–​17, 453 Palma ratio, 41 Palo Alto, 356, 358 Panama, 567, 571 Paris, 54, 221, 222, 307, 565, 571 Paris Agreement, 6, 665, 675 Parkinson disease, 295 Pascua Lama project, 724 patents, 98, 333, 339, 348, 355–​6, 361, 558, 763; patent data, 361; patenting, 325, 332–​4, 337, 339, 354, 356, 359 path dependence/​dependency, 9, 29, 113–​14, 116, 122, 127, 130–​1, 133–​5, 180, 190, 214, 219–​20, 222–​3, 237, 261, 387, 393, 577, 670, 687, 733–​4, 741, 762, 850–​3; historical processes, 214; process, 216; theories, 214, 851 Pax Americana, 25 peak oil, 752; theory of, 705–​9, 713 Pearl River Delta, 5, 29 peasants, 25, 80–​2, 90 Pebble Project, 724 Peloponnese, 124 Pennsylvania, 73, 525, 738–​9 pension(s), 44, 47, 107, 134, 591, 604, 616; benefits, 456; defined-​benefits, 543; funds, 450, 542–​3, 591, 595–​6, 598, 602, 604–​6, 614, 616, 621, 623, 634–​6, 645, 649, 652; occupational, 490; plans, 71, 604, 613; state, 201 People’s Bank of China, 586 periphery, 113–​14, 117, 120, 126, 259, 356, 368, 524, 715–​16, 721, 723, 725; centre–​periphery, 214; countries, 726; economic, 120; economies, 135; Eurozone, 130, 134; new, 715; peripherality, 114, 122–​3, 130–​1, 133, 135; peripheralness, 720; resource, 14, 726 Peru, 565, 724 PetroChina, 721 Pew Research Center, 65, 70–​1 Pforzheim, 547 Philadelphia, 525–​7, 817 Philippines, 20–​3, 25–​8, 30, 32, 34, 89, 317, 408, 436–​7, 456, 565, 665 Phoenix, 842–​3 PICO National Network, 801 Pireaus, 134

912   Subject Index planetary boundaries, 750, 755, 761 Poland, 130, 434, 436, 489, 491, 565, 708 polarization, 13, 120, 133–​4, 486, 489, 495, 670, 772;, income, 495; societal, 775 pollution, 30, 348, 668, 695, 706–​7, 709–​10; air, 30, 349; dust, 725; polluting behaviour, 231; reduced, 750; water, 30, 695; see also water Pope Francis, 41 population, 3–​4, 8, 19, 29, 41–​4, 46–​7, 53, 69, 72, 81, 84–​9, 91, 104, 106, 120, 129, 144, 147, 150, 154, 209, 248–​9, 251, 255, 347, 349, 418, 505, 703, 706, 710, 712, 736, 738, 754, 774–​8, 795, 831; changes, 254; Chinese, 772; collapse, 707; data, 92; density, 718; distribution, 44; Earth’s, 591; explosion, 755; growth, 83, 601, 670, 707, 757, 846; growth rate, 102; increases, 758; local, 91; national, 88; rural, 81, 90; size, 91; statistics, 92; urban, 81, 83–​4, 86–​8, 90, 93, 812; US, 273, 331 pornography, 277, 586 portfolio theory, 592, 594 Portland, 328 Portugal, 47, 49–​50, 58, 113, 119–​20, 122, 130–​1, 133, 135 poverty, 31, 41, 73, 82, 97–​8, 104, 106, 114, 134–​5, 144, 331, 477, 511–​12, 667, 672, 719, 750, 774, 777–​8, 780, 794–​6, 800–​1, 826, 829, 846; alleviation of, 771; asset, 802; child, 57; income, 104, 106, 108; line, 103–​4; rates, 666; reduction, 103–​4, 107–​9, 800; rising, 120; thresholds, 64; trap, 670; urban, 24 Prebish–​Singer hypothesis, 722 precariat, 495; precariousness, 13, 486–​9, 493–​5; precarity, 46, 478–​9, 486, 493, 668 price formation, 653, 655, 657–​9; commodity, 646, 