The Oxford Handbook of Africa and Economics: Context and concepts 9780199687114, 0199687110

Citation preview

T h e Ox f o r d H a n d b o o k  o f

A F R IC A A N D E C ON OM IC S Vo l u m e 1 Context and Concepts

The Oxford Handbook of

AFRICA AND ECONOMICS VOLUME 1

Context and Concepts Edited by

C É L E ST I N  M O N G A and

J U ST I N Y I F U  L I N

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 2015 The moral rights of the authors‌have been asserted First Edition published in 2015 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: 2015936455 ISBN 978–0–19–968711–4 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.

Acknowledgments

When asked by a journalist why he writes books and even manages to get other things done along the way, Wole Soyinka, Africa’s first Nobel Prize in literature, simply replied:  “the masochist in me, I  suppose.” The editors of this two-volume Oxford Handbook of Africa and Economics have not done nearly as much writing as Soyinka. Still, they can relate to his rationale for spending many long hours and sleepless nights among papers and books, trying to bring a complex but exciting task such as this one to fruition. No book of any scope gets done without deep involvement of a very large number of people, most of whom remain anonymous. By definition, a handbook of (nearly) encyclopedic ambitions and size such as this one necessarily required a lot of help from our friends. While we cannot recognize all of those who gave much-needed support, we wish to acknowledge the advice, encouragements, and guidance of a number of very special people and institutions. The idea for this book arose out of a couple of firm convictions. The first was that despite being one of the most intellectually challenging regions in the world Africa was still under-researched—or some of the cutting-edge work done about it was not sufficiently featured in the most prestigious reference books at major university presses. The second was that, despite all the media coverage of the continent (perhaps because of it), Africa’s contribution to economic knowledge had not been told. What was missing and might be of enduring value, we thought, was a kaleidoscopic presentation of some of the various ways in which economic science has tried to explain the African condition, and how the continent’s peculiarities had shaped the field of economics. Strangely, our friends Alexie Tcheuyap, Cilas Kemedjio, and Ambroise Kom, who are professors of comparative literature and philosophy—not economics—were the first to enthusiastically embrace the idea of a handbook, and to suggest the kind of intellectual strategy that could underline it and make it unique. We are grateful to their wonderful intuition and pertinent advice, which gave us the energy and confidence to develop the initial ideas into a proposal. Hippolyte Fofack, Jean-Claude Tchatchouang, Eugène Ebodé, Hervé Assah, Rémi Kini, and Geremie Sawadogo were cheerleaders from day one, and constantly reminded us why it was intellectually and symbolically important that the first handbook on Africa and economics ever produced by Oxford University Press in its 600 years of history be comprehensive and diverse, and reflect perspectives that are not always given the visibility they deserve. Paul Collier’s searching criticisms of the initial proposal forced us to clarify our argument and approach. His generous advice and marginalia at each step of the process provided a great sense of direction that helped us avoid many pitfalls. Managing top-notch researchers to deliver on a complex project in a very short period of time is a bit like herding cats. Fortunately, we had a blueprint on how to proceed with

vi   Acknowledgments minimum risks of slippage on the timetable and budget. David Malone, with whom we worked on a previous Oxford University Press project, offered strategic and tactical advice from the very beginning of the process. We were fortunate to tap into his wisdom and expertise. Elizabeth Asiedu and Mustapha Nabli could easily have been co-editors of this handbook. Their extraordinary generosity and commitment make it difficult to express adequately the critical role these friends have played in the development and completion of this project. With her well-known toughness and brilliance, Asiedu single-handedly brought to the team several high-flying researchers who rarely have time for such intellectual endeavors. Among other things, she convinced many of her colleagues from the Association for the Advancement of African Women Economists (AAAWE) to lend support to the handbook, bringing unique perspectives to economic topics where their voices are not often heard. Despite a schedule as busy as that of heads of state, Nabli provided detailed and sharp comments on various drafts of the main chapters of the handbook and generously opened his rich Rolodex to help us recruit contributors. Thanks to his stellar reputation across the entire Arab world, we were able to attract many of the top researchers in his network of friends, former colleagues, and students to this task. One of them is Hakim Ben Hammouda, who co-wrote two chapters with Nabli while working as Finance Minister in the transition government of Tunisia at the most difficult period of his country’s history. We thank him for that. Fabien Eboussi Boulaga’s and Valentin-Yves Mudimbe’s amazing analytical abilities, and their sustaining goodwill and humor, made the preparation of this handbook an exciting task carried out with some of the most fertile minds in the world. Prior to their intellectual work in the realm of philosophy, Africa has suffered from the lack of sophisticated scholarly attention. This handbook, in many ways, was produced out of our profound regard for their pioneering work on the epistemology and ontology of the continent. Joseph Stiglitz and Roger Myerson not only contributed chapters to the handbook but provided excellent advice and suggestions on its main themes. Stiglitz, whose theoretical work on a wide range of issues addressed in this handbook, deserves a special expression of gratitude for his friendship and mentorship. Olivier Blanchard, Jeffrey Sachs, Richard Joseph, Xiaobo Zhang, and Akbar Noman, colleagues and kindred spirits with whom we have interacted about various economic topics, also helped shape this book by recommending potential contributors and convincing them to be part of the journey. A team of 137 contributors from dozens of countries in all continents worked hard to complete this two-volume handbook in record time. Many of them command great attention because of their reputation, and normally would only be contacted through highly paid agents. Yet, they all agreed to work on this project without requesting honorarium or fees, with the humility which is the other side of true greatness. Our debt to all the colleagues and friends who accompanied us on this project is impossible to repay, or even to acknowledge appropriately. One of the many advantages of being an economist in the World Bank research department was the bountiful advice that one’s colleagues provide, even unwittingly. In addition to Kaushik Basu’s early and strong endorsement of the handbook project (and his stunning humility to contribute a chapter himself, with Alaka Basu), we benefitted from thoughtful discussions of various topics studied in this handbook with Shanta Devarajan, Marcelo Giugale, Asli Demirgüç-Kunt, Luis Serven, Aart Kraay, and Bertrand Badre.

Acknowledgments   vii The positive response received from Jim Yong Kim, the World Bank Group President, eliminated the bureaucratic impulses, processes, and red tape that almost always surround such a high-visibility project. A true intellectual himself, Kim shielded the handbook project from the politically correct: in just one cryptic email message to his senior staff, he saluted the initiative for the handbook and also gave all the intellectual freedom and latitude to explore issues and policies without ever wondering whether the contributors to the project shared “official” World Bank views on any topic under study. As President of the African Development Bank, Donald Kaberuka was not simply a gorgeous mind with genuine commitment to the economic development of the African continent and an uncanny ability to seize new ideas, he was the first to spontaneously offer intellectual and financial support for the preparation of this handbook. It did not occur to him that he might have been subsidizing a “World Bank” initiative. He simply asked what kind of help was needed and made it available, never expecting anything in return except a compelling volume. A true gentleman. We hope he will not be disappointed with the final outcome. Finn Tarp, Director of UNU-WIDER, whose intellectual engagement with Africa dates back decades, had a similar attitude, providing funding for the project and contributing himself (with Tony Addison and Saurabh Singhal) a seminal piece on the economics of development aid. Mthuli Ncube and Ernest Aryeetey, two of the most prominent and busiest African economists, also strongly supported this project and always found the time to meet the deadlines—even when fighting jet lag and multiple other requests. We are grateful for their help. Jean-Claude Bastos de Morais, an entrepreneur and investor who promised his Angolan grandmother to make a difference for the African continent, has been an unusual philanthropist in supporting young African economists and cutting-edge research and innovation. This project benefited from his many ideas and help. Jiayi Zou and Yingming Yang of the ministry of Finance in China, together with the leadership and the students of the National School of Development at Peking University, offered substantial financial and logistical help to organize an authors’ workshop in Beijing for the discussion of draft chapters. Xiaobo Zhang organized an unforgettable field visit that allowed many participants to have a first-hand look at concrete examples of entrepreneurship and industrialization in a dynamically growing economy. Xi Chen, Irene Jing Lu, Hongchun Zhao, and Premi Rathan Raj devoted time and energy to coordinate all events at every single phase of the project. We thank them for their generosity and abnegation. Madeleine Velguth and Gertrud G. Champe, who are not economists, heroically translated a few chapters, overcoming the obstacles of technical jargon to produce what we hope are polished and highly readable texts. Wei Guo was a creative and energetic research assistant, always coming up with several possible solutions to each problem. Togolese artist Akué Adoboé graciously offered pictures of his paintings for the covers of the two volumes. We are deeply appreciative of such help. The team at Oxford University Press, especially Adam Swallow, Michael Dela Cruz, Aimee Wright, Sarah Kain, and Sophie Song, patiently and effectively guided us through the process. Long after our own work was done, they were (and are) still spending time and energy to the handbook project. Last but not least, our families tolerated our often long absences—including when we were physically at home but locked in libraries. We would like to thank Chen Yunying, Kephren and Maélys, for being our inspiration.

Contents xiii xvii xix

List of Figures  List of Tables  List of Contributors 

Introduction: Africa, the Next Intellectual Frontier  Célestin Monga and Justin Yifu Lin

1

PA RT I   C ON C E P T S 1. Prolegomena to Economics as an African Science:  A Philosophical Meditation  Fabien Eboussi Boulaga

29

2. Households and Income in Africa  Kathleen Beegle, Calogero Carletto, Benjamin Davis, and Alberto Zezza

46

3. Transformation of African Farm-cum-Family Structures  Catherine Guirkinger and Jean-Philippe Platteau

59

4. The Economics of Marriage in North Africa: A Unifying Theoretical Framework  Ragui Assaad and Caroline Krafft 5. The Theory of the Firm in the African Context  Christopher Malikane 6. Markets and Urban Provisioning  Jane I. Guyer 7. Development as Diffusion: Manufacturing Productivity and Africa’s Missing Middle  Alan Gelb, Christian J. Meyer, and Vijaya Ramachandran 8. Employment, Unemployment, and Underemployment in Africa  Stephen Golub and Faraz Hayat

72 86 104

115 136

x   Contents

9. Inclusive Growth in Africa  Mthuli Ncube

154

10. Poverty: Shifting Fortunes and New Perspectives  Abebe Shimeles

175

11. Dimensions of African Inequality  Arne Bigsten

197

12. Inclusive Growth and Developmental Governance:  The Next African Frontiers  Richard Joseph

213

13. Economics and the Study of Corruption in Africa  M. A. Thomas

233

14. Thoughts on Development: The African Experience  FranÇois Bourguignon

247

15. The Idea of Economic Development: Views from Africa  Hippolyte Fofack

271

PA RT I I   M E T HOD OL O G IC A L  I S SU E S 16. Principles of Economics: African Counter-Narratives  Célestin Monga

303

17. Economics and Culture in Africa  Felwine Sarr

334

18. The Economics of Non-Cognitive Skills  Laura Camfield

351

19. Modeling African Economies: A DSGE Approach  Andrew Berg, Shu-Chun S. Yang, and Luis-Felipe Zanna

370

20. Measuring Economic Progress in the African Context  Morten Jerven

393

21. Measuring Structural Economic Vulnerability in Africa  Patrick Guillaumont

407

22. Measuring Democracy: An Economic Approach  Célestin Monga

427

Contents   xi

23. Measurement and Analysis of Competitiveness  Olumide Taiwo and Julius A. Agbor

453

PA RT I I I   H I STOR IC A L T R AJ E C TOR I E S A N D E C ON OM IC L A N D S C A P E 24. Africa’s New Economic Opportunities  Paul Collier 25. Tigers or Tiger Prawns? The African Growth “Tragedy” and “Renaissance” in Perspective  Christopher Cramer and Ha-Joon Chang

473

483

26. The Economic Legacies of the African Slave Trades  Warren C. Whatley

504

27. The Economics of Colonialism in Africa  Gareth Austin

522

28. Public–Private Interface for Inclusive Development in Africa  Olu Ajakaiye and Afeikhena Jerome 29. Natural Resources in Africa: Precious Boon or Precious Bane?  Ibrahim Ahmed Elbadawi and Nadir Abdellatif Mohammed

536

553

30. Volatility and Vulnerability  Xubei Luo

567

31. Africa’s Urbanization: Challenges and Opportunities  Maria E. (Mila) Freire, Somik Lall, and Danny Leipziger

584

32. Environmental and Climate Change Issues in Africa  Tomonori Sudo

603

33. Informality, Growth, and Development in Africa  Ahmadou Aly Mbaye and Nancy Benjamin

620

34. Capitalism and African Business Cultures  Scott D. Taylor

637

xii   Contents

PA RT I V   T H E E C ON OM IC S OF P OL I T IC A L T R A N SF OR M AT ION 35. The Impact of Democracy on Economic Growth in Sub-Saharan Africa, 1982–2012  Takaaki Masaki and Nicolas van de Walle 36. The Economics of Authoritarianism in North Africa  Raj M. Desai, Anders Olofsgård, and Tarik M. Yousef 37. The Potential Economic Dividends of North African Revolutions  Mustapha Kamel Nabli and Hakim Ben Hammouda

659 675

689

38. The Economics of Violent Conflict and War in Africa  Anke Hoeffler

705

39. The Causes and Consequences of Terrorism in Africa  Juliet Elu and Gregory Price

724

40. The Political Economy of the New Arab Awakening  Mustapha Kamel Nabli and Hakim Ben Hammouda

739

41. Democratic Decentralization and Economic Development  Roger B. Myerson

756

42. The Economics of Happiness and Anger in North Africa  Nadereh Chamlou

770

Name Index Subject Index

785 795

List of Figures

2.1 Do rural income patterns in Africa compare with others? 

55

2.2 As income grows, sources of income shift for African households 

56

5.1 The impact of an increase in the probability of outage on production 

92

5.2 The demand curve 

95

5.3 The equilibrium position of the firm 

96

5.4 A fragile equilibrium of the firm 

98

5.5 A low-level equilibrium trap 

99

7.1 Labor productivity and employment share for selected sectors, Zambia and Mexico 

121

7.2 Gini coefficient of weighted productivity distribution vs. average productivity based on macro data (upper panel); Gini coefficient of value added per worker and GDP per capita based on firm survey data (lower panel) 

122

7.3 Gini coefficients of income and consumption inequality vs. GDP per capita (1950–2011) 

124

8.1 The Lewis model of labor market dualism 

144

8.2 The Harris–Todaro model 

145

9.1 Africa’s middle of the pyramid: distribution of the African population by classes 

156

9.2 Africa Infrastructure Development Index, 2010 

165

10.1 Trends in extreme poverty by region 

176

10.2 Log per capita GDP and per capita GDP growth for Africa: 1960–2011 

183

10.3 Lowess estimate of proportion of African countries with at least one growth acceleration 

184

10.4 Change in middle class status on the basis of household wealth/asset for selected African countries during 1990–2010 

185

10.5 Multidimensional asset-based poverty and per capita GDP for selected countries in Africa 

187

10.6 Poverty in agriculture, services and industry sectors by the level of aggregate poverty 

188

10.7 Gini index for asset (wealth) and educational achievements for selected African countries 

192

xiv   List of Figures 14.1 GDP per capita in developing regions: 1960–2012 

249

14.2 GDP per capita in developing regions as a proportion of that of the USA: 1960–2012  249 14.3 Poverty in the world and developing regions: 1980–2010. Proportion of population below 1.25 ppp 2005 USD per person and per day 

250

14.4 Real price of oil: 1875–2010 

256

14.5 Real price of non-oil commodities: 1865–2010 

257

14.6 Terms of trade of African countries: 1980–2012 (2000 = 100) 

263

14.7 Sectoral structure of GDP in sub-Saharan Africa: 1970–2012 

265

15.1 The evolving path of development theory 

276

15.2 A framework for sustainable development in sub-Saharan Africa 

288

16.1 Maintaining subsistence income or schooling children? Cornelian tradeoffs in Burkina Faso 

311

16.2 Good jobs generate a virtuous circle of economic and social benefits 

317

19.1 Responses to an aid surge under different reserve accumulation policies 

382

19.2 Effects of two investing approaches for natural resource revenue 

386

19.3 Debt sustainability of public investment scaling-up under external commercial debt 

387

21.1 The Economic Vulnerability Index, 2005–2009 and 2011–2012 versions compared 

411

21.2 The Physical Vulnerability to Climate Change Index (PVCCI) 

416

22.1 Freedom in the World, 1972–2013 

431

22.2 Africa: Average Polity 2 Score, 1960–2010 

431

22.3 Africa’s visible and invisible political worlds 

432

22.4 A dynamic model of comparison of political well-being 

442

23.1 TWV and relative GDP per capita 1981–1990 

463

25.1 Manufacturing, value added (% of GDP) 

493

25.2 Manufacturing, value added (constant 2005 US$) 

493

25.3 Diversification of exports, Mozambique 

494

25.4 Diversification of exports, Ethiopia 

494

25.5 Roads, total network (km) 

495

25.6 Gross value added in construction 

497

25.7 Government investment 

497

25.8 (A) Gross enrolment ratio (Primary) (%). (B) Gross enrolment ratio (Secondary) (%). (C) Gross enrolment ratio (Tertiary) (%). (D) Tertiary graduates in science, engineering, manufacturing and construction. 

498

List of Figures    xv 26.1 Real British gold prices and slave prices 

508

29.1 Long run oil prices: 1861–2012 

554

29.2 (A) Median oil rents/capita vs. GDP/capita growth. (B) Median mineral rents/capita vs. GDP/capita growth. (C) Median oil and mineral exports in resource-rich Africa 

555

29.3 Political checks and balances and polity indicators in resource-rich countries 

561

29.4 Median governance and control of corruption in resource-rich Africa 

562

29.5 Government in resource-rich countries 

564

30.1 Microeconomic effect of a large negative shock on the income and consumption path of a poor and non-poor household 

571

31.1 Africa: Urbanization and GDP, 2011 

586

31.2 Output shares, SSA, and EAP 

587

31.3 Urbanization and per capita GDP across regions 

587

31.4 Comparing SSA and EAP 

588

31.5 Tanzanian cities: energy access and economic activity 

592

31.6 Africa distribution of population by size of settlement sub-Saharan Africa 

593

31.7 Paved roads in SSA cities, miles per 1000 inhabitants 

594

31.8 Carbon emissions go up with income 

596

32.1 System approach for sustainable development 

605

32.2 Ideal sustainable development pathway 

606

32.3 Unsustainable development path due to environmental capital loss 

607

32.4 Historical trends in Africa’s ecological footprint 

609

32.5 Africa’s percentage of global GHG emissions 

610

32.6 GHG emissions per capita in African countries 

611

33.1 Contribution of informal sector to non-agricultural GDP in African regions 

623

35.1 The marginal effect of democracy on growth at different lengths of democracy duration 

668

36.1 Government welfare and democracy in Arab countries, 1950–2010 

677

38.1 Global prevalence of violent conflicts, 1946–2012 

706

38.2 Global prevalence of violent conflicts, 1400–1939 

709

38.3 Prevalence of violent conflicts in Africa and Europe, 1400–1939 

710

39.1 Trend for countries with the most vital incidents 

731

39.2 GNI per capita 

732

List of Tables

2.1 Diverse income-generating activity is the norm in Africa 

54

4.1 Hypotheses on the relationship between females’ characteristics and marriage outcomes and correlations between marriage outcomes 

79

6.1 Urban growth in Africa, 1950–2010 

108

8.1 Distribution of employment by sector, selected African countries 

138

8.2 Indicators of labor costs, selected regions and countries 

140

9.1 Africa GDP per capita and GDP per capita adjusted for inequality 

164

9.2 Some indicators for inclusive growth 

170

9.3 Inclusive Growth Index (IGI) for Africa (2006–2010) 

172

10.1 Regression of growth by quintiles of initial per capita GDP in each period for of Africa 

181

10.2 Transition matrix by wealth status for African countries 

185

10.3 Multidimensional asset poverty for selected African countries 

186

10.4 Decomposition of poverty by sector of employment in Africa 

187

10.5 Two-step GMM estimate of the relationship between poverty and sectoral shares of employment 

189

15.1 Annual growth rates of GDP and per capita GDP by decade 

278

18.1 Examples of non-cognitive skills 

353

18.2 Examples of economic analyses of measures of non-cognitive skills in longitudinal datasets 

367

19.1 Baseline calibration 

379

19.2 Government consumption multipliers: baseline calibration 

383

19.3 Public investment multipliers: different investment efficiency 

384

21.1 Economic Vulnerability Index (EVI): level in 2011 and change from 2000 to 2011 

412

21.2 Components of the Economic Vulnerability Index (EVI) in 2011 

413

21.3 The Physical Vulnerability to Climate Change Index 

416

21.4 Components of the Vulnerability to Climate Change Index 

417

23.1 TWV and REER 1998–2010 

465

xviii   List of Tables 27.1 Number of working days required of an unskilled urban African to equal average annual per capita revenue in British African colonies, 1910–1938  526 27.2 Foreign investment in sub-Saharan Africa, 1870–1936 

528

27.3 Manufacturing in selected African countries, 1960 

529

27.4 Real wages as ratios of “family subsistence basket” in the capital cities of British colonies, 1900s to 1950s (peacetime decades) 

530

27.5 Infant mortality in selected African countries 

531

29.1 Key economic variables around the second oil price boom (2000–2005) 

557

29.2 Democracy checks, and balances as indicators of credibility and inter-temporal commitments 

560

31.1 Current distribution of population by type of settlement 

598

31.2 Projected urban populations 

599

32.1 Climate change trends and impacts 

610

32.2 Number of visitors to parks and game reserves in Kenya 2005–2009 

613

33.1 Informal sector as a share of non-agricultural employment in selected African countries 

624

35.1 The impact of democracy and democracy duration on GDP per capita growth 

666

35.2 Sensitivity tests 

669

37.1 Egypt and Tunisia: GDP growth and unemployment (%) 

693

38.1 Severity of violent conflicts, 1946–2012 

707

39.1 Terrorism income and number of incidence in Africa 

730

List of Contributors

Julius A. Agbor  is a political economist, currently a research associate at the Department of Economics at Stellenbosch University (South Africa). He has previously been a research fellow at the Brookings Africa Growth Initiative (Washington, DC) and an assistant professor at the University of the Western Cape (South Africa). He obtained his doctorate from the University of Cape Town (South Africa). Olu Ajakaiye  is currently Executive Chairman, African Centre for Shared Development Capacity Building (ACSDCB), Ibadan, a research and training center for shared development policy, planning, implementation and impact/outcome assessment in Africa. He is also the current President of the Nigeria Economic Society and in that capacity Honorary Adviser to the President on Economic Matters and Member of National Economic Management Team. Prior to this, he was Director of Research at the African Economic Research Consortium (AERC) Nairobi (2004–2011) and Director-General, Nigerian Institute of Social and Economic Research (NISER), Ibadan (1999–2004). Olu Ajakaiye specializes in development economics and he is well published. Ragui Assaad  is Professor of Planning and Public Affairs at the University of Minnesota’s Humphrey School of Public Affairs and director of graduate studies for its Master of Development Practice program. He is a Research Fellow of the Economic Research Forum in Cairo, Egypt, and serves as its thematic director for Labor and Human Resource Development. His current research focuses on labor markets in the Arab World, with a focus on youth and gender issues as they relate to education, transition from school-to-work, employment and unemployment, informality, responses to economic shocks, migration, and family formation. Gareth Austin  is professor of African and comparative economic history at the Graduate Institute in Geneva. After teaching at a harambee school in Kenya, he did his BA at Cambridge and PhD at Birmingham. His past employers include the University of Ghana and the London School of Economics (economic history department). He has published extensively on the economic history of sub-Saharan Africa, especially Ghana, from the late precolonial era to the present. His work includes Labour, Land and Capital in Ghana: From Slavery to Free Labour in Asante, 1807–1956 (2005); Labour-Intensive Industrialization in Global History (edited with K. Sugihara, 2013). Kathleen Beegle  is a Lead Economist in the World Bank Africa Chief Economist office. Her research interest includes the measurement of poverty dynamics, socio-economic dimensions of economic shocks, and methodological studies on household survey data collection. She was Deputy Director of the World Development Report 2013 on Jobs. As member of the World Bank Living Standards Measurement Study (LSMS) team, she has expertise in the design and implementation of household survey operations and use of household surveys

xx   List of Contributors for poverty and policy analysis. She received her PhD in Economics from Michigan State University. Nancy Benjamin  has worked as an Assistant Professor at Syracuse University, following a PhD in economics from UC Berkeley. She has worked in the Asia Department of the IMF, the Research Department of the US International Trade Commission, and published on resource-rich countries, the impact of trade policy on productivity and growth, non-tariff barriers, and on trade in services. At the World Bank, she has worked on West African and Middle Eastern countries, focusing on public expenditures, growth, governance, and informality, as well as the regulation of multi-country infrastructure. She enjoys working closely with client countries: work on the informal sector in Africa is the product of extensive collaboration with African research teams. Andrew Berg  is Assistant Director and chief of the Development Macroeconomics Division in the IMF’s Research Department. Previously, he was in the Fund’s African Department and mission chief to Malawi. He has also worked at the US Treasury and as an associate of Jeffrey Sachs. He has a PhD in Economics from MIT and an undergraduate degree from Harvard. He has published articles on sustained growth accelerations, the macroeconomics of aid, and prediction of currency crises. His current research agenda includes inequality and growth, debt sustainability, the management of natural resource wealth, and monetary policy in low-income countries. Arne Bigsten  is Professor of Development Economics at the University of Gothenburg and Director of Gothenburg Centre of Globalization and Development. His research has concerned poverty and income distribution, trade and globalization, industrial development, foreign aid, and institutional reform. Bigsten has been involved in major projects on the impact of the coffee boom in the 1970s in Kenya and Tanzania, the Regional Programme of Enterprise Development in Africa, Ethiopian households, and the impacts of new forms of aid. He has done work for the World Bank, United Nations, Sida, WIDER, OECD, African Development Bank, EU, UNIDO, and the IMF. Fabien Eboussi Boulaga  has taught philosophy and theology at the universities of Yaounde, Abidjan, and the Catholic University of Central Africa for several decades. He has also served as visiting scholar in academic institutions around the world, most notably at Harvard University’s Dubois Institute. His books include La crise du Muntu: Authenticité africaine et philosophie (1977) and Christianity Without Fetishes: An African Critique and Recapture of Christianity (1984). Eboussi Boulaga is also the founding editor of Terroirs, a Journal of Culture and Social Sciences. François Bourguignon  is professor of economics at the Paris School of Economics and at the Ecole des Hautes Etudes en Sciences Sociales in Paris. He is a specialist in public economic policy, income distribution and inequality, and economic development and has authored a large number of academic papers and books. Prior to his current appointment, he held the position of Senior Vice President for Development Economics and Chief Economist at the World Bank from 2003 to 2007. He was then appointed as the director of the newly created Paris School of Economics, an appointment he held until 2013. Along with teaching economics and supervising doctoral students, he is presently involved in several research projects. He is also active in the international development community, lecturing

List of Contributors    xxi and advising leading international agencies as well as foreign governments. The list of his recently published books includes La mondialisation de l'inégalité, Le Seuil, 2012 (translated in German and Italian), an expanded version due to appear in English (The globalization of Inequality, Princeton University Press, March 2015); Handbook of Income Distribution, Volume 2, Elsevier (co-edited with A. B. Atkinson). Laura Camfield  trained as an anthropologist, but now works collaboratively using qualitative and quantitative methods and training others in their use, most recently with the DFID-funded Young Lives longitudinal research study. Her current ESRC-funded research is on enhancing the quality of cross-national methodologies used to generate data on poverty and vulnerability throughout the life course. At UEA she directs postgraduate research within DEV, convenes the main research training module for master’s students, and teaches on courses in impact evaluation and ethnography. She has published widely on methodology, specifically in relation to mixing methods to improve the quality of surveys and measures. Calogero Carletto  is a Lead Economist and Manager of the Living Standards Measurement Study (LSMS) in the Development Research Group at the World Bank. His research interests are in the areas of poverty, food security, agriculture, and rural development, as well as data collection methods and measurement issues. Most recent research includes publications in the Journal of Development Economics, World Development, and Journal of Development Studies. He has previously worked for various UN agencies and the International Food Policy Research Institute (IFPRI). He holds a PhD in Agricultural and Resource Economics from the University of California at Berkeley. Nadereh Chamlou  is a former Senior Advisor at the World Bank, and is currently an international development advisor. Nadereh has held technical, advisory, and managerial positions in various geographic regions and diverse sectors of the World Bank Group. She has extensive experience on economic and social issues of the Middle East and North Africa region. Ha-Joon Chang  teaches in the Faculty of Economics at Cambridge University. His books include the international bestseller Bad Samaritans: The Myth of Free Trade and the Secret History of Capitalism, Kicking Away the Ladder, winner of the 2003 Myrdal Prize, and 23 Things They Don’t Tell You About Capitalism. In 2005, Chang was awarded the Leontief Prize for Advancing the Frontiers of Economic Thought. His most recent book is Economics: The User's Guide (2014). Paul Collier is Professor of Economics and Public Policy at the Blavatnik School of Government, and Director of the Centre for the Study of African Economies at Oxford University. He is a Director of the International Growth Centre. His work covers a wide range of issues in African economic development, among them the management of natural resources, urbanization, conflict, democracy, and HIV. In 2014 he was knighted and won the President’s Medal of the British Academy. Christopher Cramer is Professor of the Political Economy of Development at SOAS, University of London, where he teaches development economics and on the MSc course Violence, Conflict and Development. His research interests include rural labor markets in Africa, industrial policy, and the political economy of violent conflict. His publications

xxii   List of Contributors include Civil War is Not a Stupid Thing (C. Hurst 2006)  and with colleagues the report Fairtrade, Employment and Poverty Reduction in Ethiopia and Uganda (www.ftepr.org/ publications). He is Vice Chair of the Royal Africa Society, and Chair of the Scientific Committee of the African Programme on Rethinking Development Economics (APORDE). Benjamin Davis is the Deputy Director of the Agricultural Development Economics Division at FAO. He has served as Social Policy Advisor for the UNICEF Regional Office in Eastern and Southern Africa and as a Research and Post Doctoral Fellow at IFPRI. He holds a PhD in Agricultural Economics from UC Berkeley. Raj M. Desai  is Associate Professor of International Development at the Edmund A. Walsh School of Foreign Service at Georgetown University, and a Non-resident Senior Fellow at the Brookings Institution, Washington, DC. He previously worked at the World Bank. He earned his PhD in political science from Harvard University. His principal research interests are in the political economy of development and foreign aid. He has authored numerous articles on economic reform and poverty alleviation in political science and economics journals, and is co-author of After the Spring: Economic Transitions in the Arab World (OUP 2012). Ibrahim Ahmed Elbadawi  is currently Director of Economic Policy & Research Center at the Dubai Economic Council, and before that was a Lead Economist at the Research Department of the World Bank. He has published widely on macroeconomic and development policy, democratic transitions and the economics of civil wars and post-conflict transitions. He is a Research Fellow at the Center for Global Development; Associate Editor of the Middle East Development Journal; Thematic Research Leader for “Natural Resource Management and Economic Diversification” at the Economic Research Forum for the Middle East and member of the Advisory Board of the Arab Planning Institute. Juliet Elu  is the Charles Merrill Professor of Economics and Chair at Morehouse College in Atlanta, Georgia. Her research and teaching interests are in the areas of Quantitative Methods, Micro/Macro Economics, International Trade & Development, Gender Issues, and Economic Anthropometry in Developing Economies. Her research has been published in the American Economic Review, Journal of African Development, Journal of Third World Studies, Journal of Developing Areas, African Development Bank Review Journal, Journal of Black Political Economy, Journal of Peace Economic, and Journal of Economic Studies. She earned her BSc in Economics/Political Science and MBA/MPA from Utah State University, and completed her doctorate in Economics from the University of Utah in Salt Lake City, Utah. Hippolyte Fofack  is Afreximbank Chief Economist and Director of Research and fellow of the African Academy of Sciences. He previously worked as Senior Economist at the World Bank. He holds a PhD from the American University. Maria E.  (Mila) Freire  is a Senior Adviser of the Growth Dialogue and former staff of the World Bank. While at the World Bank she served as Senior Advisor to the Sustainable Development Network, Manager of the Urban Program for Latin America, and core member of the World Development Report, 2009. Her main field of concentration has been decentralization, climate change, public finance, and urban economics. Mila has a PhD in Economics from the University of Berkeley, California, and teaches Urban Economics at the

List of Contributors    xxiii Johns Hopkins University. Between 1992 and 1996, Mila held a managing director position in the Caixa Geral de Depositos, the largest Bank in Portugal. Alan Gelb  is a Senior Fellow at the Center for Global Development. He previously held several positions at the World Bank, including Director of Development Policy and Chief Economist of the Africa Region. His ongoing work covers the areas of African development and competitiveness, the management of natural resource countries, results-based aid, and the development applications of biometric technology and identification systems. He has a PhD from Oxford University. Stephen Golub  is Franklin and Betty Barr Professor of Economics at Swarthmore College and adjunct professor at the Wharton School of the University of Pennsylvania. He has been a visiting professor at Columbia, Yale, Berkeley, and Cheikh Anta Diop University of Dakar. He has written numerous articles and co-authored Money Credit and Capital with James Tobin. He has served as a consultant to the United Nations, the World Bank, the International Monetary Fund, the Organization for Economic Cooperation and Development, and the US Federal Reserve. Patrick Guillaumont President Ferdi, Emeritus Professor, University of Auvergne, Director of the Revue d’Economie du Développement, member of the European Development Network (EUDN), Fellow of the Oxford CSAE, has been a member of the Committee of Development Policy at the United Nations (CDP) from 1987 to 2009. He has also been a member of many advisory international committees and has worked for various international institutions and governments. He has published many books and papers in a wide set of economic journals. His recent work is mainly about public development assistance, its allocation and effectiveness, vulnerability, and the least developed countries. His last book was Caught in a Trap, Identifying the Least Developed Countries (Economica 2009). Catherine Guirkinger  is a professor of economics at the University of Namur (Belgium) and belongs to its Centre for Research in Economic Development (CRED). She holds a PhD from the University of California, Davis, in Agricultural and Resource Economics. Her main field is development economics with a focus on the economics of extended families particularly in Africa and on microfinance (credit and insurance). A lot of her work is based on first-hand data she collected in several countries (Peru, Cameroun, Mali, Burkina Faso). Jane I. Guyer  is George Armstrong Kelly Professor in the Department of Anthropology at Johns Hopkins University. Educated at the London School of Economics (BA (Soc) 1965)  and the University of Rochester (PhD 1972), she has carried out field research in Nigeria and Cameroon over a period of several decades, with primary focus on the history and organization of production and marketing systems. Her books include Family and Farm in Southern Cameroon (1984); Feeding African Cities (edited volume, 1987); An African Niche Economy. Farming to Feed Ibadan (1997); Marginal Gains. Monetary Transactions in Atlantic Africa (2004), and several other co-edited collections on African economies. She served as a member of the International Advisory Group to the President of The World Bank and the Governments of Chad and Cameroon on the Chad–Cameroon Oil Field and Pipeline Project (2001–2009).

xxiv   List of Contributors Hakim Ben Hammouda  has served as the Minister of economy and finance of Tunisia. Before being appointed Minister in January 2014, he performed senior managerial positions with various international organizations. He was appointed special adviser to the President of the African Development Bank from 2011 to 2014, and Director of the Institute of Training and Technical Cooperation with the World Trade Organization (2008–2011). Previously, he held various directorships with the United Nations Economic Commission for Africa, as director of the subregional office in Central Africa and then director of the Trade and Regional Integration Division and Chief Economist. For many years he has dealt with Africa’s development and has contributed to the implementation of major program of actions such as NEPAD. During his career, he has directly supervised key continental publications such as Economic Report on Africa, African Economic Outlook (AfDB, OECD, and UN), Assessing regional Integration in Africa. Hakim Ben Hammouda has a PhD in international economics and regularly teaches economic development at several universities. He is the author of over 20 books, several articles in international scientific journals in the field of international trade and economic development and weekly chronicles in different newspapers. Faraz Hayat is a research assistant at the University of Chicago. He graduated from Swarthmore College in 2014 with High Honors in Economics. Anke Hoeffler  is an economist at the Centre for the Study of African Economies (CSAE), University of Oxford. In her work she focuses on the macroeconomics of developing countries, the economics of conflict and political economy issues. She has published numerous journal articles and book chapters on the causes and consequences of civil war, the impact of development aid and the state of democracy in Africa. Afeikhena Jerome  currently works for DFID/SPARC as National Coordinator of the State Peer Review Mechanism instituted by the Governors of Nigeria’s 36 states. Prior to joining the Secretariat of Nigeria Governors’ Forum, he worked with the United Nations as the Coordinator for Economic Governance and Management, NEPAD/ African Peer Review Mechanism (APRM), Midrand, South Africa, from 2006 to 2012 after teaching in several African Universities for about 15 years. He has been a consultant for several international organizations including the World Bank, African Development Bank, UNECA, UNIDO, and the African Capacity Building Foundation. Morten Jerven  is the author of many works, including Poor Numbers: How We Are Misled by African Development Statistics and What To Do About It, Cornell University Press, 2013. Economic Growth and Measurement Reconsidered in Botswana, Kenya, Tanzania, and Zambia, 1965–1995, Oxford University Press, 2014, Africa: How Economists Got it Wrong, Zed Books, 2015, and Measuring African Development: Past and Present, Routledge, 2015. Morten Jerven is an economic historian, with a PhD from the London School of Economics. Since 2009 he has been working at the School for International Studies at Simon Fraser University in Vancouver, Canada. Richard Joseph is John Evans Professor of International History and Politics at Northwestern University and a non-resident Senior Fellow of the Brookings Institution. He has written extensively on African governance, democratization, and political economy. His publications include Democracy and Prebendal Politics in Nigeria (1987); Radical Nationalism

List of Contributors    xxv in Cameroun (1977); State, Conflict, and Democracy in Africa (1999), editor; Smart Aid for African Development (2008), co-editor; and many articles including “Growth, Security, and Democracy in Africa,” Journal of Democracy (2014). His recent essays and commentaries can be found at www.africaplus.wordpress.com. He is preparing studies of Nigerian politics and African governance and development. Caroline Krafft received her master’s degree in public policy from the University of Minnesota’s Humphrey School of Public Affairs and is now a PhD candidate in the Department of Applied Economics at the University of Minnesota. Her research examines issues in development economics, primarily labor, education, health, and inequality in the Middle East and North Africa. Current projects include work on early childhood development, labor market dynamics, life course transitions, human capital accumulation, and fertility. Somik Lall is a Lead Economist for Urban Development at the World Bank. He has been a core team member of the World Development Report 2009: Reshaping Economic Geography, Senior Economic Counselor to the Indian Prime Minister’s National Transport Development Policy Committee, and Lead Author of the World Bank’s flagship report on urbanization Planning, Connecting, and Financing Cities Now. He currently leads a World Bank program on the Urbanization Review, which provides diagnostic tools and a policy framework for policymakers to manage rapid urbanization and city development. Urbanization Reviews have been completed or are ongoing in over 20 countries, covering over 55 percent of the global urban population. His research and policy interests span urban and spatial economics, rural urban migration, infrastructure development, and public finance, with more than 40 publications featured in peer-reviewed journals, edited volumes, and working papers. Dr. Lall holds a Bachelor’s Degree in engineering, a master’s degree in city planning, and doctorate in economics and public policy. Danny Leipziger  is Professor of International Business and International Affairs at George Washington University and Managing Director of the Growth Dialogue. Previously, he held leadership positions at the World Bank, including Vice President for Poverty Reduction and Economic Management (2004–2009). He was Vice Chair of Spence Commission on Economic Growth. Dr. Leipziger holds a PhD from Brown University. His publications of a dozen books include Lessons from East Asia, Korea’s Distribution of Income and Wealth, Stuck in the Middle, and Globalization and Growth. His 45 journal articles and other published work span development economics and development finance. He is a regular media contributor. Justin Yifu Lin is professor and honorary dean, National School of Development at Peking University. He was the Senior Vice President and Chief Economist of the World Bank, 2008–2012. Prior to this, Mr. Lin served for 15  years as Founding Director of the China Centre for Economic Research at Peking University. He is the author of 23 books including Against the Consensus:  Reflections on the Great Recession, the Quest for Prosperity:  How Developing Economies Can Take Off, Demystifying the Chinese Economy, and New Structural Economics: A Framework for Rethinking Development and Policy. He is a Corresponding Fellow of the British Academy and a Fellow of the Academy of Sciences for Developing World.

xxvi   List of Contributors Xubei Luo is a Senior Economist at the World Bank. She has coauthored a book of Multilateral Banks and the Development Process (Transaction Publishers, 2012)  and published over 40 articles on growth, poverty, inequality, labor market, and trade facilitation. She holds a PhD in Economics from the International Development Research Center (CERDI), University of Auvergne, France, and three MAs in Project Analysis, Development Economics, and Economic Policy from CERDI, an MA in Finance and a BA in International Finance from Sun Yat-Sen University, China. Takaaki Masaki  is a PhD candidate in the Department of Government at Cornell University. His research focuses on the issues of political economy, distributive politics, development finance, and economic development. In particular, he is interested in how international factors—such as foreign aid, foreign direct investment, trade, and economic sanctions—influence domestic politics in sub-Saharan Africa. Masaki holds a bachelor’s degree in Liberal Arts from Soka University of America and a master’s degree in Conflict Resolution from Georgetown University. Ahmadou Aly Mbaye  is currently the Centre de Recherches Economiques Appliquées at Cheikh anta DIOP University (UCAD, Dakar, Senegal). He holds a Doctorate in Development Economics from the University of Clermont Ferrand (CERDI, France) and has held several international academic positions in recent years, including as a Fulbright and visiting professor and research scholar at Swarthmore College, George Washington University, and the World Bank. He has several publications in the area of development. He has also served as a consultant for many international organizations such as UNCTAD, WTO, FAO, the World Bank, UNECA, WAEMU (West African Economic and Monetary Union), and the Senegalese government. Christopher Malikane  is an Associate Professor of Economics in the School of Economic and Business Sciences at the University of the Witwatersrand, South Africa. He is also Director of the Macroeconomics and Financial Analysis Group at the School of Economic and Business Sciences. Christian J. Meyer is a doctoral student at the European University Institute in Florence. Meyer was previously with the Poverty and Inequality Team at the World Bank’s Development Economics Research Group. Before that, he was a research associate at the Center for Global Development, where he worked on poverty and inequality in Latin America and the Caribbean, fiscal incidence analysis, as well as labor markets and industrial organization in sub-Saharan Africa. Meyer previously worked with the United Nations, the European Commission, and the German federal government. He holds a BA from WHU Vallendar and a MPP from the Hertie School of Governance, Berlin. Nadir Abdellatif Mohammed is the Country Director for the Gulf Countries, Middle East and North African region, of the World Bank. He obtained MPhil, PhD (Cantab) and post-doctoral qualifications in economics from the University of Cambridge. He has also worked as a researcher/lecturer in the Universities of Cambridge, Oxford, and Addis Ababa. He publishes widely on issues of defense economics, economic development, and fiscal policy. He worked in the African Development Bank and the Islamic Development Bank before joining the World Bank Group in 1998.