658; theory of, 204 private sector, 603, 638–​9, 784, 800, 813, 815, 818 privatization(s), 133–​4, 470, 477–​9, 568, 635–​7, 811–​16, 819, 841, 854; uneven, 812 privatized, 477, 636, 638, 803; newly, 635 producer–​customer, interaction, 289–​90, 299; vertical relation, 291 product differentiation, 146, 182, 373; product life cycle (PLC) model, 370 productivity, 22–​3, 31, 34, 85–​7, 90, 98, 115, 127, 135, 143, 150, 152, 161, 163–​4, 249–​51, 255, 270, 273, 275–​6, 334, 347, 349, 352, 411, 468,

500–​1, 503, 506, 509–​10, 532, 684, 698, 736, 754, 759, 779, 792, 794, 799, 828, 846–​8; asset, 638; benefits, 349, 351; definition, 263; economic, 633; factor, 147, 629, 640; firm, 263; gains, 413; labour, 467, 505; pressures, 493; regional, 510; research, 275; TFP, 101 professional services, 181, 183, 637; clusters, 184 profits, 32, 81, 129, 149, 163–​4, 166, 374, 433, 577, 598, 656, 698, 734, 738, 795, 834; corporate, 559; for-​profits, 819; non-​profits, 810, 815, 819 property rights, 100, 103, 105, 107, 145, 232–​3, 352, 637, 638, 683, 696–​8, 754, 830; private, 751 pro-​poor growth, 776–​8, 780; definition, 776 prosperity, 12, 51, 55, 109, 113, 134, 136, 144, 148, 154, 168, 230, 331–​2, 455, 727, 750, 764, 792, 794, 826–​9, 831–​5 prospika system, 82 protectionism, 6, 65, 80, 87 Provincial Darwinism model, 108 public administration, 122, 123–​5, 148 public sector, 6, 74, 148, 151, 154–​5, 470, 477, 568, 638–​9, 800, 811–​17, 820–​1; ownership, 639; public–​private partnerships, 784, 810 Pune, 413 Punjab, 102, 104, 106 Purchase, 565 Qatar, 47 Qinhuangdao, 28 Quebec, 495, 637, 819 race, 73, 162, 197, 474, 477, 487, 511, 541, 546, 614; racialization, 487; racialized, 476–​7, 486, 488 Rajasthan, 106 Rajput caste, 107 Rakuten, 441 Raleigh–​Durham, 525–​7 Ranchi, 107 rationality, 197–​8, 207, 209, 616; bounded, 199, 214–​15; external, 199; human, 199; idealized notions of, 207; internal, 199; labour market, 164; levels of, 198; test of, 198 Reaganomics, 469; see also Reagan, Ronald

Subject Index   913 real estate, 32, 120, 122, 127, 129, 131–​3, 135, 558, 560, 564, 671, 737, 740 recession(s), 32, 63, 72, 116, 119, 122, 134–​5, 331, 491, 542, 549, 618, 758, 842–​3, 853–​4, 857; economic, 455, 853; global, 90, 815; great, 6, 64, 70, 73; national, 849; post-​recession, 488; prolonged, 113, 127, 130, 134–​5 recycling, 30, 34, 397, 802; rates, 760 Redwood City, 356, 358 regional branching, 214, 223; regional resilience, 223, 665, 844, 849, 851–​4, 856 Regional Studies Association, 11, 495 regionalism, global, 369; metropolitan, 811; social movement, 801 regulation, 6, 13, 26, 64, 68, 115–​16, 131, 133, 135, 148, 236, 277, 312, 392, 437, 439, 443, 448–​51, 453, 456–​7, 466, 469–​70, 473, 478, 486–​7, 491–​4, 551, 570–​2, 584, 586, 601, 635, 637–​8, 659, 696, 735, 741, 812, 819; financial, 557, 567, 587, 622; government, 694; industry, 571; labour-​market, 487; market, 739; national, 737–​8; new, 