List of Contributors    xxvii Célestin Monga is Managing Director at the United Nations Industrial Development Organization (UNIDO). He previously worked as Senior Advisor and Director at the World Bank and has held various board and senior positions in academia and financial services. A graduate of MIT, Harvard, and the universities of Paris 1 Panthéon-Sorbonne, Bordeaux, and Pau, Dr. Monga was the Economics Editor for the five-volume New Encyclopedia of Africa (Charles Scribner’s, 2007). His published works have been translated into multiple languages. Roger B. Myerson  taught at Northwestern University from 1976 to 2001 and has since taught at the University of Chicago’s Economics Department. His analysis of incentive constraints in economic communication introduced several fundamental ideas of mechanism design theory. Myerson is author of Game Theory: Analysis of Conflict and has used game-theoretic analysis to study political systems. He has written on resolve and restraint in strategic deterrence, on moral hazard and leadership in the foundations of the state, and on the vital importance of local democracy in development. In 2007, he was awarded the Nobel Memorial Prize in Economic Sciences. Mustapha Kamel Nabli  currently heads an independent economics think-tank in Tunisia. He was Governor of the Central Bank of Tunisia from January 2011, just after the revolution, until July 2012. He was at the World Bank from 1997 to 2010, where he was Chief Economist and Director of the Social and Economic Development Department for the MENA region (1999–2008). From 1990 to 1995 he was Minister of Planning and Regional Development in Tunisia. He held various academic positions before that at the University of Tunis and elsewhere. He holds a master’s degree and a PhD in Economics from the University of California, Los Angeles (1974). Mthuli Ncube is a Senior Research Fellow at University of Oxford, Blavatnik School of Government. Previously, he was Vice President and Chief Economist at the African Development Bank Group (AfDB), where he spearheaded economic strategic thinking on inclusive growth, which is an objective the bank is currently pursuing in activities. He was also Professor and Dean of Wits Business School, and Dean of the Faculty of Commerce, Law and Management, University of the Witwatersrand. He was also a Lecturer in Finance at the London School of Economics (LSE). He is Chairman of the African Economic Research Consortium, and has extensive experience in investment banking. He holds a PhD in Economics (Mathematical Finance) from the University of Cambridge, and has published widely in economics and finance. Anders Olofsgård  is Deputy Director of the Stockholm Institute of Transition Economics, and Associate Professor at the Stockholm School of Economics. Before that he was Associate Professor of Economics at the Edmund A. Walsh School of Foreign Service, Georgetown University. He earned his PhD in economics from Stockholm University. His primary research areas are political economy, development, and applied micro-economics, and he has published widely in both economics and political science journals. Anders has also been a visiting scholar at the research department of the IMF, and done work for the World Bank, USAID, and the Swedish Parliament. Jean-Philippe Platteau  is Professor Emeritus of economics at the University of Namur, Belgium. His main field is development economics and most of his work has been concerned with the understanding of the role of institutions in economic development, and the processes of institutional change. The influence of non-economic factors and various

xxviii   List of Contributors frontier issues at the interface between economics and sociology are a central focus of his research projects. He has written numerous articles in academic journals and published several books, including Halting Degradation of Natural Resources – Is There a Role for Rural Communities? (Clarendon Press, 1995) with J.M. Baland, Institutions, Social Norms, and Economic Development (Harwood Publishers and Routledge, 2000), Culture and Development: New Insights Into an Old Debate (Routledge, 2010). Moreover, he is presently completing a book: Religion, Politics, and Development: Is Islam a Special Problem? Gregory Price  is Professor of Economics and Interim Dean of the School of Business at Langston University. His previous appointments include Charles E. Merrill Professor and Chair, Department of Economics, Morehouse College. Director of the Mississippi Urban Research Center, Professor of Economics at Jackson State University, and Economics Program Director at the National Science Foundation. He also served as President of the National Economic Association in 2008. An applied econometrician and theorist, his current research interests include economic anthropometry, the economics of historically black colleges/universities, the effects of race on economic stratification, and the causes/consequences of slavery. His research has been published in a wide variety of journals such as Economics and Human Biology, Review of Black Political Economy, Review of Economics and Statistics, American Economic Review, and Review of Development Economics. A native of New Haven, Connecticut, Dr. Price earned his BA in economics from Morehouse College, and completed his economics doctorate at the University of Wisconsin, Milwaukee. Vijaya Ramachandran  is a senior fellow at the Center for Global Development. She works on private-sector development, food security, humanitarian assistance, and development interventions in fragile states. Prior to joining CGD, Ramachandran served on the faculty at Georgetown University and also worked in the Africa Private Sector Group of the World Bank and in the Executive Office of the Secretary-General of the United Nations. Her work has appeared in several media outlets including the Financial Times, the Guardian, The Washington Post, The New York Times, National Public Radio, Voice of America, and The Economist. Vijaya Ramachandran earned her BA, MA, and PhD in Business Economics from Harvard University. Felwine Sarr  is full Professor of economics at the University Gaston Berger of Saint Louis (Senegal). His fields of research are political economy, macroeconomics policies, econometry of times series and economic of development. Dean of the faculty of Economy and Management (2011–2014) and Director of the Faculty of Civilization, Religions, Arts and Communication (2011–2013) of the University Gaston Berger. He is author of scientific articles, novels, and short stories. Abebe Shimeles  received post-graduate and undergraduate degrees in economics from the School of Economics and Commercial Law, Goteborg University, Delhi School of Economics, and Addis Ababa University. He has several years of work experience in academia and development agencies, such as Actionaid, UNECA, World Bank, and the African Development Bank. Currently he is Acting Director of the Development Research Department at the African Development Group. He is IZA research fellow and actively works on empirical research covering a wide range of topics including poverty, income

List of Contributors    xxix distribution, labor markets, health insurance, and tax evasion in the context of African countries. Tomonori Sudo  is Senior Research Fellow of JICA Research Institute and Advisor of Office for Global Issues and Development Partnership, Operations Strategy Department of JICA. He is currently conducting research on environment and climate change issues including climate change finance. He is also serving as a member of the Bureau of Environment and Development Cooperation Network (ENVIRONET) of the OECD Development Assistance Committee (DAC). During his 20-year career at JICA and the former Japan Bank for International Cooperation (JBIC) and Overseas Economic Cooperation Fund (OECF), he has experienced ODA loan operations to Indonesia, Malaysia, Bangladesh, and China, and secondment to the African Development Bank as a Private Sector Specialist to promote private sector development in Africa, and secondment to the Institute for Global Environmental Strategies (IGES) to conduct research on climate policy and capacity building on the Clean Development Mechanism in several Asian countries. Before joining OECF, he worked for the Sakura Bank and is experienced in cooperate finance, project finance, and forex operations. Dr. Sudo received his PhD in International Studies from Waseda University, Japan, a BA in Economics from Osaka University, Japan, and MSc in Environmental & Resource Economics from the University of London (University College London). Olumide Taiwo  is a Development Economist and Director of Research at the Centre for the Study of the Economies of Africa (CSEA), Abuja, Nigeria. Prior to the CSEA, he was Africa Research Fellow at the Brookings Institution, visiting lecturer at the University of the Witwatersrand, Johannesburg, and assistant professor of economics at the American University of Nigeria, Yola, Nigeria. He holds a PhD in Economics from Brown University, an MSc in Economics and a BSc in Actuarial Science (with First Class Honors) from the University of Lagos, Nigeria. Scott D. Taylor  is associate professor and Director of the African Studies Program in the School of Foreign Service at Georgetown University. His research focuses on African politics and political economy, with a particular emphasis on business–state relations, private sector development, governance, and political and economic reform. He has been consulted by numerous international development organizations. Dr. Taylor is the author of Politics in Southern Africa: Transition and Transformation (with Gretchen Bauer); Culture and Customs of Zambia; Business and the State in Southern Africa: The Politics of Economic Reform; and Globalization and the Cultures of Business in Africa: From Patrimonialism to Profit. M. A. Thomas  is a visiting researcher at Georgetown University. She is a lawyer and political economist who has worked on corruption and rule of law issues in low-income countries as an academic and as a practitioner in the foreign aid community. She is the author of Govern Like Us: U.S. Expectations of Poor Countries (Columbia University Press). Nicolas van de Walle is the Maxwell M.  Upson Professor of Government at Cornell University in Ithaca, New York. He taught at Michigan State University from 1990 to 2004. He gained his PhD at Princeton University in 1990. In addition, van de Walle has worked extensively as a consultant for a variety of international and multilateral organizations,

xxx   List of Contributors including the World Bank, USAID, and UNDP. He has published widely on democratization issues as well as on the politics of economic reform and on the effectiveness of foreign aid, with a special focus on sub-Saharan Africa. His books include Democratic Trajectories in Africa: Unraveling the Impact of Foreign Aid (2013, with Danielle Resnick), Overcoming Stagnation in Aid-Dependent Countries (2005), African Economies and The Politics of Permanent Crisis, 1979-1999 (2001), and Democratic Experiments in Africa: Regime Transitions in Comparative Perspectives (1997, with Michael Bratton). He is also the author of over a hundred journal articles, reports, and book chapters. He is currently engaged in a book project on the practice of democracy in sub-Saharan Africa that focuses on electoral dynamics and party systems in the region since 1990. Warren C.  Whatley  is Professor of Economics, University of Michigan. His current research is on development in sub-Saharan Africa in long-run historical perspective, focusing primarily on the legacies of two major shocks to the region: the transatlantic slave trade, and colonization. Previous research focused on the economic history of African Americans in the USA: cotton slavery, post-bellum sharecropping, migration to the north, and industrial employment. His long-range research goal is to link these two research strands into an economic history of Africans in the Atlantic World. His initial paper on this area addresses the origins of African American culture. Shu-Chun S. Yang  is a Senior Economist in the IMF’s Research Department and a Professor in the Institute of Economics at National Sun-Yat-Sen University, Taiwan. Before joining the IMF, she was a Principal Analyst at the US Congressional Budget Office, an Assistant Research Fellow at Academic Sinica, an Economist at the US Joint Committee on Taxation, and an Assistant Professor at John Carroll University. She has a PhD in Economics from Indiana University and has published articles on fiscal policy, policy foresight, and macroeconomic management of natural resource revenues in developing countries. Tarik M. Yousef  is Chief Executive Officer of Silatech and Nonresident Senior Fellow at Brookings. He holds a PhD in economics from Harvard University and served between 1998 and 2008 at Georgetown University as Associate Professor of Economics and Sheikh Al Sabah Professor of Arab Studies in the School of Foreign Service. Recent publications include the forthcoming co-edited volumes:  Public Sector Reform in the Middle East and North Africa:  The Lessons of Experience (World Bank, 2014); Young Generation Awakening: Economics, Society and Policy on the Eve of the Arab Spring (Oxford University Press, 2015). Luis-Felipe Zanna  is a Senior Economist of the Development Macroeconomics Division in the IMF’s Research Department. Before joining the Fund in 2006, he worked in the International Finance Division of the Federal Reserve Board. He has a PhD in Economics from the University of Pennsylvania and undergraduate and master’s degrees from Universidad de Los Andes (Colombia). He has published articles on monetary policy, macroeconomic stability, and adaptive learning. His current research agenda is in the areas of fiscal policy, monetary policy, and debt sustainability issues, with a particular focus on low-income countries.

List of Contributors    xxxi Alberto Zezza  is a Senior Economist in the Development Research Group of the World Bank. His research interests are in the area of agricultural and rural development, rural poverty reduction, food security, and migration. He is part of the Living Standards Measurement Study (LSMS) team, where he focuses specifically on improving the availability and quality of food consumption and of livestock data in LSMS-type household surveys. Prior to joining the World Bank, he was with the Agricultural Development Economics Division of the Food and Agriculture Organization of the UN (FAO). He holds a PhD in Agricultural and Development Economics from the University of Rome, and an MA in Development Economics from the University of Sussex.

Introduction

A frica, th e Ne xt In tellectual Front i e r Célestin Monga AND Justin Yifu Lin 1 Introduction One of the most successful and memorable commercial campaigns of recent decades featured a famous tennis player promoting a new type of camera by a well-known manufacturer. A 30-second clip filled with beautiful pictures of the good-looking athlete, it ended with a simple, three-word slogan: “Image is everything!” That bold assertion gained a life for itself well beyond the advertisement world and sparked debates and counter-slogans, including some counter-narratives of the notion that “image is nothing …” Everyone familiar with Africa knows the power of image, perception, and reputation. Since the dark days of slavery, the continent’s image has never been a stellar one. With international media headlines focusing primarily on its pains (low levels of development, poverty, violent conflicts, outbreaks of scary exotic diseases, etc.), it is no surprise that survey results still depict Africa to be mostly a dangerous place, as a region “out of history,” as a major world leader recently said in a public statement, repeating perhaps unwittingly some of the old Hegelian fantasies about the continent.1 There are many different Africas, of course, starting with the obvious fact that the continent currently counts 54 independent countries. It is home to the world’s oldest civilizations and its many contributions to human knowledge will probably never be properly documented. From an economic perspective, some African countries and regions are doing

1 

In a 2007 speech in Dakar, Senegal, then French President Nicolas Sarkozy suggested that Africa had failed to embrace progress. He said: “The tragedy of Africa is that the African has not fully entered into history… This man (the traditional African) never launched himself towards the future… Africa’s problem is to stop always repeating, always mulling over, to liberate itself from the myth of the eternal return. It is to realise that the golden age that Africa is forever recalling will not return because it has never existed.” Former Malian President Alpha Oumar Konare, who was then chairman of the African Union Commission, swiftly labeled Sarkozy’s speech as “declarations of a bygone era.” Sarkozy’s words drew outrage because they were reminiscent of Hegel (1956)’s racist assertion that “the African man” was somehow sub-human and certainly devoid of the kind of consciousness that would make him part of world history.

2   Africa and Economics quite well, averaging in recent decades the highest growth rates in the world. But such news has not consistently grabbed headlines. In fact, it has often been dismissed as an anomaly because of the (bad) “neighborhood effect”: if “Africa”—whatever that means is associated with failure in the global consciousness, what good does it make to see on the front page of some magazine that Equatorial Guinea has averaged over 19 percent GDP growth for 20 years, and that Ethiopia, Angola, Uganda, and Mozambique, all achieved about 7 percent GDP growth per year for that long period? If the poor perception of “Africa” is so entrenched what difference does it make to learn that 20 sub-Saharan countries that do not produce oil averaged GDP growth rates of 4 percent or higher for two decades? Or that poverty levels are declining while education and health indicators are showing real improvements? Africa’s image has been so poor that even when it accidentally becomes the subject of “good” news, the information is given superficially, if not in caricatures. The continent is really never presented as a place where humanity is reinventing itself positively. A good illustration of the hysteresis of its poor image can be seen in comparative advertisements: New York is typically shown to be home to the grandiose Empire State Building or the Statute of Liberty; Paris is naturally depicted as the place when the human mind created the Eiffel Tower; London is represented with majestic Big Ben; African cities or countries are celebrated for its big roaring lions, leopards, and chimpanzees. This introductory chapter is not about debating stereotypes of Africa or celebrating the continent’s still unfinished economic progress. Rather, it discusses the reasons why Africa has remained neglected in economics, despite its important contributions to the discipline. Section 2 highlights Africa’s enduring intellectual influence on some of the world’s leading economists. Section 3 discusses the traditional reasons why that deep positive influence is little known and rarely acknowledged in mainstream economics. Section 4 offers a different explanation of the neglect of Africa as a rich source of economic knowledge, it suggests that economic thinking on Africa has generally mirrored the general evolution of macroeconomics and the dominant frameworks of development economics, which have been fraught with analytical sins and mimetic choices. Section 5 presents the objectives of this volume and discusses the challenges of producing relevant knowledge.

2  Africa’s Insidious Intellectual Influence: The Depardieu Theorem One of the best-kept secrets in the world of economists is the fact that some of today’s leading researchers did their cutting-edge work on Africa. Joseph Stiglitz, who pioneered Information Economics and ignited fundamental change in the prevailing paradigm within economics, acknowledges the crucial importance of Africa in opening his mind to fruitful ways of looking at economic issues. He recalls studying economics in college in Gary, Indiana (USA), a city marred by poverty, periodic unemployment, and massive racial discrimination. Yet the theories that he was taught paid little attention to poverty, said that all markets cleared—including the labor market—so unemployment must be nothing more

Introduction   3 than a phantasm, and that the profit motive ensured that there could not be economic discrimination. As a graduate student, he set out to try to create models with assumptions— and conclusions—closer to those that accorded with the world he saw, with all of its imperfections. My first visits to the developing world in 1967, and a more extensive stay in Kenya in 1969, made an indelible impression on me, he writes. Models of perfect markets, as badly flawed as they might seem for Europe or America, seemed truly inappropriate for these countries… I had seen cyclical unemployment—sometimes quite large—and the hardship it brought as I grew up, but I had not seen the massive unemployment that characterized African cities, unemployment that could not be explained either by unions or minimum wage laws (which, even when they existed, were regularly circumvented). Again, there was a massive discrepancy between the models we had been taught and what I saw. (Stiglitz 2001)

Many of the key assumptions that went into the dominant competitive equilibrium model seemed not to fit these economies well. The issues that attracted young Stiglitz’s attention had to do with the imperfection of information, the absence of markets, and the pervasiveness and persistence of seeming dysfunctional institutions, like sharecropping. “With workers having to surrender 50 percent or more of their income to landlords, surely (if conventional economics were correct), incentives were greatly attenuated. Traditional economics said not only that institutions (like sharecropping) did not matter, but neither did the distribution of wealth. But if workers owned their own land, then they would not face what amounted to a 50 percent tax. Surely, the distribution of wealth did matter.” (Stiglitz 2001.) “Again, there was a massive discrepancy between the models we had been taught and what I saw. The new ideas and models were not only useful in addressing broad philosophical questions, such as the appropriate role of the state, but also in analyzing concrete policy issues.” (Stiglitz 2001.) In other words, Africa’s puzzling economic issues led him to challenge some of the fundamentals of the economic doctrine and sparked his work on the economics of information asymmetry. Gérard Depardieu, the French actor, once said that anyone who goes to Africa never really comes back. The powerful impact that the acquaintance and encounter with Africa can have on a researcher’s intellectual development is indeed often a lasting phenomenon. Although not explicitly acknowledged, Africa’s influence pervades the work of other Nobel laureates James Tobin and Peter A. Diamond, who also happened to be pensioners at the Nairobi Institute of Development Studies in the late 1960s. Tobin’s achievements cover a broad spectrum of economic research. He made notable contributions in such widely differing areas as econometric methods and strictly formalized risk theory, the theory of household and firm behavior, general macro theory, and applied analysis of economic policy. His Nobel Prize citation and his own intellectual autobiography make no explicit reference to Africa and only focus on his analyses of financial markets and their relations to expenditure decisions, employment, production, and prices. In studying how households and firms decide to hold different real and financial assets, and, simultaneously, incur debts, Tobin showed how these decisions are governed by weighing risk and expected rate of return. Unlike many other theorists in the field, he did not confine his analysis solely to money, but considered the entire range of assets and debts. Tobin rarely

4   Africa and Economics discussed his time as a researcher in Nairobi while he was already a well-established figure in the economics discipline. One can only conjecture that his time in Kenya may have given him the distance to reflect on his portfolio selection theory, which describes how individual households and firms in industrialized economies determine the composition of their assets. Diamond too started his most cutting-edge work after spending a year in East Africa. “A hike up Mount Longonot, a dormant volcano in the Rift Valley shortly after we arrived in Kenya was practically the first time I had my feet off pavement, he writes. Both places were eye-opening, given how limited was my range of previous experiences” (Diamond 2010). Before traveling to Kenya, he was deeply grounded in general equilibrium (Arrow-Debreu) theory but also aware of its limitations, most notably the completeness of the coordination of agents that happens with complete competitive markets. Arrow-Debreu theory does not contain a mechanism or process for an economy to achieve its equilibrium allocation. In the 1960s there was ongoing work to find a hypothetical process that would converge to this equilibrium, with a focus on equations for price adjustment based on excess demands or supplies at tentative prices (referred to as tâtonnement). In his initial attempts to change the theory (Diamond 1967), he simply limited the set of available markets on the role of the stock market in resource allocation. After he came back from Kenya, he revisited the issue, convinced that the wrong question was being asked. In a second attempt at changing the theory, rather than asking whether a process could be found that would converge to a standard competitive equilibrium, he chose to look for the allocation to which a plausible process would converge. This led first to a paper that applied search theory to a retail market (Diamond 1971). He then took thinking in terms of a process in real time to the law-and-economics question of the effects of alternative rules for breach of contract (Diamond and Maskin 1979, 1981), and then to the labor market, then to the entire economy. “Dissatisfaction with (well-understood) analysis was part of the drive that led to this success, he later reflected; trying to get a more satisfactory perspective that would open up the ability to better answer economic questions was another part. And greatly enjoying the work itself mattered too.” (Diamond 2010.) Who knows what the change of scene from the USA to Kenya did to young Diamond’s research strategy? A simplistic but plausible conjecture is perhaps that the fresh air of Mount Longonot or the vertiginous beauty of the Rift Valley inspired Diamond to develop some of the methodological innovations that enriched economic theory. It is also quite plausible that by observing the behavior of households and firms and the functioning of markets and institutions in Africa, he better understood the complexity of the discipline and the need to build new intellectual frameworks that adhere more closely to reality. Africa’s intellectual influence is more clearly discernible in the works of other leading economists. Besides several future Nobel laureates, the Nairobi Institute of Development Studies also hosted in the 1960s John Harris and Michael Todaro (the celebrated authors of the Harris–Todaro Model), Gary Fields, and a few others who subsequently became household names in the field of development economics. Of course, not everyone went to Nairobi. Still, African economies have inspired a large number of influential economists to produce some of their best work. Paul Collier, who has become a rock star in the discipline with his work on Africa, notes that “Professor A. Deaton (Princeton) based the prestigious Oxford Clarendon Lectures (1991) on two data sets which best tested his propositions on the central topic of consumption behavior; one was from the United States, the other from Côte d’Ivoire. This is indeed part of a wider phenomenon, the upgrading of the economics of development.” (1993: 59).

Introduction   5 The Depardieu Theorem according to which one never escapes from Africa’s insidious intellectual influence seems to be in full swing. In recent years, the list of well-known researchers who have turned their attention to Africa and contributed to the “upgrading” of development economics includes Roger Myerson (another Nobel laureate), Stanley Fischer, Daron Acemoglu, James Robinson, Jeffrey Sachs, François Bourguignon, Jean-Paul Azam, Kaushik Basu, Dani Rodrik, Ricardo Hausmann, Nicholas Stern, Pierre-Richard Agénor, Tim Besley, Esther Duflo, Christoper Udry, Ann Harrison, Ha-Joon Chang—just to name a few. African researchers themselves have elevated their game, publishing landmark articles and books on a wide range of economic subjects, some of which are discussed below. In sum, mainstream macroeconomics has been enriched by cutting-edge work carried out on Africa on issues as diverse as the modeling of small open economies or the theory of repressed inflation. Likewise, mainstream microeconomics has benefited enormously from the study of factor markets, product markets, and household economics across the continent of Africa (Collier 1993). Yet, despite its breadth and depth, Africa’s contribution to economics remains largely untraceable. A good indication of this trend is the fact that for about eight decades (from 1900 to 1981), the total number of Africa-related documents (journal articles, books, collective volume articles, dissertations, working papers, and book reviews from the Journal of Economic Literature) published in the EconLit (which includes references to articles in economics journals from all over the world, most of which are in English or with English summaries) was insignificant, at less than 100 per year. It tripled to 310 in 1984 and reached 1050 in 1998 and 2441 in 2013. While this growth has been exponential in the past three decades, it remains negligible: Africa-related publications in 2013 still represented only 4 percent of the total of 58 649.

3  Traditional Explanations of the Benign Neglect Why has Africa’s analytical contribution to economics been (so far) largely ignored despite the substantial body of theoretical and empirical works derived from it? The typical answer to that question is often purely technical and has to do with data limitations and methodological choices made by researchers who specialized on the economic issues of the continent. During the first several decades of independence, few African countries devote enough resources to build strong and credible statistical institutions and systems. As a result, macroeconomic data remained rudimentary for a very long time. Most advanced economies recalibrate their GDP figures and sector weightings every five years. In African countries where data is sparse, statistical agencies just select a “base year”—a year when data on the economy is good enough to be used as reference. They then add on the extra data they collect each year to generate estimates of economic growth. Still, for most African countries, national accounts for instance have been incomplete and inconsistent since independence, and are still unsatisfactory (Jerven 2014, 2013; Devarajan 2013).2 This has led to the embarrassing rebasing of GDP in many important countries: in 2 

A survey of 45 African countries by IMF (2013) found that only four countries met the so-called five-year rule for the selection of the “base year.”

6   Africa and Economics 2010, Ghana opened the way when it changed its base year for GDP calculation from 1993 to 2006, which implied that, in previous estimates, about $13 billion of economic activity had gone unrecorded. GDP was thus increased by 60 percent overnight and the country was upgraded from a low-income to a lower-middle-income country. Nigeria followed suit in 2014, announcing that new numbers showed that the national economy had doubled compared to the old numbers in use just the day before. They had not updated the benchmark for GDP estimations since 1990—nearly a quarter of a century. This change alone meant that Nigeria suddenly became bigger than the South African economy, and the total GDP of Africa increased by 15 to 20 percent. These decisions have emboldened government statisticians across the continent, many of whom are working to change the sources and methods of how they estimate the size of the economy. Despite data shortages, macroeconomic and institutional analyses devoted to Africa have contributed substantial advances to economic thinking. Ndulu (1986) for instance pioneered research on previously neglected aspects of fiscal policy, highlighting patterns of interactions between public and private investment and shedding new light on the traditional notions of “crowding out” and “crowding in.” Ndulu et al. (2007, 2008) focused on the mystery of low economic growth rates in Africa during the first four decades of independence (1960–2000), a period that was one of remarkable growth and transformation in the world economy. They examined the impact of resource wealth and geographical remoteness on Africa’s growth and developed a new dataset of governance regimes covering all of sub-Saharan Africa to provide new analytical frameworks for analysis and major contribution to empirics of growth. Studies on African monetary unions by Tchundjang Pouémi (1980), Devarajan and De Melo (1991), Monga and Tchatchouang (1996), and Monga (1997) revealed some of the intrinsic issues of fiscal coordination, credibility, and growth in small open economies with a fixed exchange rate. These studies could have been of great use to researchers and policymakers in Europe, as they anticipated some of the issues that would later be observed in the crisis of the Eurozone. An area where African economists have been pioneers is the ana­ lysis of linkages between institutions, governance, and economic growth. While the precise nature or sequence of governance reforms needed to promote good outcomes and successful development is not yet well understood, African researchers have identified a clear relationship between good governance and equitable, sustainable economic growth. For instance, Nabli (2007) has highlighted the “checks and balances dimension” as the single most critical element in good governance, rather than a single branch of government such as the judiciary, civil service, or legislature. At the micro level too, the lack of quality data has long been presented as impeding economic research on Africa to the point that analyses of poverty and inequality were sometimes based on extrapolation of trends observed from surveys carried out in other countries. Most African countries have carried out at least one household survey of income/expenditure during the 2000 decade and more than two-thirds conducted a household labor force survey (with half undertaking one or more special surveys focusing on the informal sector) but Algeria, the Democratic Republic of the Congo, and Nigeria, three of the largest countries, have not carried out a population census in 20 years. Yet, the respectability and status of economic research are not always correlated with the scope and quality of available data. Even in industrialized economies, the quality and legitimacy of data sets used for research have improved only over time. Throughout much of the nineteenth and twentieth centuries, countries like Italy, Greece, the Soviet Union, and certainly most East European countries

Introduction   7 did not have credible, high-standard, national-level, micro-, and macroeconomic data. The same was true for India and even China. But these limitations did not prevent the academic community from taking seriously and being influenced by some of the analytical (both theoretical and empirical) work carried out there. So, other more convincing explanations must be given to the benign neglect of economic research in/on Africa. In fact, despite the poor quality and the inconsistency of the data, African economies have provided some opportunities for cutting-edge analytical work and the continent’s labor, credit, and land markets are slowly becoming fertile grounds for major theoretical innovation: African employment, unemployment, labor markets institutions, and migration were the topics of important advances and high-level intellectual debates in the seventies (Stiglitz 1974; Fields 1975; Harris-Todaro 1977).3 Likewise, microeconomic studies of African financial markets have led to a new understanding of the economics of rotating savings and credit associations (Besley et al. 1993). Aryeetey (1988) identified new criteria for optimal patterns of financial integration. Udry (1990) used findings from fieldwork in Nigeria to challenge conventional views of rural credit markets as using collateral and interlocking of contracts to overcome informational asymmetries between borrowers and lenders. Nguessan’s (1996) work shed new light on theories of central banking governance. Laffont and Nguessan (2000) enriched the literature on optimal collusion-proof group contracts. Focusing on adverse selection as a foundation of group lending, they developed a simple static model to show that there is no collateral effect if borrowers do not know each other. If the borrowers know each other, group lending can yield efficient lending, provided that there is no collusive behavior and transfers are not allowed between colluding partners. Studies of human capital were also stimulated by the theoretical model developed by Jovanovic and Nyarko (1996) on the beneficial effects of learning by doing and the optimal strategies for adopting new technologies—an analysis at least partly inspired by the experience of African countries. Bates’s (1981) compelling analysis of the reasons why public policies are consistently adverse to most farmers’ interests, offered a rationale for creating market distortions, skewing incentives, and burdening social welfare and long-term development. His political-economy analysis of the paradoxical features of development in modern Africa has provided convincing explanations of how governments around the world have intervened and diverted resources from farmers to other sectors of society. In recent years, Pinkovskiy and Sala-i-Martin (2010) have used a new methodology, which combines the standard Penn World Tables GDP series with a comprehensive inequality database, to support empinically the idea of an African economic renaissance. They invalidate the popular image that the poor majority in all African nations and many African nations as a whole are stuck in “poverty traps” created by unfortunate geography and calamitous history. Their work shows that across the continent, poverty fell for both landlocked and coastal countries, for mineral-rich and mineral-poor countries, for countries with favorable and unfavorable agriculture, for countries with different colonizers, and for countries with varying degrees of exposure to the African slave trade. Their study also demonstrate that the benefits of growth were so widely distributed that African inequality actually fell substantially. Likewise, Young (2012) uses measures of real consumption based on the ownership of durable goods, the quality of housing, the health and mortality 3 

See Collier (1993) for a critical and detailed review.

8   Africa and Economics of children, the education of youths, and the allocation of female time in the household to show that sub-Saharan living standards have, for two decades, increased about 3.4–3.7 percent per year, that is, three and a half to four times the rate indicated in international data sets. Again, the methodological insights from these studies extend well beyond Africa, enrich the discipline of economics, and should have changed the way development experts and the international community approach issues of global poverty and inequality. These various contributions to economics by researchers working on Africa represent only a very small sample of the intellectual portfolio that has been developed in recent decades. They show that data is not (or no longer) the binding constraint that has prevented broader validation and wide recognition. Yet, the economics of Africa still lies very much on the fringe of the discipline and few students who enroll into PhD programs in top colleges and universities around the world choose the continent as their main area of interest. One would have expected that the region that is the last development frontier and the most challenging would also attract the best minds, according to the well-known principle that potential return rises with an increase in risk. The other traditional explanation as to why economic research in and on Africa has not have a more visible impact and influence on mainstream economic thinking is methodological—the suggestion being that it is part of development economics, a sub-discipline that has not followed the most rigorous standards in vogue in the field. Krugman contends that the marginalization of development economics for many decades results from the methodological choice made by Albert Hirschman, Gunnar Myrdal, and others, to reject the drive toward rigor, to ignore the pressures to produce buttoned-down, mathematically consistent analyses, and adopt instead a sort of muscular pragmatism in grappling with the problem of development. The crisis of high development theory in the late 1950s was neither empirical nor ideological: it was methodological. High development theorists were having a hard time expressing their ideas in the kind of tightly specified models that were increasingly becoming the unique language of discourse of economic analysis. They were faced with the choice of either adopting that increasingly dominant intellectual style, or finding themselves pushed into the intellectual periphery. They didn’t make the transition, and as a result high development theory was largely purged from economics, even development economics. (Krugman 1995)

While that critique may have been valid in the early years of development economics it is a bit unfair. By today’s standards, it is indeed surprising that Rosenstein-Rodan’s (1943) paper, the founding statement of the sub-discipline of development economics, was published in a leading academic journal of the time without any formal analytical content or even reference to formalized works. But of the methodological choices made by early structuralists were dictated by the state of knowledge in the economics discipline in the early 1940s. Many important researchers did not have to use mathematical models for their work in economics to be internationally recognized. Moreover, development economics emerged with the early structuralists’ challenge to neo-classical economics—a challenge that also had methodological implications. These early structuralists neglected the use of formal models because they viewed econometrics mainly as a tool for analyzing data from actual experience. They were suspicious of it because classical statistical theory tended to impose stringent requirements on data—especially

Introduction   9 good time series—that was not always available in many developing countries. The initial proponents of development economics wanted to explain complex economic phenomena, which could not be captured realistically using models with one, two, or very few variables in so-called “reduced form.” The distinguishing feature of their framework was the notion that macroeconomics must relate to the institutional structure of an economy and the perceived behavioral patterns of households and firms, and that economics must be constructed directly in terms of aggregates such as household consumption, business investment, and total exports, not from optimizing decisions made by individual “agents.” These features stood in sharp contrast with the basic assumption of neo-classical economics—the mainstream interpretation of Adam Smith—that businesses and consumers act rationally to maximize in one case their own profits and in the other case their own welfare (Lin and Monga 2014). Even when the proponents of the so-called “late structuralism” shifted methodological gears and adopted mathematics in the exposition of their views,4 they still did not manage to get their ideas into mainstream economics. In fact, another not-often-discussed reason for the benign neglect of economic know­ ledge from Africa (and other parts of the developing word) has always been the intellectual hegemony of some people in Western academic institutions. This hegemony has been exerted in at least two ways: first, the academic community in the West has always assumed that their theories were universal, and therefore applicable without adjustment or even customization to the developing world, including Africa. Many intellectuals from developing countries accepted that hegemony. They went to study in the West to seek the sutra for the modernization of their motherland. When they returned home to work or teach, they generally tried to use the knowledge and tools learned in the West to understand and address the problems in Africa, rather than trying to understand the true causes of the problems, and propose new theories and new answers. Second, even when some economists from Africa attempted to break that hegemony and carried out research that reflected the specific conditions of the continent and yielded new theoretical insights, their works were rarely accepted and recognized by mainstream economists if the insights conflicted the prevailing dominant theories in the West. For instance, the often notable work done over several decades by African researchers such as Samir Amin and Bernard Founou at the African Institute for Economic Development and Planning (IDEP) in Dakar, Senegal, was easily ridiculed for its Marxist tone despite its rele­ vance for structural transformation and inequality. Africa-based research institutions such as the African Economic Research Consortium (AERC), the Council for the Development of Social Science Research in Africa (CODESRIA), or the Economic Research Forum (ERF), only gained respectability in the 1990s when they started accommodating mainstream approaches in their research agendas. Yet, for decades, these institutions had produced original and interesting work in various areas of the social sciences, which remained largely 4  Early structuralists devised dual economy models, also called small general equilibrium models with a large agricultural sector, to analyze the long-run development of an economy with two specific sectors, a large agricultural and an industrial or modern one and explore the path through which a poor economy does convert itself into an industrial one. A second generation of structuralists offered models that were more comprehensive (they included both real and financial sides), dynamic, sensitive to initial conditions, and responsive to the nature of the data to which they were calibrated. See Taylor (1992, 2004).

10   Africa and Economics ignored. Some of Africa’s most original economists and thinkers are alumni of these institutions.5 Likewise, the United Nations Economic Commission for Africa (UNECA) has long suffered the same type of “leftist” suspicion that clouded for a long time the reputation of its Latin American equivalent (ECLAC) in certain powerful intellectual and policy circles. In sum, although the issues of data limitations and methodological choices cannot be underestimated, other much credible explanations for the low level of attention given to economic research on Africa should be explored.

4  Economics and Africa: Analytical Sins and Mimetic Choices If the data deficit is not the main explanation for the relative marginalization of economic research on Africa, what else could explain it? Without carrying econometric analyses to answer that question, one can refer to the broader difficulties that have confronted economics a discipline since World War II (with a climatic identity crisis after the 2007–2009 Great Recession and the calls for rethinking), and intellectual mimicry (the propensity to blindly transfer to the African context all the dominant intellectual frameworks in vogue in high-income countries) as be the two most credible factors. Let’s start with the first of these two factors, namely the ups and downs of economics itself. Research on Africa could obviously not escape from the debates, challenges, and travails of economics over the decades—and most notably macroeconomics, the crown jewel of the discipline. The brutality of the Great Recession—largely unforeseen to many mainstream economists—has opened up a fierce debate on the identification of its causes, and on the validity of a lot of the economic knowledge (Blanchard 2009; Solow 2008, 2009; Akerlof and Shiller 2009; Krugman 2009; Acemoglu 2009; Stiglitz 2010; Cochrane 2011; Lin 2013). Beyond the polemical nature of some of the arguments, it is clear that the existing corpus of economics acknowledge did not (or could not) predict a crisis of the scale observed in 2007–2009. The global crisis has not invalidated everything about macroeconomics. However, it highlighted some mistakes of the discipline’s dominant intellectual framework, and reopened some of the old wounds that have marked macroeconomics since its inception. A short intellectual history of macroeconomics would start with early Keynesianism, and the widely shared belief that researchers had learned much of what mattered from the Great Depression and from the boom of the post-war period. The main if not only objective of macroeconomic policy was then the management business cycles. Keynesian economics yielded a very straightforward conclusion: market economies, central banks, and governments could use the two main macroeconomic instruments—monetary policy and budgetary policy—to freely set the unemployment rate at a desired level, provided that a certain rate of inflation is factored in. In short, macroeconomic management boiled down 5  Benno Ndulu, Ibrahim Elbadawi, Thandhika Mkandawire, Charles Soludo, Ibrahim Lipumba, Mthuli Ncube, Bernadette Kamgnia, Elizabet Asiedu, Adebayo Olukoshi, Mustapha Kassé, Mohamed Dowidar, Ahmed Galal, Ernesto Gove, or Achille Mbembe—just to name a few—are among them.

Introduction   11 to an inflation-unemployment trade-off exercise guided by the ideological, social, or policy preferences of the authorities. That initial consensus, which can be labeled Macroeconomics 1.0, led to the widely shared belief in a quasi-stable relationship between the nominal wage growth rate and the unemployment rate. The relationship was diagnosed in a famous curve (Phillips 1958). It complemented and reinforced the fixed-price Keynesian model. For a while, almost everyone was ecstatic about such a simple and elegant idea: the authoritative curve made it possible to explain inflation in a clear manner. The curve seemed rather convincing and showed visually that when the unemployment rate decreases, nominal wages increase and impact general price levels. Wage inflation explained the increase in the general level of prices. Like an infallible oracle, the curve showed that one could only decrease unemployment at the cost of an increase in inflation. Conversely, if one chose to reduce inflation, it would be necessary to put up with a rise in unemployment. Thus, the belief was that it was just a matter of trading off two priorities, resolving the dilemma between inflation and unemployment. It was however a mistake—perhaps the first in a long list of analytical sins. Although the general intuition of Philips (the author of the famous curve) and Keynes’s analysis were reasonable, they both rested on shaky assumptions, especially the idea that public policymakers can manipulate economic agents at will. Such a postulate was oblivious of the fact that nothing is permanent, and that nothing is stable. The initial framework was therefore based on rigid and exogenous behavioral assumptions (i.e. it considered the behavior of agents to be unchanged). What can be true in the short term is not always in the medium or long term. The idea that public policymakers could manipulate agents at will was naive. Satisfied with the coherence of the rather simplistic reasoning on which the analytical framework of their science was founded, macroeconomists believed for several decades they had elevated their discipline to the status “hard” science like physics or chemistry. Yet, the idea behind the Philips curve was quite simplistic. Economic policymakers could fool the agents by faulting their expectations. At no time did the emerging macroeconomic science consider that agents could react to or even anticipate circumstances and policies, and adopt unanticipated behaviors. The conservative revolution of the 1970s and the study of anticipations invalidated key Keynesian assumptions and gave rise to what may be termed Macroeconomics 2.0. It showed that the famous trade-off between unemployment and inflation was actually an illusion. One obvious reason is the fact that the unemployment rate in a market economy cannot drop below a “natural rate” reflecting the fraction of the labor force that is unable to work at any given time for various reasons. That rate depends on many factors including human capital obsolescence or the loss of job skills that worsens unemployment, the fact that some job­ seekers may drop out of the labor force, the reinforcement of the bargaining power of unionized workers against new recruitment, the existence of a generous social protection system, etc. Having understood that economic agents form their anticipations, not only on the basis of observation of past experience, but also on an intuitive and often sophisticated appraisal of the manner in which the economy operates, macroeconomists took a bold step forward. They came to the conclusion that the short-term unemployment-inflation trade-off reflected in the famous Phillips curve was not sustainable in the long term, and that a non-accelerating inflation rate of unemployment actually transforms the curve into a straight line. When the central bank embarks on an expansionist monetary policy to help the government create the feeling of artificial wealth, workers anticipate quite well the future

12   Africa and Economics implications of such a decision on the economy. Having some knowledge on how the central bank acts to control the money supply and run an optimal monetary policy, these agents are capable of observing and analyzing the decisions of the technocrats—who believe they are infallible—and in so doing, can predict the purchasing power risk they could face in the future. In countries with powerful trade unions, for example, workers immediately incorporate in their salary demands the effects of inevitable price increases. Consequently, the monetary policy decisions carefully prepared by the authorities do not ultimately have any impact on the real economy. In other words, monetary policy, as an instrument used for stimulating aggregate demand and economic growth, even in the short term, is no more than an illusion. It does not even help gain in job creation what is lost in the fight against inflation. The constant and rational anticipations formed by economic agents have a clear implication: money can never seriously influence the general level of economic activity. Any attempt at increasing money supply is neutralized immediately by a rise in inflation. When money growth has no effect on real economy (or real equilibrium), it is said to be super-neutral (Sargent 1987). The logical conclusion is obvious: the responsibility of the central bank must be limited to contain inflation, and thus to hold back the progression of money supply. Despite its appeal and the elegance of its modeling tools, neo-classical economics too had its shortcomings—and critics. In his Nobel lecture, Stiglitz summed up well what was wrong with rational expectations theories: In the 70s, economists became increasingly critical of traditional Keynesian ideas, partly because of their assumed lack of micro-foundations. The attempts made to construct a new macro-economics based on traditional micro-economics, with its assumptions of wellfunctioning markets, was doomed to failure. Recessions and depressions, accompanied by massive unemployment, were symptomatic of massive market failures. The market for labor was clearly not clearing. How could a theory that began with the assumption that all markets clear ever provide an explanation? If individuals could easily smooth their consumption by borrowing at safe rates of interest, then the relatively slight loss of lifetime income caused by an interruption of work of six months or a year would hardly be a problem; but the unemployed do not have access to capital markets, at least not at reasonable terms, and thus unemployment is a cause of enormous stress. If markets were perfect, individuals could buy private insurance against these risks; yet it is obvious that they cannot. Thus, one of the main developments to follow from this line of research into the consequences of information imperfections for the functioning of markets is the construction of macroeconomic models that help explain why the economy amplifies shocks and makes them persistent, and why there may be, even in competitive equilibrium, unemployment and credit rationing. (Stiglitz 2001)

“Neo-Keynesianism,” which could be referred to as Macroeconomics 3.0, emerged in reaction to neo-classical criticisms and tried to build macroeconomic theories from the lessons from the microeconomics of goods, labor, and capital markets. It also started with the premise that persistent unemployment and economic fluctuations are major policy issues—with recessions and depressions reflecting big market failures. But its various brands did not offer a unified vision of how the economy behaves. One school of thought within new Keynesian economics sought to explain price rigidities. Romer (1993) for instance believed that firms are imperfectly competitive, and face small barriers to nominal price flexibility. Emphasizing the complementarity between real and nominal rigidities, he showed that even small barriers can have large macroeconomic effects. Greenwald

Introduction   13 and Stiglitz (1993) suggested another brand of new Keynesian economics. They downplayed the importance of price rigidity, and focused instead on market failures in labor and capital markets. They explained market imperfections by costly and imperfect information, and also argue that small disturbances to the economy can lead to large and persistent fluctuations. Tobin (1993), a proponent of yet another view, argued that the main problem of his fellow neo-Keynesians is that they really did not incorporate old (true) Keynesianism in their thinking; he advocated stabilizing policies (or policy rules) that create and sustain regressive expectations of output and price departures from equilibrium. “In the absence of activist ‘feedback’ policies, he wrote, monetary and fiscal, flexibility may well be destabilizing, both to prices and to real macro variables. Governments and central banks should not expect disinflation or deflation alone to maintain or restore full employment” (1993: 64). The battle between various groups of Keynesian and neo-classical economists eventually led to some sort of truce, with each side painfully accepting a compromise. A dominant synthesis—defined by Blanchard (2009) as mainstream macroeconomics—emerged an attempt to mediate the methodological battles that had shaken the discipline of economics for decades. It therefore tried to reconcile the strengths of the neo-classical and new-Keynesian frameworks. On the one hand, it used the tools of dynamic stochastic general equilibrium theory, taking preferences, constraints, and optimization as a starting point and then building on these microeconomic foundations. It satisfied free market theoreticians. On the other hand, it validated the idea of nominal rigidities, which helped explain why monetary policy can have real effects in the short run. That consensus framework generally assumed that the economy is dominated by monopolistically competitive firms that change prices only intermittently, which creates price dynamics and the so-called new Keynesian Phillips curve (Blanchard and Perotti 2002). However, the elegance of the new, synthetic model did not compensate its most obvious flaw: at the core, it considers the economy—any economy—to be a dynamic general equilibrium system that deviates from Pareto optimum mainly because of sticky prices. The truth of the matter is that there are a number of other market imperfections that constantly force the economy to deviate from optimality. Moreover, the key principles for modeling the agents’ expectations, preferences, decisions, and behavior are simply too unrealistic and it is too risky to rely solely on them for policy design. The very idea of representative agents, which underlines the current consensus in macroeconomic theory, is inconsistent with the heterogeneity that is the dominant feature in almost all economies. There is no doubt that strong analytical progress has been made since the days of John Maynard Keynes. However, most of the existing mathematical models of economic systems and even business cycles are only remotely reflective of the behavior of households, firms, and governments. Their use of microeconomic tools has often remained rudimentary, and their neglect of lessons from other disciplines has been a mistake. In fact, progress in macroeconomics may have been inversely proportional to the intellectual investment in macroeconomic modeling. Mankiw has observed that “while the early macroeconomists were engineers trying to solve practical problems, the macroeconomists of the past several decades have been more interested in developing analytical tools and establishing theoretical principles” (2006: 30). Solow, one of the most important names in the history of the discipline, is even more critical in his assessment of progress in macroeconomics. In a scathing review, he challenges the

14   Africa and Economics mood of “self-congratulation” among macroeconomists who still anchor their models in the optimizing behavior of firms and consumers. Solow highlights many analytical sins committed by mainstream economists and mocks “modern macro” as macroeconomics that is deduced from a model in which a single immortal consumer-workerowner maximizes a perfectly conventional time-additive utility function over an infinite horizon, under perfect foresight or rational expectations, and in an institutional and technological environment that favors universal price-taking behavior… After all, a modern economy is populated by consumers, workers, pensioners, owners, managers, investors, entrepreneurs, bankers, and others, with different and sometimes conflicting desires, in formation, expectations, capacities, beliefs, and rules of behavior. Their interactions in markets and elsewhere are studied in other branches of economics; mechanisms based on those interactions have been plausibly implicated in macroeconomic fluctuations. To ignore all this in principle does not seem to qualify as mere abstraction—that is setting aside inessential details. It seems more like the arbitrary suppression of clues merely because they are inconvenient for cherished preconceptions. I have no objection to the assumption, at least as a first approximation, that individual agents optimize as best they can. That does not imply—or even suggest—that the whole economy acts like a single optimizer under the simplest possible constraints. So in what sense is this ‘dynamic stochastic general equilibrium’ model firmly grounded in the principles of economic theory? (Solow 2008: 243–244)

In sum, the quest for macroeconomic knowledge is still in an unfinished journey and the spectacular development of its many sub-fields only makes learning more challenging—and more exciting. Arrow and Bresnahan note that Economics is one of the least specialized of all academic fields. Any economist feels capable of at least understanding and possibly even evaluating work in many areas of specialization. But the growth of the literature has made it less and less possible to understand the frontier of a field other than one’s own (and even that is increasingly difficult). This is especially a problem for economics, because economic problems are never distinct. Microeconomic allocations and macroeconomic fluctuations and growth both affect and are affected by foreign trade and by changes in industrial organization. Nor can economic policy formation ignore theory or evidence, and empirical economists (whatever their methods) ignore theory and policy issues at their peril. In economics then, as in every field of academic endeavor, skilled researchers need access to work at the frontier in different areas of the subject. (Arrow and Bresnahan 2009)

The struggles of economics as a science and controversies within the discipline were reflected in the evolution of development economics. Not surprisingly, economic research on Africa too was influenced by these debates. This often led to intellectual mimicry, which is the second main factor behind its relative marginalization. Even today, one of the biggest intellectual problems in development economics—to which economic research on Africa obviously belong—is the assumption of similarity among all countries, regardless of their levels of development and endowment structures at a given moment in time. By ignoring the structural differences between high- and low-income economies, many “Africanist” researchers committed yet another series of analytical sins. It is easy to intuitively recognize that Burundi is not Switzerland even though they are both small and landlocked. Yet economists have struggled to systematically elaborate theoretical, empirical, and policy frameworks consistent with the most essential stylized facts.