570; of exchange rates, 568; self-​, 569; social, 471; social and economic, 635; theories, 472 Regulationists, 115 regulatory systems, 32, 637; regulatory transformation, 465, 472–​4 relational, 162, 184, 209, 217, 295–​6, 383, 385, 392–​3, 399, 417, 451–​2, 469, 621, 672, 674, 848; relational, accounts, 97; relational, approach(es), 10, 179, 180, 181, 184, 186, 189, 635; economic geography, 9, 205; know-​who, 183; perspective(s), 12, 184, 188–​9, 428; positions, 188; research, 179; research design, 180, 187, 189–​90; thinking, 187 renewables, 703–​5, 709–​11, 757–​8; renewable energy, 694, 709, 716, 742, 757, 760; renewable resources, 695, 704, 760 reregulation, 437, 438, 443, 479, 493; see also deregulation; regulation research and development (R & D), 148, 151, 251, 292, 299, 332, 334–​8, 376, 413–​15, 469, 709, 763, 799 resilience, 9, 14, 114, 116, 127, 132, 136, 223, 246, 260, 263, 296, 331–​2, 665, 667, 669–​70, 675–​6, 817–​18, 839–​40, 843–​5, 847–​54, 856–​60; adaptive, 855; building, 858; definition,

844–​6, 852; determinants of, 855; ecological, 845; ecological model of, 847; economic, 854, 857; model, 846; perverse, 859; regional economic, 859 Resolution Foundation, 56 resource curse, 66, 720, 725–​6, 735, 738–​9; theory, 719 resource peripheries, 14, 715–​19, 721–​8 restructuring, 13, 113–​14, 116, 122–​3, 126, 130–​1, 133–​5, 154, 269, 315, 465–​7 1, 473, 478–​9, 794, 851; approach, 470, 479; capital, 475; corporate, 474, 561; economic, 116, 795, 800; industrial, 69, 72, 115, 122, 466, 475; of retail banking, 619; public services, 478; sectoral, 135; studies, 469, 471; supply network, 438; systemic, 256 retail capital, 427, 432, 435, 444; retail revolution, 427 retirement, 71, 115, 201, 492, 611, 616 Rio Tinto, 721 risk aversion, 202, 651, 654 risk-​taking, 151, 154, 203, 612, 616 Riyadh, 565 robustness, 44, 101, 392, 395, 828, 833, 835, 845, 848 Rockefeller Foundation, 419, 818, 855 Romania, 120, 122, 415 Rome, 307 Rotherham, 842 Rothschild, 566, 567 Rotterdam, 551, 669 Round Rock, 333 Route 128, 235, 259, 413 Rowntree, 449 Royal Academy, 544 Royal Bank of Scotland, 571 Royal Dutch Shell, 721 Ruhr Valley, 235 rule of law, 26, 100, 233 runaway shops, 467, 471 rural migrant workers (nongmingong), 82 rural sourcing, 418, 419 Russia, 4, 20, 75, 86, 89, 561–​2, 565, 650, 708, 721, 723, 735; Russian, 163, 544; see also Soviet Union;  USSR Sahara, 707 Salt Lake City, 817

914   Subject Index Samsung, 335 San Francisco, 7, 8, 274, 328, 333, 356, 358, 565, 570 San Francisco Bay Area, 7, 155, 356, 531 San Jose, 328, 333 San Marino, 47 San Precario movement, 487 San Ramon, 356 Santa Clara, 356 São Paulo, 413, 565 satisficing, 199 Saudi Arabia, 565 savings, 24–​5, 66, 70–​1, 545, 549, 586, 602, 604, 612, 614, 617–​18, 635–​6, 649, 659, 802–​3; and loan crisis, 544; behaviour, 208; savings, cost, 412, 813; labour cost, 410 Scandinavia, 42, 115, 315, 793 schistosiamsis, 98 Schlumberger, 741 Schumpeterian, 300, 852; entrepreneurship, 248, 256; momentum, 287; theory, 851; view, 150 Schwarz, 429, 430 Scotland, 143, 488, 735, 819–​20; see also Britain; United Kingdom Scottish National Party, 820 sea-​level rise, 665–​6, 668–​70, 675 Seattle, 328, 333; Battle in Seattle, 473 securities, 558–​60, 563–​4, 566, 578, 594, 652, 658; firms, 564–​6, 570 Securities and Exchange Commission (SEC), 598, 740 securitization, 542, 546, 561, 563–​4, 604, 612, 614; insurance-​linked, 673 segmentation, 179, 275, 288, 360, 476, 486; economic, 512; labour-​market, 487 segregation, 472; class, 511; economic, 512; geographical, 511; occupational, 476; socio-​ economic, 500; Selby, 842 Seoul, 11, 390, 565 Seoul National University, 208 service class, 504; service providers, 306, 408–​11, 413, 415–​20, 565, 595; service sector, 22–​3, 33, 88, 466, 471–​3, 511 Service Employees International Union, 529 Seven & I, 430, 443 sexual revolution, 8

SFMade, 818, 822 shale, 14, 707–​8, 733–​9, 741; boom, 740; development, 736–​7, 739–​41; extraction area, 737; gas, 708, 734–​6; oil, 14, 733, 736–​7, 739–​40, 759; plays, 736–​7, 740; reserves, 740; revolution, 737; see also gas extraction; natural gas Shanghai, 5, 24, 28–​9, 32, 87–​9, 565 Sheffield, 215 Shell, 450, 741 Shenzhen, 5, 28, 82 short termism, 604, 649 Sialkot, 453 Silicon Valley, 152, 235, 259, 325, 328, 335, 349, 351, 353, 356, 358–​9, 386, 413–​15, 510, 565, 570, 798 Singapore, 7, 11, 25, 27, 30, 49–​50, 58, 67, 99, 100, 307, 390, 485, 562, 565, 567, 571, 615, 623; Singaporean, 49, 614 skill acquisition, 151; skill development, 314, 520–​2, 524–​7, 529–​32; skill-​biased technological change (SBTC), 65–​8, 75 Slovakia, 434, 440 Slovenia, 47, 49–​50, 120, 489 small and medium-​sized enterprises (SMEs), 114, 122–​7, 129, 131–​2, 134, 390 Small Business Act of Europe (SBA), 126–​7, 129 Smithian, 167; see also Smith, Adam social: class, 166, 169, 316; cohesion, 39, 295; exclusion, 12, 120, 774–​5; mobility, 43, 109; overhead capital, 630; reproduction, 393, 476–​9; services, 82–​4, 90–​1, 628, 779, 802, 812, 814 Social Impact Bond, 821 socialization, 169, 296, 472, 476 software development, 25, 408, 410, 413–​14, 416 solar, 756; energy, 675, 694, 765; fuels, 757; modules, 762; panels, 694, 709; power, 5, 708, 732, 763; vs coal and nuclear, 763 Solow model, 102 Sony, 78 South Africa, 43, 416, 419, 438, 443, 453, 455–​6, 562, 565, 721, 724, 773, 784; Southern Africa, 794 South America, 416, 724; South American, 26 South Carolina, 73 South Gobi, 715; South Gobi Desert, 715, 723