Introduction   15 Knowledge development can be sparked by imitation. French philosopher René Girard has suggested the concept of “mimetic desire” to explain how humans, the species most apt at imitation, often manage to put themselves in trouble. He highlights the fact people also imitate other people’s desires, and depending on how this happens, it may lead to conflicts and rivalries. Girard (1977) distinguishes “imitation” as (the positive virtue of reproducing someone else’s behavior) from “mimesis” (the negative aspect of rivalry, which also refers to the deeper, instinctive response that humans have to each other). Economists working on Africa and policymakers who dream to transform low-income economies into vibrant industrialized powerhouses through a rapid process of modernization have too often fallen into the trap of adopting wholesale thinking and policy frameworks from abroad, which are not suited to the prevailing specific conditions (i.e. the comparative advantage determined by the “endowment structure”) of the country contexts in which they operated. In such circumstances, imitation has often turned into mimesis and generated tensions, false rivalries, and disappointment. Lin (2009) offers a theoretical account of such analytical sins. Examining the development models that dominated the development literature in the 1950s and 1960s, he mostly finds “a comparative-advantage-defying (CAD) strategy.” In the aftermath of World War II when development economics rose to some prominence, the main objective was to convert previously colonized countries and territories into successful economies with industrial structure that looked just like the most advanced ones. The dominant mode of thinking among leading development economists was to elaborate theories and policies relying on central planning of investment, state ownership, import substitution (that is, protectionism), capital and exchange controls, and other interventionist policies, to achieve such a mimetic goal. As Nobel laureate Gunnar Myrdal observed: “The special advisers to underdeveloped countries who have taken the time and trouble to acquaint themselves with the problem … all recommend central planning as the first condition of progress” (1956: 201). The drive to catch up to developed countries as quickly as possible and the appeal of nation building led economists and leaders in many underdeveloped countries to disregard the economy’s “endowment structure”—that is, “the relative abundance of capital, labor, and natural resources”—and steer resources away from labor-intensive toward capital-intensive industries (such as iron and steel). Trade barriers were intellectually justified as necessary tools to protect infant manufacturing industries, while favored domestic firms were given cheap loans and other forms of government assistance. The goal was to build the “commanding heights” of the economy (heavy industries). Conventional wisdom held that the existence of market failures justified interventionist policies to protect even “nonviable” firms—that is, firms that could not survive in a free market. Although some Latin American countries managed to perform temporarily well with policy prescriptions from early structuralism (Ocampo and Ros 2011), the results across the developing world were generally disappointing. The perceived general failure of the first wave of early structuralist economics (Development Economics 1.0), especially in its stated aim of structurally transforming poor countries’ production pattern, led to the return of neo-classical orthodoxy in development thinking in the 1970s and 1980s. In the aftermath of the first oil shock and a radical questioning of Keynesian economics, a group of development experts—mainly within the World Bank and the International Monetary Fund—redefined sustained growth and structural changes in poor countries as comprising macroeconomic stabilization and structural

16   Africa and Economics adjustment policies. That new framework, which came to be known as the Washington Consensus, represented the second wave of development thinking—Development Economics 2.0. It focused on government failures. While it underwent some variations, it became the blueprint for economic transformation in many developing countries with the aim to establish well-functioning market institutions similar to those in high-income countries. Unfortunately, its results were also disappointing (Lin and Monga 2014). Parallel to the dismissal of early structuralism and the re-emergence of neo-classical economics, by the end of the 1980s, the development economics research community was witnessing the end of an era dominated by research based on cross-country regressions, which attempted to identify growth determinants (Barro 1997). Yet again policy prescriptions stemming from such regressions did not produce tangible results. That approach focused on independent and marginal effects of a multitude of growth determinants. This led to the linearization of complex theoretical models. Later on it became clear that growth determinants interact with each other and that to be successful, even the “right” policy reforms must be implemented with other reforms. Taking the issue of growth and poverty reduction from a completely different angle—policy results on the ground—another popular addendum to neo-classical development economics has gained much publicity in recent years. It is the quest for rigorous impact evaluation of development projects and programs and can be labeled Development 2.5. It has generated a new, micro approach to development led by economists at the MIT Poverty Lab, whose goal is “to reduce poverty by ensuring that policy is based on scientific evidence” (Banerjee and Duflo 2011). Proponents of these micro recipes suggest applying the simplicity and robustness of randomized controlled trials (RCT) techniques—the same approach used by the medical industry to determine if a drug or treatment does what it was designed to do—to poverty interventions to determine whether or not a program is effective. However, assessing the impact of specific projects, no more than one at a time, without taking into consideration the many sources of heterogeneity relevant to behavior and the interaction effects—the fact that each such project is only one component of a development portfolio that often cuts across sectors—can lead to biases (Ravallion 2009). For economists engaged in research on structural change, the main questions should be why and how some countries succeeded and others failed to fundamentally transform their economies, so that the countries that remain trapped into poverty can derive policy insights from past experiences to avoid mistakes and start successfully a sustained, dynamic growth in their countries. RCT studies are not designed to answer such macro questions. For development economics in general and economic research on Africa in particular, an important lesson from these debates and from the evolution of the discipline is perhaps the need to go back to Adam Smith’s methodology, that is, to refocus the work on inquiries into the nature and causes of the wealth of nations. A third wave of development thinking is currently underway, which builds on both early structuralism and neo-classical economics. Drawing lessons from history and economic analysis, it aims at reconciling insights from previous brands of development knowledge and to provide policymakers in all low-income countries with practical frameworks for identifying sectors and industries that are consistent with their comparative advantage (Aghion et al. 2012; Lin 2012a, b; Lin and Monga 2014, 2011; Monga 2013), and facilitating the process of structural change (Rodrik 2006, 2007; Hausmann et al. 2008; Hidalgo and Hausmann 2009). The third wave also draws lessons from history to identify factors that help or hinder the reallocation of resources from low- to

Introduction   17 high-productivity sectors. It recommends a new distribution of roles between governments and markets (Aghion and Howitt 2009). It is worth noting that various regions of the developing world have been the focus of development economics at various moments in time, often influencing that sub-discipline quite heavily. Looking simply at the volume of books and articles in English, one could argue that after the publication of Rosentstein-Rodan’s (1943) landmark article, which analyzed “problems of industrialization of Eastern and Southeastern Europe,” the development experience of the Asian Subcontinent was more often studied by (future) leading researchers who produced seminal contributions to mainstream economics in the 1950s and 1960s—such as Gunnar Myrdal, Amartya Sen, Theodore Schultz, Jagdish Bhagwati, or Dwight Perkins—while the economic stories of Latin America only took center stage in the discipline and dominated development thinking and policy during the 1980s and 1990s, thanks to the works of Rudi Dornbusch, John Williamson, Sebastian Edwards, and many others. The real story is obviously more complex: Latin America’s development experience has always been the subject of debates in economics and researchers such as Arthur Lewis, Raul Prebisch, Albert Hirshman, or Celsio Furtado made their marks on development thinking in the 1950s and 1960s, even if their works were not published by the leading economic journals at that time. Ocampo and Ros (2011) offer several good reasons for such dominance: the region’s traditionally high levels of inequality in the distribution of income in comparison to other developing regions and the developed nations; the debates on the importance of geography versus institutions in development, which have often concentrated on Latin America’s colonial legacy; the region’s controversial experiences with import substitution industrialization in the post-war period and subsequently with market reforms and the Washington Consensus policies; and the abundance of natural resources in the region and the resulting specialization of many countries in primary exports. Yet, on many important research and policy questions such as the “Dutch disease,” the “resource curse” hypothesis, the effects of the pattern of trade specialization on economic growth and inequality, or the appropriate balance of state and markets in different stages of economic development, the most readily available general books in English on the economics of the region are the institutional reports published by multilateral organizations with their well-known strengths and weaknesses. “Furthermore, most of the literature on Latin American economics is generally published in Spanish and Portuguese, not English, and the literature available in English is biased towards certain conceptual frameworks, and therefore tends to leave aside analyses by the school that is broadly known as Latin American structuralism (and neo-structuralism)” (Ocampo and Ross 2011). The same could be said about Asia’s economic development experiences—including China’s unpredicted but spectacular growth and poverty reduction—which are still understudied given their potential contributions not only to development economics but also to mainstream economics and to the broader field of the social sciences. The discussion of the travails of economics in general and development economics in particular underlines the need for humility but also the excitement that should guide economic research on Africa—a field that has already yielded important insights in the works of some of the best minds in the field. Because the continent is still the world’s region with the slowest pace of convergence and the most difficult economic challenges, the potential intellectual payoffs of research there are the highest.

18   Africa and Economics

5  Explorations at the Development Frontier: The Challenge of  Relevant Knowledge The title of this handbook, which has puzzled some of the reviewers of the initial proposal, deserves some explanation. “Africa and Economics” may sound elusive. True, we thought of the title as the main channel for conveying the objectives and content of the Handbook and also a marketing tool to arouse intellectual curiosity and entice potential readers. In doing so we acknowledge the philosophical risks of some exteriority report between the two concepts, one in which economics could be seen as a foreign reality. We are also well aware of the potential traps of essentialism, which lurk behind any intellectual enterprise associated with geography and ethnicity. It should be noted, however, that such accusations tend to be more readily made about intellectual projects devoted to Africa. After all, no one seriously suspects Burda and Wyplosz (1993) for falling into essentialism despite the fact that they adopted a deliberately “ethnic” perspective in their well-known classic volume titled Macroeconomics: A European Text. Likewise, there have been some important books about “Chinese” or “American” philosophy. The Oxford Handbook of Africa and Economics does not advocate some sort of “African economics” but examines ways in which the science of economics has sought to decipher a continent where economic progress has been slow for centuries, and how insights from the study of African societies have enriched or could enrich economics. The Handbook emphasizes the relation between Africa and Economics. One of its goals is precisely to highlight areas in which the study of African economies makes economics a “domestic” intellectual reality and enriches knowledge. Africa’s economic experience has been peculiar in many ways that are highlighted throughout the Handbook. Even today, the continent’s development trajectories are marked by its unique economic history and unique ways in which agents have internalized the memories of its past. However, in an era of globalization and transformation, these facts do not justify any form of intellectual determinism. They simply deserve some consideration in a book that attempts to study the symmetrical relationship between Africa and economics. The Oxford Handbook of Africa and Economics has two main purposes. One is to examine how the various branches of economics have been applied to Africa’s experience and how well they have helped understand some of the specific issues there. It therefore aims at shedding light on the evolution of economic knowledge on a continent that remains the last development frontier despite some progress in recent years. Unlike other social sciences disciplines such as anthropology or sociology, economics has historically challenged the significance of particularities and assumed the universal validity of formal constructs, concepts, theories, and general assumptions. This two-volume handbook identifies the central themes, issues, questions, and methods of analysis of economics, and discusses how they have been approached in the African context over time. In doing so, it highlights the relevance of the particular socio-historical context of Africa, the various waves of research, the main currents, the debates—including on the results of economic policies—and the relationship with the evolution of thinking in the mainstream economic literature.

Introduction   19 A second purpose of the Handbook is to review and document—often indirectly—how the study of African societies has contributed to and shaped various fields of the discipline of economics. Although the rigorous study of African economies was long considered both difficult, because of data shortages, and of low quality, it has always posed exciting conceptual and policy challenges to economic theorists. The study of African economies no longer lies in the hand of a small number of peripheral or isolated academics, as was the case half a century ago. Nor does it still lie on the fringe of mainstream analyzes. An increasingly large number of researchers are deriving some of their most powerful analytic insights from work on Africa. In fact, mainstream macroeconomics has already been enriched by cutting-edge work carried out on Africa on issues as diverse as the modeling of small open economies or the theory of repressed inflation. Likewise, mainstream microeconomics has benefited enormously from the study of factor markets, product markets, and household economics across the continent of Africa. Perhaps the domain where Africa’s contribution to economic knowledge has been the most obvious is that of political economy, for obvious reasons: beyond its large number of nation-states, the region’s long and intense socio-political history has been a unique laboratory for various experiments in economic policymaking and political management. Because knowledge from African economic issues is now widely acknowledged as significant and persuasive as knowledge from any other part of the world—no less, no more—The Oxford Handbook of Africa and Economics chronicles some of the ways in which economics as a discipline has applied to the particulars of the African context, and how the study of the continent has helped define economics. By systematically exploring that two-way relationship it contributes to the efforts of the global intellectual community to: heighten the awareness of students, researchers, and policymakers and development practitioners, of global issues and possibilities; encourage and share learning experiences; promote the understanding of the challenges to what becomes conventional thinking; and foster the perpetual quest for universal truths. The ultimate goal of the editors of this volume is to provide useful references and resources to an audience of scholars, students, development practitioners, and policymakers. Following King Lear’s admonition to always “express our darker purpose,” (Shakespeare), it is appropriate to clarify whether we had any other hidden intellectual agenda. The short answer is no. True, the two editors believe that development economics, which appeared after World War II with the intention of helping developing countries industrialize their economies, reduce poverty and narrow their income gaps with advanced countries, has generally failed to fulfill its promises. The developing countries that blindly followed the recommendations of its various waves to formulate their development policies did not achieve the intended goals. We therefore believe that it is time for a new phase of development thinking focused on structural change, driven by changes in endowment structure and comparative advantages and drawing useful lessons from successful developing, not just developed, countries and other disciplines (Lin 2012a, b; Monga 2011, 2013). The market should be the fundamental institution for resource allocation and the state should play a proactive facilitating role in the process. This aggiornamento is called New Structural Economics (NSE). However, the editors’ economic philosophy does not shape all the content of this volume. In conformity with the long tradition of the Oxford Handbook series, The Oxford Handbook of Africa and Economics aims at offering authoritative and state-of-the art surveys of know­ ledge, debates, and current thinking and research on Africa and economics. It is designed

20   Africa and Economics to bridge the gap between the sometimes overly technical contributions seen in academic journals, and the more concrete intellectual needs of policymakers and development practitioners. While we have proactively sought contributors from diverse backgrounds and intellectual horizons, we did not feel compelled to achieve political or ideological equilibriums. Our decision was also justified by the fact that the Handbook covers the entire continent of Africa, not just sub-Saharan Africa. As economic entities and from a purely analytical perspective, Morocco and Algeria are not fundamentally different from Cameroon and Angola. The Handbook is a two-volume project.6 This volume focuses on “Context and Concepts,” and includes 44 chapters organized into four sections: the first one deals with conceptual issues and examines the epistemological challenges surrounding the relationships between Africa and economics. It also surveys the uses of basic economic concepts such as households, marriage, families, farms, firms, and markets in the African context, and presents reflections on employment, unemployment and underemployment, inequality, governance, corruption, the continent’s growth, and new economic physiognomy. It ends with thoughts on African development from various perspectives. The second section discusses some methodological questions. It includes chapters on a reassessment of the traditional principles of economics, the importance of non-cognitive skills, the modeling of African economies (with a focus on the DSGE approach), and the debates over measurement of political progress, economic vulnerability, and competitiveness. The third section is devoted to critical reviews of the historical trajectories that led to Africa’s economic landscape. Arguments are made about the sustainability of Africa’s recent growth experience. Legacies of the African slave trades and colonialism are evaluated, together with issues of economic vulnerability and volatility, urbanization, informality, the public–private interface, capitalist business cultures, and the continent’s love affair with monetary unions. The fourth section explores the deeper economic trends underlying socio-political changes. It includes a political economy analysis of the new Arab awakening; discussions of the economics of authoritarianism and the potential economic dividends of North African Revolutions; an empirical analysis of the impact of democracy on economic growth in sub-Saharan Africa; an analysis of the economics of violent conflict and war; an exploration of the causes and consequences of terrorism; reflections on the relationship between democratic decentralization and economic development, and on the economics of happiness and anger in North Africa. In a letter to seek employment written to his uncle Lord Burleigh, in 1592, Francis Bacon expressed his enthusiasm and humility by emphasizing his thirst for learning: “I have taken all knowledge to be my province,” he noted (quoted by Evans 1969: 368). He did not get the job, perhaps because his uncle felt he was bragging too naively about what the human mind could achieve. Still, he later asserted that “knowledge is power” (De heresibus), “For all knowledge and wonder (which is the seed of knowledge) is an impression of pleasure in itself.” (The Advancement of Learning I.) While most people would agree that knowledge is an intrinsically good asset to possess, the challenging questions are about what knowledge 6 

Volume 2 is devoted to “Policies and Practices”.

Introduction   21 actually is, how to generate it, when it is useful and relevant to address social and policy issues, and how to make the best use of it under various circumstances. From ancient times philosophers and historians have grappled with such basic questions, which still confront economists and other social scientists. By asserting confidently in Theaetetus that “knowledge is justified true belief ” or “believing what is true and having sufficient reasons for it,” Plato thought he had solved the main problem. Yet, the criteria for validating such a sophism actually opened more questions than it provided answers: one had to believe, find the truth, and provide valid justification for it. Beliefs themselves have many features: they imply assumptions about perception, preconceptions, memory, interpretation of facts, mental models, etc. Not surprisingly, philosophers and psychologists have disagreed on how strongly one must believe in something for it to qualify as knowledge.7 Simply defined as the accuracy of an account of the world, the notion of the truth is even more complex. Proponents of the so-called correspondence theory consider truth to be a real alignment of what is believed and independent reality. Yet again, there are many subjective obstacles to validating such a statement. Coherence theory, which defines truth more humbly as the coherence of our beliefs with each other, does not solve the subjectivity problems either. Pragmatic theory, another philosophical school, maintains that the truth of beliefs is just a matter of their usefulness, which may be a practical way of approaching the problem but is also mired in relativism. Moreover, “true belief is not sufficient for knowledge. We can have beliefs that are true, yet still not have knowledge. Something more—a justification for what is believed—is needed. The most obvious reason for this is that the believer needs some way to recognize that his beliefs are true” (McInerney 1992: 41). But that third criterion of knowledge—its justification—also raises serious conceptual and theoretical problems: supporting evidence for true beliefs are not easy to find or to validate. Debates constantly rage between proponents of absolute certainty and those who contend themselves with whatever they consider “sufficient” conditions. Not surprisingly, these (still largely unresolved) philosophical problems have translated into debates in the social sciences. The two dominant but contrary views about the sources of knowledge reflect different intellectual traditions: rationalists, who draw inspiration from Descartes, Spinoza, Leibniz, and others, strongly believe that knowledge can be attained through reasons, and a process of deductive proofs based on self-evident first principles. Theorems can thus be proved from basic axioms and procedures and using mathematical models. By contrast, empiricists whose approach follows the tradition of John Locke and David Hume, consider knowledge to be the result of sense perception and “inner perception” of the operations of the mind. Although they too have strong faith in the power of reasoning through mathematical models to understand how ideas are related to each other, they contend that the way these ideas apply to the outside world is mainly through experience. In fact, no consideration of economic development strategies can take place outside an at least implicit theory of truth that underlines it. As Krieger reminds us, “our choice is not 7  René Descartes, for instance, claimed that one must be absolutely certain of something in order to know it and that anything less than absolutely certain belief is not knowledge, while John Locke thought that strong belief is sufficient for knowledge. For a critical analysis of these ideas, see McInerney (1992, Chapter 4) and Carr and O’Connor (1982).

22   Africa and Economics between having a theory or not having one; for have one (or two or three or more incompatible ones) we must. Our choice is rather between having an awareness of those theoretical issues which our criticism inevitably raises or going along without such an awareness” (1976: 7). However, there is always a limit beyond which the drive to theorize brings fewer and fewer knowledge benefits and more and more dangers of distraction. The problems posed by theory’s hegemony have been well recognized in economics (Sen 1977). They are even more acute when one realizes that “theory and assumptions are synonyms” and that “other synonyms of assumption are hypothesis, premise, and suppositions” (Manski 2013: 11). But in attempting to move in the opposite direction, researchers may have gone too far. The surge in empiricism is reflected in the wide reliance on ran­domized control trials as the dominant tool of analysis in development studies has also led to an almost religious belief in the intrinsic value of number games. Yet these so-called evidence-based methods, often postmortem evaluations of projects whose lessons are not transferable from one area to another, still fail to establish causation rigorously, or to inform policy choices because they do not really enhance the ability to predict whether government programs will be effective (Cartwright and Hardie 2012). They often lead to misleading or useless certainties. In some ways, they lead us back to David Hume’s skeptical empiricism. Where does this leave researchers seeking for economic truths and relevant knowledge in the African context? The authors assembled for this handbook fall into all sides of these old debates. Regardless of their philosophical background, they all aim at contributing to global learning, and to provide relevant knowledge to policymakers and practitioners of development in Africa and elsewhere. Some of them may define themselves primarily as theorists, while others would happily refer to themselves as empiricists. More importantly, they all share the humble belief that economic truths can only be reached by working from multiple angles. We invite the readers to keep this admonition in mind.

References Acemoglu, D. (2009). The crisis of 2008: Structural Lessons for and from Economics. Mimeo, Cambridge, MA: Massachusetts Institute of Technology, January 11. Aghion, P., Dewatripont, M., Du, L. et al. (2012). Industrial Policy and Competition. NBER Working Paper 18048, National Bureau of Economic Research, Cambridge, MA. Aghion, P., and Howitt, P. (2009). The Economics of Growth. Cambridge, MA: MIT Press. Akerlof, G.A., and Shiller, R.J. (2009). Animal Spirits:  How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism. Princeton, NJ:  Princeton University Press. Arrow, K.J., and Bresnahan, T. (2009). Preface. Annual Review of Economics, 1(1): doi: 10.1146/ annurev.ec.1.081709.100001. Aryeetey, E. (1988). Financial Integration and Development in Sub-Saharan Africa. London, Routledge. Banerjee, A., and Duflo, E. (2011). Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. New York: Public Affairs. Barro, R.J. (1997). Determinants of Economic Growth: A Cross-Country Empirical Study. Cambridge, MA: MIT Press.

Introduction   23 Bates, R.H. (1981). Markets and States in Tropical Africa. Berkeley:  University of California Press. Besley, T.J., Coate, S., and Loury, G. (1993). The economics of rotating savings and credit associations. American Economic Review, 83(4):792–810. Blanchard, O.J. (2009). The state of macro. Annual Review of Economics, 1(1):209–228. Blanchard, O., and Perotti, R. (2002). An empirical characterization of the dynamic effects of changes in government spending and taxes on output. Quarterly Journal of Economics, 117:1329–1368. Burda, M.C., and Wyplosz, C. (1993). Macroeconomics: A European Text. New York: Oxford University Press. Carr, B., and O’Connor, D.J. (1982). Introduction to the Theory of Knowledge. Minneapolis: University of Minnesota Press. Cartwright, N., and Hardie, J.  (2012). Evidence-Based Policy:  A  Practical Guide to Doing It Better. New York: Oxford University Press. Cochrane, J.H. (2011). Understanding policy in the great recession:  some unpleasant fiscal arithmetic. European Economic Review, 55:2–30. Collier, P. (1993). Africa and the Study of Economics, in R.H. Bates, V.Y. Mudimbe, and J. O’Barr (eds), Africa and the Disciplines: The Contributions of Research in Africa to the Social Sciences and Humanities. New York: Oxford University Press, pp. 58–82. Devarajan, S. (2013). Africa’s Statistical Tragedy. Review of Income and Wealth, 59:S9–S15. Devarajan, S., and de Melo, J. (1991). Membership in the CFA Franc Zone: Odyssean Journey or Trojan Horse? in A. Chhibber and S. Fischer (eds), Economic Reform in Sub-Saharan Africa. Washington DC, World Bank, pp. 25–33. Diamond, P. (2010). Biographical. http://www.nobelprize.org/nobel_prizes/economic-sciences/ laureates/2010/diamond-bio.html. Diamond, P. (1967). The role of a stock market in a general equilibrium model with technological uncertainty. American Economic Review, 57:759–776. Diamond, P. (1971). A model of price adjustment, Journal of Economic Theory, 3:156–168. Diamond, P., and Maskin, E. (1979). An equilibrium analysis of search and breach of contract, i: steady states. Bell Journal of Economics, 10:282–316. Diamond, P., and Maskin, E. (1981). An equilibrium analysis of search and breach of contract, ii: a non-steady state example. Journal of Economic Theory, 25:165–195. Evans, B. (1969). Dictionary of Quotations. New York: Wing Books. Fields, G.S. (1975). Rural-urban migration, urban unemployment and underemployment, and job-search activity in LDCs. Journal of Development Economics, 2(2):165−187. Girard, R. (1977). Violence and the Sacred. Baltimore: The Johns Hopkins University Press. Greenwald, B., and Stiglitz, J.E. (1993). New and old Keynesians. Journal of Economic Perspectives, 7(1):23–44. Harris, J.R., and Todaro, M.P. (1977). Migration, unemployment, and development: a two-sector analysis. American Economic Review, 60:126–141. Hausmann, R., Rodrik, D., and Velasco, A. (2008). Growth diagnostics, in N. Serra and J.E. Stiglitz (eds), The Washington Consensus Reconsidered: Towards a New Global Governance. New York: Oxford University Press, pp. 324–354. Hegel, G.W.F. (1956). The Philosophy of History. New York: Dover. Hidalgo, C.A., and Hausmann, R. (2009). The building blocks of economic complexity. Proceedings of the National Academy of Sciences, 106(26):10570–10575.

24   Africa and Economics Jerven, M. (2014). Economic Growth and Measurement Reconsidered in Botswana, Kenya, Tanzania, and Zambia, 1965-1995. New York: Oxford University Press. Jerven, M. (2013). Poor Numbers: How We Are Misled by African Development Statistics and What to Do about It. New York: Cornell University Press. Jovanovic, B., and Nyarko, Y. (1996). Learning by doing and the choice of technology. Econometrica, 64(6):1299–1310. Krieger, M. (1976). Theory of Criticism:  A  Tradition and Its Systems. Baltimore, MD:  Johns Hopkins University Press. Krugman, P. (2009). How did economists get it so wrong? New York Times Magazine, September 2. http://www.nytimes.com/2009/09/06/magazine/06Economic-t. html?pagewanted=all&_r=0. Krugman, P. (1995). The Rise and Fall of Development Economics. Cambridge, MA:  MIT, mimeo, http://web.mit.edu/krugman/www/dishpan.html. Laffont, J.J., and N’Guessan, T.T. (2000). Group lending with adverse selection. European Economic Review, 44:773–784. Lin, J.Y. (2013) Against the Consensus:  Reflections on the Great Recession. Cambridge, UK: Cambridge University Press. Lin, J.Y. (2012a) New Structural Economics:  A  Framework for Rethinking Development and Policy. Washington, DC: World Bank. Lin, J.Y. (2012b). The Quest for Prosperity: How Developing Economies Can Take Off. Princeton, NJ: Princeton University Press. Lin, J.Y. (2009). Economic Development and Transition:  Thought, Strategy, and Viability. New York: Cambridge University Press. Lin, J.Y., and Monga, C. (2014). The evolving paradigms of structural change, in B. Currie-Adler, R. Kanbur, D. Malone, and R. Medhora (eds), International Development: Ideas, Experience, and Prospects. New York: Oxford University Press, pp. 277–294. Lin, J.Y., and Monga, C. (2011). Growth identification and facilitation: the role of the state in the dynamics of structural change. Development Policy Review, 29(3):259–310. Mankiw, N.G. (2006). The macroeconomists as scientist and engineer. Journal of Economic Perspectives, 20:29–46. Manski, C.F. (2013). Public Policy in an Uncertain World: Analysis and Decisions. Cambridge, MA: Harvard University Press. McInerney, P.K. (1992). Introduction to Philosophy. New York: HarperCollins. Monga, C. (2013). Winning the jackpot: job dividends in a multipolar world, in J. Stiglitz, J.Y. Lin, and E. Patel (eds), The Industrial Policy Revolution II: Africa in the 21st Century. New York: Palgrave Macmillan, pp. 135–172. Monga, C. (2011). Post-macroeconomics: lessons from the crisis and strategic directions ahead. Journal of International Commerce, Economics and Policy, 2(2):277–304. Monga, C. (1997). A currency reform index for Western and Central Africa. World Economy, 20(1):103–125. Monga, C., and Tchatchouang, J.-C. (1996). Sortir du piège monétaire. Paris, Economica. Myrdal, G. (1956). An International Economy: Problems and Prospects. New York, Harper. Nabli, M.K. (ed.) (2007). Breaking the Barriers to Higher Economic Growth: Better Governance and Deeper Reforms in the Middle East and North Africa. Washington, DC: World Bank.

Introduction   25 Ndulu, B.J., O’Connell, S., Bates, R.H. et al. (eds) (2007). The Political Economy of Economic Growth in Africa, 1960-2000. vol. 1. New York: Cambridge University Press. Ndulu, B.J., O’Connell, S.A., Azam, J.P. et al. (eds) (2008). The Political Economy of Economic Growth in Africa, 1960-2000, vol. 2 Country Case Studies. New York: Cambridge University Press. Ndulu, B.J. (1986). Investment, output growth and capacity utilization in an African economy:  the case of manufacturing sector in Tanzania. Eastern African Economic Review, 2:14–30. Nguessan, T. (1996). Gouvernance collégiale de banque centrale et politique monétaire: Enjeux, fondements et modalités pour les pays africains de la zone franc. Paris, L’Harmattan. Ocampo, J.-A., and Ros, J. (eds) (2011). The Oxford Handbook of Latin American Economics. New York: Oxford University Press. Philips, W. (1958). The Relationship between Unemployment and the Rate of Change of Money Wages in the United Kingdom 1861–1957. Economica, 25: 283–299. Pinkovskiy, M., and Sala-i-Martin, X. (2010). African Poverty is Falling … Much Faster than You Think! NBER Working Paper 15775. Ravallion, M. (2009). Evaluation in the practice of development. The World Bank Research Observer, 24(1):29–53. Rodrik, D. (2007). One Economics, Many Recipes: Globalization, Institutions, and Economic Growth. Princeton, NJ: Princeton University Press. Rodrik, D. (2006). Goodbye Washington Consensus, hello Washington confusion? A review of the World Bank’s economic growth in the 1990s: learning from a decade of reform. Journal of Economic Literature, XLIV:973–987. Romer, D. (1993). The new Keynesian synthesis. Journal of Economic Perspectives, 7:5–22. Rosenstein-Rodan, P. (1943). Problems of industrialization of Eastern and South-Eastern Europe. Economic Journal, 111:210−211. Sargent, T.J. (1987). Dynamic Macroeconomic Theory. Boston, MA: Harvard University Press. Sen, A.K. (1977). Rational fools: a critique of the behavioral foundations of economic theory. Philosophy & Public Affairs, 6(4):317–344. Solow, R.M. (2009). How to understand the disaster. New York Review of Books, 56:4–8. Solow, R.M. (2008). Comments: the state of macroeconomics. Journal of Economic Perspectives, 22(1):243–246. Stiglitz, J.E. (2010). Freefall:  America, Free Markets, and the Sinking of the World Economy. New York: W. W. Norton & Company. Stiglitz, J.E. (2001). Information and the change in the paradigm in economics. Nobel Prize lecture, Oslo, December 8. Stiglitz, J.E. (1974). Alternative theories of wage determination:  the labor turnover model. Quarterly Journal of Economics, 88(2):194–227. Taylor, L. (1992). Structuralist and competing approaches to development Economics, in A. Dutt and K. Jameson (eds), New Directions in Development Economics. Brookfield: Edward Elgar, pp. 35–56. Taylor, L. (2004). Reconstructing Macroeconomics: Structuralist Proposals and Critiques of the Mainstream. Cambridge, MA: Harvard University Press.

26   Africa and Economics Tchundjang Pouémi, J. (1980). Monnaie, servitude et liberté: la répression monétaire en Afrique. Paris: Jeune Afrique Conseil. Tobin, J. (1993). Price flexibility and output stability: an old Keynesian view. Journal of Economic Perspectives, 7:45–65. Udry, C. (1990). Credit markets in northern Nigeria: credit as insurance in a rural economy. World Bank Economic Review, 4(3):251–269. Young, A. (2012). The African growth miracle. Journal of Political Economy, 120(4):696–739.

Pa rt  I

C ON C E P T S

Chapter 1

Prolegome na to Ec onomics as a n African Sc i e nc e A Philosophical Meditation

Fabien Eboussi Boulaga

1.1 Introduction “Africa” and “economics” are most commonly associated with performance measured according to a selection of indicators of global development that are applicable and are applied to all nations. From this point of view, economic Africa is the sum and the average of growth rates, ratios, and other measures of the national entities that compose it, assigning to them a place, rank, and value on the homogenized scale of indicators of global development. Specifically, economic performances are the principal indicators in this selection: they come from the field of “economic science” and growth theory. Of course they do not cover the entire field of development, which includes other explicit goals, such as “reduction” of poverty and improvement in the quality of life, health, and education. But they are decisive and logically primary for these other performances. This statement of fact is often simply accepted, with ad hoc counterbalances. But one can choose to emphasize the apparently contradictory situations that turn up in the standard presentation of economic science as a “science of contraries,” of development and underdevelopment. Gillis and others acknowledge the existence of a problem in the preface of their treatise Economics of Development: “While this book draws extensively on the tools of classical and neoclassical economic theory, development involves major issues for which these economic theories do not provide answers, or at best provide only partial answers.” They go on to point out important areas that either escape the notice or fall outside the competence of theory (Gillis et al. 1983: xv–xvi). Other authors such as Cabin proposed the challenge of a dilemma. Adopting development models supposedly based on economic theories is not always successful, while ignoring them does not necessarily lead to failure. They conclude: “The countries that have done the best … have drawn from all

30   Concepts available sources without worrying about doctrinal models. On the contrary, all the states that applied pure models paid dearly for it.” A  lesson of skepticism and common sense emerges from this fable: “Moral, draw on all theories and hold exclusively to none” (Cabin 2000: 186). These conflicting approaches highlight a great many paradoxes of the discourse on economic development. Let us begin, as with an “initial situation” to be changed, with its contradictory injunctions or constraints. The starting hypothesis is the following:  first, the reflexive and critical reception of development and its contradictions and paradoxes is the counterpart of economics, of its scientific aim. Second, under the appearance of a pragmatically eclectic approach, this type of reflexive rationality draws from all theories all kinds of data and technologies that serve as a vehicle for development doctrines and practices. The mechanics of the principles of selection, construction, and deployment are discovered a posteriori. They are in anticipation of its immanent normativity, in principle open to a future systematic thematization. Third, beyond its function of designation and identification, the proper noun Africa also functions as a distinct location, an ideality inseparable from its space and as a “total part”. It is simultaneously a quotient of depth and a gradient of opening, obliging one to use it as a starting point to deploy the totality of all other locations as possibilities of existence or human habitat. This chapter highlights the rational mechanics that underline the formulation of propositions in conformity with the logic of linking economic production to a geographic location. Section 2 starts with a discussion of the relationships between the general dynamics of economics, history, and society and explores the ways in which economic ideas and propositions are received in the specific context of Africa, what interactions and transformations take place (i.e. according to principles of selection), and what arguments and dynamic justifications are used to reinterpret and even “recreate” the continent as an economic entity. From this will progressively emerge some characteristic traits describing and specifying the “African reference,” as both contingent and necessary—an a posteriori necessity. Section 3 examines the philosophical foundations of development economics. Section 4 discusses the autonomy of the pragmatics of development. Section 5 offers some concluding thoughts and stresses the need for conceptual reconstructions.

1.2  Economics, History, and Society 1.2.1  The asceticism of historicity and the 

contemporaneousness of social spaces/times How can the African particular be set opposite the presumed universal of economic science? The concept of a reflexive initial situation implies that one does not position oneself outside it; one already finds oneself engaged and implicated in it. One can therefore not have an immediately distanced relationship with it, in which the particular is simply subsumed under the universal by a “knowing rational subject.” Economics does not appear, in this context, as an idea, proceeding from an essence from which one would merely have to deduce basic concepts, theorems, and criteria of its scientificity, and the validity of its

Economics as an African Science    31 propositions and rules for their use or practical application. Its reception assumes that it be set back into the place that received it, into a social totality, that of its emergence as an effective instrumentality, as well as a symbol and value. Its historicity is rooted in (i) the economy of life formed by systems of production by which a society attempts to solve the problem of its subsistence, to provide biological, material, and symbolic existence for individuals; and, in the sphere of informational systems of life that ensure its functioning, by serving as “a base for the life of significations,” valorizations, and ultimate motivations and justifications. The sense of historicity is the fruit of a culture by which we apprehend the particularity of our situation and our distance from this act of the emergence of economics in the economy. Reasoned anecdotal history, consisting of fragments stored away in our memory and used in our argumentations and in producing concepts, is oriented by our position in the system of actual and possible relations in the economic universe. That is why, following Granger (1967: 1018–1053): 1. We cannot totally “ignore” the ideas and macro-analysis of the physiocrats, making of land the foundation of the economy and of the economic subject. 2. We note the birth of economics “intended as scientific truth,” under the influence of the mechanicist view, according to which economic phenomena are governed by “laws as certain and necessary as the laws of nature” and the capitalist-type company is the only economic subject that counts. 3. We see the scientific aim strengthened when economics is then apprehended as a mathematical object made up of formal connections and equations of its concepts, forms, and operations. The economic subject becomes the individual, considered as a consumer and a producer, and economic activity is reduced to the confrontation of individual subjects whose satisfactions are measurable. 4. With Marx and Marxism, the subject of economics is society as a whole, considered at the level of the relationships of production and their underside of social relationships. Economics thus becomes a theorization of the totality of the mechanisms of class domination of workers. The economic system is not the expression of universal economic laws, but a functioning of these laws in a historically determined structure of social relationships of production. 5. With and after Keynes, we return to the global aim of economics in principle if not in fact, with the world and all humanity as its field of reflection, intervention and effective action. The economic subject is no longer above all “the individual engaged in exchange, or the competitive firm.” But it is understood as the “totality of a body” (to be specific, the nation), viewed in its functioning as producer, consumer, and investor. Theory is no longer a neutral description of facts assumed to be intrinsically governed by rational laws, but the display of a general, strategic, operating mechanism, enabling effective interventions in aggregates of facts with the aim of implementing ideal conceptions. What lessons can be drawn from this picture? This overview can certainly be accepted for the idea of variations that it introduces into the very definition of economics, its scientific basis, and the determination of its subject. The overview is just as important for what it reveals of the elective affinities, the expectations and feelings of the interpreter of what

32   Concepts constitutes economic matters. It must be noted that the temporality determining proximity and distance is not calendar time, nor that of evolutionism, which sets up a hierarchy and downgrades the stages already gone through by considering them as dialectically replaced by something better. In different forms, the contemporaneousness of heterogeneous social space–time interactions should be viewed as a situation typical of development and of the reflexivity of self-implication.

1.2.2  Economy, social life, and social sciences: of social meaning and practices Economics thus finds itself confronted with the “polysemy” of “economic facts,” and it moves towards integrating them into other facts where they will interact with these facts in the complexity of human life and action. It is no longer the mirror of nature, but a construction of operating models that measure themselves against reality to influence it or simply to protect the basic economic mechanisms that define the field of economics as such. It is more than a theory; it is an action strategy in the sphere of the production, distribution, exchange, and consumption of “material” (measurable) goods. The significant dimension of economic activity and the status of economics as a social science are obvious when one favors another current definition lacking the prestige of others that are more “classic” and more “learned” (e.g. economics is the science of the allocation of scarce goods or the science that studies the processes of allocation of scarce means among multiple, alternative ends). More down to earth, it enumerates, by spelling them out, the differentiated acts and processes making up economics. It describes economics as “the science that studies the phenomena of the production, distribution, and consumption of goods and services.” With their operational technologies, these fundamental acts open right onto the social practices of economic activity and its “ultimate” motivations of meaningful activity, in connection with kinship, work, play, celebration, and the symbolizations of freedom with its powers, limits, and dilemmas: Each form of social life is in a way an ultimate form; each one, is a presentification of totality, a form of recapitulation. But on the other hand, each one is only a mediator, in the sense that it is useful only in making possible the emergence of another form … Each form includes a dimension of rootedness, which expresses man’s dependence on his natural environment, and a dimension of meaning, which opens to the total horizon of his existence. The function of social life, in its various forms, is precisely to link these two dimensions to each other, each time according to a specific modality. (Ladrière 1972: 46)

It is in the practices of the other forms of social activity, in their universes of signification and their mediation, that one can find interpretive categories which remain economically intelligible as its principal and regulating ideas, corresponding to second-order criteria and to the reflexivity of self-involvement. In the properties and values of this sphere of life as human, the economic form will find its ultimate justifications and motivations. It will be at once instrument and symbol, fact and value, mediation and finality.

Economics as an African Science    33 Conventional economic theory believes in what it considers to be “a bedrock of truths about the human condition.” These seem to be ordinary psychological commonplaces, tinged with anachronism and set up as articles of faith: Among these truths are: 1) the overwhelming majority of men and women are naturally and incorrigibly interested in improving their material conditions; 2) efforts to repress this natural desire lead only to coercive and impoverished polities; 3) when this natural desire is given sufficient latitude so that commercial transactions are not discouraged, economic growth does take place; 4) as a result of such growth, everyone does eventually indeed improve his condition, however unequally in extent and time; 5) such economic growth results in the huge expansion of the property-owning middle classes—a necessary (though not sufficient) condition for a liberal society in which individual rights are respected. (Kristol 1981: 218)

Kristol’s final sentence in that contribution sets out a veritable confession of faith: “This is not all we need to know, but it is what we do know, and it is surely not asking too much of economic theory that in its passion for sophisticated methodology it not leave this knowledge behind” (218). It is surprising that such considerations of simplistic-seeming psychology are tenable, invested with such weight of absolute “knowledge,” and credited with a universal motivational power, long after the advent of comparativism in the social and anthropological sciences. The ground so far covered results in the following propositions: 1. The scientific aim of economics is to quantify economic phenomena so as to structure them into conceptual variables representative of economic life, able to be analyzed and explained in terms of equations, their connections, and the play of their formal mechanisms. 2. It is made more rigorous by the pluralization of its levels and scales where its axioms, concepts, and propositional and argumentative patterns are modalized in different ways, without making of one of these levels the whole that includes both the others and itself, or more concretely, without making of the individual, the firm, the nation, or a larger unit the ideal form of the others, which would then be mere avatars. 3. Whatever its tools and their degree of formal and conceptual sophistication may be, current economic analysis increasingly proceeds by the construction of models as a step necessary for its effectiveness, for testing its ability to take on experience, for resolving the conflicts of interest, and implementing the passage between formal thought and experience, between the local and the global. In order to have applicability, economic science is thus largely “determined in the choice of its variables by the practical conditions of its action, which it transforms in return.” 4. The relationship between theory and practice is no longer seen as a one-way, ordered relationship. Economic techniques are no longer considered as tools of theory, like empirical recipes for its “applications.”

34   Concepts 5. Economics gets its signification or meaning from being closely tied to life, thus translating itself as a social activity into a systemic interoperability with other social activities. Economics can be understood only as integrated into practices, as an “economic field” in interaction with other disciplinary fields. The totality of their actual or possible practices expresses life that produces itself, reproduces itself, and transmits itself as an ideality inseparable from its human phenomenon. 6. It is the anthropogenic content of these practices that determines the signification and pertinence of all sciences, their axiomatic choices, and, in the final analysis, the scientificity and validity of their formalizations and constructions. Considered in this way, the following definition loses its vagueness of pre-scientific triviality to ring epistemologically sound and strong: “Economics is a study of man in the ordinary business of life.” It is by Alfred Marshall, in his work Principles of Economics (1890). The true sense of economics is therefore only a specification of the human sense that belongs to social life, to the totality of the “ordinary business of life.” The African perspective as a construction will include, in its social practices, this requirement of integral, concrete life, each form of which is an end in itself and a mediation for the others, at once value and means.