Subject Index   915 South Pacific, 669 sovereign: debt, 113, 118, 131–​4, 552; funds, 591, 596; wealth funds, 168, 572, 614, 623, 635–​6, 645, 711 Soviet, 82, 86; Soviet Union, 754; see also USSR space economy, 159, 161–​6, 168–​7 1, 189, 854 Spain, 7, 44, 47–​53, 56, 58, 86, 113, 117–​20, 122, 129–​30, 133, 135, 430, 485–​6, 489, 494, 565, 631; Spanish, 47, 99 spatial, 512; equilibrium, 146, 152, 350, 352, 509, 850, 851; impossibility theorem, 166; spatiality of finance, 9, 557 Special United Nations Fund for Economic Development (SUNFED), 630 specialization, 123, 129, 131–​2, 151, 167, 180, 185, 187, 189, 219, 250, 259, 326, 328, 332–​3, 339, 409, 798, 827; dual specialization correlation (DSC), 333; economic, 123; functional, 88; geographical, 383; horizontal, 859; increased, 275; industrial, 114–​16, 122, 130–​1, 134–​5, 384; knowledge-​ intensive, 135; narrow, 249; of cities, 274; sectoral, 126, 130; shallow, 127; smart, 223; spatial, 116, 120; strong, 127; vertical, 390 speciation, 246, 256, 261, 264; industry, 261 speculation, 32, 579, 647–​8, 651, 653, 656–​7, 659, 739; definition, 653; land, 740; thesis, 656 Springfield, 333 Sri Lankan, 455 stabilization, 636–​7, 853 stakeholders, 217, 450, 598, 783, 787, 796, 801, 830 Stalinist, 80–​1 Stamford, 73 Standard & Poor’s, 740 standard employment relationship (SER), 485–​8, 490, 492, 495 State Street, 565 status quo bias, 200, 203 STEM occupations, 334, 337 Stern Review, 6, 750; see also Stern, N. stock-​flow consistent approach, 549 Stockholm, 565 store expansion, 442; international, 444 strategic coupling, 10, 383, 385, 387–​91, 394, 398, 561

structural adjustment, 135; policies, 494; programmes, 434 structuralism, 162; post-​structuralism, 478 structuration, 162, 190 subcontracting, 120, 129, 131–​2, 352, 384, 390, 471, 529 subprime: credit facilities, 620; crisis, 6–​8, 12, 540, 544, 546, 550–​1, 569–​70, 612; lending, 13, 540, 542–​3, 546, 551; mortgage crisis, 606, 614 sub-​Saharan Africa, 419, 436 substitutability, 751, 758–​60, 764 Summer Institute in Economic Geography, 11 sunk cost(s), 289, 311, 314, 317–​18, 637, 733–​4 super rich, 39, 42, 45–​8 supply and demand, 708; curve, 688; law of, 231, 465, 614, 645, 652–​8, 688–​9 supply chains, 432, 453, 469, 561, 666, 671, 762, 856; domestic, 28; financial, 543; global, 25, 236, 278, 420, 449, 451, 454–​5, 576, 671, 831; local, 23, 434; management of, 451; resource, 683 sustainability, 2–​3, 5, 8, 10, 12, 55, 119, 134, 456, 594, 596, 603, 612, 636, 659, 704, 716, 732, 749, 772, 780–​1, 783, 786, 815, 817–​18, 840, 845; environmental, 724; social, 557; see also under development Suzhou, 28 Sweden, 46, 49–​51, 120, 315, 430, 491, 505, 565; Swedish, 452 Switzerland, 49–​51, 99, 560, 562, 565, 567; Swiss, 543, 614 Sydney, 317, 565 T. Rowe Price Group, 565 Tacoma, 333 Taft–​Hartley Act, 524 Taipei, 390, 565 Taiwan, 22, 24–​5, 389–​90, 434, 440, 565 Tamil Nadu, 102, 104, 106 Tangshan, 28 Tanzanian, 454 tax(es), 47, 56, 64, 75, 495, 689, 711, 751, 759, 814; avoidance, 586; corporate, 101; evasion, 569; favouritism, 800; havens, 567, 571; sales, 277

916   Subject Index taxation, 57, 64, 75, 129, 571, 629, 636, 827; low, 567 Tech Shop, 818 Tel Aviv, 565 Tennessee, 73 territorial dynamics, 385, 387–​8; intra-​, 391 territoriality, 7, 395 Tesco, 429, 430, 440–​2 Tesla, 335, 763; Tesla Motors, 335 Texas, 410, 561, 721, 735–​6, 739 Texas Instruments, 410 Thailand, 20–​3, 25, 27–​30, 32, 34, 389–​90, 434, 436–​8, 444, 565, 671, 685, 772 Thatcherism, 469; see also Thatcher, M. thermal energy, 694; see also energy transitions Thessaly, 124 Tianjin, 28–​9 time and space, 74, 197, 200, 203, 205–​6, 209, 545, 630, 821 Tmall, 441 Tokyo, 5, 24, 29, 89, 307, 565 Toronto, 307, 565 Toshiba, 78 tourism, 120, 122–​5, 132, 134–​5, 305, 451, 671, 697, 736, 738 Toyota, 763 trade, 1, 3–​5, 8, 11, 25, 27–​31, 33–​6, 64, 67, 101, 120, 122–​4, 126–​7, 129, 131–​2, 135, 215, 230, 245, 270, 278, 312, 398, 524, 549, 559, 562, 578–​9, 587, 614, 647–​9, 652, 654, 656–​7, 685, 697, 718–​19, 826–​7, 831, 834; agreements, 434, 637; associations, 857; barriers, 437, 850; cocaine, 576; costs, 146, 373–​4; credit, 566; cross-​border, 452; ethical, 452, 454–​5; fairs, 187, 189, 218, 299; flows, 74; free, 163, 167; gains, 275; global, 667, 720–​2, 795; in finished goods, 394; in intellectual property, 832; in services, 279; internal, 797; international, 373–​5, 382, 386, 427, 568; intra-​firm, 382; liberalization, 65, 147, 449–​50; mysteries, 245, 263; networks, 83, 716; policies, 65, 75, 667, 735; promotion, 324; retail, 123, 124, 125, 135; sanctions, 761; South–​North, 427; South–​South, 427, 456; terms of, 104; theory/​theories, 368, 383; value-​added, 396; wholesale, 123, 125, 135

Trans Pacific Partnership (TPP), 36 transnational corporation(s) (TNC/​TNCs), 10, 382, 387, 390, 428, 434–​8, 450, 561–​2, 567, 735; retail, 431–​2, 434–​40 transnational retailing, 429, 443 Transparency International, 32, 100 transport/​ation, 81, 133, 161, 163–​6, 172, 185, 270–​1, 309, 312, 349, 356, 359, 416–​17, 501, 510, 575, 634, 717, 726, 732, 738, 741–​2, 797, 814; costs, 143, 165, 253, 269–​70, 276, 278–​9, 373–​4, 394, 499, 717, 850 trickle down effect, 43, 771, 775, 792, 800–​1 troika (IMF/​EC/​ECB), 113, 119, 133, 135; see also European Commission (EC); European Central Bank (ECB); International Monetary Fund (IMF) Tropic of Cancer, 99 tropics, 98 trypanosomiasis, 98 Tulsa, 410 Turkey, 490; Turkish, 544 Typhoon Haiyan (Yolanda), 665 Uber, 279, 492 Ukraine, 86 Ulaanbaatar, 715 unemployment, 83, 90, 114–​15, 117–​18, 130, 132, 134–​5, 150, 197, 331, 489, 524, 568, 614, 620, 780, 795, 800, 846, 848; benefit system, 853; local, 841; long-​term, 117–​18; youth, 114, 118 UNESCO, 305 union(s): anti-​union, 467, 474, 493; de-​ unionization, 529; non-​unionized, 68, 476, 493; unionism, 473–​5; unionization, 73; unionized, 524, 794; see also labour/​trade union(s) United Arab Emirates, 47, 565 United Front, 101 United Kingdom (UK), 1, 5, 7, 9, 34, 40–​59, 86, 99, 115, 203, 215, 234, 237, 310–​11, 315–​17, 355, 391, 430, 433, 440–​2, 449, 452–​3, 455, 467–​70, 487–​91, 493–​5, 541–​3, 546, 551, 564–​5, 567, 570–​1, 616–​19, 622, 637–​8, 640, 673, 737, 740, 793–​4, 811, 813, 819–​22, 841, 843, 857–​8 United Nations (UN), 5–​6, 30, 50, 89, 119, 451, 456, 630, 703–​4, 751, 855