1.3  Development Economics 1.3.1  Phenomenology of a dualism and reflexivity From a reflexive and critical perspective, economics does not come to mind as a “science.” When presented as development it is perceived as the expression of a kind of natural selection or history that eliminates modes of life and collective behaviors considered as inadequate material and institutional technologies. These failed or were downgraded when confronted with adversity or with the competition of technologies current in societies that achieved their industrial and scientific transformation. Reflexivity immediately reveals development as an asymmetrical, historic social relationship, a hierarchical relationship between groups having different technologies. These technologies are both a material force and a symbolic power. They enable one to understand and legitimize the fact that cognitive, scientific, and technical products can be recognized and accepted as such, without reducing them to the effect of something forcibly imposed or to a relativized history to be left behind or in the process of being left behind. However, in the case of “Africa,” the incitement to reflexivity comes from the fact that the continent is assumed to be heterogeneous and foreign to this science by its “culture” or cultures that are its formal negation. Nevertheless, all that matters are the social encounters and interactions by which everything begins and ends. They act on the living, cognitive, and symbolic social structures of the recipient. They change his self-image. He sees and defines himself as he was never able to see or define himself before being placed, by this relationship, into a common dualistic space, into a world that includes on one hand development and, on the other, non-development or underdevelopment. What happens then? Reflexivity is faced with two possible paths: the first path is that of identity and difference. It is that of mimetic reflexivity, of assimilation to the other and/or dissimilation from him,

Economics as an African Science    35 of vindication and/or denigration of the self or of all that one takes for one’s other. Mimetic reflexivity in the case of development is the royal road to contradictions and failures. Why? The idea and credo of development are not challenged, any more than are the beliefs and convictions concerning identity and difference that they introduce into the discourse legitimizing its economics. The second path approaches the subject from the viewpoint of economic science that explains or justifies the duality or scission of its universe of “discourse” into development and underdevelopment. It establishes, between these two “halves,” a relationship of knowledge to lack of knowledge, of “science” to ignorance and, consequently, of the power and ability of knowledge to the “impotence” and “inability” of lack of knowledge or ignorance. This “intrusive” dichotomy problematizes and dialecticizes economic science that is called upon to say in what and of what it is science and to draw internal limits for its scientificity and validity, as well as their necessary and/or sufficient conditions. It thus introduces, as its reflexive effect or indirect consequence, an alteration in the very understanding of economics. A distinction can then be made between the (primary) criteria of scientificity and validity of the propositions of economics and some other “second-order criteria.” The historicity or locality function of these criteria make them meta-criteria in the creation of the meaning that objective and constructed knowledge takes on in social life. This function is its structure of reception and legitimization, in the form of a theory or doctrine of practices integrating economics into the totality of social life. The latter could be understood either as the direct “application” anywhere of economic theory with its prescriptions and standard formalisms and calculations; or as its actualization or adaptation to the context and changes, with ad hoc adjustments; or, finally, as a testing of theory by practice, capable of disqualifying or rectifying some of its postulates, concepts, and arguments. This brings us to the critique of the development/underdevelopment duality, preparing the way for other alternatives and for the choices of the African perspective, for a counter-attack of creative reflexivity. Its attack angle focuses more narrowly on the way in which development manifests itself. Presented as development, economics is seen as the expression of a kind of natural selection, eliminating behaviors that have failed in the course of a history of progress. Reflexivity, however, reveals development as an asymmetrical, historic social relationship, a relationship of order (of progress and evolutionary hierarchy) between groups having different technologies. Both a material force and a symbolic power, these technologies make it possible to understand why and how cognitive and scientific products can be recognized and accepted in their translation or incarnation in what the ignorant or “lay person,” the non-specialist, assumes to be heterogeneous to the objective structure of science. Defining economics as development thematizes the economic life of industrial societies, tautologically defined as societies where work is organized by the sciences. Theory, at this point, appears as the symbol or secret of material power. Its scientificity is an object of credit on the grounds of the technologies that have demonstrated their effectiveness in the organization of production, distribution, exchange, and consumption as judged by the armament and transportation industries, cornerstones of this belief in science as their supposed foundation stone. The mimetic reflexivity of identity and difference ratifies this duality or dichotomy and the assumption that the pole of development is destined to triumph, due to its irresistible expansion and absorption of what remains outside it. The reproduction taught by the doctrine of development economics helps shorten the reprieve of what has already been condemned

36   Concepts to death. It settles epistemological questions on the nature of economic science, its status, unity, or plurality. Too easily conceded, this dichotomy carries with it a series of oppositions between rational and irrational, natural and not natural. They then underpin the opposition or distinction between economics as such and development economics. The latter is set up as a “separate science,” inferior and linked to the economic climate. Indeed, the idea of progress entails the “scientific” underqualification of development economics, reduced to pedagogical functions of catching up. The definition of underdevelopment as the other or the negative of economics makes of it the totality of what the subject—African to be specific—does not have or is not, of what it lacks in order to have development or be developed. It is transformed into a self-fulfilling prophecy. For when “economic actors” describe themselves as “underdeveloped,” “poor,” or “poor countries,” they actually become such, accepting all the indicators that categorize them as being in the depths of poverty, illness, corruption, “backwardness or total failure,” nearly inhuman. That being the case, the mimetic reflexivity of catching up, whether submissive or react­ ive and rebellious, is a defeat that intelligence inflicts upon itself, endorsing, in the form of equivalence and repetitive expansion of the identical or even inversion, all the incoherence, all the contradictions of a partial, regional knowledge that admits its improbable generalization by simple diffusion. Science, even economic science, cannot be prescribed: it is every­ where or it is nowhere. “Universality” is that from which particularities are thought and grasped as particularities and enter into a chorus that gives it its “faces,” its proven phenomenality, the modulations of its generalization in the “unidiversality” of the earth-world. Another path is that of reflexivity resulting from having accorded, with the unity of science, primacy to theory over practice, and the one-way relationship going from the first to the second, making of the latter an “application.” This passage involves many preliminary conditions. They boil down to the current economic conditions of industrial societies, which, by definition, are not those of non-industrial societies. The implication of this choice of the unity of economics as guide and norm of development science that is most fraught with consequences seems to be the favor given to one model among several others, products of different times and contexts, for incommensurable, heterogeneous functions, but coexisting on the battlefield of doctrines. It is set up as invariable, resisting all the changes, contradictions, and refutations that circumstances can inflict upon it. This suggests that it has been transformed into an ideological doctrine that still claims to be a science. Despite all their differences regarding objects of research, methods, and centers of interest, economists profess a common point of view on certain “givens,” fundamental beliefs or practices, particularly when it comes to “teaching” development economics. In the introduction to their textbook, Gillis et al. write: “First of all, this text makes extensive use of the theoretical tools of classical and neoclassical economics in the belief that these tools contribute substantially to our understanding of development problems and their solution.” These authors immediately add that their “text does not rely solely or even primarily on theory, but on comparative studies, decades of varied case studies from many, different countries around the world, and the contributions of historians and specialists in development economics” (Gillis et al. 1983, xv). But the remarks concluding their introduction suggest that the discipline of economics cannot be generalized by simple extension and application, even if it is in principle universal. Would it then be like the Platonic Idea to which matter opposes opacity, traversed only by violent domination? A background of customary beliefs fosters an insurmountable misunderstanding between economics as a logico-deductive science and the political economy of

Economics as an African Science    37 life as an activity of socio-cosmobiological exchanges. It is the opposition between a logic of truth and a logic of credibility, between the true and the transmissible.

1.3.2  Pragmatism and reflexive logic: truth, credibility, and transmissibility 1.3.2.1  Logic of credibility These reflections take us to an intriguing question: How does economics first “come to mind” in the African perspective, in the lens of development? The question can be interpreted in several different ways. One of them leads us towards the following seemingly theoretical and problematical explication. Given economic science as fact, does or even can the subject presupposed by development, as its condition of possibility or its cognitive vehicle correspond to the subject outlined in Kristol’s “five truths” or characteristics? Perhaps a more appropriate and useful way to formulate this question is in the following general manner: Would an empirical, pragmatic approach, based on activities and practices, selecting, keeping, and organizing those that are successful and lasting, striving to generalize and formalize them, lead back to economics as it has become established? Would it recognize, after using the theoretical tools of classical and neoclassical economics, the same “rational” and “universal” economic subject (the famous homo economicus)? Would the economic subject be identical to the one embodied in the “bedrock of truths about the human condition” that constitute the “what we do know” of dominant economists, designating what can be called the sphere of their ultimate motivations and justifications from which are drawn the secondary or meta-criteria that make it possible to talk about their discipline in a sensible way, and preserve the primary coherence and scientific criteria? These are epistemological paradoxes that must be examined, concerning the connection of economics to development economics (from a deductive or quasi-deductive method to a method of learning or application) and the equivalence of the logic of truth (transfer of the truth of premises to the conclusion) to a logic of credibility (an effective ascent from the experience of individuals to general laws or rules). It does not stand to reason that behind or beneath appearances there is a unique unitary knowledge, a “science” organizing all these multiple and effective activities, institutions, organizations, operations, and agents. Economics makes its entrance by means of the dynamics working within the structures and functioning of particular sectors, by means of the sharp side of financial and economic engineering, of operational, industrial, and managerial technologies, which are fruits of experience, codifications, and “good practices.” Economics, an established science, sets itself up in a normative position demanding a logico-deductive approach rather than that presupposed by the position of learning appropriate to underdevelopment or its cognitive subject in the form of tabula rasa or negativity. Experiences of economic transformation and industrialization have taken place outside of or contrary to the ex ante prescriptions of some development economics treatises. Does a single counterfactual event of this kind mean the refutation of normativity constituted a priori, putting a stop to all the scholasticism of necessary and/or sufficient conditions to be fulfilled, of stages to be passed through? “Only development makes sense.” In this peremptory maxim, Deng Xiaoping appeals to the pragmatic rationality that asserts itself only ex post, in its results and success. It is not known in advance what will happen in this passage from

38   Concepts underdevelopment to development: it is the sort of thing that emerges randomly, like the human phenomenon in general. This anthropogenic fact can be used to get a clearer idea of the various fittings and “joinings” of economics, its effectiveness by regionalization (thanks to diffusion, to the anthropological dialectics of tendency and fact, and to borrowing).

1.3.2.2  Transmissibility versus truth Armed with such cautions and methodological and epistemological “expectations,” it is right to go forward, not only to rethink development, but, in this roundabout way and mediated by its teaching, to rediscover economics on another scale, the scale of historical anthropology. If one admits that it is the scale that creates the phenomenon to be constructed in analyzable and explicable facts, then we could renew its discipline without repudiating its assets, but in transmuting them, by virtue of the very ordinary miracle or paradox of all successful learning, which is to discover that truth (scientific or living) resides in transmissibility.

1.4  Autonomy of Development, Practice of a Finality without end 1.4.1  Historical discourses Development conveys trends and doctrines. Effective, differentiated use is made of it according to situations, encounters, and the level and volume of activities. Its history is not linear, where theories succeed and supplant one another, the successor having a more powerful and more comprehensive rationality than the predecessor. They coexist, reinventing themselves in non-homogeneous space, at qualitatively differentiated positions (society, habitat and environment, culture) but in relationships and options open onto doctrines once supposedly situated at steps or stages of unilaterally homogeneous time and space. This time and this space are in fact equivocal, referring to the time and space of the development of industrial capitalism and, at the same time, to a timeless conceptual self-begetting of economics. But history suggests that entry into development and therefore into economics can take any of a number of access roads offered by doctrines and trends. Their value is then one of position and can present itself as technology, as an end that is at the same time a means, symbol, and instrument, or merely as a condition with and/or against which action must reckon. The category of developing countries includes countries with nothing in common, even if it is convenient (without sure cognitive and functional gain) to classify them in groups according to their past capacities empowering them to more or less rapid and substantial modern economic growth. This done, one must nevertheless conclude: “Fortunately there is no standard list of barriers that must be overcome or prerequisites that must be in place before development is possible. Instead, as the economic historian Alexander Gerschenkron pointed out, for most presumed prerequisites there are usually substitutes” (Gillis et  al. 1983: 18). In the midst of such uncertainties and divergences and possibly of such an outcome, the concept of position seems far more illuminating and welcome.

Economics as an African Science    39 These facts of development suggest two directions for exploring temporal representations, which are hypothetically more adequate or specific to economics. The first direction is that of anthropological transmission of knowledge and techniques. Emphasis is specifically placed on the modalities according to which groups behave with respect to techniques that are initially “foreign” to them and how they appropriate them, and even more on the ways in which this “knowledge” is transmitted. One way in which development is useful is to call the attention of economic agents in Africa to one of the systems of rationality. It does not have truth as its absolute or transcendental horizon, but the transmissibility of its objects, of its productions, by undefined replications or metamorphoses. The time/space of transmissibility is not that of the generalization (transfer of the “truth” of the premises to the conclusions, expansion of the identical, isomorphic, and totalizing) that supports the imaginary of the universality of science and its mechanisms. It is that of the internal environment of the economy of a group as a living human organism: “By general laws of proportions, because there is no life possible without a coherent organism, the internal environment tends not to dilate, but to concentrate; civilizing expansion is not a real function of the group, but an aspect of its concentration. It is the expression of the necessity that the group feels to particularize itself more profoundly by increasing its means of action” (Leroi-Gourhan 1945: 452). In the normal condition of the group, enough internal tension must be preserved to assimilate external contributions, when and as much as they are needed. When these contributions from the outside pass through the internal environment, they are necessarily transformed into utility, a part of a system of utilities or functions of utilities or techniques. This should be understood strictly as regarding beliefs, knowledge, and representations of all kinds, including those of its past. Every living group is oriented toward the future, and its past has meaning only as it appears as a utility that can be integrated into the living process of transmissibility. The second direction is that which compares the process of development with that of co-evolution. The coexistence of socioeconomic times and of technico-ecological environments can be taken on without prophesying their homogenization or contributing to or accelerating it. How can this diversity be imagined without hierarchizing it on the scale of an ascending evolution of growth, progress, and improvement? How can one not see as unsatisfactory the dichotomy between economics inscribed as growth theory in the structure of a capitalist, mercantile, industrial, and free market society and all the rest, past, present, and future? How can one believe that economics can escape the hypothesis that the reflexivity of its interactions with this rest, and the retroactive effects of the transformations related to it, redefine the rules of the game and the roles of the actors and factors of production? A reflexive and critical economics should consider that the possibility of the collapse and disappearance of the zone of “non-economics,” of its so-called “externalities,” is one of the fundamental problems for the economics of globalization. The globalization of financial flows and information and the expansion of object systems as well as the power of the firms and companies “managing” them make it important to attempt to understand globalization as the resumption of the ways of life that have fashioned and continue to fashion the earth as the habitat of the human race. The hypothesis of co-evolution should be examined for a co-development that presents itself as a task of the present turned towards the future. Development heralds this when it integrates ecological imperatives.

40   Concepts

1.4.2  Strategies and practices Economics approaches Africa as necessary and sufficient in development in the form of more or less codified “good practices,” in the translation of its doctrines and theories in terms of technologies of the economic activities of a society, and in more or less strong interactions with other societies. It can be postulated that what succeeds is, by definition, in conformity with current economic science (whose principles, laws, and rules apply) or with that to come, which will not be discontinuous in nature from what is presently called economics. If such is the case, it would not be unwarranted to see things from the point of view of business and management, which consider the economy as an “cohort of companies” and a body of plans, programs, and portfolios that are initiated, organized, and managed according to generally accepted and recognized norms and processes constituting the most complete acquaintance with “the firm’s environmental factors” (including its culture, structure and local organizational processes, political climate, etc.) that might compromise—or not—the desired results. From the viewpoint of business and management economics appears or reveals itself as technology. This technology can be considered “as an immaterial good represented by operational knowledge and expertise, that is, it is specifically devoted to solving concrete, generic, or repetitive problems and, in the end, to satisfying the real needs of man and society” (Babissakana 2006: 332). This can be expressed in a more detailed fashion, adapted to the specificity of the economic field: “Technology represents the combination of codified or explicit knowledge and expertise and/or of tacit or implicit expertise capable of being mobilized in an activity of conception and implementation aimed specifically at solving problems of production, distribution, exchange and consumption in the fields of goods and services” (Babissakana 2006: 332). It will be noted that in this field “technological knowledge in the form of inductions, procedures, rules and blueprints, capable of being transmitted or reproduced in printed or digital form”—but of great importance—is “the tacit or implicit technological knowledge, formalized or informal, held by man and transmitted through practices, on site and on the job, from man to man and through apprenticeship” (332). In this context, the market itself is presented unhesitatingly as a priority “technology” that “the state must absorb, internalize, appropriate, and use in an effective manner,” if it wants to provide the nation with “an adequate industrial and economic organization,” in keeping with the level of knowledge and techniques currently available. To paraphrase a Hegelian aphorism, one does not free oneself from the market economy, but through the market economy. To accomplish this, a state must be a strategist, stimulating, coordinating, piloting, and ensuring leeway to reach fixed goals. It is the active guardian of effective convergence technologies, those of sense as direction and shared social value, which form strategy:  Strategy is a technology enabling one to define objectives within a credible amount of time … and calibrate the actions and means to be implemented in order to achieve this. Strategy must be global, (a totality of the entity and the instruments), dynamic (a period of time) and discriminating (selectivity of objectives, actions and means). Strategy is embodied in men and institutions that found its pertinence, foreseeable nature, operability, credibility and, in the end, its success. Strategy is not simply a document whatever its graphic or editorial quality may be. (Babissakana 2006: 432)

Economics as an African Science    41 There is, for the African observer, a historically confirmed model of this autonomy of development conceived as a totality of technologies appropriate to solving problems concerning mastery of the environment (more broadly, of “nature”), the production of goods and equipment, objects of consumption and services, distribution, and exchange. It is the path opened by Japan and followed by all the Tigers and Dragons of Southeast Asia, among others. It raises questions, some of which seem to have no definitive answers: • Does a society’s access to “science” assume a prior degree of development of productive forces or of wealth as a necessary condition? • Are the ways and means to reach it perhaps not necessarily linked to economics? Is economics a discipline of prestige, the humanist “point of honor,” the “solemn complement” of what is nothing but a war continued with other means? • The economic success of a society, its “development,” might be only a matter of a combination of favorable forces, of strokes of luck, sometimes of a reversal of fortune and of strategy, this military art of tying together the various favorable “friendly” factors and outflanking or shattering the “enemy” obstacles to reaching the desired goal. • Do successes always “prove” that the economic system supported and projected by economics is reproducible? Are they rather perceived and expressed as effects by a rationality of judgment relying on observations of particular practices whose observable consequences give credibility to the general premises that they induce or marshal for their justification? In its allegiance to sciences having a deductive methodology or a logico-formal construction, economics expresses a logical rationality that transmits values of truth, from its principles and premises to the theorems and “laws” deduced from them. Autonomous technological development relies on a logic of credibility. It moves from results or consequences to the horizon of their premises or principles, transmitting only values of credibility through “trustworthy objects.” It leads one to posit that if the other kind of rationality, that of the logic of truth, were universal, it would be generalizable in its diffusion and effects. How many planets as large as the earth would be necessary so that everyone could consume as much as today’s largest consumers, according to a rate of “indefinite growth”? This may not constitute a refutation by the absurd (reductio ad absurdum) but certainly by vertigo (ad vertiginem).

1.4.3  Prospective studies and plans The typical strategy of economic development is to use the rationality of transmissibility and express it through forecasts and a never-ending temporality. Activities incorporate themselves into one another and synchronize themselves in the present. Development is then intended to project their increased and transformed interdependence into the future. It is the attraction of the future that makes the living being, reorganizes his past, and directs the present, so as to support or maintain uninterruptedly the elimination of shortages and hunger and the prevention of calamities. Note should be taken of the growing tendency to carry out global forecasting, over a significant length of time for national entities, but also increasingly for each of the principal regions composing Africa and even for the entire continent. It is possible to carry out inventories of what has been done, fill in gaps, and coordinate their

42   Concepts results. The validity of the analyses, diagnoses, and proposals will lead to drawing up criteria of the pertinence and convergence of the “African perspective”—a perspective that aims to go beyond the horizon of even long-term predictions. The latter tend to be only an extrapolation of the present by a maximal aggrandizement of dominant ways of living, thinking, producing, and consuming, merely their “revised edition.” The paradox of development is to commit us to transmitting what we do not possess, which is inscribed neither in our “genes” nor in our “cultural heredity,” and without “preexistence” in the assets of the other, which would make of him a great “predecessor,” unsurpassable but imitable. What we have to do will always be what the other is lacking and our positions will be unique and not redundant. They will be complementary to make or unmake a world. It is in forecasting that we can see the beginnings of one of the models of the rationality of credibility and transmissibility, of which no one, by definition, can have either the monopoly or the exclusivity. Anthropological temporality can be understood only as the past that survives in a present insofar as this present projects itself into the future.

1.5  Anthropology of the Inhabited Earth and Conceptual Reconstructions The detour through development economics was essential as a return to the ground of all reception of theories, their human inscription, which is their source and end. That is why it functions as a critical location of their more or less selective and complete acceptability. It is from this viewpoint, that of their transmissibility, that Africans are thus “summoned” to rethink and reconstruct economics. A reappropriation, which is carefully considered and reflected upon, amounts to actual invention. It would be recommended for its effectiveness in solving problems and dilemmas that are hindrances. It ought to seem sensible to speak of economics as an “African science” using this approach. In it, acceptability, appropriation, and transmissibility are defining operations or features of scientific knowledge, approached in an “anthropological manner,” that is, from a location where the proper noun, the name itself, is a nub of objective mutual and reciprocal relationships. The topics “anthropo-historical topic” and “objectivity of position” suggest a suitable outline.

1.5.1 Anthropo-historical topic Economics, through the mediation of development economics, belongs to the domain of action, decisions, and choices. It produces and carries through the models that inscribe its principles of objective value, growth, and utility in a given location. Its agents carry out these conversions according to an operational and technological mode of expressing situations, individual, or collective “preferences.” If one confers on “Africa” the status of “collective decision-maker,” useful exchange there interacts with reciprocity, mutuality, and gift, and each of these elements is unable to subsist without the others. Abstracting exchange from this complex turns it into an abstraction by conferring on it the absolute status of an ideality.

Economics as an African Science    43 Despite strict formalisms and strong logico-deductive constructions, the principle of greed imagined as the fundamental law of social nature, like universal gravitation, still remains what it has always been in humanity, a source of disorder and shame. In an oft-cited article, entitled “Economic Possibilities for our Grandchildren,” Keynes, faced with this aberration considered rational, expressed this wish: “I see us free, therefore, to return to some of the most sure and certain principles of religion and traditional virtue—that avarice is a vice, that the exaction of usury is a misdemeanor, and the love of money is detestable … We shall once more value ends above means and prefer the good to the useful” (Keynes 1972: 331). The return to these principles is via that rational altruism inscribed in the autochthony of our species, dictating the reciprocity of our perspectives and the supportive mutuality of our respective positions in view of the system of reference Earth, bearer of the absolute conditions of the emergence and reproduction of human life.

1.5.2  Objectivity of position The axes of the reconceptualization of economics (so that it can be called “scientific” because it is “African”) have already been suggested by the characterization of its specific scientificity. It denotes transmissibility, rationality of credibility, and its “field’s” linking of the instrumental and the symbolic. Its temporality, where the past is apprehended in the present as a future perfect is translated in terms of a spatial location within a topology or science of positional relationships, in the appropriate register of rationality and “objectivity.” For it is the position that is the fundamental category of a conceptual reconstruction beginning from a topos, as the relational place or “perspective” on the earth taken as the referential system. This is the ground for a reconstruction plan that we must define and describe succinctly. At this stage, the patronage of a few eminent precursors can be freely called upon. Aristotle shows the way by his distinction and intermeshing of the categories of place (topos) and position (keisthai), which respond to questions of localization (where?) and posture or attitude. Leibniz opens the exploration of place for reasons other than the needs of quantification, measurement (dimension), and positioning (the magnitudes of position, coordinates, vector calculus), thus tilting it into other categories. Euler and his followers transform his “presentiment” into an inquiry on the spatial relations between objects, taken as such, the issue being “the possibility of a purely conceptual thought of place itself,” thus taking it out of the “non-status” of an object of thought. This orientation leads to topology. The social sciences use weak versions of this, quasi-formal metaphoric adaptations or transpositions. Thus one can study transformations leading from one physico-social or textual configuration to another and the constants that emerge. How can a place be deformed and “transpose” itself while retaining something of its structure as a spatial configuration? Position is thus from the outset approached from the viewpoint of transposition. It leads to reasoning about things by placing them in relationship with each other, and grasping each of them by its position relative to others. By inscribing the African position or perspective in a topological economics, one adopts the epistemological posture of a reflexivity that is critical of development in order to construct its immanent conceptual normativity. Africa here has reality and is an object of rational knowledge and reasonable action only as determined as a reflexive or self-reflexive relationship, that is, as making itself and acting. It is a place in the dialectics of places that make up the world.

44   Concepts Does the category of position keep its value of rationality and objectivity when these claim to be “the view from nowhere”? Thomas Nagel characterizes objectivity in these terms: “A view or form of thought is more objective than another if it relies less on the specifics of the individual’s makeup and position in the world, or on the character of the particular type of creature he is” (Nagel 1986: 5). The validity of this criterion, in the perspective of the present analysis, is a necessary condition to shatter the illusions of points of view that do not know they are such, being unaware of their partiality and unduly setting it up as universality. But this classical conception of objectivity, focusing on independence with regard to position, is inappropriate for realities that are the exclusive resultant of positions in interaction, notably in the reflexivity of self-implication. In opposition to Nagel, Amartya Sen emphasizes the pertinence of the objectivity of position, as regards the law, when the effects of positional variability are tested in order to eliminate bias and make appropriate corrections or improvements (Sen 2010: 202). The error would be to identify position with “subjectivity,” with the accidental property of an ontology of substance and the deceptive appearance of a dualistic and mentalist metaphysics: An observational utterance is not necessarily an utterance on the specific functioning of some mind. It identifies a phenomenon that also has physical qualities independent of anyone’s mind … The little mass of the Moon can darken the big mass of the Sun from the specific point of view of the earthling—and it would be difficult to claim that a solar eclipse has “its source in the mind.” If our work consists in predicting eclipses, the important thing with regard to the relative sizes of the Sun and the Moon is the coincidence of their positional projections seen from the Earth and … not directly their respective real sizes. (Sen 2010: 203)

Such an obvious fact ought to invalidate the discourses that make of monadic and antagonistic cultures or civilizations, of “identities” and mentalities essentially either closed or open, the primordial agents and determiners of the future of Africa and the world. In this eminently sublunar science of interactions and exchanges that is political economy, it is the objectivity of position that prevails, with its type of specific rationality of credibility: strategic, reflexive, and self-implicative. It frames the other forms that take on the value of tactics, mediations, isotopes, and counterparts.

References Babissakana (2006). “Les Débats économiques du Cameroun et d’Afrique,” Les Cahiers des Notes d’Analyse Technique N° 2, Prescriptor (La firme de l’Intelligence pour l’Investissement en Afrique), Yaounde, Cameroun. Cabin, P. (ed.) (2000). L’économie repensée. Paris: Editions Sciences Humaines. Gillis, M., Perkins, D.H., Roemer, M. et al. (1983). Economics of Development. New York: W.W. Norton. Granger, G.-G. (1967). Epistémologie économique, in J. Piaget (ed.), Logique et connaissance scientifique, Encyclopédie de la Pléiade. Paris: Gallimard, pp. 1018–1055. Keynes, J.M. (1972) Economic Possibilities for our Grandchildren. The Collected Writings of John Maynard Keynes, vol. 9. London: Macmillan, pp. 321–332.

Economics as an African Science    45 Kristol, I. (1981). Rationalism in economics, in D. Bell and I. Kristol (eds), Crisis in Economic Theory. New York: Basic Books, pp. 201–218. Ladrière J. (1972). “Sur la signification de l’économie,” in Technique économique et finalité humaine—Synthèse des Travaux d’un Séminaire de Philosophie sociale (Centre d’études et de recherche universitaire de Namur). Gembloux, Belgium: Editions Duculot, pp. 33–66. Leroi-Gourhan, A. (1945). Milieu et Techniques. Paris: Editions Albin Michel. Nagel, T. (1986). The View from Nowhere. New York: Oxford University Press. Sen, A. (2010). L’idée de justice. Paris: Flammarion.

Chapter 2

Househol d s a nd Inc om e in A fri c a Kathleen Beegle, Calogero Carletto, Benjamin Davis, and Alberto Zezza

2.1 Introduction The concept of income is more or less closely related to just about anything that economists think, talk, and write about. Despite its ubiquity, or perhaps because of it, it is a difficult concept to define, and even more difficult to measure. One of the major issues to emerge during the discussions of the Canberra Group was the existence of two traditions of household income measurement.1 The macro-approach has its roots in national accounts and in particular the standards laid out in the System of National Accounts (SNA) (Commission of the European Communities et al. 1993). On the other hand, the micro-approach has roots in microeconomics and particularly the study of poverty and its relationship among different socioeconomic groups within society. At the macro-level, households are just one of the institutional sectors involved in the generation of income. In microeconomics, the unit of analysis is most often the household, or the firm, and sometimes an individual. While microand macroeconomists have developed different terminologies and conventions, and use different data sources and levels of analysis, both are dealing with essentially the same concept of household income (Canberra Group 2001: 5–6). In practice, a major part of the theoretical and empirical work on household income has to do with the study of welfare, income distribution, and inequality. Traditionally, the majority of countries in Latin America base their poverty estimates on household income. Countries in Eastern and Central Europe and the Balkans are also slowly moving towards adopting household income as a measure of welfare—partly to meet statistical requirements for full 1 

The Canberra Group on Household Income Statistics was an international expert group established in 1996 at the initiative of the Australian Bureau of Statistics. This effort produced a handbook in 2001 to address common conceptual, definitional, and practical issues facing national statistical offices in producing household income statistics. A second edition of the handbook, with updates to the original, was released in 2011.

Households and Income in Africa    47 accession to the European Union. In other regions of the world like Africa, income is less frequently used as the basis to measure poverty and welfare. Nonetheless, the usefulness and policy relevance of good income statistics are hardly ever questioned. Measurement of household income in developing countries is notoriously fraught with problems. While it is difficult to establish the extent of the problem, there is a widely shared consensus that income is under-reported (McKay 2000). While some fraction of under-reporting might be purposive efforts by households to hide income, in low-income settings, which encompasses the majority of the African continent, attention is typically focused on other causes. The high level of informality and the large share of households in smallholder agriculture makes it difficult to solidly grasp concepts and measures of income. This is one of the main reasons why most African countries still rely on consumption-based measures of welfare (McKay 2000; Deaton and Zaidi 2002). It is notable that the three major international efforts towards systematizing work on household income data—the Luxembourg Income Study, the Canberra Group, and the Wye Group—all share a focus on high- and middle-income countries, and have little to say that draws on the experience of the African continent, with the possible exception of South Africa. Irrespective of its limitations, however, the collection of quality income data in Africa remains important. Besides welfare, income serves as a key dimension to study the sectoral composition of the economy in microeconomic analyses: how households derive their livelihoods, and the productivity and returns to assets and economic activities. A large body of literature has emerged since the 1980s on rural non-farm income diversification of the rural economy (Lanjouw and Feder 2001; Lanjouw and Lanjouw 2001; Haggblade et al. 2007; Davis et al. 2010) and on micro-enterprises and self-employment (Vijverberg and Mead 2000; Fox and Sohnesen 2012, specific to Africa). In this chapter, we explore the concept and measurement of income from a micro-perspective as it applies to African households. We first look at the conceptual and def­ initional issues around some key components of household income (section 2). The thorny issues around the theory and practice of income measurement are discussed in section 3, followed by a short discussion on empirical evidence on the structure of African household income in section 4. The key messages emerging from this review are summarized in section 5.

2.2  Features of the Concept of Income for African Households There is a long and well-documented approach to measuring household income in household surveys. The definition commonly accepted as the international standard is laid out in 2003 in the resolution of the 17th International Conference of Labour Statisticians (ICLS) concerning household income and expenditure statistics which defines income as follows: Household income consists of all receipts whether monetary or in kind (goods and services) that are received by the household or by individual members of the household at annual or more frequent intervals, but excludes windfall gains and other such irregular and typically onetime receipts. Household income receipts are available for current consumption and do

48   Concepts not reduce the net worth of the household through a reduction of its cash, the disposal of its other financial or non-financial assets or an increase in its liabilities. (ILO 2003)

As noted in the Canberra Group Handbook (2011), the ICLS definition of household income is consistent, to the extent possible, with the definition used in national accounts (SNA 2008). Based on the ICLS definition, income can be considered as formed by all receipts that (i) recur regularly; (ii) contribute to current economic well-being; and (iii) do not arise from a reduction in net worth. These three criteria are embodied in each of the components of income; as such, irregular payments such as lottery earnings or inheritances; investments and savings and the value of durables are not normally included in the estimation of income. In an effort to harmonize household income data for developing countries, Covarrubias et  al. (2009) define seven components of household income:  agricultural wages, non-agricultural wages, non-agricultural self-employment, crop production, livestock production, transfers, and other income. These income components are to be net of costs (such as business expenditure). They also include in-kind receipts (such as wage payments in the form of food). Following the SNA definitions, the value of own-produced goods which are not marketed are included (such as subsistence farming output which is consumed by the household and the value of housing repair). In many African settings, income estimates from non-marketed goods will constitute a large share of household income. On the other hand, under the SNA, own-produced services are not included. Cooking, cleaning, childcare, and other services done by household members within the household are excluded. There is a vibrant debate as to whether these activities should be included (Young 2000). Wage income. Labor economists use measures of individual wage income to investigate basics such as labor participation and economic returns in the labor market. Wage labor is usually the easiest income component to collect, not only in Africa but also elsewhere. The deduction of input costs is not necessary, and for many, the receipts tend to be more regular than other income sources. The measure should cover all wage-income, regardless of the traits of the work—that is, whether the job is a primary or secondary job, full-time or part-time, seasonal or permanent, labeled as casual work, or considered formal employment (whether the job has a contract or benefits). Further analyzing wage income will, however, depend on these traits of the work. And so most surveys collect these details of the job in addition to the income from wage work. Agricultural income.  Agriculture accounts for a large portion of overall economic activity in African countries. Smallholder (household) farming accounts for the vast majority of agricultural production on the continent. Moreover, it is a major source of livelihood for the poor, particularly in rural areas. It is unfortunately perhaps the hardest income component to measure, because of its complex production process, high-level informality, seasonality, and variability linked to weather patterns and agronomic shocks. It encompasses not only crop production but also livestock and fisheries.2 The estimation of agricultural income should account for total production (which could be sold, home-consumed, and given out as gifts) net of expenditures incurred in realizing 2  Livestock has the added complication that animals are in fact partly capital assets, partly products. It is very difficult to separate the “stock” from the “flow” components in practice when computing livestock income.

Households and Income in Africa    49 the production (agricultural inputs such as seeds, pesticides, fertilizers, and farm labor payments). A major challenge is the choice of the prices to value this production both because prices fluctuate during the year and because some local foods are not marketed sufficiently to have price data (i.e. missing prices). Self-employment. Just as in agriculture, a majority of the non-agricultural production units in Africa, as in the rest of the developing world, are household enterprises (Vijverberg and Mead 2000; Gindling and Newhouse 2014). Household enterprises share with agriculture many of the characteristics that make agricultural income so hard to measure: informality, seasonality, and no record keeping, among others.3 Household purchases should be attributable to household consumption or the negative side of the household enterprise balance sheet. This is difficult in practice. For instance, the enterprise can often use parts of the dwelling where the household resides. Fuel and other expenditures on transport (car, motorbike) may be used for both the private or commercial activities of the household. Transfers.  Transfers are associated with non-labor income from public and private sources. While typically not a large share of income in Africa, these are emerging sources as migration (Lucas 2006) and social protection programs (Ellis et al. 2009) expand in the region. Private transfers include remittance income from individuals and the value of benefits obtained from private organizations—gifts and contributions not associated with the performance of a job or the provision of a service. Public transfers include state-funded pensions and social benefits, which include unconditional or conditional cash transfers. Other sources.  All other household income sources that do not fall into the aforementioned categories are accounted for in this last grouping, mainly rental income (including rented land and assets). In middle- and high-income countries, capital and rental income tends to be substantial, and is then classified as a separate item component (see for instance the classification adopted in the Luxembourg Income Study Database). In household surveys from Africa, these sources of income are small and rare.

2.3  Measuring Household Income in Africa in Practice In Africa, because of the particular composition of household income, the task of realizing the theory of income measurement is particularly challenging. Here we discuss briefly some of the issues that arise in the practice of applying theoretical concepts to data collection of household income in Africa. The first, seemingly most basic stage of measuring household income is defining the membership of households over some specific time period. The basic approach outlined in Glewwe (2000) for defining household membership in household surveys are exacerbated in many African contexts (Guyer 1981; Randall et al. 2013). Beaman and Dillon (2012) studied this issue in Mali, where extended families often reside in common areas and income activities are, to some extent, jointly undertaken. They found that household 3  Joshi et al. (2011) focus on this from the perspective of measuring informality in the economy at large.

50   Concepts rosters, and therefore household measures, were especially sensitive to inclusion of income-generating and production keywords in the survey questionnaire. Related to the income focus here, although levels of agricultural production and inputs are not affected, per capita measures differ significantly. Another dimension of this problem is the fluidity of household membership over time, which has not yet been carefully quantified in African settings.4 In the best practice approach, individual household members would self-report their income-generating activities and earnings. In the case of joint activities—the source of a large share of household income in African households–the most knowledgeable member would report. In practice, however, this does not always happen during survey implementation. This issue can take on a strong gender dimension. In many parts of Africa, it has long been observed that there are gender concerns with reporting income, possibly due to the nature of women’s income activities, and to husbands and wives undertaking distinct and separate income activities. Sometimes this takes the form of gender-specific crop production (see discussion and references to ethnographic literature from Cote d’Ivoire in Hoddinott and Haddad 1995). But it also extends to non-farm income sources. In Ghana, there is direct evidence from surveys that spouses do not have reliable information about profits from the other’s enterprises (Goldstein 2000). Fisher, Reimer, and Carr (2010) showed that husbands in Malawi underestimate their wife’s income. Using data from Tanzania, Bardasi et al. (2011) showed that women’s income-generating work may be under-reported when reported by other household members. They also showed that some forms of income generation may not be reported by African households due to wording on questionnaires. Collecting comprehensive metadata on the interview process and survey implementation, while not solving the problem, can help in better understanding the sources of potential bias. Beyond the gender dimension, there may be other reasons for underreporting of earnings. Much of the discussion of underreporting is focused on households at the top end of the distribution, where both non-response rates and under-reporting has been shown to be largest in developed countries (Lakner and Milanovic 2013). In countries with functioning tax and other administrative systems in place, it is possible to complement income data coming from household surveys. However, in most of Africa this is not an option. Yet, contrary to results from the developed world, Azzarri et al. (2010), drawing on data from five African economies and 12 others, show that, in fact, under-reporting is more serious for low income households—especially those households who receive most income from agriculture. The problem of under-coverage of households in the lower tail of the income distribution may also be due to poor survey implementation, since poorer households tend to live in more remote and less accessible areas. Imputation techniques are often used to fill the gaps but the task is often hampered by poor underlying data and low analytical capacity.

4 

Halliday (2010) explores this with data from the USA and El Salvador. In Papua New Guinea, Gibson (2001) shows that a correlation between the two reports of household size for households interviewed seven months apart is only 0.65 in urban areas and 0.75 in rural areas. There is less evidence from Africa, due to the lack of panel surveys.

Households and Income in Africa    51 Reporting income from activities which fluctuate both within a year and across years is difficult for households. Constructing an annual income measure for African households from a single survey visit is difficult; however, interviewing households several times in a year is seldom an option due to prohibitive costs. Choosing a long recall period raises concerns about recall bias due to memory lapse (Beegle et al. 2012, focus on such recall bias in regards to farm production and expenditure reporting; much more has been written on recall bias in consumption reporting). This may be particularly true for income which is generated in small, irregular amounts, such as is the case of the highly informal labor markets in Africa. Production of crops not subject to distinct seasons (and specific harvest periods), such as cassava and banana, are a challenge for households to report accurately (Deininger et al. 2012). Specific to measuring agricultural income, the use of non-standard measurement units for the quantification of production is particularly common in Africa. These units change across regions in the same country. The move to standard measurement units is likely to go hand-in-hand with development and globalization; thus, it is likely that over the next decades African countries will converge to using units which are more easily quantifiable. But that may be decades away. In some surveys, this is a hidden problem when the units are standardized on the questionnaire—resulting in an informal and likely error-prone impromptu conversion from a local unit to a standard one by interviewers and respondents. Beyond the difficulties of properly quantifying farm production, putting a monetary value to it is perhaps a greater challenge. Available price information is deficient in a number of dimensions. Household surveys may not include a separate price survey. The unit price of products sold can stand-in for price data to value home-consumed production. But for areas with mainly subsistence or near-subsistence farmers who sell little or no product on the market, this data will be thin. Some surveys ask farmers to value the share of production that is consumed by the household, but this is a practice not to be recommended particularly for autarkic smallholders. The issue of poor price data also comes up when using prices or unit values of a transaction(s) done over a short reference period to annualize income. The availability of good price data, at finer level of geographic and temporal resolution, would help solve the problem and contribute to better quantify a component of income, which in most African countries still represents a sizable share of household income. Similar concerns are raised in regards to non-farm self-employment, also very common in Africa, but we have little research on the scale of this problem. De Mel et al. (2009), in their Sri Lanka study, attribute the observed underreporting (to the tune of 30 percent) to poor valuation of goods and services used for home-consumption and the mismatch between revenues and expenses. Studying data from Thailand, Samphantharak and Townsend (2012) offer a number of specific suggestions for questionnaire design to measure household non-farm enterprise income. Other components of income are particularly challenging to measure in some African settings, with thin or non-existent markets. This is the case, for instance, when trying to quantify the net value of owner-occupied household services, also referred to as imputed rents. While much has been written on the subject for developed countries in terms of the importance and implications of imputing rents (The Canberra Group 2011; Frick and Grabka 2002), the proposed methods are not easily applicable to Africa, particularly to the rural areas where no sale or rental housing market exists. Better imputation methods and underlying data are badly needed.

52   Concepts

2.4  Patterns of Household Income in Africa A widely accepted tenet of the development literature is that, in the process of structural economic transformation that accompanies economic development, the farm sector declines as a country’s GDP grows (Chenery and Syrquin 1975). In rural areas, this implies that a shrinking agricultural sector and expanding rural non-farm (RNF) activities should be viewed as likely features of economic development. The available empirical evidence points to the existence of a large RNF economy, as well as an increasing role for RNF activities. In urban areas, development is generally accompanied by an increasing share of the formal economy in total output, and in an increase in the proportion of wage workers in the labor force (Gindling and Newhouse 2014). The patterns of pluri-activity (i.e. individuals performing different jobs at the same time or seasonally) also tend to undergo structural changes with development. Reflecting on these theories, in this section we offer a glimpse into the patterns of household income sources from recent African household data. We draw on the most consistently harmonized data currently available. We briefly focus on three areas: overall patterns of income sources, whether African household income follows the development story of other regions, and finally patterns of wealth and rural income sources in Africa. The agricultural sector is still a major source of livelihood (Table 2.1). Agriculture (which includes the value of crop and livestock production, plus agricultural wages) accounts for anywhere between 36 percent (Ghana) and 71 percent (Malawi) of household income.5 The share of non-agricultural income is higher in three out of eight countries.6 Transfers tend to be less than 10 percent of household income, but are as high as 18 percent in Kenya. In urban areas, as one would expect, non-agricultural income has the lion’s share of household income. The split between wages and self-employment varies from country to country. In rural areas, agriculture accounts for more than 50 percent of household income in all the countries in the sample and up to 80 percent in Malawi. The shares of livestock and agricultural wages vary across countries, linked as they are to country-specific agro-ecological conditions and rural institutions. The share of transfers is often lower in rural than in urban areas. One question that emerges when African survey data are pooled with data from other developing regions, is whether there is an African specificity, with African rural areas being more agricultural dependent compared to rural areas in countries at similar stages of development. Figure 2.1 (borrowed from Davis et al. 2014) illustrates the case. In the figure African countries are represented with black dots and countries in other regions with white dots. In terms of shares of income, overall, the share of non-agricultural income (Figure 2.1A) among rural households increases with increasing levels of GDP per capita in Africa as elsewhere. The decreasing importance of farm income over GDP can be seen in Figure 2.1B. But it is also true that agricultural sources of income are particularly important for the countries

5  These are the means of income shares from national representative surveys of households in the country. 6  These crude categories mask considerable variation within sectors. For both agricultural and non-agricultural income-generating activities, there may be a high-productivity/high-return sub-sector, confined mostly among privileged, better-endowed groups in high-potential areas. These high-return

Households and Income in Africa    53 from sub-Saharan Africa. With increasing levels of GDP, on-farm sources of income are replaced by the increasing importance of non-agricultural wage income (Figure 2.1C) and public and private transfers (Figure 2.1D). The countries in the African sample show a tendency towards on-farm sources of income—they have higher shares of on-farm income, and lower shares on non-agricultural wage income, than countries of other regions, including those at similar levels of GDP. The African countries generally have less income from agricultural wage labor (Figure 2.1E), reflecting low levels of rural landlessness. There is no discernable difference between African and non-African countries in terms of income from non-agricultural self-employment (Figure 2.1F), nor by GDP. Finally, we turn briefly to the relationship across countries between rural income-generating activities and wealth; for each country we examine activities by expenditure quintile. Figure 2.2 charts income shares by expenditure quintile. Focusing on on-farm activities, the darkest color, we see a sharp decrease in the share of on-farm income with increasing levels of wealth, dropping from around 50 percent of income in the poorest quintile in most countries, to less than 20 percent in the wealthiest quintile. The drop in on-farm sources of income is made up by the increasing importance of off-farm (non-agricultural wage and self-employment) sources of income for wealthier rural households. With the exception of those countries that have negligible agricultural labor wage markets, poorer rural households tend to have a higher rate of participation in agricultural wage employment. Similarly, the share of income from agricultural wage labor is more important for poorer households in these countries, and the relationship holds regardless of the level of development.

2.5 Conclusions Household income data are crucial to gain a better understanding of household and individual livelihoods in Africa and to inform policy for growth and poverty reduction. While from a conceptual standpoint, clear and unambiguous definitions are available, translating these concepts into practice continues to be a challenge, particularly in African countries where the large shares of informality and irregularity of income sources compounds measurement problems. Although dramatic improvements in the availability of data have occurred in recent years, the lack of adequate standards fully adaptable to developing countries’ contexts, combined with low in-country capacity, continue to hamper improvements in the quality and comparability of income data and, consequently, in the policy relevance of the analysis. Initiatives activities usually have significant barriers to entry or accumulation in terms of land, human capital, and other productive assets (Reardon et al. 2000). In contrast, a low-productivity segment usually serves as a source of residual income or subsistence food production and as a refuge for the rural poor. This covers activities such as subsistence agriculture, seasonal agricultural wage labor, and various forms of off-farm self-employment. The observed dualism also often appears to be drawn along gender lines, with women more likely to participate in the least remunerated agricultural and non-agricultural activities. Given the existence of both low- and high-return rural income-generating activities, with varying barriers to access, previous empirical studies have shown a wide variety of results in terms of the relationship of rural income-generating activities, and in particular RNF activities, to poverty in Africa (Senadza 2011).

Table 2.1  Diverse income-generating activity is the norm in Africa Group I

Group II

(1)

(1) + (2) + (3)

(4) + (5) + (6) + (7)

(1) + (2)

NonOn-farm farm total total

Transfers and Off-farm other total

(2)

(5)

(6)

(7)

Group III

(3)

(4) Non-farm wage employment

Non-farm selfemployment Transfers Other

Agricultural total

Nonagricultural total

(4) + (5)

(6) + (7)

(3) + (4) + (5) + (6) (7) +

Country and year

Agriculture- Agriculture— crops livestock

Agricultural wage employment

Total Ghana 2005 Kenya 2005 Madagascar 2001 Malawi 2011 Nigeria 2010 Niger 2011 Tanzania 2009 Uganda 2009/10

32.0 24.9 47.3 51.4 38.0 40.7 41.3 37.5

1.8 12.1 7.0 5.3 4.0 7.3 10.8 9.7

2.1 5.9 4.6 14.7 1.1 2.8 3.7 6.3

18.4 24.4 19.1 13.6 17.7 8.5 13.7 16.7

33.6 12.2 13.3 8.4 36.1 29.6 20.7 20.0

11.8 17.8 5.6 5.8 1.1 10.8 9.7 7.9

0.3 2.8 3.1 0.9 2.0 0.1 0.0 2.0

35.9 42.9 58.9 71.4 43.1 50.9 55.8 53.4

64.1 57.1 41.1 28.6 56.9 48.9 44.2 46.6

33.8 37.0 54.3 56.6 42.0 48.1 52.1 47.2

52.0 36.6 32.4 22.0 53.8 38.0 34.5 36.7

12.1 20.5 8.7 6.8 3.2 10.9 9.7 9.9

66.2 63.0 45.7 43.4 58.0 51.8 47.9 52.9

8.3 3.2 17.8 10.4 9.4 6.4 8.3 11.5

0.3 1.5 2.9 1.3 1.5 0.1 3.2 2.5

1.4 2.0 4.9 13.1 1.4 1.1 2.1 1.9

31.7 54.3 41.1 44.9 29.9 29.8 35.7 32.8

43.2 21.0 20.6 21.8 51.9 48.2 43.3 32.1

14.7 13.4 9.9 4.0 2.0 13.2 7.3 13.9

0.3 4.7 2.8 4.5 3.9 0.3 0.0 5.2

10.1 6.6 25.7 24.8 12.3 7.6 13.6 15.9

89.9 93.4 74.3 75.2 87.7 91.5 86.4 84.1

8.6 4.7 20.7 11.7 10.9 6.5 11.5 14.0

74.9 75.4 61.6 66.6 81.9 78.0 79.0 64.9

15.0 18.0 12.7 8.6 5.8 13.5 7.4 19.2

91.4 95.3 79.3 88.3 89.1 92.6 88.5 86.0

49.5 32.0 56.5 59.1 46.2 48.1 52.6 45.8

3.0 15.6 8.2 6.0 9.4 8.9 13.3 11.9

2.5 7.2 4.5 15.0 1.1 3.2 4.3 7.7

8.6 14.6 12.3 7.4 10.5 3.9 6.5 11.5

26.5 9.3 11.0 5.8 31.3 25.6 12.7 16.2

9.6 19.2 4.3 6.1 0.5 10.4 10.5 5.9

0.3 2.1 3.2 0.2 1.2 0.0 0.1 1.0

55.0 54.8 69.3 80.1 56.6 60.1 70.2 65.4

45.0 45.2 30.7 19.9 43.4 39.9 29.8 34.6

52.5 47.6 64.8 65.1 55.6 56.9 66.0 57.7

35.1 23.9 23.2 13.7 41.8 29.5 19.2 27.7

9.9 21.4 7.5 6.3 1.6 10.4 10.5 6.9

47.5 52.4 35.2 34.9 44.4 43.0 34.1 42.3

Urban households Ghana 2005 Kenya 2005 Madagascar 2001 Malawi 2011 Nigeria 2010 Niger 2011 Tanzania 2009 Uganda 2009/10 Rural households Ghana 2005 Kenya 2005 Madagascar 2001 Malawi 2011 Nigeria 2010 Niger 2011 Tanzania 2009 Uganda 2009/10

Households and Income in Africa    55 Share of non-agricultural income

(a)

90% 80%

80% R² = 0.4822

70%

70%

60%

60%

50%

50%

40%

40%

30%

30%

20%

20%

10%

10%

0%

6.0

6.5

7.0

7.5

8.0

8.5

Share of non-agricultural wage income

(c)

0%

9.0

70%

40%

60%

35%

6.0

6.5

7.0

7.5

8.0

8.5

9.0

Share of transfer income

50%

30%

40%

R² = 0.337

25% 20%

R² = 0.3315

30%

15%

20%

10%

10%

5% 6.0

6.5

7.0

7.5

8.0

8.5

0%

9.0

6.0

6.5

7.0

7.5

8.0

8.5

9.0

Share of agricultural wage income

(e)

Share of non-agricultural self-employment income

(f)

30%

40%

25%

35%

20%

30% 25%

15%

20%

R² = 0.0209

10%

15%

R² = 0.003

10%

5% 0%

R² = 0.4286

(d)

45%

0%

Share of on-farm income

(b)

90%

5% 6.0

6.5

7.0

7.5

8.0

8.5

9.0

0%

6.0

6.5

7.0

7.5

8.0

8.5

9.0

Figure 2.1   Do rural income patterns in Africa compare with others? like the Luxembourg Income Study, the Canberra Group, and the Wye City Group have made important contributions to advancing the agenda but their focus on developed countries has limited their usefulness in Africa where there are radically different realities. Initiatives as the Rural Income Generating Activities (RIGA) program described in Covarrubias et al. (2009) are a step in the right direction. Household income in Africa is diverse. While agriculture continues to be the most important source of income for the majority of poor households, diversification into non-farm sources is the norm and not the exception even in the poorest countries of Africa. Economic development is bound to result in a shrinking agricultural sector and the growth of more formal, non-farm activities. As such, household income sources will also shift. While the shift will make international statistical income standards more applicable, in the medium run, the range of measurement challenges described here will remain relevant. Understanding the fundamentals of development in Africa and policies that improve economic well-being will require continued investment in the theory and measurement of household income.

56   Concepts Share of total income from main income-generating activities by expenditure quintiles

Shares of income (%)

1 .8 .6 .4 .2 0

N

IG

10

04 IG

05

09 N

TA N

N KE

A

11

05

G U

A H G

G

A

06

98 U

A H G

AL

11

01 M

AD

92 M

A H G

AD

04 M

AL

11 M

ER IG N

93

12345 1234 5 12345 12345 12345 12345 12345 12345 1 2345 12345 12345 12 345 12345 12345

On-farm activities

Agricultural wages

Transfers and other non-labour sources

Non-farm activities

Figure  2.2   As income grows, sources of income shift for African households. Note: 1. Surveys sorted by increasing per capita GDP 2. Expenditure quintiles move from poorer to richer

References Azzarri, C., Carletto, C., Covarrubias, K. et al. (2010). Measure for Measure: Systematic Patterns of Deviation between Measures of Income and Consumption in Developing Countries. Evidence from a New Dataset. Paper presented at the Fifth International Conference on Agricultural Statistics, Kampala, Uganda. Bardasi, E., Beegle, K., Dillon, A., and Serneels, P. (2011). Do labor statistics depend on how and to whom the questions are asked? Results from a survey in Tanzania. World Bank Economic Review, 25(3):418–477. Beaman, L., and Dillon, A. (2012). Do household definitions matter in survey design? Results from a randomized survey in Mali. Journal of Development Economics, 98(1):124–135. Beegle, K., Carletto, C., and Himelein, K. (2012). Reliability of recall in agricultural data. Journal of Development Economics, 98(1):34–41. Canberra Group. (2001). Expert Group on Household Income Statistics:  Final Report and Recommendations. Ottawa. Canberra Group. (2011). Handbook on Household Income Statistics. Geneva: United Nations. Chenery, H.B., and Syrquin, M. (1975). Patterns of Development 1950-1970. London: Oxford University Press. Commission of the European Communities (1993). System of National Accounts. Brussels/ Luxembourg, New York, Paris, Washington, DC. Covarrubias, K., De la O Campos, A., and Zezza, A. (2009). Measuring Household Income using Multitopic Household Surveys in Developing Countries. Washington, DC: FAO Project.

Households and Income in Africa    57 Davis. B., Di Giuseppe, S., and Zezza, A. (2014). Income Diversification Patterns in Rural Sub-Saharan Africa: Reassessing the evidence. Washington, DC: World Bank and FAO. Davis, B., Winters, P., Carletto, C., et al. (2010). A cross country comparison of rural income generating activities. World Development, 38(1):48–63. Deininger, K., Carletto, C., Savastano, S., and Muwonge, J. (2012). Can diaries help in improving agricultural production statistics? Evidence from Uganda. Journal of Development Economics, 98(1):42–50. Deaton, A., and Zaidi, S. (2002). Guidelines for Constructing Consumption Aggregates for Welfare Analysis. Washington, DC: The World Bank. de Mel, S., McKenzie, D. J., and Woodruff, C. (2009). Measuring microenterprises profits: must we ask how the sausage is made? Journal of Development Economics, 88(1):19–31. Ellis, F., Devereux, S., and White, P. (2009). Social protection in Africa. Cheltenham: Edward Elgar Publishing. Fisher, M., Reimer, J. J., and Carr, E. R. (2010). Who should be interviewed in surveys of household income? World Development, 38(7):966–973. Fox, L., and Sohnesen, T. P. (2012). Household Enterprises in Sub-Saharan Africa: Why They Matter for Growth, Jobs, and Livelihoods. World Bank Policy Research Working Paper 6184, World Bank, Washington, DC. Frick, J., and Grabka, M. (2002). The Personal Distribution of Income and Imputed rent. A Cross-national Comparison for the UK, West Germany and the USA. German Institute of Economic Research Discussion Paper # 271, DIW Berlin, German Institute for Economic Research. Gibson, J. (2001). Measuring chronic poverty without a panel. Journal of Development Economics, 65(2):243–266. Gindling, T., and Newhouse, D. (2014) Self-Employment in the Developing World. World Development, 56: 313–331. Goldstein, M.P. (2000). Intrahousehold Allocation and Farming in Southern Ghana. PhD Dissertation, University of California, Berkeley. Glewwe, P. (2000). Household roster, in M. Grosh and P. Glewwe (eds), Designing Household Survey Questionnaires for Developing Countries. Washington, DC: The World Bank. Guyer, J.I. (1981). Household and community in African studies. African Studies Review, 24(2/3):84–137. Haggblade, S., Hazell, P., and Reardon, T. (eds) (2007). Transforming the Rural Nonfarm Economy. Baltimore: Johns Hopkins University Press. Halliday, T.J. (2010). Mismeasured household size and its implications for the identification of economies of scale. Oxford Bulletin of Economics and Statistics, 72(2):246-262. Hoddinott, J., and Haddad, L. (1995). Does female income share influence household expenditure? evidence from Cote d’Ivoire. Oxford Bulletin of Economics and Statistics, 57(1):77–96. ILO. (2003). Household Income and Expenditure Statistics. Seventeenth International Conference of Labour Statisticians, International Labour Organization, Geneva, Switzerland. Joshi, K., Amoranto, G., and Hasan, R. (2011). Informal sector enterprises: some measurement issues. Review of Income and Wealth, 57:S143–S165. Lakner, C., and Milanovic, B. (2013). Global Income Distribution: From the Fall of the Berlin Wall to the Great Recession. World Bank Policy Research Working Paper 6719. Lanjouw, J., and Lanjouw, P. (2001). The rural non-farm sector: issues and evidence from developing countries. Agricultural Economics, 26(1):1–23.

58   Concepts Lanjouw, P., and Feder, G. (2001). Rural Nonfarm Activities and Rural Development: From Experience Towards Strategy. World Bank Rural Development Strategy Background Paper #4. Lucas, R.E.B. (2006). Migration and economic development in Africa: a review of evidence. Journal of African Economies, 15(2):337–395. McKay, A. (2000). Should the survey measure total household income? in M. Grosh, and P. Glewwe (eds), Designing Household Survey Questionnaires for Developing Countries. Washington, DC: The World Bank. Randall, S., Coast, E., Compaore, N., and Antoine, P. (2013). The power of the interviewer. Demographic Research, 28(27):763–792. Reardon, T., Taylor, J.E., Stamoulis, K. et al. (2000). Effects of non-farm employment on rural income inequality in developing countries: an investment perspective. Journal of Agricultural Economics, 51(2):266–288. Samphantharak, K., and Townsend, R. M. (2012). Measuring the return on household enterprise: what matters most for whom? Journal of Development Economics, 98(1):58–70. Senadza, B. (2011). Does non-farm income improve or worsen income inequality? Evidence from Rural Ghana. African Review of Economics and Finance, 2(20):104–121. Vijverberg, W. P., and Mead, D. C. (2000). Household enterprises, in M. Grosh, and P. Glewwe (eds), Designing Household Survey Questionnaires for Developing Countries. Washington, DC: The World Bank. Young, S.A. (2000). Income from Households’ non-SNA Production:  A  Review. Geneva: International Labour Organization.

Chapter 3

T ransform at i on of Af rican Farm-c um- Fa mi ly Stru cture s Catherine Guirkinger and Jean-Philippe Platteau

3.1 Introduction The analysis of slavery or serfdom by Evsey Domar (1970) suggests that, in a two-factor economy based on land and labor alone, it is impossible to have free land, free peasants, and non-working landowners simultaneously. Any two elements of this triad can nevertheless coexist (see also Boserup 1965: 73; Mathur 1991: 47–49; Binswanger et al. 1995: 2670–2673). Peasant societies correspond to the configuration in which free peasants and free land exist simultaneously while a class of non-working landowners is conspicuously absent. By contrast, under systems of slavery, serfdom, debt peonage, or indentured labor, there is free land and non-working landowners, yet not free peasants. In conditions of land abundance, therefore, the workers need to be enslaved, enserfed, indentured, or tied for a landowning class to be able to exist, and this requires special political conditions, such as they have been found in Latin America, Eastern Europe and Russia, the southern part of North America, or colonial plantations in parts of Asia and Africa. In sub-Saharan Africa, ruling families, lineages, or larger social groups did try hard to “capture peasantries” or develop systems of property rights in man. Yet, the presence of strongly hierarchical societies stratified along caste lines has always been limited to certain parts of the region.1 When trying to explain why many “capturing” attempts were unsuccessful in Africa in general, and in Guyana in particular, Mathur (1991) pointed out that people “knew too well the byways of the forest and hills in the country; whenever any attempt was 1  Well-known examples are the Hausa and the Songhaï-Zarma in Niger and Mali, the Bambara in Mali, the Wolof, Toucouleur and Soninké in Senegal, the Maures in Mauritania, the Tutsi-Hutu in Rwanda-Burundi, and the Peul aristocracy in Guinea, northern Nigeria, and northern Cameroon.

60   Concepts made to use them, they would in a few days escape into the jungle and could not be brought back.” This solution was not available to indentured laborers brought from outside since they could be “kept out of the forests by the local tribes who would not hesitate in murdering such intruders” (48–49). We have to add that the possibility of acquiring membership in stranger communities (migration was a common phenomenon) combined with the rudimentary character of cultivation techniques opened many “exit options” to the hunted-down cultivators who wished to escape the grip of oppressive rulers or chiefs, particularly so in the dense forest areas of humid central Africa (Platteau 2000: Chapter 5). The absence of a landowning, non-working class does not imply, however, that small peasant family farms predominate in the African countryside. In fact, the social structure in Africa has long been characterized by clans and extended families, so that the model of a free peasantry is not exactly vindicated, calling for a refinement of Domar’s framework. Such a refinement should allow for the possible existence of intermediate forms between free and tied labor that may be combined with free land and non-working landholders. Indeed, in many parts of sub-Saharan Africa (e.g. Burkina Faso, Gambia, Senegal, or Mali), land is cultivated by large patriarchal families where strong hierarchy prevails and many adult workers, males and females, are placed under the authority of a single head acting as the ultimate manager of both land and labor within the household. From an analytical point of view, it then makes sense to treat the patriarchal family as a functional equivalent of the feudal land estate. In various countries, interestingly, this family organization dominated even in conditions of plentiful land resources: we thus have simultaneously family farms, free land, and non-working patriarchs. However, the tying of the labor force to the strong authority of the patriarch broadly reconciles this observation with Domar’s framework. Empirical observations show that these farm-cum-family structures are gradually disappearing or transforming themselves into forms closer to the more individualized structures found in Europe and Asia. Furthermore, it appears that the strong power of the patriarch is eroding as land becomes scarce, suggesting that when the frontier closes more individual farms emerge, in accordance with Domar’s intuition. Africa therefore offers a unique laboratory for studying the forces that drive such a transformation process. When discussing this issue, we need to carefully distinguish between two different types of individualization of farms and families: (i) the emergence of mixed farm structures in which the family remains whole yet adult members receive private plots that they can use for their own benefit during limited periods of time, and (ii) the splitting of the stem household into branch households that goes hand in hand with the division of the land and the granting of pre-mortem inheritance. The first type, that maintains collective fields on which all members of the extended family work as well as collective kitchens and meals in which the same take part, is obviously a less advanced form of individualization than the second type. Economic theory supplies us with several possible explanations of the individualization of farm-cum-family structures, but most of them actually deal with only one of the two aforementioned forms. Thus, the theory of Ester Boserup (1965), which puts emphasis on the growing role of labor quality, the theory of Andrew Foster and Mark Rosenzweig (2002), which focuses on the diminishing importance of the public good character of consumption, and a risk-based theory put forward by Michael Carter (1987) and by Matthieu Delpierre et al. (2013) are motivated by the desire to explain the splitting or division of large households. By contrast, the theory of Marcel Fafchamps (2001), which draws attention to a commitment problem on the side of the household head, and that of Elizabeth Sadoulet (1992),

Transformation of African Farm-cum-Family Structures    61 which assumes a problem of limited liability on the side of the worker, aim at explaining the awarding of private plots within a collective or family structure. Only the theory proposed by Guirkinger and Platteau (2011) purports to explain both types of individualization of farm-cum-family structures. Furthermore, precisely because it gives a key role to resource endowments (and outside opportunities), it may be articulated with Domar’s framework yet in a manner that necessitates a new substantial adjustment. As land becomes scarce (the land frontier is closed), the typical mechanism leading to individualization of farm-cum-family structures is the “market path.” This means that the decline of the value of labor relative to land gives rise to a situation where landowners no longer need to “capture” peasants (or to run after escaping workers) because land-hungry farmers readily offer their labor to the landed class. Increasingly active land and labor markets makes inequality in land distribution persist as a result of economic forces alone, so that support from political rulers is no longer required. Guirkinger and Platteau, on the other hand, suggest, in a way that will become clear later, that an alternative “non-market path” may exist in which individualization obtains in the absence of land and labor markets. In this chapter, we proceed in three steps. In section 2, we review the partial theories of individualization, bringing their arguments into light and discussing their strengths and weaknesses with special reference to the African context. Section 3 begins by expounding the central argument behind the more general theory of Guirkinger and Platteau (2011), which explains family-cum-farm structures in terms of the magnitude of land endowment and outside opportunities. It then moves to the empirical evidence that has been harnessed in support of this theory. Finally, in section 4, we offer some conclusive discussion regarding the contribution of economic theory regarding our understanding of evolution of farm-cum-family forms in sub-Saharan Africa, in particular.

3.2  Partial Theories of Individualization of Farm-cum-Family Structures So far, economists have proposed few theories of the evolution of the farm-cum-family structure, and the available theories aim at explaining either the shift from the collective farm to the mixed form in which individual and collective fields coexist, or the break-up of the collective farms into individual units. Fafchamps (2001) offers an example of the former since he tries to explain the decision of the household head to allocate individual plots to family members. His explanation rests on the assumption that there exists a serious commitment problem inside the family: the head is unwilling or unable to commit to reward the work of other family members on the family field after the harvest, and the latter are therefore tempted to relax their labor efforts or to divert them to other income-earning activities. To solve this commitment failure problem, the head decides to reward his wife and dependents by granting them access to individual plots of land and the right to freely dispose of the resulting produce. Such a commitment problem, however, can only exist if the short-term gain for the household head when deviating from cooperation (reneging on the promise to reward workers for their efforts on the collective field) exceeds the long-term flow of benefits ensuing from a smooth relationship

62   Concepts between him and the working members. As Fafchamps himself admits, this condition is restrictive since the game played within the family is by definition of a long (and indeterminate) duration, and the discount rate of future benefits typically low (future cooperation among close relatives matters a lot). Moreover, even assuming that Fafchamps’ hypothesis is valid, it remains unclear why there should be a tendency over time for collective farms to transform themselves into mixed farms, as we seem to observe in reality. Other authors have tried to explain the coexistence of collective fields and individual plots in agricultural farms, yet they explicitly refer to agricultural producer cooperatives or quasi-feudal set-ups rather than extended family farms. Regarding producer cooperatives, emphasis is typically put on the existence of economies of scale for certain types of activities, or on the need for insurance and the role of income-pooling (Chayanov 1991; Swain 1985; Putterman 1983, 1985, 1987, 1989; Putterman and DiGiorgio 1985; Carter 1987; Meyer 1989). Interestingly, the latter argument has been recently extended to the family context by Delpierre et al. (2013). As in Carter (1987), the analysis focuses on a trade-off between efficiency and insurance considerations. The trade-off arises because working in common on a collective field and distributing the output equally among participant members insures them against idiosyncratic risks, but joint farming also entails efficiency losses owing to the moral-hazard-in-team problem (itself caused by the impossibility to measure individual contributions and reward them accordingly). Unlike in Carter, however, Delpierre et  al. assumed that family members may make reciprocal transfers between themselves for the purpose of smoothing idiosyncratic variations in income. In spite of that generous assumption in favor of complete individualization, they show that the optimum may correspond to the mixed farm regime, where a collective field subsists. As for relationships between estate owners and workers, limited liability constraints and the demand for insurance are the main motives prompting the adoption of the mixed farm structure. The idea behind the first explanation (Sadoulet 1992) is that landlords worry about the possibility that their tenants are unable to pay the entire amount of their land rents or shares because of a wealth constraint. Awarding them a private plot under a labor exchange arrangement—the landlord combines direct cultivation with the help of wage workers on a portion of his property and the distribution of individual plots for private use on the remaining portion of the estate—is the best way for the landlords to solve the dilemma between reducing the rent charged to lower the occurrence of default and increasing it to capture the full surplus that the tenant can obtain from utilizing his family labor. This labor–service contract (the exchange of free labor for use on the landlord’s field against free access to a private plot of land for personal use by the tenant), indeed, enables the landlord to impose an optimal level of insurance and, thus, efficient resource use on the tenant.2 Another justification for the same system, proposed by Allen (1984), lies in risk considerations: the labor exchange arrangement is equivalent to a sharecropping contract that would be applied to the whole farm area and may therefore be motivated by risk sharing. Yet, underlying Allen’s argument is the restrictive assumption that labor effort on the estate owner’s 2  Such a system has been widely observed, for example in the post-Carolingian manors of medieval Europe, in American plantations using slave labor and in Russian boyar estates using serf labor (Van Zanden 2009: 56, fn. 13; Blum 1957, 1961; Kolchin 1987), in feudal Japanese farms during the Tokugawa era (Smith 1959), or among estate landlords of Latin America, such as those employing Inquilino laborers in Chile after the middle of the eighteenth century (Bauer 1975; de Janvry 1981).

Transformation of African Farm-cum-Family Structures    63 field can be monitored at no cost. If monitoring is imperfect, the equivalence result does not hold anymore: granting sharecropping contracts to risk-averse tenants on the whole estate domain is more efficient than a system in which individual plots coexist with the landlord’s field. As a result, the functional equivalent of the collective sector in a producer cooperative may not come into existence. Let us now turn to the strand of theories purporting to explain farm break-ups. One of the key references here is Foster and Rosenzweig (2002).3 The main intuition of the authors is that an extended family may decide to split if the benefit of sharing public goods by co-residing is smaller than the loss of efficiency due to decreasing returns to scale in production. The trade-off is therefore between advantages in joint consumption and efficiency in production. If this line is followed, we can detect two different ways to explain the increasing incidence of individual farms: (i) growing disinterest of younger generations in the sort of public goods jointly produced and consumed on the collective farm, and (ii) rising importance of decreasing returns to scale as a result of the shift to more land-intensive agricultural techniques. Clearly related to the latter proposition is the work of Boserup (1965), who attributes the rise of peasant farms to growing land scarcity and the consequent intensification of agricultural techniques. The underlying argument has enjoyed a wide resonance among development economists who have helped express it in the language of modern information theory (Binswanger and Rosenzweig 1986; Binswanger and McIntire 1987; Pingali, Bigot, and Binswanger 1987; Binswanger, McIntire, and Udry 1989; Hayami and Otsuka 1993). It can be stated as follows. As land pressure increases, farmers are induced to shift to more intensive forms of land use, which implies that they adopt increasingly land-saving and labor-using techniques. An important characteristic of these techniques is that labor quality, which is costly to monitor, assumes growing importance. Given the incentive problems associated with care-intensive activities (sometimes labeled “management diseconomies of scale”), the small family or peasant farm in which a few co-workers (spouses and their children) are residual claimants appears as the most efficient farm structure. Although Boserup’s story is undeniably appealing, both theoretically and empirically, it cannot apparently account for situations in which evolution towards more individualized forms of family-cum-farm structures takes place in the absence of noticeable technical progress. Thus, in Russia during the seventeenth to nineteenth centuries, a shift from large and complex agricultural households (married brothers stay together at least till the death of the father) to smaller and more simple ones (married brothers part with each other while the father is still alive, but a household may remain multigenerational) occurred, a change which historians generally ascribe to the expansion of non-agricultural opportunities rather than to the adoption of new agricultural techniques (Worobec 1995; Moon 1999). In the old cotton zone of southern Mali (West Africa), collective farms appear to be increasingly replaced by mixed farms and small farms born of the break-up of large family farms, despite persisting technological stagnation (Guirkinger and Platteau 2014a). In addition, it is a striking feature of the countryside in Mali and other West African countries that when private plots coexist with a collective family field, household members continue to take their meals 3  They do not allow for individual plots since in their framework co-residence implies collective farming only.

64   Concepts in common, and the preparation of the meals continues to be based on a rotation between women belonging to the different constituent households.

3.3  A More General Approach to Changing Family-cum-Farm Structures 3.3.1 Theory Guirkinger and Platteau (2014b) have proposed a different theory to account for the gradual individualization of family-cum-farm structures, understood as the growing incidence of both private plots within mixed farming units and splitting of stem households into branch households. Like Boserup, they put primary emphasis on the role of changing land/labor ratios yet, unlike her, they do not refer to technological change as the key mechanism through which the influence of land pressure is being felt. Moreover, their explanation does not rely on the diminishing value of joint consumption, as their observations in West Africa indicate that individualization of big family farms in the form of private plots does not end the practice of common kitchens and collective meals. Finally, they do not need to allow for risk aversion to justify the existence of collective farms. The central mechanism that operates in their framework relies on the existence of a strong patriarchal authority inside the extended household. It is, indeed, because the household head does not act as a benevolent despot fully identified with the interests of the members that a trade-off arises between efficiency and rent capture considerations. When deciding whether to give private plots to members and how large they should be, the head weighs two factors. For one thing, production is more efficient on private plots than on the collective field where cultivation is plagued by the moral hazard-in-team problem. Since the head must ensure that family members agree to stay on the family farm while they have outside options available to them, awarding individual plots allows him to more easily satisfy their participation constraints. For another thing, because the head’s income entirely comes from the produce obtained on the collective field owing to unenforceable transfers from the private plots, competition between the two types of plots for the allocation of work effort by the members is bound to cause a fall in the head’s income. It is evident that, if transfers from private plots were enforceable by the head, he would earmark the whole family land for private use by individual members since he would thus maximize his own income (members are put at their reservation utilities). In other words, the head acting as a principal would maximize efficiency. It is clearly the non-enforceability of transfers from members to the head that cause efficiency losses. There is another decision that the patriarch has to make, that is, whether to maintain the family and the farm whole (with or without private plots) or to allow a split of the stem household and the concomitant division of the family land. The extent of the split itself is to be decided since the number of (male) members authorized to leave may vary. In the case of a pre-mortem split, the total labor force available for work on the collective field decreases, which harms the patriarch, yet it is no more incumbent on him to provide for the needs of

Transformation of African Farm-cum-Family Structures    65 the departed members, which favors him. Depending on the relative importance of the various effects at work, he may prefer a mixed regime with private plots to the collective regime, or he may choose to split the family. How does the family-cum-farm structure evolve when land becomes more scarce, or when outside opportunities improve for the members? The general answer provided by Guirkinger and Platteau is that if a change occurs it will be in the direction of increasing individualization. As land pressure increases (or as outside opportunities improve), the patriarch may decide to transform a collective farm into a mixed farm or into smaller independent units. The initial organizational form is always the collective farm, which is optimal when land is sufficiently abundant. Which individualized form will first succeed the collective one is a complex issue. The reason is that there actually exist many possibilities depending upon the number of (male) members authorized to leave, and upon whether private plots are granted to the remaining members when some of them have left with a portion of the family land. The precise sequence in which different forms succeed each other as land pressure increases is the following: collective farm; partial split with no private plots; mixed farm (private plots but no split); partial split with private plots. Note that an important assumption underpinning the whole above framework is that adjustment to rising land pressure is easier to achieve through change in the family-cum-farm structure than through demographic change and fertility reduction, or through land markets. While fertility reduction requires a long-term horizon, land markets are highly imperfect owing to large transaction costs or because the fear of losing land prevents the supply side of the market from being activated (Basu 1986; Boucher et al. 2008; Platteau 2000: Chapter 4). In this context, any change in land allocation is the outcome of a decision regarding the organization of the family farm.

3.3.2 Evidence In support of their theory, Guirkinger and Platteau (2014a) and Goetghebuer, Guirkinger, and Platteau (2014) have adduced three different pieces of evidence based on their fieldwork in central Mali.

3.3.2.1  Qualitative insights The first type of evidence is made of qualitative insights obtained from open questions or semi-structured interviews with household heads and other family members. Some of these insights validate central assumptions on which the theory is built. First, questions aimed at unveiling the decision-making structure inside the household unambiguously point to a strong patriarchal authority. This is true not only with respect to land allocation (whether to award private plots or to let members leave the household with their rightful share of the family land), use of the household’s farm implements on private plots and income distribution (the head distributes the proceeds from the collective field) but also with respect to authorization of credit if members want to borrow from an external source. Heads see themselves as acting on behalf of the family and responsible for its well-ordered functioning. They assume that they would have to bear responsibility if a loan taken by a member were to be defaulted on, hence their perceived right to approve any loan taken by members.

66   Concepts There is one domain, however, in which household heads admit that their power is limited. Revealingly, this is with respect to consumption choices made by children who have independent incomes (from private plots) and claim the right to spend them according to their own preferences. Second, heads have often expressed concerns about their children’s proclivity to think of their own interests instead of the general interest of the whole family and those of their parents. Concern has thus been explicitly voiced regarding allocation of work effort between private plots and the collective field when the former do exist: according to many patriarchs, household members do not do their best while working on the collective plots, thereby causing yields to fall. For example, one of them said that “more effort is applied to the individual plots and when members work on the collective plot, they are tired.” Another one complained that when they work on the collective field, his sons “are prone to keep energy in reserve for their individual plots” (“ils se réservent”) (Guirkinger and Platteau 2014a: 212). A lot of interviewed household heads also mentioned that a better quality of labor would increase the collective output. A second type of insight comes in support of the conclusions reached by the theory. Thus, when queried about the reasons underlying the trend toward growing individualization, whether in the form of mixed farms or broken-up households, most heads pointed to increasing land pressure and consumption needs, particularly among the younger generations. As land becomes scarce, so the first argument runs, family heads find it increasingly difficult to provide for the subsistence of all their members from the collective fields. On the other hand, the main reasons given by the heads of branch households to explain why they themselves broke away from the stem household are rising land pressure in the stem household (34% of interpretable answers), and the eruption of conflicts within the family, most often involving their brothers or uncles (again 34%). Other reasons include low production in the stem household, and the existence of special needs that could not have been satisfied if the member had stayed with the whole family (expensive medicine to cure a wife, for example). Note that village or community-level conflicts, including intra-family disputes, are often caused by acute scarcity, real or anticipated, of available land assets (see André and Platteau 1998, for Rwanda, and Haugerud 1993: 162–176, for Kenya). There may thus be a significant overlap between the two dominant motives alleged to lie behind household splits.

3.3.2.2  Quantitative evidence 1: determinants of  private land allocation The first sort of quantitative evidence harnessed in support of the endowment-based theory is based on an econometric model attempting to highlight household characteristics responsible for the presence or absence of private plots (Guirkinger and Platteau 2014a). The estimated model is a simple probit model in which the dependent variable is interpreted as the probability to grant private plots to (male) members. In accordance with expectations, the results show that the household head is more likely to distribute private plots when land per man is lower and when the family is larger. Regarding the latter, the implication of the theory must be borne in mind: when the size of the workforce on the collective field is larger, the scope of the moral-hazard-in-team problem increases, which enhances the relative attractiveness of private plots where no efficiency problem arises. The two key explanatory

Transformation of African Farm-cum-Family Structures    67 variables—land availability and size of family workforce—have been instrumented with the help of historical data. In this way, the potential endogeneity of current land availability and household size—residential choices, and therefore household size, are likely to be directly influenced by land allocation—is taken care of.4 The second main result reached in the study of the determinants of private plot allocation was much less expected. When the family is decomposed into married men and other members, only the first category appears to have a significant influence on private plot allocation, and the effect is strongly significant. Moreover, the magnitude of the effect is far from negligible: thus, an increase of one unit in the number of married men increases the probability of individual plots by almost 10 percentage points. This result suggests that the standard moral-hazard-in-team argument needs to be refined. As usually stated, this argument implies that the magnitude of the efficiency loss increases with the number of team members considered as equivalent units. What the above shows is the assumption of an undifferentiated impact of group size is not applicable to the context of an extended or complex family. Why is it that free riding on other members’ efforts in collective cultivation is observed when several married men work together, and not when unmarried ones do? Two types of explanations come to mind here. First, being strangers, daughters- or sisters-in-law tend to make the household more heterogeneous: they are not tied to the household by the same emotions and feelings of loyalty as their husbands. But the weakening of solidarity may also arise from the behavior of the sons or nephews if, once they get married, they tend to identify with their nuclear family more than with the extended family. As a result, they may not feel as loyal as before to the large household unit, thereby fostering feelings of competition and rivalry. Second, when the families of married men are of unequal size, the sharing rule is bound to look arbitrary to at least some couples. Thus, if the sharing rule provides for equal incomes to all married adults regardless of the size of their family, parents with more children feel discriminated. On the contrary, if shares are proportional to family size, parents with fewer children feel exploited because they work partly for the benefit of larger conjugal units. Interestingly, these two weaknesses of complex households are also stressed in anthropological and historical literature (see, for example, Worobec 1995: 81, for pre-Communist Russia). It is striking that the aforementioned results continue to hold if what is explained is not the presence or absence of private plots but the share of the family land that is earmarked for individual cultivation. In particular, the stronger the land pressure or the smaller the land endowment of the household the larger the share of it that is allocated to private plots.

3.3.2.3  Quantitative evidence 2: comparing land productivities between collective and private plots In another paper, Guirkinger and Platteau, together with Goetghebuer, have tested their theory through a different angle, that is, they have compared land productivity levels between collective fields and private plots (Goetghebuer et al. 2014). In other words, they have put to 4

  More precisely, endogeneity would be present if sons are prone to leave the family farm when no individual plots are awarded by their father. The absence of individual plots would then appear, spuriously, to arise from small families and land abundance.

68   Concepts a quantitative test their assumption that effort is more efficiently applied to the latter than to the former. This is a tricky exercise since it requires that a variety of possible confounding factors (including variations in land quality, intensity of use of modern inputs, and crop choices) are well controlled for. When this is done, results prove to be in keeping with the theory: private plots are more productive than collective plots, and there is strong evidence that productivity differentials can be attributed to substantial variations in labor effort applied to cultivation. The second finding deserves attention because it provides indirect support for the incentive-based mechanism behind the theory: the productivity advantage of private plots exists for care-intensive crops yet not for care-saving crops. Because of the minor role of labor quality in the production process of care-saving crops, these crops are less or not vulnerable to the moral-hazard-in-team problem. There is a third central finding, and it confirms the key role of household composition highlighted in the previous empirical study. It turns out, indeed, that inefficiency on the collective fields is more serious when there are more married men in the working team. Finally, the study brings out an unconventional type of evidence that combines qualitative information with descriptive statistics. The idea is that something can be learned by relating the degree of strictness of time allocation rules laid down by the patriarch and the size or composition of the family workforce. The rules are ranked by decreasing order of severity: (i) male members are allowed to work only one or two day(s) a week on their individual plots; (ii) they may work before sunrise and after sunset, and sometimes also one day a week; and (iii) they may work five or six days a week, or whenever they want. The results bear out the hypothesis that rules are stricter when the number of participants in collective cultivation is higher and when there are more married men among them or the family workforce is more heterogeneous. The effect of heterogeneity is reflected in the finding that in horizontally extended families (with brothers of the head and nephews), the most severe time allocation rule is applied in a large majority of the cases (about 70%), while it is applied much less often in other types of families (in about 40% of the cases). Further confirmation is obtained if an index of family cohesion based on the Hamilton rule (the ratio of the Hamiltonian weighted sum of workers to the total number of workers) is used: the time allocation rule imposed by the head is significantly more severe when cohesion is lower.

3.4 Discussion A deep historical approach to family systems brings to light a huge variety of complex forms and a dynamic that resembles anything but a linear path of transformation (see, e.g., Stone 1979; Goody 1983; Todd 2011). In the light of these scholarly works, economic theories of farm-cum-family structures and their transformation appear to be based upon an utterly mechanistic view of human societies and an outrageous simplification of their complex realities. This said, it must also be reckoned that despite the appearance economic theories do not necessarily point to a simple, linear path of transformation. The Guirkinger–Platteau theory shows that outside opportunities play a role analogue to that of land endowment: individualization of farm-cum-family forms is predicted to

Transformation of African Farm-cum-Family Structures    69 unfold both as outside opportunities improve and land pressure increases. Therefore, if land becomes more scarce while outside opportunities deteriorate (say, due to economic crisis or the closure of migration outlets), the two forces will work in opposite directions and, if the second effect dominates, more collective forms may emerge. It must be added that the main merit of theory building is that it brings a rigorous structure on a problem, implying in particular that the precise assumptions on which it is based are clearly spelt out. In the case of the same theory, the influence of land scarcity on the individualization of farm-cum-family forms ceases to operate if land and labor markets function relatively well, or if fertility behavior is adapted relatively quickly to the evolving factor endowments. On the other hand, it is evident that a theory is misleading if it fails to grasp an important relationship or mechanism, and if it omits critical variables. Hence the necessity to put theories to serious empirical testing and to moreover check whether an alternative explanation might not account for the same outcome. One thing economists are not comfortable with is preference changes. Thus, one cannot rule out the possibility that individualization occurs as a result of the growing desire of younger generations to emancipate themselves from erstwhile patriarchal authority. It is revealing, in this respect, that patriarchs lament the growing consumption needs of the younger generation. In this eventuality, the transformative process is initiated by the pressure of household members rather than the decision of the head, which means that the former have somehow acquired strong bargaining power unconnected with their outside economic opportunities. This hypothesis is not to be taken lightly even though one may argue that outside opportunities is the main driver of the members’ ability to compel the head to take their viewpoint into account. It is not easily testable, though, owing to the difficulty to measure preference changes. One key issue, here, is that people may not reveal their true preferences in face-to-face interviews, especially because they may be reluctant to confess individualistic proclivities in a traditional environment permeated with collective norms and community- or family-centered values. Another problem arises from the fact that several processes—increasing land pressure, improving outside economic opportunities and associated risk diversification possibilities, and growing individualism of younger generations—may be at work simultaneously, thus making the task of disentangling them extremely arduous. It remains true, however, that in African cases of advanced individualization of farm-cum-family forms, such as in Rwanda, Burundi, Kenya (around Lake Victoria), Tanzania (Arusha area), and Malawi, where the small family farm is well established, land pressure seems to have played a major role. As a matter of fact, it is in regard of land endowments that other African areas which did not go as far on the way toward individualization mainly differ from the above countries or regions: population densities are markedly lower in less individualized areas.

References Allen, F. (1984). Mixed wage and rent contracts as reinterpretation of share contracts. Journal of Development Economics, 16(3):313–317. André, C., and Platteau, J.P., (1998). 1998. Land relations under unbearable stress: Rwanda caught in the Malthusian trap. Journal of Economic Behavior and Organization, 34 (1):1-47.

70   Concepts Basu, K. (1986). The market for land—an analysis of interim transactions. Journal of Development Economics, 20:163–177. Bauer, A. (1975). Chilean Rural Society from the Spanish Conquest to 1930. Cambridge: Cambridge University Press. Binswanger, H.P., Deininger, K., and Feder, G. (1995). Power, Distortions, Revolt, and Reform in Agricultural Land Relations, in T.P. Schultz, and J. Strauss (eds), Handbook of Development Economics, Elsevier and North-Holland, Vol. 3, Part B, Chap. 42:2659-2772. Binswanger, H., McIntire, J., and Udry, C. (1989). Production relations in semi-arid African agriculture, in Bardhan, P. (ed.), The Economic Theory of Agrarian Institutions. Oxford: Clarendon Press, pp. 122–144. Binswanger, H., and McIntire, J. (1987) Behavioural and material determinants of production relations in land-abundant tropical agriculture. Economic Development and Cultural Change, 36(1):73–99. Binswanger, H., and Rosenzweig, M. (1986). Behavioral and material determinants of production relations in agriculture. Journal of Development Studies, 22(3):503–539. Blum, J. (1957). The rise of serfdom in Eastern Europe. American Historical Review, 63:807–836. Blum, J. (1961). Lord and Peasant in Russia From the Ninth to the Nineteenth Century. Princeton: Princeton University Press. Boserup, E. (1965). Conditions of Agricultural Growth. Chicago: Aldine Publishing Co. Boucher S., Carter, M., and Guirkinger, C. (2008). Risk rationing and activity choice in moral hazard constrained credit market. American Journal of Agricultural Economics, 90(2):991-1004. Carter, M. (1987). Risk sharing and incentives in the decollectivization of agriculture. Oxford Economic Papers, 39:577–595. Chayanov, A. (1991). The Theory of Peasant Cooperatives. London: I.B Tauris & Co Ltd. de Janvry, A. (1981). The Agrarian Question and Reformism in Latin America. Baltimore: The Johns Hopkins University Press. Delpierre, M., Guirkinger, C., and Platteau, J-P. (2013). Risk as an Impediment to Individualization of Land Tenure? The Role of Collective Fields in Extended Agricultural Households. Working Paper, Center for Research in Development Economics (CRED), University of Namur, Belgium. Domar, E. (1970). The causes of slavery or serfdom: a hypothesis. Journal of Economic History, 30(1):18–32. Fafchamps, M. (2001). Intrahousehold access to land and sources of inefficiency: theory and concepts, in A. de Janvry, G. Gordillo, J.P. Platteau, and E. Sadoulet (eds), Access to Land, Rural Poverty and Public Action. Oxford: Oxford University Press. Foster, A., and Rosenzweig, M. (2002). Household division and rural economic growth. Review of Economic Studies, 69:839–869. Goetghebuer, T., Guirkinger, C., and Platteau, J.P. (2014). Productive inefficiency in extended agricultural households:  evidence from Mali. Working Paper, Centre for Research in Economic Development (CRED), University of Namur. Goody, J. (1983). The Development of the Family and Marriage in Europe. Cambridge: Cambridge University Press. Guirkinger, C., and Platteau, J.P. (2014a). The effect of land scarcity on farm structure: empirical evidence from Mali. Economic Development and Cultural Change, 62 (2):195-238. Guirkinger, C., and Platteau, J.P. (2014b). Transformation of the family under rising land pressure: a theoretical essay. Journal of Comparative Economics (forthcoming).