Subject Index   917 United Nations Conference on Trade and Development (UNCTAD), 307, 369, 375, 382, 384, 396, 398, 399, 407, 645–​6, 648, 650–​2, 654–​5, 658, 660 United Nations Development Programme (UNDP), 65–​6, 74, 106, 785 United Nations Economic Commission for Europe, 631 United Nations Environment Programme (UNEP), 30, 749 United Nations Framework Convention on Climate Change, 6 United States Geological Survey (USGS), 753, 754 United States of America (US/​USA), 1, 4–​6, 8–​9, 12, 25, 27, 29, 33–​6, 40–​4, 46, 48–​55, 57–​8, 63–​75, 86, 89, 99, 149, 151, 154, 164, 168, 197, 202, 215, 220, 234, 254, 259, 263, 273, 278–​9, 306, 310–​11, 317, 325–​6, 328, 331–​4, 336, 339, 352–​3, 355, 362, 383, 408, 410, 413, 415–​16, 419, 428–​31, 433, 441, 443, 450, 452, 467–​8, 470, 489, 492–​4, 502, 505, 509–​11, 519, 525–​6, 531, 541–​2, 544, 546–​9, 551, 561, 564, 566–​7, 570–​1, 593, 598, 604, 613, 616, 618–​20, 622–​3, 650, 653, 665, 668, 672, 684, 694, 705, 707–​8, 720, 733, 735–​41, 753, 759, 762, 774, 792–​5, 801–​2, 811–​13, 815–​20, 822, 833, 842–​3, 859 upward mobility, 795, 801–​3 Urban Manufacturing Alliance, 818 urban planning, 14, 632, 839 urbanization, 4–​5, 7, 24, 29–​30, 71, 78–​9, 83–​6, 90–​2, 146, 308–​9, 314–​15, 317, 330, 361, 506, 572, 601, 716, 774, 781, 797, 803; advantages, 184; economies, 798; hyper-​, 31; incomplete, 83; literature, 91 Uruguay, 832 US Cluster Mapping Project (USCMP), 326, 331–​2, 339 use value, 685, 689–​93, 695–​9 user–​producer interaction, 287–​8, 293 USSR, 719; see also Russia; Soviet Union utilitarianism, 149 Uttar Pradesh, 104, 106 UTV, 312 value chain(s), 28, 33–​4, 180–​3, 286, 298, 332, 335–​8, 369, 375–​6, 384, 398, 456, 601; geographical concentration of, 275;

linkages, 183, 335; rent theory, 453; see also global value chain(s) (GVC/​GVCs) value shop framework, 183 Vancouver, 565 Venezuela, 735 venture capital, 234, 315, 543, 565, 577, 601; urban, 309; venture capitalists, 543, 596 Vietnam, 8, 20–​3, 26–​8, 30, 33, 34, 437, 453, 565; Vietnam War, 8, 68–​9, 707 virtual currencies, 575, 580–​3, 585, 587 vulnerability, 36, 90, 113, 122, 127, 130, 133, 207, 332, 579, 665–​8, 670, 674–​6, 716, 721, 830, 844, 848–​9, 854–​5; climate, 666; definition, 675; economic, 114, 125–​6, 134–​5, 487, 666–​7, 672–​3; environmental, 725; household, 672; of Asia, 32; regional, 116, 667, 671, 673; regional economic, 668–​9; research, 675; risk of, 848; to flooding, 670; to shocks, 843; to storms, 669 wage(s), 65, 69–​70, 90, 150, 164–​6, 168–​9, 276, 279, 334, 347, 349, 351–​3, 356, 412, 418, 455–​6, 465, 472, 476, 485, 503, 506, 512, 520–​1, 527–​8, 531–​2, 669, 674, 690, 759, 780, 793, 795, 798–​9, 803; adequate, 489; adjustments, 67, 134, 149; appropriate, 163; bargaining, 567; better, 532; bill, 