Transformation of African Farm-cum-Family Structures    71 Haugerud, A. (1993). The Culture of Politics in Modern Kenya. Cambridge:  Cambridge University Press. Hayami, Y., and Otsuka, K. (1993). The Economics of Contract Choice—An Agrarian Perspective. Oxford: Clarendon Press. Kolchin, P. (ed.) (1987). Unfree Labor—American Slavery and Russian Serfdom. Cambridge, MA: The Belknap Press of Harvard University Press. Mathur, P.N. (1991). Why Developing Countries Fail to Develop? London: Macmillan. Meyer, C.A. (1989). Agrarian reform in the Dominican Republic: an associative solution to the collective/individual dilemma. World Development, 17(8):1255–1267. Moon (1999). The Russian Peasantry, 1600-1930—The World the Peasants Made. London: Longman. Pingali, P., Bigot, Y., and Binswanger, H.P. (1987). Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa. Baltimore: The Johns Hopkins University Press. Platteau, J.P. (2000). Institutions, Social Norms, and Economic Development. London: Routledge. Putterman, L. (1989). Agricultural producer co-operatives, in Bardhan, P. (ed.), The Economic Theory of Agrarian Institutions. Oxford: Clarendon Press, pp. 319–339. Putterman, L. (1987). The incentive problem and the demise of team farming in China. Journal of Development Economics, 26:103–127. Putterman, L. (1985). Extrinsic versus intrinsic problems of agricultural cooperation: antiincentivism in Tanzania and China. Journal of Development Studies, 21:175–204. Putterman, L. (1983). A modified collective agriculture in rural growth-with-equity: reconsidering the private, unimodal solution. World Development, 11(2):77–100. Putterman, L., and Digiorgio, M. (1985). Choice and efficiency in a model of democratic semi­ collective agriculture. Oxford Economic Papers, 37:1–21. Sadoulet, E. (1992). Labor-service tenancy contracts in a Latin American context. American Economic Review, 82:1031–1042. Stone, L. (1979). The Family, Sex and Marriage in England 1500-1800. London: Penguin Books. Swain, N. (1985). Collective Farms Which Work? Cambridge: Cambridge University Press. Todd, E. (2011). L’origine des Systèmes Familiaux, Tome 1: l’Eurasie. Paris: Gallimard. Van Zanden, J. (2009). The Long Road to the Industrial Revolution—The European Economy in a Global Perspective, 1000-1800. Leiden: Brill. Worobec, C. (1995). Peasant Russia—Family and Community in the Post-Emancipation Period. Dekalb: Northern Illinois University Press.

Chapter 4

The Ec onomi c s of M arriag e i n North A fri c a  A Unifying Theoretical Framework

Ragui Assaad and Caroline Krafft

4.1 Introduction Marriage is the single most important transaction and transition in young North Africans’ lives. Marriage defines the basic economic and social unit—the household—and joins two families together. Marriage marks the transition to adulthood (Hoodfar 1997; Singerman and Ibrahim 2003). Even the words used to describe females pivot on marriage. Females are girls until they marry, and then women (Sadiqi 2003; Singerman 2007). Adult roles, including engaging in sex, childbearing, and independent living are essentially exclusively reserved for married individuals (El Feki 2013; Hoodfar 1997; Singerman and Ibrahim 2003). The quantity of resources invested in this vital transition often exceeds any other inter-generational transfer, including inheritances (Singerman and Ibrahim 2003). Who young people marry will shape their social and economic experience for decades to come, making matrimonial decisions extremely high stakes. Despite the fact that marriage is the single largest transaction and most important contract undertaken by young North African men and women, there has been very little rigor­ous research on the economics of marriage in North Africa, and what research exists has been fragmentary. This chapter reviews what is known about the economics of marriage in North Africa, focusing on Egypt, Morocco, and Tunisia. We compare and synthesize theoretical perspectives on the economics of marriage to generate a unified framework for future work. We also propose a research agenda for future work on the vitally important, but under-researched topics within the economics of marriage. In Assaad and Krafft (2015), we empirically test hypotheses generated from our theoretical framework and demonstrate the potential for meaningful empirical work, even with existing data, on this important topic.

The Economics of Marriage in North Africa    73

4.2  Economic Models of Marriage Although marriage has always been one of the most important contracts and opportunities for wealth transfer in an individual’s life, it was not traditionally the subject of economic analysis. Becker’s theory of marriage (1973, 1974) was the first to apply economic theory to the institution of marriage. This framework of a marriage market, in which utility-maximizing individuals make choices resulting in a market equilibrium, extended the basic concepts of neo-classical economics—rational choice and markets—to the institution of marriage (Becker 1973, 1974a; Grossbard-Shechlman 1995). The gains from marriage, based in part on complementary spousal labor, also encompass the quantity and quality of children resulting from the union. Alternatives in the marriage market, and in other markets (such as engaging in wage work instead of household labor) shape marriage market outcomes. Sorting, based on complementary or substitutable traits, plays an important role in maximizing marital output. The division of output in the resulting household is also linked to matching in the marriage market (Becker 1973). Becker’s theory additionally recognizes the inherent uncertainty in selecting a spouse and search costs (Becker 1974a). Becker’s work on the economics of marriage has been applied to issues such as polygamy, polygyny, divorce, fertility, labor force participation, and wages (Angrist 2002; Becker, Landes, and Michael 1977; Becker 1974a; Dougherty 2006; Grossbard 1976; Grossbard-Shechtman 1986; Light 2004). For instance, Becker et al. (1977) use an expected utility framework and uncertainty at marriage to explain marital dissolutions as the result of lower than expected utility in marriage. Grossbard-Shechtman (1986) examines how monogamy, compared to polygyny, creates a downward bias in the income elasticities of fertility. Dougherty (2006) tests whether the married male wage premium is the result of specialization of labor within marriage, as Becker had proposed. Angrist (2002) uses variation in immigration as a natural experiment to study the effect of sex ratios (the relative supply of potential spouses) on the marriage and labor markets. The framework originally proposed by Becker has been and continues to be applied to a wide variety of questions in the economics of marriage. An important alternative to Becker’s framework, in understanding the economics of marriage, is the game-theoretic approach. Game theory has often been applied to understanding issues of allocation within households and marriages (Lundberg and Pollak 1996, 2003; McElroy 1990; Udry 1996) in sharp contrast to models that treat the household as a single unit, perhaps with a benevolent head (Becker 1974b). An important feature of these models is their focus on the bargaining power of different parties and bargaining behaviors, which can illuminate both processes and outcomes in the marriage market. One implication of a game-theoretic approach is that outcomes are not necessarily efficient. Udry (1996) demonstrates inefficiency in the allocation of resources across plots of agricultural land controlled by different members of the household and Lundberg and Pollak (2003) illustrate this potential inefficiency for the case of couples’ location decisions. These models also extend into the search for a partner in the marriage market (Adachi 2003; Bergstrom and Bagnoli 1993; Smith 2006). Adachi (2003) proposes a model of the marriage market based on two-sided matching with search costs. Marriage (and equilibrium) occurs when a match surpasses an individual’s reservation utility. This model is

74   Concepts contingent upon the assumption that individuals meet each other according to some kind of random process: a matching framework that is unlikely to apply in contexts where women’s social circles are limited. Bergstrom and Bagnoli (1993) explain age differences between spouses in the context of a game of incomplete information. Because of the sexual division of labor, the qualities of females, such as the ability for childbearing, become apparent earlier, while those of males, such as earning power, only become apparent over time. Spouses match on expected quality, so males with good qualities wait until these qualities become apparent, and marry high-quality younger females. These additions to modeling the marriage market add important real-world features and recognize the imperfect information and game-theoretic nature of marriage markets. In this chapter, we draw on both Becker’s framework and game-theoretic approaches to understand the economics of marriage in North Africa. We draw on Becker’s understanding of how individuals’ traits affect their marriage outcomes, and draw on a game-theoretic understanding of how marriage contracts are negotiated.

4.3  The Institution of Marriage in North Africa In North Africa, marriage is clearly defined through practice, religion, and law (Hoodfar 1997). For instance, in Egypt marriages unfold in a series of stages (Sieverding 2012; Singerman 2007). First, there is the informal engagement (qira’at al-fatiha). This is followed by the formal engagement (khutuba). The signing of the marriage contract (katb al-kitab) is when legal marriage occurs. The actual wedding (dukhla) follows the legal marriage and is when the husband and wife cohabitate and begin their married life. The length between the different stages can vary, and more than one stage may occur at the same point in time. For instance, the legal marriage and actual wedding may take place at the same time (Singerman 2007). In North Africa, marriage marks the transition to adult roles, particularly for women. Regardless of her age, an unmarried woman is a girl (bint) and only upon marriage does she become a woman (sit) (Singerman 2007). Independent living, engaging in sex, and childbearing are only permissible within the context of marriage (El Feki 2013; Hoodfar 1997; Singerman and Ibrahim 2003). Cohabitation of men and women outside of a legally recognized marriage is extremely rare, and almost all births occur within marriage (El-Zanaty and Way 2009). Within marriage, gender roles are strictly defined. There is a clear sexual division of labor within marriages. Women are responsible for housework and childcare, while men are the breadwinners (Hoodfar 1997). The institution of marriage in North Africa has a number of important features that shape the economics of marriage. Marriage outcomes are determined through bargaining between two families rather than two individuals. Asymmetric rights within marriage, grounded in tradition and religion, favor men once the marriage has taken place (Hoodfar 1997). While engagements can be broken off by both sides, divorce, although uncommon, is easily initiated by men, but more difficult to obtain for women. If it occurs, divorce is much more harmful for women, both socially and economically (El Feki 2013; Hoodfar 1997). Moreover,

The Economics of Marriage in North Africa    75 with the exception of Tunisia, Muslim men are able, at least in theory, to take up to four wives, substantially reducing women’s bargaining power within marriage. Because of how marriages are structured, the bride side’s bargaining power is greatest up front, before the couple is married. Contracts detailing marriage conditions are negotiated up front. The families of the bride and groom, and particularly the parents of the bride, play a large role in the negotiations (Hoodfar 1997). Marriage outcomes, in detail down to the level of kitchen utensils, are agreed to in conjunction with the marriage contract (Amin and Al-Bassusi 2004). Not only are material living conditions negotiated, but so are many financial and behavioral outcomes. Families can negotiate up front over issues such as whether the couple will have one meal or two a week that contains meat, as well as issues such as where the couple will live, whether the bride will work, and whether the bride should use contraception in the first year of marriage or have a child and then use contraception (Hoodfar 1997). Thus, in North Africa, the marriage contract is of profound importance to the economic and social arrangements of young people. In line with their socially recognized role within marriage as the main breadwinners, grooms and their families are also primarily responsible for the costs of marriage. Although the bride’s side contributes specific components to the costs of marriage, the groom and his family bear the primary financial burden (Hoodfar 1997; Singerman 2007). There are a number of different components to marriage costs in North Africa. In Egypt, these include the bride price (mahr), the jewelry gift given to the bride (shabka), housing, furniture and appliances, the trousseau and home furnishings (gihaz), and the wedding celebration itself. In negotiating their marriage contracts, young people face a variety of trade-offs. For instance, it can take years for potential grooms to accumulate the resources and goods necessary for marriage, and brides therefore have to decide whether to marry earlier, but before all these resources are acquired, or wait until their material conditions were secured to marry (Hoodfar 1997). Alternatively, similar to the situation analyzed by Bergstrom and Bagnoli (1993), young women may decide (or be pressured) to marry older men who have already accumulated the necessary resources, rather than wait for an otherwise more desirable younger man to do so.

4.4  Economic Analyses of Marriage in North Africa Despite the profound importance of the institution of marriage in social and economic arrangements, there is limited and fragmentary literature on the economics of marriage in North Africa. A few topics receive particular attention in the literature. Economic demography has a robust but primarily descriptive literature, focusing on trends such as the age at marriage, which has been increasing, and prevalence of marriage, which has been relatively universal in the region, especially in contrast to areas such as South Africa (Eltigani 2000; Mensch 2005; Nosseir 2003; Rashad, Osman, and Roudi-Fahimi 2005; Salem, forthcoming). While the median age at marriage has been rising, this phenomenon has been met with mixed feelings, as delays in marriage also delay adult roles, and create a period of ‘wait adulthood’ or “waithood” (Dhillon, Dyer, and Yousef 2009; Singerman 2007). Although there has

76   Concepts been a popular portrayal of marriage as a declining and increasingly expensive institution in the region (El Feki 2013; Salem, forthcoming), there is limited empirical support for such claims. The costs of marriage have also received some, primarily descriptive, attention in the literature (Salem, forthcoming, 2011; Singerman and Ibrahim 2003; Singerman 2007). For instance, Singerman (2007) finds that real costs have been decreasing over time and have been relatively flat in the decade preceding 2006; however, we question the accuracy of the cost recall method used to identify this trend. Singerman (2007) also confirms empirically that the burden of marriage costs falls primarily on the groom’s side, which bears nearly 70% of costs. The same study breaks down cost shares by various socioeconomic characteristics, and in an ordinary least squares regression for costs, finds decreasing costs with increasing education of the bride, but increasing costs with the education of the groom. This suggests that bride education may not be considered an asset in the marriage market and may not increase her family’s bargaining power when negotiating the marriage contract. The household structure and place of residence of newly married couples have also been examined (Amin and Al-Bassusi 2004; Elbadawy 2007; Salem, forthcoming; Singerman 2007). While traditionally married couples had lived with the husbands’ extended family, increasingly nuclear households are becoming the norm in North Africa. Modernization trends, such as urbanization and education have been credited with driving this pattern (Nosseir 2003). The trend towards nuclear residence has been linked to increased costs of marriage and delayed marriage (Amin and Al-Bassusi 2004; Salem, forthcoming; Singerman 2007), as the costs of setting up an independent household are substantial. Finally, the high consanguinity levels in North Africa have received some attention in both the economic demography and health literatures (Ben Halim et al. 2013; Casterline and El-Zeini 2003; Elbadawy 2007; Mensch 2005; Mokhtar and Abdel-Fattah 2001). Consanguinity, that is marriage between individuals who share a (known) common ancestor, is a common practice in Arab countries (Casterline and El-Zeini 2003; Mensch 2005; Rashad et al. 2005). Marriage between first cousins is the most common form of consanguinity practiced in the region (Casterline and El-Zeini 2003). A variety of explanations have been proposed for consanguineous marriages, including the economic rationales that it will be lower cost (Casterline and El-Zeini 2003; Hoodfar 1997; Singerman 2007), or helps maintain family property, but these arguments do not have strong empirical support (Casterline and El-Zeini 2003). There are also a number of other rationales, including that there is less uncertainty about spouse qualities (since information issues are substantial in the marriage market), that the husband and wife and their families will be more compatible, that wives will be treated better, and that there will be less marital conflict and greater marital stability (Casterline and El-Zeini 2003; Hoodfar 1997; Sholkamy 2003). Consanguinity has been associated with traditional and arranged marriages, and modernization theorists expected substantial declines in consanguinity over time, declines which have not materialized in Egypt (Casterline and El-Zeini 2003; Singerman 2007). This may be because, although the debate is contentious, consanguinity is on the whole beneficial for women (Casterline and El-Zeini 2003). While young men and their families consider consanguineous marriages appealing because of lower costs, women consider this a disadvantage. Kin marriages essentially reduce the uncertainty around a spouse’s characteristics. Additionally, they help protect women against domestic violence (Hoodfar 1997).

The Economics of Marriage in North Africa    77 The anthropological and sociological literature (Amin and Al-Bassusi 2004; El Feki 2013; Hoodfar 1997) on the institution of marriage tends to be much more thorough than the economics literature. The limited economics of marriage literature for North Africa is primarily descriptive in nature. Only a handful of papers attempt to estimate the determinants of various marriage outcomes in a multivariate framework. Several focus on the transition to and timing of marriage. Assaad, Binzel, and Gadallah (2010) examine how the transition from school to work affects the transition to marriage for young men in Egypt. They demonstrate the importance of young men finding a “good” job in order to get married. Assaad and Ramadan (2008) examine the role of housing policy reforms in curbing the delays in marriage for young men, also in Egypt. Assaad and Zouari (2003) examine how the timing of marriage and fertility affect the level and type of female labor force participation in urban Morocco. They find that marriage and childbearing decrease women’s participation in work, particularly private wage work. A few papers examine marriage outcomes such as consanguinity, costs, and bargaining power within marriage. Casterline and El-Zeini (2003) perform simulations of the effect of reductions in family size on consanguinity, and find that reduced family size is unlikely to substantially reduce consanguinity. Elbadawy (2007) examines the returns to education for females in the marriage market in Egypt, in terms of its association with spousal education, nuclear residence, consanguinity, and marriage costs. Highly educated women are found to have improved chances of marrying husbands with higher education. Educated women are more likely to live independently upon marriage, and to marry outside their families. Sieverding (2012) examines how wage work affects young Egyptian women’s marriage outcomes, and finds that young women who work contribute more to the costs of marriage and have higher total costs, but do not marry more rapidly. Salem (2011) examines how women’s marriage assets and wage work affect their bargaining power in Egypt. The findings demonstrate that, for women, greater wages before marriage lead to greater contributions to marriage costs, which in turn increases bargaining power within marriage. Given the limited body of research, there is very little that can be said in terms of consistent findings or controversies in the literature. There is a clear need for substantially more research on the determinants of marriage outcomes, as well as the economics of marriage generally in North Africa.

4.5  A Unifying Framework In order to enhance the state of research on the economics of marriage in North Africa, and to set a framework and agenda for future work, we offer a unifying framework for considering the economics of marriage. We propose that marriage outcomes be considered the result of a bargaining process between families. This is consistent with the findings of the anthropological literature (Hoodfar 1997) and encompasses the descriptions and findings of much of the economics-oriented literature as well (Elbadawy 2007; Salem 2011; Sieverding 2012). While other perspectives such as “modernization” hypotheses can and have been considered, the empirical evidence provided elsewhere (Assaad and Krafft 2015; Salem 2011) is not consistent with modernization but is consistent with a bargaining framework.

78   Concepts As they search for and contract with spouses, young people’s traits (and those of their families) determine a number of different marriage outcomes that will shape their adult lives. We specifically consider age at marriage, consanguinity, nuclear residence, total costs of marriage, bride’s side share of costs, and the age difference between the bride and groom as outcomes of the matching and bargaining process. A potential wife seeks a smaller age difference for a more equitable marriage, optimal timing of marriage, a nuclear household, high costs to ensure a higher standard of living, more choice of groom (less cousin marriage), and a lower bride’s side share. Total costs in particular are likely to be a sign of bride side bargaining power. A potential husband may in fact desire a larger age difference, optimal timing of marriage for himself, may want a nuclear household but be deterred by the costs, want moderate costs (as his side pays a higher share), want more choice of bride (less cousin marriage), and a higher bride’s side share. Ultimately, these outcomes are affected by bargaining power, ability to pay, and the bride’s and groom’s characteristics. There are potential trade-offs among different outcomes. For instance, accumulating the savings necessary to form a nuclear household may cause a delay in age at marriage. The prevalence and timing of marriage particularly depends on both male and female side bargaining power and the characteristics of individuals and their families, especially the ability to pay. Employment and housing options are expected to be particularly important for men’s marriage prospects. Certain characteristics—such as a more educated partner—are more desirable. Ability to pay and an individual’s characteristics may interact to determine their side’s share of marriage costs and marriage outcomes. When one side has a low ability to pay, and bad characteristics, they are likely to contract a marriage with a low-cost share but bad outcomes, such as non-optimal timing of marriage and a non-nuclear household. If one side has a low ability to pay but good characteristics, they will continue to have a low share, but may delay marriage in order to negotiate for potentially better outcomes. If one side has a high ability to pay, but bad characteristics (for instance, relatively low education for their wealth level), they will pay a higher share of costs and will experience mixed outcomes. Households with a high ability to pay and good characteristics can generally expect good outcomes, but are also likely to pay a high share of costs to obtain such outcomes. Table 4.1 delineates the different hypotheses we would expect in terms of how women’s characteristics affect bargaining power and ultimately marriage outcomes, all else being equal. We present empirical tests of some of these hypotheses using data from Egypt, Morocco, and Tunisia in Assaad and Krafft (2015). Better own education is expected to delay age at marriage because it increases a woman’s expectations on marriage outcomes. Own education also increases nuclear residence, total costs, and decreases the age difference and chances of a consanguineous marriage. Parents’ education and father’s employment status are expected to lead to a more optimal age at marriage, a higher bride’s share and similar effects as own education on consanguinity, nuclear residence, total costs, and age difference. We expect parental wealth to behave similarly, but in particular to increase optimal timing of marriage. We expect there to be a number of cohort effects associated with modernization, including a delay in age at marriage, increased nuclear residence, total costs, and bride share, and a decrease in the age difference and chances of a consanguineous marriage. The cost of housing has been identified as a substantial barrier to marriage in the region (Assaad et al. 2010; Assaad and Ramadan 2008; Dhillon et al. 2009; Singerman 2007), and we expect

The Economics of Marriage in North Africa    79 Table 4.1  Hypotheses on the relationship between females’ characteristics and marriage outcomes and correlations between marriage outcomes Female char.

Age at marriage

Consanguinity

Nuclear residence

Total costs

Bride share

Age difference

Own education Cohort Parents’ education Father’s employment Parental wealth Rental housing

+ + + then – + then – + then – –

– – – – – –

+ + + + + +

+ + + + + –

? + + + + ?

– – – – – ?

+ –

+ then – – +

+ ? + +

– – ? ? –

Expected correlations between marriage outcomes Age at marriage Consanguinity Nuclear Total costs Bride share

–

increases in the supply of rental housing to decrease the age at marriage and total costs, increase nuclear residence, and decrease consanguineous marriages. We expect substantial interactions between different marriage outcomes, as they are the result of a complex negotiation process with numerous trade-offs (Table 4.1). We expect a later age at marriage to be related to a lower chance of consanguineous marriage and a higher chance of nuclear living arrangements. Later age at marriage is expected to increase and then decrease total costs on the theory that optimal age at marriage maximizes women’s bargaining power and her family’s ability to negotiate a larger contract. Older age at marriage is also expected to increase the bride share, but decrease the age difference. Consanguineous marriages are expected to be negatively associated with nuclear living arrangements, have lower costs, and less of an age difference. Nuclear household living arrangements at marriage will increase costs and the bride share. Total costs and bride share will be positively correlated, and the bride share will be negatively correlated to the age difference. Additionally, we expect that Christian women have greater bargaining power than Muslim women, since a Christian marriage contract is more symmetric with neither side having the right to repudiate the other and the husband has no right to take additional wives. We also expect that countries’ demographic structures have a substantial impact on the timing of marriage. North African countries tend to have experienced a substantial youth bulge, and commonly have a large age gap between husbands and wives. Together, these factors will affect the prevalence and timing of marriage in North Africa. The demographic transition is also expected to affect the child quantity/quality trade-off. As the demographic transition occurs, specifically as early mortality declines, parents can be more certain and secure that investments in their children will carry through to returns from adults. They therefore desire higher child quality. One way men can get higher child quality is by investing in a higher quality wife, and therefore we expect that the premium to female education will increase (Schultz 2008). Child quantity is primarily ensured by a younger wife, and the

80   Concepts de-emphasis of child quantity may be one of the factors contributing to rising age of marriage among women. There are a variety of specific characteristics, outcomes, and trade-offs in Table 4.1, and the model can easily be extended to consider additional outcomes and characteristics by assessing how characteristics are likely to affect bargaining power, and how bargaining power in turn affects the outcome. For instance, sex ratios have been theorized to have an impact on the marriage market (Angrist 2002). Presumably, a greater supply of females relative to males would reduce females’ bargaining power, and in turn have negative impacts on females’ marriage outcomes. Education matches could also be considered, as they have been elsewhere (Elbadawy 2007; Sieverding 2012). In order to obtain a relatively more educated husband, the bride side might have to pay a greater share of costs or accept extended household living.

4.6  Implications for Local and Global Studies of Marriage Markets Our unifying theoretical framework is driven by a recognition of the asymmetric rights of men and women within marriage in North Africa, and the strong incentives women and their families have to bargain for increased resources up front. A number of features of North African marriage markets are unique to their context. Yet the overarching framework of a marriage market, utility-maximizing individuals, potential uncertainty and search costs, and bargaining driven by individuals’ characteristics and ability to pay has wider implications for the economics of marriage. The clearly defined customary, religious, and legal structure of marriage, with specific stages that define marriage, the detailed marriage contract, and asymmetric rights within marriage, is relatively unique to North Africa, specifically, and the rest of the Islamic World more generally. The North African framework can easily be extended to a number of other Arab and Islamic countries, which share common cultural, legal, and religious practices around marriage (El Feki 2013), including the same asymmetric rights and upfront bargaining. Similar economic consequences are felt in other Arab countries. For instance, young people in the West Bank, Jordan, and Saudi Arabia face similar struggles to accumulate the resources necessary for a wedding (Salem 2014; Singerman 2007). Where different rights and customs around marriage prevail, even within Africa, the basis of our framework—the clearly defined asymmetrical rights within marriage and strong incentives for upfront bargaining—is absent. However, the practice of a bride price for marriages, a global practice that is particularly common in sub-Saharan Africa (Anderson 2007), could be examined from a similar perspective as the result of a bargaining process between two families. The economic struggles of young people in sub-Saharan Africa, delaying or forgoing marriage as they work to gather brideprice and obtain housing (Shapiro and Gebreselassie 2013), are akin to the challenges young people face in North Africa. The overarching framework of the economics of marriage has a common global element. Young people seeking spouses generally face a marriage market, potential uncertainty and search costs, and will bargain over their partner and marriage outcomes. The North African context, with clearly defined institutions around marriage and extremely detailed marriage

The Economics of Marriage in North Africa    81 contracts, provides an important opportunity to explicitly study the economics of marriage. Understanding how North Africans’ characteristics and bargaining power shape their marriage outcomes in this well-defined context can provide insights into the economics of marriage for global settings with less clearly defined marriage processes and outcomes.

4.7  Conclusions and Directions for Future Research 4.7.1  Previous literature The body of literature on the economics of marriage in North Africa is quite fragmentary, and primarily descriptive, focusing on patterns such as trends in the prevalence of marriage and age at marriage (Eltigani 2000; Mensch 2005; Nosseir 2003; Rashad et al. 2005; Salem, forthcoming), costs (Salem, forthcoming, 2011; Singerman and Ibrahim 2003; Singerman 2007), nuclear residence (Elbadawy 2007; Nosseir 2003; Salem, forthcoming; Singerman 2007), and consanguinity (Ben Halim et al. 2013; Casterline and El-Zeini 2003; Elbadawy 2007; Mensch 2005; Mokhtar and Abdel-Fattah 2001). The anthropological literature has actually tended to take the most holistic approach to the economics of marriage (Amin and Al-Bassusi 2004; El Feki 2013; Hoodfar 1997). Only a few forays have been made into multivariate analyses of topics such as the timing of marriage (Assaad et al. 2010; Assaad and Ramadan 2008; Assaad and Zouari 2003), consanguinity (Casterline and El-Zeini 2003; Elbadawy 2007), costs (Singerman 2007), and the relationship between women’s work and marriage (Salem 2011; Sieverding 2012; Assaad and Zouari 2003). Marriage has not traditionally been the subject of economic analysis, and globally the framework of a marriage market, utility-maximizing individuals, potential uncertainty and search costs, and a resulting market equilibrium is relatively recent (Becker 1973, 1974a; Grossbard-Shechlman 1995). Even more recent are applications of game theory to the economics of marriage (Lundberg and Pollak 1996, 2003; McElroy 1990; Udry 1996) and the search for a partner (Adachi 2003; Bergstrom and Bagnoli 1993; Smith 2006), which can help explain both the processes and outcomes of the marriage market.

4.7.2  This chapter’s contributions In this chapter we drew on both the classical, Becker approach to the economics of marriage and game-theoretic approaches to offer a unifying framework for understanding the economics of marriage in North Africa. Owing to asymmetric rights that favor men once marriage has taken place, the bride’s side bargaining power is greatest up front, and so detailed marriage contracts are negotiated to secure potential benefits and agree on trade-offs. We offer a unifying framework for understanding marriage outcomes, recognizing that ability to pay and individuals’ characteristics create bargaining power and shape outcomes, and that there are trade-offs between different outcomes. Given the nascent state of the literature, many important questions on the relationship between characteristics, ability to pay,

82   Concepts bargaining power, and marriage outcomes are unanswered, as are questions about the relationships and trade-offs between marriage outcomes, such as consanguinity and costs. We undertake a preliminary empirical investigation of some of these issues in Assaad and Krafft (2015).

4.7.3  An agenda for future research One of the most important elements for progressing research on the economics of marriage in North Africa (and elsewhere) is improving the quantity and quality of information available on marriage outcomes and processes. Currently, the Egypt Labor Market Panel Survey1 (ELMPS) is the only survey in the region with a detailed module on marriage, although in the Middle East Jordan has comparable data, and a planned labor market survey in Tunisia similar to the ELMPS will include a module on marriage, allowing for future comparative work. Similar survey modules should be incorporated into other surveys in the region. Since marriage is a substantial expenditure, often the most substantial intergenerational transfer in young people’s lives (Singerman and Ibrahim 2003), this topic ought to be incorporated into household income and expenditure surveys, which could collect data on savings for marriage and cost components for recent marriages. Questions on a number of topics, such as consanguinity and living arrangements at marriage, could be incorporated into surveys focusing on demographic or health issues. Surveys specifically on marriage—including the search for a spouse for those not yet married—would open up substantial new areas of research into search and bargaining behaviors, as well as the economics of the marriage market, including issues such as information problems and uncertainty. Although new data will be an important component to future research on the economics of marriage in North Africa, there are numerous areas that merit future research, and for which there is at least some data currently available. What determines age differences between spouses, and how this affects bargaining power within the household merits further investigation. The relationship between employment and marriage for men and women is a vital topic for understanding transitions to adulthood. The evidence to date indicates men need employment to secure a successful marriage, and that “good” employment speeds marriage for women (Assaad and Krafft 2015), but it is also clear that married life and employment are difficult to reconcile for women (Assaad and El-Hamidi 2009; Assaad and Zouari 2003; Hoodfar 1997). Particularly for men, the role of migration in enabling or delaying marriage merits further research. How the assets individuals bring to marriage and their bargaining positions and processes in negotiating the marriage contract affect bargaining power and gender roles within the marriage has important implications, particularly for the well-being of women and children. Marriage represents the most important transition and transaction in young people’s lives and shapes their economic and social trajectory. Previously, the literature lacked a unifying theoretical framework to understand the economics of marriage in North Africa, where 1  This survey, fielded in 1998, 2006, and 2012, with a detailed module on marriage in the 2006 and 2012 waves, is publicly available at http://www.erfdataportal.com. See Assaad and Krafft (2013) for additional information on the survey.

The Economics of Marriage in North Africa    83 asymmetric rights in marriage have led to substantial upfront bargaining. This chapter has provided a new theoretical framework that will hopefully spur future research. To date, there has been only little, primarily descriptive, empirical research on the economics of marriage. While we begin to address the shortage of empirical research in Assaad and Krafft (2015), much more needs to be done to understand the economics of marriage in North Africa.

References Adachi, H. (2003). A  search model of two-sided matching under nontransferable utility. Journal of Economic Theory, 113(2):182–198. Amin, S., and Al-Bassusi, N.H. (2004). Education, wage work, and marriage: Perspectives of Egyptian working women. Journal of Marriage and Family, 66(5):1287–1299. Anderson, S. (2007). The economics of dowry and brideprice. The Journal of Economic Perspectives, 21(4):151–174. Angrist, J. (2002). How do sex ratios affect marriage and labor markets? Evidence from America’s second generation. The Quarterly Journal of Economics, 117(3):997–1038. Assaad, R., Binzel, C., and Gadallah, M. (2010). Transitions to employment and marriage among young men in Egypt. Middle East Development Journal, 2(1):39–88. Assaad, R., and El-Hamidi, F. (2009). Women in the Egyptian labor market: an analysis of developments, 1988-2006, in R. Assaad (ed.), The Egyptian Labor Market Revisited. Cairo, Egypt: The American University in Cairo Press, pp. 219–258. Assaad, R., and Krafft, C. (2013). The Egypt labor market panel survey: introducing the 2012 round. IZA Journal of Labor and Development, 2(8):1–30. Assaad, R., and Krafft, C. (2015). An empirical analysis of the economics of marriage in Egypt, Morocco, and Tunisia, in C. Monga and J.Y. Lin (eds) The Oxford Handbook of Africa and Economics, Vol. 2. Oxford: Oxford University Press. Assaad, R., and Ramadan, M. (2008). Did housing policy reforms curb the delay in marriage among young men in Egypt? Middle East Youth Initiative Policy Outlook. Washington, DC. http://www.shababinclusion.org/content/document/detail/1226/ [accessed July 14, 2014]. Assaad, R., and Zouari, S. (2003). Estimating the impact of marriage and fertility on female labor force participation when decisions are interrelated: evidence from urban Morocco. Topics in Middle Eastern and North African Economies, 5:1–37. Becker, G.S. (1973). A theory of marriage: part I. Journal of Political Economy, 81(4):813–846. Becker, G.S. (1974a). A theory of marriage: part II. Journal of Political Economy, 82(2):S11–S26. Becker, G.S. (1974b). A  theory of social interactions. Journal of Political Economy, 82(6):1063–1093. Becker, G.S., Landes, E., and Michael, R. (1977). An economic analysis of marital instability. Journal of Political Economy, 85(6):1141–1187. Ben Halim, N., Ben Alaya Bouafif, N. et al. (2013). Consanguinity, endogamy, and genetic disorders in Tunisia. Journal of Community Genetics, 4:273–284. Bergstrom, T., and Bagnoli, M. (1993). Courtship as a waiting game. Journal of Political Economy, 101(1):185–202. Casterline, J.B., and El-Zeini, L.O. (2003). Consanguinity in the Arab Region: current patterns and prospects for change, in Institutions, Ideologies, and Agency: Family Change in the Arab Middle East and Diaspora. Chapel Hill: University of North Carolina.

84   Concepts Dhillon, N., Dyer, P., and Yousef, T. (2009). Generation in waiting: an overview of school to work and family formation transitions, in Generation in Waiting: The Unfulfilled Promise of Young People in the Middle East. Washington, DC: The Brookings Institution, pp. 11–38. Dougherty, C. (2006). The marriage earning premium as a distributed fixed effect. Journal of Human Resources, 41(2):433–443. El Feki, S. (2013). Sex and the Citadel:  Intimate Life in a Changing Arab World. New York: Pantheon. El-Zanaty, F., and Way, A. (2009). Egypt Demographic and Health Survey 2008. Cairo, Egypt: Ministry of Health, El-Zanaty and Associates, and Macro International. Elbadawy, A. (2007). Education Returns in the Marriage Market:  Does Female Education Investment Improve the Quality of Future Husbands in Egypt ? Cairo, Egypt:  Population Council. Mimeo. Eltigani, E.E. (2000). Changes in family-building patterns in Egypt and Morocco: a comparative analysis. International Family Planning Perspectives, 26(2):73–78. Grossbard, A. (1976). An economic analysis of polygyny:  the case of Maiduguri. Current Anthropology, 17(4):701–707. Grossbard-Shechlman, S. (1995). Marriage Market Models, in M. Tommasi and K. Ierulli (eds), The New Economics of Human Behavior. Cambridge: Cambridge University Press, pp. 92–111. Grossbard-Shechtman, A. (1986). Economic behavior, marriage, and fertility. Journal of Economic Behavior and Organization, 7:415–424. Hoodfar, H. (1997). Between Marriage and the Market: Intimate Politics and Survival in Cairo. Berkeley, CA: University of California Press. Light, A. (2004). Gender differences in the marriage and cohabitation income premium. Demography, 41(2):263–284. Lundberg, S., and Pollak, R.A. (1996). Bargaining and distribution in marriage. The Journal of Economic Perspectives, 10(4):139–158. Lundberg, S., and Pollak, R.A. (2003). Efficiency in marriage. Review of Economics of the Household, 1:153–167. McElroy, M. (1990). The empirical content of Nash-bargained household behavior. Journal of Human Resources, 25(4):559–583. Mensch, B.S. (2005). The transition to marriage, in C.B. Lloyd (ed.), Growing up Global: The Changing Transitions to Adulthood in Developing Countries. Washington, DC: The National Academies Press, pp. 416–505. Mokhtar, M.M., and Abdel-Fattah, M.M. (2001). Consanguinity and advanced maternal age as risk factors for reproductive losses in Alexandria, Egypt. European Journal of Epidemiology, 17(6):559–565. Nosseir, N. (2003). Family in the new millenium:  major trends affecting families in North Africa, in Families in the Process of Development. New York: United Nations Publications, pp. 188–199. Rashad, H., Osman, M., and Roudi-Fahimi, F. (2005). Marriage In The Arab World. Washington, DC: Population Reference Bureau. Sadiqi, F. (2003). Women, Gender, and Language in Morocco. Leiden, The Netherlands: Koninklijke Brill NV. Salem, R. (2011). Women’s economic resources and bargaining in marriage: does Egyptian women’s status depend on earnings or marriage payments? Gender and Work in the MENA

The Economics of Marriage in North Africa    85 Region Working Paper Series: Poverty, Job Quality, and Labor Market Dynamics. Population Council: Cairo, Egypt. Salem, R. (2014). Trends and differentials in Jordanian marriage behavior: timing, spousal characteristics, household structure and matrimonial expenditures, in R. Assaad (ed.), The Jordanian Labour Market in the New Millennium. Oxford: Oxford University Press, pp. 189–217. Salem, R. (forthcoming). Imagined crises: assessing evidence of delayed marriage and never-marriage in contemporary Egypt, in K. Celello and H. Kholoussy (eds), Domestic Tensions, National Anxieties: Global Perspectives on Marriage Crisis. Oxford: Oxford University Press. Schultz, T.P. (2008). Population policies, fertility, women’s human capital, and child quality. Handbook of Development Economics, 4(7):3249–3303. Shapiro, D., and Gebreselassie, T. (2013). Marriage in sub-Saharan Africa:  trends, determinants, and consequences. Population Research and Policy Review, 33:229–255. Sholkamy, H. (2003). Rationales for kin marriages in rural Upper Egypt. Cairo Papers in Social Science, 24(1–2):62–79. Sieverding, M. (2012). Gender and Generational Change in Egypt. University of California, Berkeley. Singerman, D. (2007). The Economic Imperatives of Marriage: Emerging Practices and Identities Among Youth in the Middle East. Middle East Youth Initiative Working Paper. http://www.shababinclusion.org/content/document/detail/559/. Singerman, D., and Ibrahim, B. (2003). The costs of marriage in Egypt: a hidden variable in the new Arab demography, in N.S. Hopkins (ed.), Cairo Papers in Social Science, volume 24 numbers 1/2: The New Arab Family. Cairo, Egypt: American University in Cairo Press, pp. 80–166. Smith, L. (2006). The Marriage Model with Search Frictions. Journal of Political Economy, 114(6):1124–1144. Udry, C. (1996). Gender, agricultural production, and the theory of the household. Journal of Political Economy, 104(5):1010–1046.

Chapter 5

The Theory of t h e Fi rm in the African C ont e xt Christopher Malikane

5.1 Introduction African firms often find themselves operating in extremely challenging environments. Working in economies that have little or no intermediate input sectors, African firms find themselves confronted with high levels of volatility of costs. In addition, the African environment is often characterized by uncertain supplies of basic public goods and services such as water, electricity, and the rule of law. As pointed out by Bigsten and Söderbom (2006), African firms have to contend with volatile prices stemming from unstable exchange rates, volatile cash-flows due to fluctuations in demand, irregular customer payment, unreliability of infrastructure, corruption, and so on. Such uncertainty raises the risk premium that is attached to the activities of African firms and increases the threshold rate of return that is necessary to trigger enterprise investment. These challenges are partially summarized by Biggs and Shah (2006), who state that among African firms, “technical and management skills are low on average, and absenteeism and acts of employee pilfering are numerous. Product standardization is relatively low. Businesses operate under conditions of considerable uncertainty. Financial and insurance markets are severely underdeveloped, limiting access to credit and insurance. And market exchange is underpinned by weak public institutions of property rights and contract, poor governance, and poor infrastructure services.” In addition to these challenges, African firms operate in environments that have weak financial markets and very limited long-term borrowing and lending (Gwatidzo and Ojah 2009, 2014). This limited access to credit, while operating differentially between small and large firms, and may also be sensitive to firm location and the industry in which the firm operates, affects the ability of firms to expand supply capacity. With limited supply capacity, African firms are largely constrained from taking advantage of favorable market developments, such as the opening up of new export markets or favorable movements of the exchange rate. This chapter discusses the theory of the firm in the African context. There is now a significant amount of microeconometric research, spawned by the World Bank’s initiative to

The Theory of the Firm in the African Context    87 collect extensive firm-level data, which studies the behavior of firms, particularly in the African context. The novelty of these studies is that they use rich firm-level data sets in order to answer key questions about firm behavior and performance. Aggregate data tend to hide the heterogeneity among firms and may provide a misleading picture about developments at the point of production, in response to macro- and microeconomic shocks. Given the vast microeconometric literature on the behavior of African firms that has accumulated, it is useful to construct a unified theoretical framework that sheds some light on the seemingly diverse findings of the empirical literature. This chapter provides a review of some of the empirical findings and on that basis, formulates a basic microeconomic theory that can be used to provide a coherent understanding of firm behavior in the African context. The main proposition put forward here is that a significant number of African firms are trapped in a vicious cycle of stagnation, or a low-level equilibrium. Poor access to indirect inputs such as electricity raises costs and inhibits firm competitiveness and output expansion. With limited output expansion, firms are unable to fully exploit economies of scale and to raise their total factor productivity as the basis to launch into the export market. With limited access to export markets, African firms are also unable to hedge against currency fluctuations, which create volatility primarily on the cost side. The combination of these factors generates poor performance, which makes it difficult for these firms to access credit. Simultaneously, this poor performance also makes it difficult for African firms to extend trade credit as a means to extend domestic demand. The general situation is therefore one where African firms produce lower levels of output than they would like, which prevents them from achieving high levels of efficiency. This is partly because of poor supply of indirect inputs and partly because of insufficient demand. However, demand is insufficient partly because of the inability of African firms to extend and access trade credit. Access to credit is, in turn, to a large extent determined by the level of performance of firms. Firms that attempt to break this cycle through export markets appear to be large and exploit economies of scale to achieve high levels of efficiency. Firms that attempt to break out of this cycle by expanding domestic demand use trade credit as an instrument (see, for example, Ramachandran and Shah 1999). However, this is often limited to close-knit social networks that are largely based on ethnicity. The remainder of this chapter is structured as follows: Section 2 provides a brief review of the literature on the constraints faced by African firms. Section 3 builds a theoretical framework that incorporates the insights that are derived from empirical literature on the behavior and performance of African firms. Section 4 offers policy recommendations and section 5 some concluding thoughts.

5.2  Review of the Literature on Constraints Faced by African Firms The constraints faced by African firms are to a large extent similar to those faced by firms that operate in a developing country context, generally. However, the impact of those constraints on African firms seems to be different. This chapter focuses on the following constraints: (i) access to credit; (ii) limited domestic goods markets; (iii) access to indirect inputs such as

88   Concepts electricity, water, and transport infrastructure; and (iv) weak institutional environment, for example, corruption, poor legal environment, and skills constraints. One of the most researched problems faced by African firms is access to credit. Malik et al. (2006) found that, in the case of Nigerian manufacturing firms, for example, the credit constraint is cited as the second most important constraint perceived by firms; 47 percent of the firms placed access to credit among their top three problems. Van Biesebroeck (2005) found that access to formal credit is the largest problem faced by firms that employ fewer than 50 people. However, he also reports that this problem afflicts firms that do not export far more than those of similar size, but which participate in exports. Therefore, even controlling for size, firms that have limited export-orientation tend to be more credit-constrained than firms of similar size with higher levels of export orientation. One possible explanation is that exporters may be earning foreign exchange, which is then used to hedge against the rising costs brought about by currency depreciation. There is also strong evidence that access to credit depends on firm size. Bigsten et al. (2003) found that of the firms that have applied for a loan, about 25 percent received a formal loan. They also found that small firms are less likely to get a formal loan than large firms. Bigsten and Söderbom (2006) summarized the findings on credit access by noting that two-thirds of small firms appear to be credit constrained while only 10 percent of large firms are credit constrained. However, it should be noted that these findings may understate the problem, since some firms may choose not to apply for a loan because they anticipate their application to be unsuccessful. Access to bank credit is important, particularly in the context where intermediate inputs are imported. Because they are so limited to access credit, African firms miss out on opportunities to raise productivity arising from importing intermediate inputs, as found by Kasahara and Rodrigue (2008) in the case of Chilean manufacturing firms. Access to bank credit also facilitates the acquisition of new technologies that are embodied in new capital equipment. Bank credit is one form of credit that firms use to smooth their production, another form is trade credit. However, trade credit works well if there are well-developed legal institutions to resolve disputes and social networks to enforce contracts that are based on trust. In Africa, the role of networks in facilitating market exchanges in the context of high transactions costs and poor civil services such as courts has been noted by Fafchamps (2001) and Ojah et al. (2010). The empirical significance of social networks in facilitating trade credit in the African context is well documented. Kuada (2009) found that Ghanaian women entrepreneurs cultivate social networks as a means to lessen the credit constraints they face in formal credit markets. Fafchamps (2000) found some evidence of ethnic and gender bias in the allocation of trade credit in Kenya and Zimbabwe. Ramachandran and Shah (1999), for example, confirmed that informational and financial networks created by minority entrepreneurs provide access to credit, information, and technology for members of these networks. Along the same lines, Biggs et al. (2002) and Biggs and Shah (2006) found that although ethnicity does not play a significant role in determining access to credit among Kenyan firms, belonging to an ethnic group, particularly of Asian-Kenyans, explains access to informal sources of finance, such as trade credit. They propose raising the levels of education of African entrepreneurs as a way to mitigate the exclusionary effects of such close-knit networks. Access to trade credit is an important means of expanding the demand for firms. In the case of Nigerian manufacturing firms, Malik et al. (2006) found that insufficient demand

The Theory of the Firm in the African Context    89 ranks as the third most important constraint to firms’ expansion. This point has been noted by van Biesebroeck (2005), who observed that, because of the difficulty of enforcing contracts in Africa, expanding sales through trade credit is particularly risky. In addition, as pointed out by Bigsten et al. (2003), payments are often late, leading to disputes and renegotiations. Domestically oriented firms in particular, because they cannot hedge against currency fluctuations, are often the worst performers. This in turn limits their ability to extend trade credit. In this way, African firms seem to be caught in a vicious cycle of stagnation in which under-performance limits their ability to extend trade credit, while trade credit is an important factor in driving output expansion. Access to quality infrastructure is an important determinant of the business environment. However, as pointed out by Bigsten and Söderbom (2006), there is little analysis available on the impact of infrastructure on manufacturing firm productivity. Public infrastructure can be viewed as an indirect input in the production process. Eifert et al. (2008) noted that although the costs associated with indirect inputs are low and relatively stable in advanced economies, in developing countries they are very significant. They noted that most studies on firm performance in the African context ignore the role of these costs. Yet, in most poor African countries, indirect costs account for 20–30 percent of total costs. They further argue that the nature and magnitude of indirect costs suggests that poor infrastructure and poor quality of public services constitute a significant barrier to the competitiveness of African manufacturing firms. For example, these authors found that, given their sample of firms, on a yearly basis power outages translate into average losses in sales ranging from 3 percent to 7 percent of total sales. Mozambique has an average of 192 power outages per year, Eritrea has 94, Kenya has 80, Madagascar has 78, Uganda has 74, Tanzania has 67, Zambia has 40, and Senegal has 29 power outages per year. These observations have been confirmed by a number of studies. In a survey of the experience of South African firms in doing business on the rest of the continent, Games (2004) noted bad roads, poor infrastructure, unreliable power, supplies, and lack of telecommunications as part of the logistical problems of doing business on the African continent. Another case related to Nigerian manufacturing firms, which rank problems with physical infrastructure as the most pressing. Malik et al. (2006) found that many Nigerian firms, particularly medium and large firms, supplement their electricity supply with the use of private generators, which is three times more expensive than electricity from the national grid. They also found that small firms lose as much as 24 percent of their annual output as a result of power outages. Because of unreliable power supply, 69 percent of Nigerian firms reported power outages as the main reason for their underutilization of capacity. These authors found that on average, Nigerian manufacturing firms utilized only 44 percent of their capacity. Among indirect costs, it seems that it is access to electricity that is a problem. Malik et  al. (2006) reported that in the case of Nigerian manufacturing firms, 57  percent reported the poor supply of electricity as a major problem. They further noted that “survey evidence points towards a growing dissatisfaction with the country’s dilapidated physical infrastructure, in particular the unreliable and irregular power supply” (p. iii). A similar observation was reported by Gelb et al. (2007) in the case of Ugandan firms, where 87 percent of firms rated electricity supply as a severe constraint. In addition Bigsten and Söderbom (2006) drew similar conclusions in their summary of a decade of manufacturing surveys in Africa. Besides disruptions of production arising from

90   Concepts electricity outages, poor logistics infrastructure locks firms in localized small markets, constrains their demand growth, and inhibits them from fully participating in international trade. As a result, African firms find themselves operating at output levels that are far from efficient.

5.3  Production, Outages, and Export Orientation 3.1  Production with exogenous probability of power outage Assume a representative firm that produces at the level of efficiency A. Following Söderling (2000), Bigsten et al. (2003), and van Biesebroeck (2005), we postulate that this efficiency level is positively determined by the export orientation of the firm and the flow of the indirect input, say electricity or water. Let γ be the fraction of output that is exported by the firm and X be the level of the indirect input. In the African context, the flow of the indirect input is a random variable that assumes a value of zero with probability q. In other words, there is a positive probability that there may be an outage in the flow of the indirect input. Therefore the expected flow of the indirect input is (1–q)X. As pointed out by Eifert et al. (2008), the costs associated with indirect inputs, let alone those associated with their poor quality, deserve explicit consideration in modeling firm behavior and performance. The unreliability of the supply of the indirect input and its low quality in the African context is documented in a number of manufacturing surveys, for example, Malik et al. (2006) and Gelb et al. (2007). The next step is to incorporate the insights regarding the determinants of export orientation γ. Empirical literature suggests that this parameter is endogenously and positively determined by the size of the firm (see, for example, Valodia and Velia 2006; Rankin et al. 2006), which we measure by the level of output Q. Van Biesebroeck (2005) found that, for African firms, half of the export effect is realized by exploiting scale economies, assuming these scale economies are identical for all firms. It therefore makes sense to link export orientation to the level of production. Furthermore, Söderbom and Teal (2003) showed that low levels of efficiency and size are important factors that limit the ability of African manufacturing firms to export. In our formulation, we assume that there exists Qe > 0 such that γ (Qe) = 0 for all Q < Qe. For concreteness, let:

γ = − γ 0 + γ 1Q, (1)

which implies that Q e =

γ0

γ1

. As pointed out by Eifert et al. (2008), the indirect input is a

complement to other factors of production. Given the above assumptions, we posit that:

(

)

A = A0 + α x (1 − q )bx + α γ γ 1 Q − α γ γ 0 ,

(2)

The Theory of the Firm in the African Context    91 where the parameters αx and αγ capture the response of production efficiency to changes in the flow of the indirect input and export orientation respectively, and bx is the input requirement coefficient for the indirect input. Our assumption in equation (2) is that when there is disruption and there is no export orientation, production efficiency falls to a minimum A0. The equation summarizes the findings by Bigsten et al. (2003), that there are significant gains in efficiency among African firms as a result of exporting, that is, αγ is not zero. In addition, as noted by van Biesebroeck (2005), in the sub-Saharan African context, exporters are found to be more efficient than domestically oriented firms. However, van Biesebroeck noted that the difference between African exporting firms and those in other regions is that African exporters improve their relative performance after they enter foreign markets. However, a similar observation was reported by Blalock and Gertler (2004) in the case of Indonesian manufacturing firms. Equation (2) has an important implication. As output increases, production efficiency rises. This means that in the African context, an increase in the level of production for exports will shift downwards average and marginal costs and thereby reinforce the competitiveness of firms. Three implications arise from equation (2). Firstly, a firm that is in a position to exploit scale economies through expansion of markets, for example, a high level of export orientation, will experience increased levels of production efficiency. This result was empirically confirmed by van Biesebroeck (2005), when he noted that firms that engage in exporting experience a 25 percent to 28 percent shift in their production functions. Secondly, a more reliable flow of the indirect input, reflected in low levels of q, will enjoy high levels of production efficiency. Thirdly, there is a level of output at which production efficiency collapses to zero. This level of production is given by:

Q** =

A0 + aγ γ 0

(a (1 − q )b x

x

+ aγ γ 1

)

. (3)

As the probability of outages of the indirect input rises, the critical level of production that a firm must achieve in order to avoid total efficiency collapse rises. This “hurdle” level of production is further pushed up if the firm has no export orientation, that is, if γ’s are zero. Note that equations (1) and (2) capture the idea that there is a mutually reinforcing relationship between total factor productivity and export orientation, as found by Söderling (2000) in the case of Cameroon. The production function of the firm is:

Q = AF ( Z , N ) , (4)

where Z is the imported intermediate input and N is the labor input. There is no substitution between the imported intermediate input and labor. Therefore F(Z, N) = min(Z, N), which is Leontief. Our assumption of a Leontief production function follows the observation by Eifert et al. (2008) that indirect inputs are broadly complementary to other factors of production. Besides simplifying our analysis, the Leontief assumption seems appropriate in the African context where appropriation of new technologies may be limited by insufficient access to credit finance. Assume that F(Z, N) = bzZ, that is, the intermediate imported input is a constraint on production.

92   Concepts By substituting equation (2)  into equation (4)  we obtain the reduced-form level of production: Q=



(A

0

)

− α γ γ 0 bz Z

. 1 − α x (1 − q )bx + α γ γ 1 bz Z

(

(5)

)

Equation (5) shows that, for a firm that has an export orientation and is supported by a reliable flow of the indirect input, an increase in the level of the direct inputs will lead to a more than proportionate increase in the level of production. It is also clear from equation (5) that, despite a Leontief technology, production can exhibit increasing returns, since as Z approaches the level: Z* =



1

(α (1 − q )b x

x

)

+ α γ γ 1 bz

. (6)

The level of output rises exponentially. We can also ascertain that changes in the probability of outage will affect both the “hurdle” level of production and the upper limit of the imported input. For example, Figure 5.1 illustrates the impact of the increase in the probability of outage on the level of production. Note that the production function does not start from the origin, since there is a level of output Q** at which production efficiency is zero. What Figure 5.1 shows is that as the probability of outage rises, the “hurdle” level of production rises. At low levels of output, the firm is forced to raise the productivity of its direct inputs in order to make up for the fall in efficiency brought about by the increase in the probability of outage. However, the increase

Q

Q1**

Q0**

Z0*

Z1*

Z

Figure  5.1   The impact of an increase in the probability of outage on production.

The Theory of the Firm in the African Context    93 in the maximum level of the direct input also means that the marginal product of this input decreases.

5.3.2  Production with endogenous probability of outage When firms anticipate outages in the flow of the indirect input, they tend to be cautious in the choice of production levels. In Africa particularly, where disruptions in production due to outages are common, firms would tend to under-utilize capacity in order to limit the losses they may suffer in case of outages. For example, when there is an outage, workers are not working and they will have to be paid their wages regardless. Similarly, in case the firm has debt, the firm has to service its debt regardless of the production disruption. In some instances the causality moves in the opposite direction, where under-utilization of capacity is a result of outages, that is, firms cannot operate certain equipment because of the insufficient supply of the indirect input, for example, electricity. This situation constrains firms from accessing new technologies that improve production efficiency. Building on this insight, we assume that in general, when the level of aggregate production in the economy rises, it puts pressure on the supply of the indirect input, which then leads to outages. Consequently, it may be useful to posit a causal relationship between the level of output of the representative firm and the probability of outage, that is, q = q(Q), where q′ > 0.  Specifically, assume that: q = q0 + δQ. (7)



This simple relationship implies that, although the firm can attempt to exploit its economies of scale by expanding its export markets, or by issuing more trade credit, there are limits to this strategy. The increase in output will initially raise production efficiency. However, beyond a certain point, the increase in production will put pressure on the national supply grid of the indirect input and therefore trigger outages. This can be observed by inserting equation (7) in equation (2) to get:

(

)

A = A0 + α x (1 − q0 − δQ )bx + α γ γ 1 Q − α γ γ 0 , (8)

which is quadratic in Q. The level of output that is consistent with maximum production efficiency under conditions of endogenous outages is:

Q =

(α (1 − q )b x

0

x

α x bx δ

+ αy γ1

) . (9)

Equation (9) shows that without export orientation, the firm has limited scope to exploit economies of scale. In addition, the exogenous component of the probability of outage also plays an important role in determining the range of output levels that do not trigger outages. ∼ Empirical literature suggests that African firms operate at levels of output that are below Q, ∼ despite the fact that Q is already low because of high levels of the exogenous probability of outage. For example, Bigsten et al. (2003) and van Biesebroeck (2005) found significant

94   Concepts productivity effects from exporting. This literature finds that an increase in productivity in the production function of 7–8 percent translates to value-added productivity gains of 20–25 percent in the short run and up to 50 percent in the long run.

5.3.3  Demand and export orientation There are a significant number of African firms that cite limited access to markets as the reason for limited expansion. For example, van Biesebroeck (2005) found that, when asked directly, 16 percent of the owners or managers mention that insufficient demand is the principal limit to their expansion. Furthermore, he noted that almost a third lists it among the top three problems. He concluded on the basis of this that many firms may be operating at a smaller scale than they would like. In the case of Nigerian manufacturers, Malik et al. (2006) reported that insufficient demand is cited as the third most severe constraint on firms’ expansion. To model this situation, we assume the firm’s price level is a weighted average of domestic and foreign price, all denominated in domestic currency. This means that:

P = (1 − γ ) P d + γEP f . (10)

Let Pd = d0 – d1Q. At this stage, we could explicitly incorporate the effects of sales based on trade credit, particularly in considering domestic demand. However, we chose to focus our attention explicitly on the exchange rate and export orientation in order to simplify our analysis. It should be noted though that trade credit effects on domestic demand are implicit in the parameter d0, and if trade credit is endogenously determined by efficiency and firm size, it would also affect the magnitude of d1 as well. It follows from equation (10) that the demand curve faced by the firm is:

P = (1 + γ 0 − γ 1Q )(d0 − d1Q) + ( − γ 0 + γ 1Q ) EP f , (11)

where we have substituted in the endogenously determined export-orientation γ. Note that when there is no export-orientation, the demand curve would shift inwards. Graphically, the demand curve can be illustrated as shown in Figure 5.2. What Figure 5.2 shows is that there exists a level of output Qm above which the demand curve is perverse, or above which the equilibrium of the firm is unstable (and hence unsustainable). It is possible that this point corresponds to an unviable, negative price level. In which case, the downward-sloping portion of the curve would be the only relevant portion in the firm’s decision-making. Figure 5.2 illustrates for an example, the impact of the depreciation of the exchange rate on the demand curve, which is to shift the curve from D0 to D1. ∼ Note that it is possible to have a situation where Q > Qm , in which case the level of output that corresponds to maximum efficiency lies on the perverse part of the demand curve. This would imply that, as long as firms operate on the downward-sloping part of the demand curve, they will not achieve maximum production efficiency. If they push production to achieve maximum efficiency this would not be sustainable because the resultant equilibrium would be unstable.

The Theory of the Firm in the African Context    95 D1

P

D0

Qm

Q

Figure  5.2   The demand  curve. ∼

To ensure that Q > Qm , that is, maximum efficiency is attained on the downwardsloping portion of the demand curve, firms must be relatively large because γ0 must be sufficiently large. The size of the firm, in so far as it is measured by the level of production, will ensure that Q > Qe. Exporting therefore ensures that maximum efficiency can be attained at a stable equilibrium position for the firm. Another, less attractive factor that can lead to a stable equilibrium with maximum efficiency, is the increase in the exogenous probability of outage. However, this leads to capacity under-utilization and other costs to firms.

5.3.4  Optimal production Applying the standard profit maximization rule, under the assumption that African firms have some pricing power in domestic markets, but not in export markets, we can write the revenue function as:

(

)

R = (1 + γ 0 − γ 1Q )(d0 − d1Q ) + ( − γ 0 + γ 1Q ) EP f Q, (12)

where, as before, Pd = d0 – d1Q describes the domestic demand curve faced by the firm, E is the nominal exchange rate, and Pf is the foreign price level. Since γ = –γ0 + γ1Q, it can be observed that the marginal revenue function of the firm is quadratic, with the negatively sloped part of the curve being the relevant portion.

(

MR = (1 + γ 0 − γ 1Q )(d0 − d1Q ) + ( − γ 0 + γ 1Q ) EP f

)

− γ 1 (d0 − d1Q )Q − d1 (1 + γ 0 − γ 1Q )Q + γ 1 EP f Q.



(13)

96   Concepts In line with the demand curve, the marginal revenue function shows that there is a level of output above which it is sub-optimal for the firm to produce. The cost function faced by the firm is given by the following equation:

(

)

TC = EPzf bz + Wbn + Px bx (1 + q0 + δQ )Q + F , (14)



where F is average fixed costs and we have exploited assumption that all inputs used are complements, bj is the input requirement coefficient for input j, technology is Leontief. The term EPzf bz + Wbn + Px bx (q0 + δQ )Q represents that part of total cost that will be incurred as a result of the outage in the indirect input. Note that equations (12) and (14) capture the fact that firms engage in both exports of goods they produce and imports of some inputs, as noted by Valodia and Velia (2006). From equation (14), it follows that marginal cost is:

(

)

(

)

MC = EPzf bz + Wbn + Px bx (1 + q0 + 2δQ ). (15)



Optimal production is achieved when marginal revenue equals marginal cost. We can illustrate the equilibrium position of the firm with Figure 5.3. In Figure 5.3 the optimal level of production is at Q*. The exogenous level of the outage raises marginal cost and does not affect marginal revenue, but it affects marginal costs. If q0 rises to levels that are sufficiently high, the firm may not be able to find an optimal position in which it can make positive profits. Figure 5.3 is similar to the standard diagram of the theory of the firm. However, it does shed some light as to why, in conditions where there is a high probability of outage, firms would tend to under-utilize capacity. The exogenous probability of outage shifts the marginal

P MC

P*

ATC

B

P0

D A

Q*

MR Q0

Figure  5.3   The equilibrium position of the  firm.

Q

The Theory of the Firm in the African Context    97 cost and average total cost curves equally upwards. The firm would move up the marginal revenue curve. The result is that the gap (Q* – Q0) which measures the degree of excess capacity or productive inefficiency will widen. This is the case for 69 percent of Nigerian manufacturing firms. One of the key questions about the performance of firms in Africa is the impact of the exchange rate on firm behavior. Söderling (2000) found that the exchange rate plays a significant role in the performance of manufacturing firms in Cameroon. Specifically, he found that an overvalued exchange rate tends to negatively affect the performance of the manufacturing sector. A similar result is reported by Sekkat and Varoudakis (2000), for the case of manufactured exports in major sub-Saharan African economies. Within this framework, we can also show that exchange rate depreciation can have contractionary effects on the firm’s output. Empirical literature on the effect of exchange rate depreciation on the performance of African firms is mixed. This is the case because for a depreciation to have a positive impact on the firm’s output, it must raise marginal revenue more than it raises marginal cost, that is: ∂MR ∂MC > . ∂E ∂E

This implies that: Q>



(

(1 + q ) P

(

z

0

f

bz

2 γ 1 P − δPzf bz f

)

.

)

Under the assumption that γ 1 P f − δPzf bz > 0 , it can be observed that the increase in the exogenous probability of outage, increases the level of output above which a depreciation leads to an expansion in output. The implication is that at high levels of the probability of outage, firms have to be “large,” that is, they have to produce a sufficiently high level of output in order for the depreciation of the exchange rate to have an expansionary effect. Many African firms are relatively small to medium enterprises. They suffer from capacity under-utilization because of lack of markets and high probability of outages of the indirect input. It is therefore likely that they will fail to meet the output threshold that is required for the exchange rate depreciation to be expansionary.

(

)

The alternative case, where γ 1 P f − δPzf bz < 0, suggests that firms must produce below the above-mentioned threshold in order for the exchange rate to be expansionary. In this case it is likely that it will be small firms that would benefit from an exchange rate depreciation. This is found to be the case by Valodia and Velia (2006). Large firms are found to be both importers of machinery and at the same time they are exporters. A depreciation of the exchange rate tends not to have significant impact on the level of production for large firms. There are some interesting cases where the stability of the firm’s equilibrium position may not be robust. Firms may be forced to operate in ways that do not maximize profits because there is no equilibrium. Standard economic theory suggests that when marginal revenue is above marginal cost, the firm will have to increase production. However, there may be cases where this behavior is not a viable option. Figure 5.4 illustrates this problem.

98   Concepts D

P

MC

ATC

MR

P* P0

B

A

Q*

Q0

Q

Figure  5.4   A  fragile equilibrium of the  firm. Figure 5.4 shows a situation where the equilibrium position of the firm is in a form of a tangency. For example, a sufficiently low level of the exogenous probability of outages and prices of inputs will make marginal revenue to be above marginal cost for all levels of output. In this case, there is no price level that can satisfy demand. According to the standard adjustment rule, output must be permanently increasing without bound. Note that at point A, it is also possible that the firm may be making losses, provided the demand curve is sufficiently low. Nevertheless, in the African context, where the exogenous probability outage is relatively high it may be argued that the scenario depicted in Figure 5.4 is unlikely. However, the likely scenario, which depicts the marginal cost curve sufficiently above the marginal revenue curve, seems to capture the prevalent situation. In African context such as Mozambique, where outages occur on average in 192 days of a year, q is already high. The mar0

ginal cost curve is likely to intersect the marginal revenue curve in two places, thereby generating two equilibria; one unstable and the other stable. Figure 5.5 illustrates this situation. Figure 5.5 illustrates the situation of a low-level equilibrium trap, which is characterized by significant under-utilization of capacity, which is consistent with the finding by Malik et al. (2006) in the case of Nigerian firms. To alleviate this problem, the firm will have to raise its demand, either by seeking new markets or by extending trade credit to its customers. This scenario is likely to capture the situation that many African firms find themselves in. Exchange rate shocks or shocks from other costs can drive the firm further towards under-utilization of capacity and, hence, productive inefficiency. As the arrow illustrates, the adjustment would be for the firm to persistently reduce output until it shuts down. Input price shocks and exchange rate shocks may occur simultaneously with upward shocks to the exogenous probability of outage. In this case, Figure 5.5 shows that it is possible that marginal costs can rise far more than marginal revenue, thereby driving point A further up along the marginal revenue curve. Therefore, it could be that as firms open new markets and raise their marginal revenues, the supply of indirect inputs can face excessive strain, because of

The Theory of the Firm in the African Context    99 D P

MC

P*

ATC

MR

P0

B

A

Q*

Q0

Q

Figure  5.5   A  low-level equilibrium  trap. low investment in public infrastructure, which would then reverse the efficiency gains emanating from expanded markets. Another plausible scenario is that the capacity of firms to expand the export market may be limited by factors such as road, rail, and port infrastructure. The capacity to expand domestic markets may be limited by access to credit. Although trade credit has been found to play a significant role in the expansion of markets and securing stable demand for firms, the absence of strong legal institutions to enforce contracts in cases of conflict, limits firms from accessing, and supplying, trade credit. Consequently, there is a tendency for the demand curve to be either stagnant or to shift downwards. The effect of this is to reduce output and to erode the firm’s profit margins. Furthermore, as equation (9) demonstrates, the increase in the exogenous probability of outage limits the scope for the firm to exploit economies of scale as the basis to raise its production efficiency. As amply demonstrated by van Biesebroeck (2005), exporting plays an important role in raising the total factor productivity. However, the capacity to export also depends on the ability to exploit economies of scale, that is, to increase the level of output. Consequently, as the probability of outage rises, not only does the capacity under-utilization gap (Q* – Q0) increase, the cost functions also shift upwards from two sources. Firstly, they shift upwards because of the direct costs associated with outages. Secondly, they shift upwards because of the indirect effects arising from the decrease in total factor productivity.

5.4  Policy Recommendations The empirical literature suggests that high-performing firms are exporters. They are characterized, among other factors, by high levels of capital intensity, they invest in their productive

100   Concepts capabilities, they appropriate new technologies frequently, and they are large (see Rankin et al. 2006; van Biesebroeck 2005). Domestically oriented firms tend to have limited access to credit, they are small, they cannot fully exploit economies of scale because of insufficient demand, they have limited capacity to appropriate new technologies, and they are stagnant. Unfortunately, the latter are characteristic of many African firms, who may account for a large portion of the continent’s employment. Our analysis of firm behavior suggests that a large number of African firms may be in a low-level equilibrium trap, which is characterized by a vicious cycle of stagnation. The key factors that affect the performance of these firms are access to credit, insufficient demand, and devastating shocks that emanate from poor and unreliable supply of indirect inputs such as electricity. It seems that unit labor costs are not a major constraint on African firms, since empirical literature suggests that dynamic firms which are engaged in exporting pay on average more than domestically oriented firms. The first policy recommendation follows from Eifert et al. (2008). Given the substantial weight of indirect input costs in the operations of African firms, it is incumbent on African governments to improve the quality and reliability of supply of indirect inputs, especially electricity. As Malik et al. (2006) have found in the case of Nigerian firms, unreliable supply of indirect inputs causes firms to operate below capacity for fear of large output losses in the case of outages. This adds to the inefficiency of African firms, since it prevents them from fully exploiting economies of scale. Ensuring a reliable supply of indirect inputs and high quality of public service requires upgrading of physical infrastructure and its constant maintenance. In this regard, it is important that the activities of the Pan African Infrastructure Development Fund be closely monitored. The second policy recommendation relates to access to credit. There is ample empirical literature which suggests that social network, which may be exclusionary, has replaced public institutions as a means to facilitate the flow of trade credit. The obvious recommendation that flows from this observation is that it is once again incumbent upon African government to improve institutions that uphold the rule of law, improve their turn-around times and lower the costs associated with using them, especially when it comes to civil matters such as disputes over private contractual rights and responsibilities (see Ojah and Mokoteli 2010; Ojah et al. 2010). Since there may be problems of adverse selection and moral hazard if governments undertake to set up institutions to directly extend credit, it may be prudent for governments to set up institutions that bring together firms that operate along the same value-chains. In this way, public institutions can provide a stable platform to initiate sustainable networks that facilitate the private flow of trade credit among firms (see Ojah and Kodongo, Chapter 55, Handbook, Volume 2, for financial development-based solutions). Not only can such facilitation of trade credit expand the domestic market, and improve the efficiency of domestic-orientated firms, it will also improve access to raw materials and improve capacity utilization. Besides problems with public infrastructure and insufficient demand, a significant number of African firms are constrained from fully utilizing their capacity because of the shortage of raw materials. The third policy recommendation is for governments to expand domestic markets. This can be done by improving transport infrastructure. Firms should not be trapped in small localized markets which in turn reinforce low capacity utilization rates. Secondly, governments must prudently use their resources to support local firms through improvements in procurement

The Theory of the Firm in the African Context    101 policies. Specifically, firms that must be targeted for support must exhibit large multipliers on other sectors. Simultaneously, an effort to broaden consumer education and to set up institutions that protect consumers’ right on the African continent is important. Empirical literature generally finds that domestically oriented firms are not innovative and are stagnant. Consumer education may provide pressure on African firms to improve innovation on their product offering. The fourth policy recommendation is for government to improve their regulatory frameworks and to improve capacity to efficiently implement policies. This is particularly important when it comes to import and export activities. Africa has very limited capacity to produce intermediate inputs. Governments must lower the costs of importing intermediate inputs and capital equipment. This will facilitate easier acquisition of new technologies that improve productivity. Because many firms on the African continent are small and may be operated by entrepreneurs with low levels of education, it may be advisable for governments to offer some reliable “brokerage service” that would facilitate exporting of goods produced by local firms and importing intermediate inputs and capital equipment. Lastly, management of the exchange rate appears to be a key factor in determining the performance of manufacturing firms on the African continent. Many African currencies are commodity currencies. Therefore, it may be prudent for macroeconomic policies to provide a more stable exchange rate. An active policy that minimizes exchange rate misalignments at the very least would go a long way in providing a cushion for manufacturers. However, it appears that the most sustainable way to cushion firms from exchange rate fluctuations is to lay the basis for them to enter the export market and earn foreign exchange. This will provide for them a natural hedge against currency fluctuations. We argue that this can be done by governments providing reliable indirect inputs and quality public service.

5.5 Conclusion African firms operate in an extremely challenging business environment. They face domestic markets that are shallow and their capacity to expand into export markets is limited by a number of factors. Their competitiveness is compromised by poor access to public infrastructure and poor public service. They face high indirect costs which are related to state capacity to provide basic goods and services. They operate at low levels of capacity utilization, which constrains them from exploiting economies of scale, which further inhibits them from improving their productive efficiencies and productivity levels. These factors make African firms perform poorly. Their poor performance in turn negatively affects their prospects to access formal credit lines. In the context where many of the inputs that they use are imported, African firms find themselves with chronic shortages of raw materials and are not able to appropriate new technologies that are embodied in new capital equipment. In addition, because of their poor performance, they are not able to extend trade credit to their customers. This, over and above the problem of weak social networks and weak public institutions to enforce contracts, further constrains their ability to expand since trade credit would alleviate the problem of insufficient demand. In this way, African firms find themselves in a vicious cycle of stagnation.

102   Concepts Our theoretical analysis of firm behavior shows that an important factor that has to be improved is access to a reliable flow of indirect inputs. If governments can provide a stable supply of inputs such as electricity, water, and transport infrastructure, this will go a long way in improving current rates of capacity utilization, it will improve competitiveness and productivity, shift down marginal and average costs, and establish healthy profit margins for African firms. Improvements in the provision of indirect inputs may lower costs to a point where African firms can enter export markets and earn foreign exchange. This is the most direct and natural way in which firms can provide a hedge for themselves against currency fluctuations. It will also allow firms to access new technologies, improve capacity utilization and raise formal employment levels on the African continent. As the empirical literature indicates, firms that export pay higher wages than those that do not export. The increase in employment through the expansion of export markets will raise the standard of living and significantly reduce poverty levels on the African continent.

References Biggs T., Raturi M., Srivastava P. (2002). Ethnic networks and access to credit: Evidence from the manufacturing sector in Kenya. Journal of Economic Behaviour and Organisation, 49:473–486. Biggs T., and Shah M.K. (2006). African SMEs, networks and manufacturing performance. Journal of Banking and Finance, 30:3043–3066. Bigsten A., Collier P., Dercon S. et al. (2003). Do African manufacturing firms learn from exporting? Journal of African Economies, 12:104–125. Bigsten A., and Söderbom M. (2006). What have we learned from a decade of manufacturing enterprise surveys in Africa? World Bank Research Observer, 21:241–265. Blalock, G., Gertler, P.J. (2004). Learning from exporting revisited in a less developed setting. Journal of Development Economics, 75:397–416. Eifert, B., Gelb, A., Ramachandran, V. (2008). The cost of doing business in Africa: evidence from enterprise survey data. World Development, 36:1531–1546. Fafchamps, M. (2000). Ethnicity and credit in African manufacturing. Journal of Development Economics, 61:205–235. Fafchamps, M., (2001). Networks, communities and markets in sub-Sharan Africa: Implications for firm growth and investment. Journal of African Economies, 10 (AERC Supplement 2): 109–142. Games, D. (2004). The experience of South African firms doing business in Africa. Research Project Report 1, South African Institute of Race Relations. Gelb, A., Ramachandran, V., Shah, M.J., and Turner, G. (2007). What matters to African firms? The relevance of perceptions data. Policy Research Working Paper 4446, World Bank. Gwatidzo, T., and Ojah, K. (2009). Corporate capital structure determinants: Evidence from five African countries. African Finance Journal, 11:1–23. Gwatidzo, T., and Ojah, K. (2014). Firms’ debt choice in Africa: Are institutional infrastructure and non-traditional determinants important? International Review of Financial Analysis, 31:152–166.

The Theory of the Firm in the African Context    103 Kasahara, H., and Rodrigue, J. (2008). Does the use of imported intermediates increase productivity? Plant-level evidence. Journal of Development Economics, 87:106–118. Kuada, J. (2009). Gender, social networks and entrepreneurship in Ghana. Journal of African Business, 10:85–103. Ramachandran, V., and Shah, M.K. (1999). Minority entrepreneurs and firm performance in sub-Saharan Africa. Journal of Development Studies, 36:71–87. Malik, A., Teal, F., and Baptist, S. (2006). The performance of Nigerian manufacturing firms: Report on the Nigerian Manufacturing Enterprise Survey 2004. Center for the Study of African Economies. Ojah, K., Gwatidzo, T., and Kaniki, S. (2010). Legal environment, finance channels and investment: Evidence from the East African Community. Journal of Development Studies, 46:724–744. Ojah, K., and Kodongo, O. (2015). Financial markets development in Africa: Reflections and the ways forward, in C. Monga and J. Yifu Lin (eds), The Oxford Handbook of Africa and Economics, Volume 2. Oxford: Oxford University Press. Ojah, K., and Mokoteli, T. (2010). Possible effective financing models for entrepreneurship in South Africa:  Guides from microfinance and venture capital finance. African Finance Journal, 12:1–26. Rankin, N., Söderbom, M., and Teal, F. (2006). Exporting from manufacturing firms in sub-Saharan Africa. Journal of African Economies, 15:671–687. Sekkat, K., and Varoudakis, A. (2000). Exchange rate management and manufactured exports in sub-Saharan Africa. Journal of Development Economics, 61:237–253. Söderbom, M., and Teal, F. (2003). Are manufacturing exports the key to success in Africa? Journal of African Economies, 12:1–29. Söderling, L. (2000). Dynamics of export performance, productivity and real exchange rate in manufacturing: the case of Cameroon. Journal of African Economies, 9:411–429. Valodia, I., and Velia, M. (2006). Macro–micro linkages in trade: Trade, efficiency and competitiveness of manufacturing firms in Durban, South Africa. Journal of African Economies, 15:688–721. Van Biesebroeck, J. (2005). Exporting raises productivity in sub-Saharan African manufacturing firms. Journal of International Economics, 67:373–391.

Chapter 6

Markets and U rba n Provisi oni ng Jane I. Guyer

6.1  Introduction: Marketplaces, Markets, and the Relevance of Africa In an article that is still relevant today, Polly Hill (1989) drew attention to the importance of keeping the “market place,” as an “authorized public concourse of buyers and sellers of commodities,” in full view, alongside “market principles,” in the abstract sense of competitive price-making institutions and processes. The marketplace, she argued, is a widespread and ancient institution, predating capitalism by millennia, and manifest in very large gathering places and supervisory structures from China to Mexico. The Greek agora—literally a gathering place—was a public space for debate, commerce, military exercises, and any activity that required the citizens to gather. Aristotle discusses money and trade. Chinese marketplace systems developed a complex nested pattern, around central places, over centuries of development. Hill argues, however, that the rise of market principles has resulted in a situation where “economists have displayed little interest in marketplaces, as distinct from market principles, having condescendingly passed the subject to economic historians” (1989: 238). In Africa, marketplaces were left largely, although not entirely (see Jones 1972), to geographers and anthropologists. The result has been a detailed record on the organization of networks, associations, budgeting, and so on, but less on price mechanisms themselves. Indeed, it would be worth studying closely what economic analyses have been done with the localized price monitoring figures that come out of the consumer price index (CPI) process.1 Examination of how competitive price-making works at different scalar levels could enrich interdisciplinary work, especially in areas such as Africa where marketplaces are significant in both wholesale and retail sectors, and where production conditions are localized 1   In my book An African Niche Economy (1997), I devoted Appendix B to tracing out a history of the relevant marketplace prices over 20 years, “Towards a History of Prices and a Consumer Price Index” (that is, for the farmers in that area, to estimate real income changes).

Markets and Urban Provisioning    105 but the distribution of many commodities, and not only ecological specialties, crosses several scalar thresholds. The modern capitalist sense of a “market economy,” where anything and everything that is legally transactable through non-localized institutions, where commodities are submitted to an integrated, monetized competitive price mechanism, on the basis of national currencies and financial institutions, is comparatively recent in human history and not yet universal. In fact, certain new “local” movements, all over the world, are attempting to put the universalizing genie back into the bottle. While the current financial institutions of capitalism go back much further than the present century, the fuller development of global financialized capital and international currency markets gained momentum only after World War II. The assumption that markets should be free, and that their patterns of transaction will then reflect a basic human rationality, in the form of choice, that can be modeled, is particular to our own era of history. For application of rationality to market systems that are institutionalized in these localized forms, we have to create the intellectual space for price dynamics that do not—indeed cannot—index to shareholder value and corporate profit. A historical and anthropological approach opens up any and all monetary transactional systems, and commodity distribution systems, to empirical, institutional, and comparative analysis of what may lie behind the concept of “utility,” which has been generally a derived, rather than an externally indexed and measured, quality. This would then address the kinds of cognitive and cultural concerns that are entering into economics through the relatively new field of behavioral economics. I would add that we also need an appreciation of the intellectual conventions through which unwritten accounting and transactional commitments have worked, as they did across vast trade networks as well as localized marketplaces in Africa in the past (Guyer 2004). The lack of a profuse and deep written record of transactional conventions, except in the largely unanalyzed Arabic-script contexts of the Sahel and North Africa, is another reason why these systems have been consigned to “subsistence,” although we know from old sources and current ethnography that people had concepts and memory techniques for numerical information and calculation. Indeed, the sophistication and familiarity of trusted transactional regimes without written records may well go towards explaining Africa’s unusual alacrity in the uptake of the new electronic technologies, including monetary innovations such as M-Pesa. So this image of African economies as somehow “subsistence” is a logical entailment of the record, of a narrow interpretation of the referent of “commercial” as necessarily large enterprises, as well as of modern assumptions about markets and logics in corporate, financialized, intricately recorded economies. It is certainly inaccurate as an empirical fact and not helpful in thinking in informed disciplinary terms about how “localization” and trust might work in the various futures that people are currently attempting to envisage, where scalar nesting is a prominent feature.2 Following the conventional modernist logic, in the twenty-first century Africa is being written about as if it is a “frontier” (Kilel 2014), even an “ultimate frontier” (Mataen 2012) for present investors. This is probably true for at least some kinds of corporate and financial presences, manufacturing investments and foreign interests in “natural resources,” under 2 

In fact, Africa has not only had market distribution and monetary transactions for centuries, but it has been argued that the Africa trade was the crucible in which certain of the corporate and financial institutions of capitalist institutions, labor organization, trade insurance, corporate companies with purchasable shares, and so on, were forged (Inikori 2002; Mintz 1985; Levy 2012).

106   Concepts the institutional systems that have developed in the latter half of the twentieth century: the international financial institutions (IFIs), the money market, the financialization of assets, the globalization of economic life after 1989. The corresponding assumption, however, that African economies did not have important market infrastructures in the past, cannot be drawn as a valid generalizing inference from the dearth of disciplinary economic studies of marketplace dynamics. To bring the current widespread and wide-ranging marketplaces of Africa out of eclipse would contribute significantly to the understanding of several economic dynamics that are of great importance, especially to analysis of income and expenditure aspects of standards of living amongst African populations of all income strata, and the organizational capacity to respond to new opportunities. New opportunities have presented themselves in the colonial and post-colonial eras, when it has largely been the longstanding practices of the popular economy—that is, the economy of the majority of the people—that have developed further, to provision the rapidly growing cities. As cities grew, and where the popular economy was fostered through transport and infrastructure development, a high proportion of the provisions for African urban populations passed through marketplaces that grew from indigenous organizational templates, and thereby accounted for much in the way of people’s livelihood: in both the sense of employment provision and the sense of consumption standards.3 These same templates of occupational associations, trade networks, and money management for market organization have been elaborated to apply to new products and provisions, such as charcoal, automobile spare parts, imported beauty products, second-hand clothing (Hansen 2000), electronics, and many other goods that are distributed through the vast open-air markets and equally vast distribution networks that build on the institutional foundations through which the more basic commodities have been distributed for a long time. Studies of pricing in these new marketplaces and networks may be limited by commodity, and by particular crisis moments, such as described by the collection edited by Guyer, Denzer, and Agbaje (2002; published 2004 in Nigeria) on currency devaluation in urban Southern Nigeria 1986–1996. However, their extension and comparative combination, to address the implications of price turbulence in many localized and commodity-specific contexts, may be of general relevance.

6.2  A Brief History of African Markets Africa’s agriculture and animal husbandry are often presented as if they had only very recently been drawn into the market. In fact, African markets for local and regional redistribution of goods have a very long history, although one interrupted by large forces such as insecurity in the interior in the era of the slave trade (Inikori 2002), and sudden fluctuations in the value of the currency, such as the cowrie inflation in West Africa in the eve of colonial rule (Law 1995). During conditions inimical to trade, people could withdraw temporarily into localized production and distribution systems, which were termed “subsistence systems.” But this was not necessarily production only for the subsistence of very small 3 

Many studies describe these organizations, e.g. Clark (1994) on Ghana, Robertson (1997) on Kenya, and Guyer (1997, Ch. 14), on Western Nigeria.

Markets and Urban Provisioning    107 and localized units. Pastoral–agricultural exchange took place all across the ecological interface zones between pasture and cultivation. The cotton harvest, and the manufactures from it, were distributed across wide areas where the crop itself could not be grown, and cloth strips were also manufactured in what is now central Nigeria as a kind of currency (Johnson manuscript). Historical accounts indicate large-scale food production for supplying passing ships along the coast, and substantial inland markets followed coordinated periodic calendars in some places. On his journeys between 1854 and 1858, the missionary William Clarke (1972) described a Yoruba market “with hundreds of caravans … from towns and villages for a diameter of more than one hundred miles” with huge mounds of cotton and artisanal goods so various as to be “too tedious to mention.” Even earlier, the famous explorer Hugh Clapperton (1829) described seven different markets held every evening in Old Oyo, where numerous agricultural products were for sale (both quoted in Guyer 1997: 18). Beyond the simple fact of old marketplaces, other evidence of the existence of many different currency goods, political–geographical–pilgrimage currency zones, and regimes for counting and creating price equivalence, testifies to the development of price mechanisms, specialist roles in market supervision and trade, transport systems, and classic institutions of market exchange, such as calendrics (Hodder 1965). Harms (1981) and Van Leynseele (1983) described the riverine trades of the Congo and its tributaries. Some of these kinds of complexes served interecological functions: for example, cotton for cloth-making, straw for roofing, dried fish, firewood, and manufactures such as iron and copper goods. The following is based on a description, by Americo-Liberians visiting in 1860, of the markets in an area now designated as a forest reserve in Guinea: settlements had daily and weekly markets, as did all the major towns, distant some eight to 10 kilometers one from another, and these traded in foodstuffs, livestock, cash crops (such as cotton and kola), and artisanal goods of every description. The region was not economically or geographically marginal, but central to busy and long-established forest- savanna trade routes. (Fairhead and Leach 1995: 1030)

Daily urban food supply for political elites in pre-colonial cities such as Kumasi (Ghana) and Kano (Nigeria) derived—at least in part—from their rural domains that were farmed by workers who were in one status or another of dependence and constraint (kin, subservient status). The poorer people and the many specialist artisans were fed from markets, supplied by merchants and intermediaries who managed storage, porterage, and the other functions that make trade possible. It has been the rapid growth of urban population in the twentieth century that has fostered a corresponding growth of locally specific institutional responses and technical innovations, working from these long historical traditions. The evidence base for the early colonial period can be a challenge. Michael Watts (1987) has raised important questions about the sources. Writing of Kano, he finds that failures and famines were under-reported when government was avoiding responsibility:  either for causing the shortages or for failing to mitigate them. And this may well be the case for areas beyond the dry savannah. Colonial-era price issues, relative to salaries, in Yaoundé (Cameroon) were resolved by straightforward requisition through the chiefs, but were somehow configured by the French government as due to a failure of markets (Guyer 1987). This, however, is a high rainfall area with substantial farms and old trade routes, although not primarily for crops, possibly due to disadvantageous terms of cost–benefit for transport;

108   Concepts even dependent or servile labor for transport has to be fed. So evidence about food systems for the past needs to be treated with caution. The one thing that is clear is that pre-colonial Africa certainly had marketplaces, currencies, and pricing systems, and also large settlements in some places, well before the demographic growth of cities and the revolution in transport that took off during the late colonial period.

6.3  The Demography of Urban Growth and the Provisioning Infrastructures Africa used to be represented as a rural continent. The cities started to grow significantly after World War II, and have continued until, in some countries—Nigeria for example—the urban population is now over 50 percent of the total, distributed over about two dozen cities with populations of over 100,000, and many that would qualify as small towns rather than villages. Over 60 years, the proportion of the African population living in cities has tripled, as the absolute numbers of the urban population have grown by a factor of around 12: from around 33 million in 1950 to 400 million in 2010. Both the proportion and the numbers are expected to continue to rise (Table 6.1). In a study carried out in the 1960s, soon after independence, W.O. Jones wrote: “if tropical African markets for basic foodstuffs worked less well, we should probably know a great deal more about them. The truth of the matter is that they have done a remarkably good job of their first task, which is the provisioning of cities and towns and a few large mining developments” (1972: 18). Although food imports did begin during this time, especially Asian rice into West Africa (Pearson 1981) and wheat products (Andrae and Beckman 1985), it appears that most African cities were provisioned from their own hinterlands. The food supply of some cities had been controlled and/or politicized, through requisitioning (Guyer 1987), preferential support to white farmers and parastatal organizations (see Bryceson 1987; Mosley 1987), and by direct attempts to control the marketplace, as in the military

Table 6.1  Urban growth in Africa, 1950–2010. Percentage population residing in cities Africa as a whole Eastern Middle Northern Southern Western

1950

2010

14.4 5.5 14.0 25.8 37.7 9.7

39.2 23.3 40.9 51.2 58.5 44.3

Source: United Nations Department of Economic and Social Affairs, Division of Population. World Urbanization Prospects, the 2011 revision.