68; competition, 67; compression, 519; curve, 349, 351–​2; demands, 528; differentials, 33; divergence, 31; falling, 68; fixed, 165; flexible, 119, 134; gains, 273; growth, 273; high, 54, 116; higher, 505, 520–​1, 523–​4, 530, 532; hourly, 164; increases, 799; industrial, 92; inequality, 511; laws, 72; living, 473, 800–​1; local, 273; low, 56, 67, 83, 133, 494, 511, 520, 525, 527–​33, 672, 794, 798, 802; medium-​level, 131; minimum, 64, 68, 72–​3, 456, 494, 531–​2, 793, 801; price spiral, 568; rates, 27, 67–​8, 73, 83; real, 165, 478, 619; reductions, 119; regional, 506; rising, 794; social, 477; stagnant, 65, 75; undercutting, 525; wage/​productivity curve, 349, 351, 353; wage/​productivity relationship, 351 Wales, 488; see also Britain; United Kingdom Walgreens Boots Alliance, 429 Wall Street, 34, 349, 351, 359, 492, 548, 595, 605 Walmart, 276, 429–​30, 439–​40, 452 War on Terror, 722

918   Subject Index Warsaw, 565 Washington, 197, 398, 473; Washington, DC, 565, 820 Washington Consensus, 197, 398, 473 waste management, 451, 684 water, 30, 134, 161, 330, 417, 628–​30, 640, 665, 669, 671, 683, 689, 695, 697, 699, 706, 710, 725, 727; availability, 724; clean, 723; contamination, 724; scarcity, 716; security, 330, 724; surface, 737; see also pollution wealth, 5–​6, 31–​2, 40–​2, 46, 51, 58, 63–​4, 69–​75, 144, 147, 168, 273, 331, 477, 506, 511, 599, 615, 668, 751, 800, 802–​3, 829; accumulation, 71; advantages of, 772; creation, 612; definition, 69; distribution of, 70–​1, 152, 545, 724; family, 560; geography of, 270; global, 721; inequality, 495; inherited, 794, 798; management, 620; national, 656; polarization, 495; resource, 720, 726, 727; societal, 230; see also under inequality welfare, 14, 56, 81, 84, 85, 114, 120, 152, 198, 450, 470, 479, 616, 771; agencies, 208; benefits, 612; costs, 826; dependence on, 841; economic, 150; human, 145; policies, 620; programmes, 857; shortages, 119; social, 82, 85, 198, 512, 804, 812; spending, 119; stagnating, 149; state, 120, 131urban, 82; see also welfare state welfare state, 471, 477, 492, 568, 618, 793; see also welfare Wellington, 307 Wells Fargo, 7 West Bengal, 102, 106 West Conshohocken, 565

White House, 579 Williams College, 54 wind power, 5, 732, 763 Wipro, 408, 410 Witbank, 724 working class, 143, 504, 613; traditional, 502 World Bank, 2, 4, 11, 30, 33, 36, 41, 91, 104, 106, 120, 145, 152, 374, 382, 398, 545, 591, 630, 710, 721, 749, 751, 770, 775, 785–​6, 796, 800, 854 World Economic Forum, 65, 417, 418 World Free Zones Organization, 399 World Trade Organization (WTO), 67, 382, 384, 396, 398–​9, 437, 561, 832 World War I, 314 World War II, 6, 25–​6, 67, 311, 314, 495, 567, 632, 707, 792 World Wide Web, 269 Wuhan, 82 Wuxi, 28 Wyoming, 74 xenophobia, 6 Xi Jinping, President, 33 Xstrata, 727 Yadav caste, 107 Yangzhou, 28 Yorkshire, South, 215 Zambia, 1, 784 zero-​hours contracts, 487, 495 Zhenjiang, 28 Zhuhai, 5, 29 Zürich, 565