Markets and Urban Provisioning    109 government’s destruction on Makola Market in Accra, Ghana in 1979, in the name of “revolution” (Robertson 1983). This demolition was followed by others in Sekondi, Kumasi, and Koforidua. The evidence is strong, however, that small producers and widespread networks of transport, transformation, and trade continued to function in many places. Gracia Clarke has written extensively on Kumasi; Claire Robertson on Nairobi; and many French researchers on the filières of trade (see Guyer 1987). This expansion was especially true after each of the transport revolutions: the lorry in the colonial period, and the pick-up truck in the 1970s. Farmers still head-loaded crops from farm to wholesale market until the 1970s, but very substantial quantities of food have supplied the cities through the organizations of traders and transporters at each stage of the process, from wholesale bulking to breaking bulk at the retail end. If we take Western Nigeria as just one example, certain characteristics can be seen together that are variously distributed elsewhere throughout Africa. First of all, the provisioning sector employs a high proportion of women. The primary processing of certain products, such as gari (a cassava product), have been almost entirely in the artisanal sector, controlled by women (Wan 2001). Many of the bulk traders and even financiers are women, as are the retail sellers in the great urban markets. Some of the crops that benefit from organization by specialist women’s organizations are crucial to the diet:  such as beans in East Africa (Robertson 1997). Secondly, and to some considerable degree a function of the gender factor, most of the functionaries in the food trade are organized into associations. These have membership, designated positions, rules of operation, and meetings about prices and other conditions of the market, such as security and dealing with the authorities. Most—although not all—of them are gender-specific, which makes for a certain smoothness of operation. Whether gender-specific or mixed, all these associations have a social life of meetings, savings and loan functions, and support in case of celebration and misfortune, which ensure a dense mutual knowledge and life-implication. Thirdly, this kind of system has been commodity-responsive. The template for the overall organization is applicable to any new product, whether a new crop grown in the rural areas, or an import traded in from elsewhere. The fact that most markets work in routinely predictable ways, except under negative macro-conditions such as energy shortages and political dislocations, is the result of this long tradition of mutual accommodation and negotiation amongst these relatively small and flexible organizations, peopled by practitioners with long personal histories and training in the commodities and functions for which they are responsible and from which they earn their livelihood. Gracia Clark’s title for her book on the Kumasi women traders, Onions Are My Husband, expresses the kind of long, responsive commitment that such a system cultivates. In my ethnography of food production in an urban hinterland (Guyer 1997), I depicted this kind of system as a “niche economy,” where specialization was organized in a competitive, but highly coordinated, way. All participants had learned their particular skill from an established practitioner, so learned the material and moral standards of the trade as if in an apprenticeship. Indeed, new skills and materials were brought in through particular master-craftsmen, who then trained apprentices and granted them their “freedom” to practice independently: in both competition and collaboration with respect to the total

110   Concepts workload. I witnessed this process with the introduction of private tractor hire, where both tractor mechanics and tractor tire repair became new specialist skills, reaching out into regional and even national networks for spare parts supply and finance. All this was achieved through the same associational template as existing specialties. And new crops came into the repertoire: particularly perishable crops such as watermelon, along with specialist transporters and traders, accustomed to the particular conditions of handling and storage. One cannot underestimate the cumulative effectiveness of thousands of such organizations in provisioning the great cities of Africa in certain of the staples, all of the fresh foods, many other commodities—such as timber, charcoal, soap, and meat, the latter through ethnically based divisions of labor—and much of its infrastructural repair services. Of course, the local food products and services do not compete out the instant noodles (Errington 2013), imported wheat flour products, Asian rice, tinned milk, and other global products of the urban diet. But local and regional foods still figure very prominently in the kind of diet to which people gravitate, especially for collective meals and purchased individual sustenance during a day of work, as distinct from quick snacks and instant gratification. The imported products which have very little local competition are organized in similar ways. For example, the automobile spare parts market in Ibadan is a vast landscape of objects, organized by function, make, and type of vehicle.

6.4  Key Economic Functions of  African Markets African markets perform several key economic functions: first, it should be noted that the hinterlands, and urban agriculture, are the only sources for fresh vegetables and fruits, of the kind increasingly insisted upon by nutrition specialists, worldwide. Under African conditions, efficient marketing, storage, and distribution are crucial, and can probably only be achieved through large numbers of relatively small producers and traders, networked together in trusted coordination. These longstanding market systems then join up with the worldwide local food movement. There is a detailed literature on the trajectories of change in rural and urban malnutrition, but in urban areas nothing suggests that lack of physical availability, or even local price, play significant roles. It may be that urban malnutrition is more marked in cities that lack such marketing infrastructures, such as the post-apartheid cities of South Africa, than in those that benefit from them: possibly due to the food sector as income-generating as well as expenditure-demanding. Second, one of the greatest economic contributions of Africa’s market systems is in generating employment on a continent where there has never been an industrial revolution to create regularly paid jobs, at however low a wage. In studies of the so-termed “informal sector,” it is important, for certain analyses, to separate out the urban supply system because all logic suggests that it will be a continuing frontier of growth: of innovation in commodities traded and of publics reached, and thereby generative of demand (e.g. for transport and energy services) and of a monetized value whose velocity of circulation will be very high. If money capital is put aside by individual practitioners, all evidence suggests that the cash itself circulates

Markets and Urban Provisioning    111 within the same regional networks, through the savings and loan group functions, rather than being syphoned upwards into financialized centers. In the context of new institutions of micro-credit and other formal credit opportunities, the people’s own commentaries, and even formal study, suggest that the flexibility in the timing of repayment, which is readily accommodated within these old modes of niche organization, are better suited to the necessary seasonality, fluidity, and uncertainty of the urban provisioning system than formalized, dated debt systems. Third, the proximity of the huge urban markets, with their increasingly multi-ethnic popu­lations, is a stimulus to innovation in the regional cropping and transformation systems. More could be achieved by encouraging the artisanship of transformation with respect to new commodities, given African farmers’ long history of domestication of imported cultigens (cassava, maize, tomatoes, cocoyam, and others). The demand for inputs and small manufacturing facilities in the rural areas may be a considerable unmet need: such as different kinds of fertilizer, oil-seed preparation and bottling, drying and packing facilities for crops that can be stored and then prepared for consumption from the dried form, packaging, and so on. Fourth, the intense daily practice of varied forms of collaboration and coordination is a sine qua non of economic growth in the twenty-first century. Not all the possible forms of the old combination of patron–client/apprenticeship and association membership will necessarily be politically and socially beneficial under all circumstances. But their forms and implications may be very varied: by country, product-line, security conditions, and concomitant money management practices and institutions. Skills in organization are a very significant form of human and social capital in African marketing systems, and in the dynamics of innovation everywhere: all the more so if scalar variation and inter-linkage become increasingly characteristic of the economies of the future. Fifth, these human and social capital systems can be extrapolated into completely new lines of business: beyond the food, fuel, and artisanal goods supply functions of the past; the tools, household goods, pharmacy goods, and spare parts and repairs of the twentieth century; and the rapidly growing electronics, recycled goods, entertainment, and other functions of the present.

6.5  Conclusion: Dangers for the Future The urban supply and marketing systems of Africa need recognition and study as institutions with a future. Passing over them, even simply by assimilating them to “the informal sector”, tends to downplay all these important functions that they fulfill in favor of highlighting their still-undefined place in the formal, monetized and financialized, regulated and taxed sectors of the economy. Energy prices necessarily have a strong effect, because all goods in regional marketing systems travel by vehicles that require petrol. Shortages and high prices bring great difficulties: by leaving perishable goods to rot, by raising the retail price to the urban consumer, by wasting large amounts of time for drivers and other transport personnel and thereby dislocating the necessary coordination in the market network, and by putting pressure on the intricate credit mechanisms of deferred payment and circulating capital.

112   Concepts Security on the roads is an absolute necessity for regional market systems to work. The marketplaces themselves usually have traditional roles for supervision. Unrepaired roads and bridges, highway predators, and official (and unofficial) roadblocks all have their effects on the confidence with which regional trade is undertaken. I have witnessed traders organizing amongst themselves to travel in convoy, at a regulated speed, in order to provide safety-in-numbers and to deal with any difficulties arising, as a group rather than in isolated vehicles. Finally, there is the question of land. Large changes in the land tenure, and thereby the cropping systems, in urban hinterlands can have an impact on whole systems of production, transformation, marketing, and consumption. The present changes in land ownership and use, and their orientation to new markets, is controversial and will remain an important research theme (see Deininger 2013; Peters 2013a, 2013b). Possibly the largest impact of such changes would be the diminution of the generation of livelihoods and skills through the provisioning sector, especially for women (Koopman 2009). Long-term anthropological and historical research has also documented the damage done by inconsistent land policy (Berry 1993; Cliggett 2014). It is therefore important to recognize, and to study closely, the ramifications of the market systems—as marketplaces and as economic markets—as they now function, and as they cultivate skill-sets whose social organization builds on the great achievement of urban food supply in earlier centuries and, in the twentieth century, with what may be the fastest urban growth rates in human history.

References Andrae, G., and Beckman, B. (1985). The Wheat Trap: Bread and Underdevelopment in Nigeria. London: Zed Books. Berry, Sara. (1993). No Condition is Permanent. The Social Dynamics of Agrarian change in Sub-Saharan Africa. Madison WI: University of Wisconsin Press. Bohannan, Paul (1955). Some Principles of Exchange and Investment among the Tiv. American Anthropologist (New Series), 57, 1:60–70. Bryceson, D.F. (1987). A century of food supply in Dar es Salaam: from sumptuous suppers for the sultan to maize meal for a million, in J.I. Guyer (ed.), Feeding African Cities. Studies in Regional Social History. Manchester: Manchester University Press, pp. 154–202. Clark, G. (1994). Onions Are My Husband. Survival and Accumulation by West African Market Women. Chicago: Chicago University Press. Clarke, W.H. (1972). In J.A. Atanda (ed.), Travels and Explorations in Yorubaland, 1854-1958. Ibadan: Ibadan University Press. Cliggett, Lisa (2014). Access, Alienation and the Production of Chronic Liminality: Sixty Years of Frontier Settlement in a Zambian Park Buffer Zone. Human Organization, 73, 2:128–140. Deininger, K. (2013). In Joseph E. Stiglitz, Justin Lin Yifu, and Ebrahim Patel (eds.), The Industrial Policy Revolution II: Africa in the Twenty-first Century. International Economic Association, 386–411. Errington, F. (2013). The Noodle Narratives:  The Global Rise of an Industrial Food into the Twenty-First Century. Berkeley: University of California Press.

Markets and Urban Provisioning    113 Fairhead, J., and Leach, M. (1995). False forest history, complicit social analysis: rethinking some West African environmental narratives. World Development, 23, 6:1023–1035. Guyer, J.I. (1987). Feeding Yaoundé, capital of Cameroon, in J.I. Guyer (ed.), Feeding African Cities: Studies in Regional Social History. Manchester: International African Institute and Manchester University Press, pp. 112–154. Guyer, J.I. (1997). An African Niche Economy:  Farming to Feed Ibadan, 1968-88. Edinburgh: Edinburgh University Press and the International African Institute. Guyer, J.I. (2004). Marginal Gains. Monetary Transactions in Atlantic Africa. Chicago: University of Chicago Press. Guyer, J.I., Denzer, L., and Agbaje, A. (eds) (2002). Money Struggles and City Life. Devaluation in Ibadan and Other Urban Areas in Southern Nigeria, 1986–96. Portsmouth, NH: Heinemann, Bookbuilders, Nigeria. Hansen, K.T. (2000). Salaula:  the World of Secondhand Clothing and Zambia. Chicago: University of Chicago Press. Harms, R. (1981). River of Wealth, River of Sorrow: the Central Zaire Basin in the Era of the Slave and Ivory Trade, 1500-1891. New Haven: Yale University Press. Hill, P. (1989). Market places, in J. Eatwell, M. Milgate, and P. Newman (eds), The New Palgrave. Economic Development. New York: W.W. Norton, pp. 238–342. Hodder, B.W. (1965). Distribution of markets in Yorubaland. Scottish Geographical Magazine, 81(1):48–58. Inikori, J. (2002). Africans and the Industrial Revolution in England: a Study in International Trade and Economic Development. Cambridge: Cambridge University Press. Johnson, M. (manuscript.). Personal communication. Cloth strip currencies. Jones, W.O. (1972). Marketing Staple Food Crops in Tropical Africa. Ithaca:  Cornell University Press. Kilel, Beatrice. (2014). Africa. Last Investment Frontier: A Middle Class Approach. Frederick, Maryland: Zapphire Publishing. Koopman, J.E. (2009). Globalization, gender and poverty in the Senegal valley. Feminist Economics, 15(3):253–285. Law, R. (1995). Cowries, gold and dollars: Exchange rate instability and domestic price inflation in Dahomey in the eighteenth and nineteenth centuries, in J.I. Guyer (ed.), Money Matters. Instability, Values and Social Payments in the Modern History of West African Communities. Portsmouth, NH: Heinemann, pp. 53–73. Levy, J. (2012). Freaks of Fortune. The Emerging World of Capitalism and Risk in America. Cambridge MA: Harvard University Press. Mataen, D. (2012). Africa—The Ultimate Frontier Market: A guide to the business and investment opportunities in emerging Africa. Petersfield: Harriman House. Mintz, (1985). Sweetness and Power. The Place of Sugar in Modern History. New York: Viking. Mosley, P. (1987). The development of food supplies to Salisbury (Harare), in J.I. Guyer (ed.), Feeding African Cities. Studies in Regional Social History. Manchester: Manchester University Press, pp. 203–224. Pearson, S.R. et  al. (1981). Rice in West Africa. Policy and Economics. Stanford:  Stanford University Press. Peters, Pauline E. (2013a). Conflicts Over Land and Threats to Customary Tenure in Africa. African Affairs, 112/449: 543–562. Peters, Pauline E. (2013b). Land appropriation, surplus people and a battle over visions of agrarian futures in Africa. The Journal of Peasant Studies, 40:3, 537-562.

114   Concepts Robertson, C.C. (1983). The death of Makola and other tragedies. Canadian Journal of African Studies, 17(3):469–495. Robertson, C.C. (1997). Trouble Showed the Way: Women, Men, and Trade in the Nairobi Area, 1890-1990. Bloomington IN: University of Indiana Press. United Nations Department of Economic and Social Affairs, Division of Population (2011). World Urbanization Prospects. New York: United Nations. Van Leynseele, P. (1983). Les Libinza de la Ngiri: L’anthropologie d’un people des marais du confluent Congo-Ubangi. Leiden: African Studies Center. Wan, M.Y. (2001). Secrets of success: uncertainty, profits and prosperity in the gari economy of Ibadan, 1992–94. Africa, 71(2):225–252. Watts, M. (1987). Brittle trade: a political economy of food supply in Kano, in Jane I. Guyer (ed.), Feeding African Cities. Studies in Regional Social History. Manchester: Manchester University Press, pp. 55–111.

Chapter 7

Devel opme nt as Diffu si on  Manufacturing Productivity and Africa’s Missing Middle

Alan Gelb, Christian J. Meyer, and Vijaya Ramachandran

7.1 Introduction This chapter considers the economic development of sub-Saharan Africa1 from the perspective of slow convergence of productivity, both across sectors and across firms within sectors. Why have “productivity enclaves,” islands of high productivity in a sea of smaller low-productivity firms, not diffused more rapidly to create the “middle” so frequently observed to be missing in Africa’s economies? Why have productive firms not led a process of equalizing growth, absorbing land and labor to employ a larger share of the workforce? At Africa’s current rates of formal job creation, low-productivity service and largely informal household businesses will need to keep absorbing workers indefinitely (Fox and Gaal 2008). Development economists have long recognized that large cross-country differences in output per capita between rich and poor countries cannot be explained simply by varied accumulation of physical and human capital. Instead, they largely reflect differences in an unmeasured productivity residual, usually expressed as total factor productivity (TFP). Efforts to explain large differences in levels and change of this residual have moved beyond analysis at the aggregate level to include studies of productivity differences between sectors and, more recently, between heterogeneous firms within single sectors.

1 

Throughout the rest of this chapter, “Africa” refers to sub-Saharan Africa.

116   Concepts In a productivity-centered view of economic development, dating back at least to Clark (1940) and Lewis (1954), progress is marked by a process of convergence. Production factors, notably labor but also capital and land, migrate from “traditional” low-productivity sectors and firms towards “modern” or higher-productivity activities. In this chapter, we take stock of the literature and see how African economies have fared in this process. We focus on the manufacturing sector and on firms within the manufacturing sector, because its growth has long been argued to play a special role in this convergence process (Kaldor 1966; Cornwall 1977; Rodrik 2013). That does not mean, however, that firms in other high-productivity sectors, particularly services, should be neglected. Seeing economic development simply as a process of factor allocation assigns a somewhat passive role to the firms and entrepreneurs expected to grow the modern parts of the economy. Convergence towards the upper end of the productivity spectrum can equally be seen as a process of diffusion, wherein the economic agents driving the high-productivity parts of the economy actively extend the reach and scope of their businesses until they encompass the bulk of the economy—and ultimately increase economic growth and output per capita. Who are these agents? How do they obtain the capital and the capabilities, including skills, knowledge, and management capacity to build and grow high-productivity firms? How do the business environment and the political economy of state–business relations in Africa affect their incentives to compete, grow, and invest—domestically and across borders in Africa? Can these factors help explain the “missing middle” of Africa’s economies? We approach these questions with a mix of firm survey results, country case studies, and more anecdotal evidence. Section 2 provides an overview of literature relating economic development to productivity convergence across both sectors and firms. Section 3 considers factors from the literature in the context of Africa. While every country is different, some features are widely shared. In particular, many economies in Africa are still poor, unequal, and slow to initiate a process of convergence at the sector level. Across firms, we find that the manufacturing sector tends to have a high dispersion of labor productivity. Section 4 considers three sets of factors that have contributed to slow productivity convergence. They are interrelated, with roots in Africa’s distinctive geography as well as the historical processes that have shaped the polities and economies of today. First, we analyze distortions caused by a poor business climate, which can constrain the allocation of factors between sectors and firms in ways that can delay productivity convergence. In extreme cases, an economy can be reduced to a combination of subsistence activities and a few high-productivity enclaves, such as offshore oil wells that operate within their own security and regulatory environments. Second, we address the political economy of the complex and often difficult relations between government and business in small concentrated African economies. In this context, we consider the incentives to reform the business climate. On the side of government, one strand of analysis argues that the distinctive historical process of African state formation has not created strong incentives for states to develop the social contracts needed to underpin effective states or to acquire the capabilities needed for effective management of the economy. As a compounding factor, the small size of most African economies has led to high concentrations of market power and to powerful groups, often with close relationships to governments and an interest in preserving the status quo. This has resulted in high de facto barriers to entry and expansion and has thus slowed efforts to reform.

Development as Diffusion    117 A third influence on convergence is “agency,” the characteristics of the few leading firms and the processes through which they have acquired their knowledge of market opportunities and production. History has left many African countries with a highly unequal distribution of domestic management and commercial expertise that is often heavily concentrated within particular groups. This imbalance has further complicated state–business relations and slowed the diffusion of modern business methods and technologies. Section 5 concludes.

7.2 Literature Review The consensus view in development economics is that the large cross-country differences in output per worker, most commonly measured in gross domestic product (GDP) per capita, cannot fully be explained by the accumulation of physical and human capital. The remaining productivity residual has been approached at three levels: at the macroeconomic level, through examining sectoral patterns, and by considering the causes of productivity differences between individual firms. At the macroeconomic level, the standard neo-classical growth framework attributes the remaining output differences to an unmeasured residual called TFP or multi-factor productivity (MFP). The relative importance of factor accumulation and productive efficiency in explaining whether a country is rich or poor has been the subject of much debate.2 The change and levels of this “measure of our ignorance” (Abramovitz 1956: 11) has been conceptualized in a range of different models, including as the adoption of new production technology, international knowledge spillovers, and technological diffusion (Grossmann and Helpmann 1991; Aghion and Howitt 1992; Parente and Prescott 1994).3 By comparing cross-sectoral differences between poor and rich countries, we can relate the productivity literature to classic models of structural change, in which output and employment slowly shift away from the primary sectors (traditional agriculture, fishing, and mining) into the manufacturing sector and finally into services (Clark 1940; Lewis 1954; Chenery 1960; Kuznets 1966; Baumol and Bowen 1966). The characteristics of this process can be summarized by a few well-established empirical facts.4 The share of agriculture in 2  The consensus view appears to be that the productivity residual typically accounts for 50 percent or more of cross-country differences. Hall and Jones (1999) decompose differences in the level of output per worker between the United States and Niger in 1988 and largely attribute the 35-fold difference to a productivity residual. More recently, Jones and Romer (2010) revisited the “Kaldor facts” (Kaldor 1961) and found that less than half of cross-country differences in per capita GDP can be explained by measured inputs. 3  Klenow and Rodríguez-Clare (1997) summarized some of the literature that has emerged in response to the “neo-classical revival”, most importantly following the seminal empirical contribution by Mankiw, Romer, and Weil (1992), who found that a Solow model augmented with human capital can explain 78 percent of income per capita differences across countries. Easterly and Levine (2001) review the growth accounting literature and establish central empirical facts. Caselli’s (2005) development accounting exercise offers an excellent literature survey and provides a useful analytical framework. Finally, Hsieh and Klenow (2010) pointed to some of the open questions in the literature. 4  Herrendorf, Rogerson, and Valentinyi (2013) offer a recent synthesis of the structural transformation literature.

118   Concepts economic output or employment tends to decrease with economic development; that of services tends to increase. The output and employment shares of the manufacturing sector tend to follow an inverse U-shaped curve, first increasing and then falling as income rises. The movement of economic activity from agriculture to higher-productivity manufacturing tends to be associated with episodes of faster economic growth. In this process of structural transformation, the composition of a country’s exports plays an important role. Johnson et al. (2010) compared African countries today with historical cases of countries that were similarly poor or institutionally weak, but managed to sustain rapid growth. In almost all cases, they found that growth is associated with an increase in manufactured exports. The potential importance of the manufacturing-export link is mirrored in Hausmann et al. (2007), who measured the productivity level associated with a country’s export basket and found a positive relationship between the initial level of export sophistication relative to income and the subsequent rate of economic growth. Lederman and Maloney (2012) have challenged the interpretation of this result as supporting highly selective policy, arguing that including the share of investment in GDP or a measure of export concentration eliminates the impact of export sophistication on growth. Nevertheless, the confluence of these factors suggests that increasing shares of progressively more advanced manufactured exports has been associated with sustained growth processes. Considering countries at higher income levels, where service sector growth dominates, Bernard and Jones (1996a, b) examined the role of intersector productivity differences in aggregate productivity convergence. In 14 OECD countries, they find significant differences in TFP growth rates between sectors. They argued that productivity convergence in the service sector, combined with a declining share of manufacturing in total value added, likely contributes to aggregate productivity convergence. Even within narrowly defined industries, productivity differences across firms and across countries appear to be large and persistent, and more recent research has relaxed the assumption of homogenous firms within sectors. Baily and Solow (2001) summarized studies of industrialized countries conducted by the McKinsey Global Institute, which found a large degree of variation even within narrow manufacturing industries across six countries. For the USA, Syverson (2004) found that even at the narrow four-digit Standard Industrial Classification level, total factor productivity for plants at the 90th percentile of the productivity distribution is on average almost two times as high as for plants at the 10th percentile. Bartelsman et al. (2013) analyzed firm-level data from the USA and seven other countries in Western and Central Europe. They confirm large within-industry dispersion in total factor productivity and establish an even larger dispersion in labor productivity.5 Research has started to bridge the gap between firm-level productivity differences and aggregate TFP.6 Restuccia and Rogerson (2008) modeled how distortions in the allocation of production factors across heterogeneous firms can have an impact on aggregate output. Hsieh and Klenow (2009) argued that the higher dispersion of productivity across firms in single industries in China and India reflects greater misallocation of resources rela­ tive to the USA. The dispersion in allocative efficiency across firms can be an important

5 

Syverson (2011) surveys the empirical literature on measured productivity differences across firms. Restuccia and Rogerson (2013) provide a recent survey of the literature that links aggregate productivity to the allocation of production factors across firms. 6 

Development as Diffusion    119 driver of aggregate productivity because factors are essentially “bottled up” within less productive firms. Drawing on the seminal contribution of Lucas (1978), which considers the effect of different levels of entrepreneurial and management “talent” on the size of firms, Bartelsman et al. (2013) analyzed the relationship between firm size and productivity and the impact of firm-level distortions on aggregate output. In the absence of firm-specific distortions, factors of production will be allocated to the most productive firms. The largest firms in an industry will be the most productive, causing the weighted average level of productivity to be greater than average firm-level productivity. The size–productivity relationship is stronger in more advanced countries, which they suggest is a result of a less distorted business environment. If, however, policy-induced distortions are abundant, they will impede the factor allocation process and affect aggregate productivity through firm selection. Distortions have an even stronger negative impact if they are correlated with firm size and visibility, encouraging productive firms to “fly beneath the radar screen.” However, if some firms with privileged positions can negotiate special deals that reduce distortions, the result could also be a “missing middle” with smaller, less productive firms coexisting with a set of more productive, yet constrained firms. The literature suggests several factors that resonate with conditions in Africa’s manufacturing sector and that could hold back a more rapid convergence of productivity. We turn to some of them now.

7.3  Is Africa Different? Africa does not lack productive sectors and firms, even in its low-income economies. Some are in the extractive sectors and reflect specific natural resource endowments of mineral and hydrocarbon deposits. Some other sectors with high labor productivity, such as public utilities and finance, have low employment generation potential. However, Kenya’s agribusiness firms rank among the world’s most competitive—in an industry that requires sophisticated production technology and logistics. The country’s M-Pesa payments system has become a world leader in cell-phone banking. Even economies with very problematic business environments have a limited number of highly productive firms. The SAB-Miller beer factory in South Sudan is reportedly very efficient and employs a workforce of over 400, almost all South Sudanese nationals. It is also the only larger-scale private firm in the whole country. As further described, in many countries firms with apparently high productivity (apparently because measured productivity may partly reflect monopoly profits) coexist with sectors and subsistence enterprises with very low productivity.7 Most of Africa’s economies still have large shares of output and employment in the agricultural sector, which has far lower average productivity than the agricultural sector in rich countries (Timmer et al. 1988). Global agricultural productivity growth has accelerated

7  Modern, high-productivity business is not confined to manufacturing. Tourism, for example, has been Africa’s fastest-growing export. The obstacles to its growth resemble those for the manufacturing sector, including costly logistics, heavy regulation, and limited access to land (Christie et al. 2013).

120   Concepts over the last decades (Block 2010) but Africa’s overall agricultural productivity has lagged far behind the rest of the world. Over the last decades, emerging economies such as Brazil, Mexico, and Indonesia have experienced positive structural change where large increases in agricultural labor productivity coincided with a falling share of agricultural employment. With the exception of South Africa, many economies in Africa have seen their average agricultural labor productivity either increase slowly or stagnate (Senegal and Zambia). Agriculture’s employment share increased in Zambia and Nigeria between 1990 and 2005 (McMillan and Rodrik 2011). The gap in agricultural labor productivity to the rest of the world has increased even further. Despite the reforms that African countries have undergone since the structural adjustment phase of the late 1980s, McMillan and Rodrik (2011) found that between 1990 and 2005, Africa’s structural change ran counter to the expected pattern of structural convergence. In contrast to rapidly growing Asian countries, labor moved from high- to low-productivity activities. They also show that economies with a revealed comparative advantage in extractives are at a disadvantage. The larger the share of natural resources in exports, the smaller is the scope of productivity-enhancing structural change. Even though “enclave” extractive sectors may have high labor productivity relative to the rest, they cannot absorb the surplus labor from agriculture; much of this has thus moved into low-productivity services. More recent data suggest that there may be a turnaround, with a positive growth contribution from structural change over 2005–2010 (McMillan 2013). Even so, over the long term, globalization appears not to have fostered a desirable pattern of structural change in Africa. Figure 7.1 illustrates the diverging longer-term trajectories of sectoral productivity and employment shares in Zambia and Mexico. This convergence failure has left African countries with high levels of inequality in intersectoral productivity. To estimate the dispersion in sectoral labor productivity, we calculate a Gini coefficient of sectoral productivity based on national accounts data, weighted by sector employment shares. We find a high sectoral dispersion in Africa. Excluding Mauritius, the Gini coefficient averaged about 0.5 in 2005, relative to about 0.35 for other regions.8 Figure 7.2 (upper panel) illustrates the strong negative correlation between economy-wide productivity and inequality of intersectoral productivity. Comparable firm-level data on the evolution of productivity distributions over long periods of time is lacking for African countries, but cross-sectional comparisons based on recent data suggest a high degree of dispersion, both within the formal manufacturing sector and between the formal manufacturing sector and the rest of the economy. As an indication of the dispersion within the formal sector, we use firm-level data from the World Bank Enterprise Surveys project to calculate a Gini coefficient of value added per worker.9 As would be expected, intrasectoral dispersion is lower than intersectoral dispersion, but firm-level productivity Gini coefficients are still relatively high for the surveyed African 8 

Africa’s Gini would be even higher if allowance is made for the exceptionally high level of unemployment in South Africa; including the unemployed as a zero-productivity sector boosts South Africa’s productivity Gini from below 0.3 to over 0.4, relatively high for a middle-income country (see Figure 7.2, upper panel, point ZAF-A). Data for these calculations comes from McMillan and Rodrik (2011). 9  Calculations in this paragraph are based on a harmonized data set of 15,108 firms in 41 countries across sub-Saharan Africa, surveyed between the years 2006 and 2011. Data comes from the World Bank Enterprise Survey project and covers formal manufacturing firms with more than five employees.

Development as Diffusion    121 0.8 agr

Zambia (2005) agr

Sectoral employment share

0.6 agr

Mexico (1950)

0.4

Zambia (1950)

0.2

man

agr man man man

0 6

7

8

min min

9

Mexico (2005) min

min 10

11

Log sectoral labor productivity (2000 PPP dollar)

Figure  7.1   Labor productivity and employment share for selected sectors, Zambia and Mexico. Notes: agr = agriculture; man = manufacturing; min = minerals. Source:  Authors’ calculations, based on McMillan and Rodrik (2011).

countries (Figure 7.2, lower panel).10 To a large extent, Africa’s formal manufacturing sector appears to be dominated by a limited number of larger firms with higher labor productivity that coexist with a “long tail” of lower-productivity firms. Further evidence of productivity dispersion comes from comparing manufacturing labor productivity and labor cost per employee with overall levels of income, measured by GDP per capita. Given their low levels of income, we would expect African economies to exhibit both low labor productivity levels and low levels of labor costs relative to richer comparators. We find that this is generally the case, and especially in Ethiopia where labor costs are low relative to other African countries. At the same time, firm-level survey data suggest that formal manufacturing firms in Africa are generally both more productive in terms of value-added per employee and higher paying than firms in other regions after adjusting for levels of GDP per capita.11 Kenya and 10 

While the pattern conforms to a general tendency for productivity dispersion to be higher for manufacturing sectors in lower-income countries, much of that tendency is due to the overwhelming concentration of African countries in the low-income range. African Gini coefficients remain high after removing a number of countries where outlier observations of implausibly high-productivity firms might have caused the Gini to be excessively large. 11  See Gelb et al. (2013) for further details. They find that labor costs are 84 percent higher than expected on the basis of GDP per capita and including a range of size and sector dummies. Introducing labor productivity as an additional variable reduces the “Africa effect” by half but still leaves a substantial and significant markup relative to other developing regions.

0.7

Sub-Saharan Africa Rest of the world

Gini coefficient of sectoral labor productivity (weighted by employment share)

ZMB NGA

0.6 MWI

SEN

0.5

CHN THA

IND

VEN KEN

0.4 ETH

GHA

0.3

BOL

IDN PER BRA PHL COL

0.2 7

8

ZAF-A KOR TURMYS

ZAP SWE CHL HKG USA MEX CRI ARG GER SGP NED FRA DEN MUS FIN ITA

9 10 Log mean labor productivity

11

Gini coefficient of value added per worker

1.0

0.8

0.6

0.4 Sub-Saharan Africa Rest of the world

0.2 5

6 7 8 9 Log GDP per capita (constant 2005 US$)

10

Figure 7.2   Gini coefficient of weighted productivity distribution vs. average productivity based on macro data (upper panel); Gini coefficient of value added per worker and GDP per capita based on firm survey data (lower panel). Note:  Point ZAF-A calculated by adjusting South Africa’s productivity distribution for unemployment, assuming zero labor productivity for the unemployed. Source:  Authors’ calculations, based on McMillan and Rodrik (2011), Statistics South Africa, and World Bank Enterprise Surveys.

Development as Diffusion    123 Bangladesh, for example, are not too dissimilar in their levels of GDP per capita, but both labor productivity and labor cost per employee are far higher in Kenyan firms than in those in Bangladesh.12 Söderbom and Teal (2004) considered a similar question for Ghana. They show that firms face a steeply upward-sloping labor cost schedule as they grow larger. This constrains their growth and forces workers to find employment in small informal firms, which in turn contributes to a more dualistic economy. One explanation could be the scarcity of skills needed to work in large firms; another could be a tacit agreement to share part of the productivity rents with employees to maintain industrial peace. Gelb et al. (2013) did not find a distinctive size–labor-cost effect when comparing formal African manufacturing firms with those in comparator countries, but find that employee costs increase somewhat faster in response to increased labor productivity in African countries than in others. This, together with generally high levels of manufacturing labor productivity in Africa relative to GDP per capita, confirms the general thrust of the Söderbom–Teal hypothesis that the manufacturing sector (or at least parts of it) demonstrates productivity enclave-like characteristics.13 This is confirmed by Iacovone, Ramachandran, and Schmidt (2013) who found that African manufacturing firms, at any age, tend to be about 20–24 percent smaller than firms in other regions of the world. African firms that start small remain small, rather than converging towards the “missing middle.”14 High firm productivity dispersion should, ceteris paribus, be reflected in the labor market through wages.15 In the absence of strongly redistributive fiscal policy, high wage dispersion could then lead to higher measured income (and consumption) inequality. At the macro-level, this association seems to hold for Africa. Cross-section data on inequality conform reasonably well to an inverted U-shaped curve when African countries are excluded. Inequality is higher in middle-income countries, notably those in Latin America, than in either rich or poor countries. Africa does not conform well to this global pattern. Measured inequality is higher than in other low-income countries, with Gini coefficients comparable to those in Latin America despite Africa’s far lower income levels (Figure 7.3).16, 17

12

  The difference in real employee costs per worker relative to poorer comparators reduces somewhat when account is taken of the higher price levels in Africa as shown by purchasing power parity indices, but this still leaves the formal manufacturing sector as a productivity and pay enclave relative to the rest of the economy. 13   These patterns could of course reflect large sampling differences between the Enterprise Surveys in African and other countries, if the former selected larger firms. However, in every size category the average African firm is smaller than its counterpart abroad. It is therefore unlikely that this constitutes the reason for the observed differences. 14   Firm-level survey responses on the business environment, including access to finance and land, and the availability of power supply and skilled labor, have some explanatory power in explaining this difference, as do foreign ownership, export status of the firm, startup size and the size of the market. However, even after controlling for these variables, about 60 percent of the size gap between African firms and those in other countries remains unexplained. 15   Faggio, Salvanes, and van Reenen (2010) provide a recent empirical study of the link between wage inequality and productivity dispersion for a panel of firms in the United Kingdom. 16  African distributions are also usually measured on the basis of consumption expenditures, which tend to reduce levels of inequality relative to measures based on income as in Latin America. 17  Even South Africa, with its large formal economy, is notably unequal. Its formal economy coexists with a very low productivity “survivalist” informal economy and unemployment estimated at around 30 percent of the labor force. Its structural characteristics are reflected in very high levels of income

124   Concepts Africa Asia Latin America West

Gini coefficient

80

60

40

with Africa without Africa

20 4

6 8 10 Log GDP per capita, PPP (constant 2005 international $)

12

Figure  7.3   Gini coefficients of income and consumption inequality vs. GDP per capita (1950–2011). Note: All years shown; Eastern Europe excluded. Line is a locallyweighted regression plot (bandwidth = 0.6). Data source: Milanovic 2013.

7.4  Why Do Factors of Production Not Move to Reduce Productivity Imbalances in Africa? We consider three sets of factors that slow the diffusion of productivity. We start with the business climate, which imposes high external costs that may not be uniformly distributed across firms. Then we move on to an examination of the often complex business–government relations in Africa’s small markets. Finally, we consider the role of firm ownership structures and the unequal distribution of management and technical capacity.

7.4.1  The business climate Empirical research, anecdotal evidence, and the perceptions of firm owners and man­ agers suggest that the business climate matters to productivity, as well as to the survival and growth of firms. inequality, especially when measured on the basis of pre-tax market income and excluding highly redistributive transfers that account for about 3 percent of GDP and two thirds of the income of the poorest quintile of the population (Woolard, 2010). Its market income Gini coefficient has been estimated at 0.77 (Finn et al. 2012), above comparably measured Ginis for other countries known for their high inequality.

Development as Diffusion    125 Constraints imposed by the business climate such as power outages and the burden of regulation are recognized as “major” or “serious” by most African firms. Self-reported losses associated with power outages can amount to more than 10 percent of sales in some countries. Concern over power supply is no less in larger firms because of the very high cost of self-generated power. Behind power, bad transport networks emerge as a second infrastructure concern. Around one third of firms cite transportation as a major or severe constraint. Firms also report having to pay bribes to get things done. On average across firm surveys in Africa, around 40 percent of firms confirmed that these practices were common, with fewer in South Africa and more in other countries, including Kenya where the share exceeded 60 percent. Transport costs not only reflect problems in the supply of physical infrastructure, but also more complex political economy issues related to the business environment. In many African ports, cargo dwell times are about two weeks, compared to under a week in Asia, Europe, and Latin America (Beuran et al. 2012). Long dwell times are not just the result of port inefficiency but may be linked to businesses’ inventory management practices: cargo is sometimes purposely left at the port when the cost of clearing it immediately is high. There are storage-cost savings, especially when firms cannot immediately sell their imported goods. Terminal operators may also benefit from longer dwell times by receiving informal payments. Positive examples exist: in Durban, a strong domestic private sector with global trade interests and a government willing to support the business sector have helped to reduce dwell times. Several studies indicate the adverse impact of business climate distortions for productivity. Eifert et al. (2008) distinguished between “factory-floor” productivity and overall productivity. They define “gross” value added as sales less the cost of raw materials, and “net” value added excluding external costs such as power, transport, licensing fees, and bribes. African firms appear substantially less productive, relative to firms in comparator countries, when these “indirect costs” are included. Kenyan firms, for example, have about the same factory floor productivity as firms in China but only about half of the overall productivity. Harrison et al. (2012) concluded similarly that the productivity of African firms is not less than that of firms in other countries once allowance is made for the quality of the business climate. All of these studies should be qualified to the extent that, as already discussed and in the next section, many larger firms probably reap monopoly rents because of high shares in small domestic markets. Costly business climates, if they impact similarly on all firms, will reduce investment and growth in general, both for small and large firms and for all sectors. They can be very costly, but there is no reason why they should create significant differences between firms. The long-run equilibrium productivity and size distributions could be either less dispersed or more dispersed. However, to the extent that inputs of non-traded goods reflect difficulties in providing essentials like power, transport, security, and enforceable contracts, Africa could be relatively worse at producing non-traded goods and services than other regions. Transactions-intensive firms are therefore crowded out. When costs imposed by the business climate are very high, there is a tendency for economies to degenerate into a large number of subsistence enterprises and very few productive enclaves that are able to survive. One extreme example for this would be Equatorial Guinea, which ranks 166 out of 189 in the 2014 World Bank’s Doing Business composite

126   Concepts index but has a highly dynamic offshore oil sector that is not dependent on non-traded goods and services.18 Policy uncertainty may be as problematic a factor for firms as the average level of the business climate. Hallward-Driemeier et al. (2010) found high degrees of variability in firm-level responses to business surveys, suggesting that intracountry variation in business conditions can be larger than intercountry differences. Rather than dealing with predictable—if costly—de jure or de facto policies, firms face a series of unpredictable deals that create high levels of uncertainty and reduce growth. Smaller firms are less likely than large firms to believe that the implementation of policies is consistent and predictable. Firms with the capacity to become large or more capital intensive are therefore likely to do so if they have preferential access to deals to accommodate the uncertainty of policy implementation. In sum, the burden of a bad business climate does not fall uniformly on all sectors or businesses within sectors. The manufacturing sector is more crowded out than other sectors that are less dependent on non-traded inputs. Within the manufacturing sector, it appears that there is a “missing middle” as well. Subsistence firms are immune while very big firms have the capacity and bargaining power to “deal”. To the extent that this is the case, the formal middle will be squeezed out.19

7.4.2  Business–government relations in  Africa’s small markets Africa’s often ambivalent relations between business and government have been shaped by factors with interrelated roots in its geography and history. Populations are sparse, with inhabitants often scattered far from the coast. Its small economies are still sparser, with output per square kilometer only about 8 percent that of India or China.20 Africa’s states are also relatively new, with artificial borders set in the colonial period. Following the decision of the Organization of African Unity to endorse existing borders rather than open up a contentious process of redrawing them, they have been maintained almost without exception since independence. Herbst (2000) argued that the process of African state formation, a very different one from the classic European model as set out by Tilly (1990) and others, has reduced incentives to invest in state capacity. Throughout a thousand-year struggle to survive, European states had to develop effective institutions to raise fiscal revenue to fund armies and defend their territory. Taxation led to representation 18  Within the formal manufacturing sector, there is also a tendency for the ratio of value added to sales to be higher in Africa, suggesting that firms are less able to use the market to improve efficiency through subcontracting and outsourcing. 19  This is not only true for the manufacturing sector. Services similarly may have a missing middle: “Investment in [Zambian] tourism has been stifled by high costs in terms of both time and money and the lack of predictability of licensing and administrative requirements to open and operate a tourism business. [As a result] two thirds of hospitality establishments remain unclassified and many inhabit the informal economy” (Cattaneo 2007: 214). 20  Even in small, densely populated countries like Rwanda or Burundi income density does not approach that of India or China. It is of course higher in small island economies like Mauritius and Cape Verde.

Development as Diffusion    127 and social contracts of mutual accountability between states and their citizens. With low population density, open land frontiers, and frozen borders, African states have not faced such a Darwinian struggle to survive and so have not developed comparable institutions to underpin state capacity and accountability. While the nuances in this thesis can be debated, including the contributions of their respective colonial experiences (Acemoglu et al. 2001; Robinson 2002), the combination of small market size and low capacity has reduced the attractiveness of African countries to potential investors outside resource sectors. Combining rankings for GDP and business climate (measured by the Doing Business composite index), only eight African states make it into the top 100 of 173 countries. Of the bottom 50, 38 are African; the rest are mostly microstates or countries with very problematic governance conditions and special circumstances such as Afghanistan. Not surprisingly, competition is limited in many African product markets. In World Bank Enterprise Survey data for the formal sector in Kenya, the five largest firms accounted for 58 percent of total value added. In the Mozambique survey, even after excluding the five largest firms in the sample, the next five accounted for 47 percent of residual value added. Older Enterprise Survey data asked firms to classify themselves as “influential” or “not influential” in terms of their relationships with government. The former self-reported market shares for their main products as around 40 percent but even the latter group reported substantial market shares, probably because of limited market integration within individual countries. Detailed enterprise maps constructed for Tanzania, Zambia, Ghana, and Ethiopia (Sutton and Kellow 2010; Sutton and Kpentey 2012; Sutton and Olomi 2012; Sutton and Langmead 2013) analyzed the origins of key capabilities by focusing on the 50 most significant firms in the economy. Typically, each sub-sector is dominated by a handful of firms with a very small number responsible for the bulk of exports in every significant product category. Weak capacity and monopoly power bear on the question raised by Bräutigam et  al. (2002) of why business–government partnerships to foster growth are so rare in Africa. Echoing the conclusion of Himbara (1994) for Kenya, they found that the capabilities of the state matter a great deal for the ability to implement a pro-growth strategy. Even a pro-business, democratically elected government in Zambia in the early 1990s did little to improve business–government relations, while in Zimbabwe the presence of an authoritarian government meant that only “state elites” could survive in the private sector.21 It is not clear that consultative mechanisms have helped to improve business climate conditions for firms in general. Page (2013) offers an assessment of the Presidential Investors’ Advisory Councils established in 2001 with the support of the World Bank and International Monetary Fund (IMF). These public–private coordination mechanisms were expected to let leaders hear from successful businesses, identify constraints to investment, generate recommendations for action, and reinforce and accelerate policy reform. While the assessment is not entirely negative, on balance the Councils have advanced little in their main objectives, failing to secure sufficient engagement and initiative from either the government or investor side. In those few cases where the process led to specific recommendations, these have not been taken up. 21  Bräutigam et al. (2002) note the counter-example of Mauritius, where business and government came together to boost economic performance. For a survey of state–business relationships and the (limited) empirical research on Africa, see contributions in te Velde (2010) and (2013).

128   Concepts These studies suggest that many countries have been locked into a low-level business climate equilibrium sustained by the incentives faced by key participants. On the side of firms, small markets and monopoly rents confer an additional advantage on the big players, with bargaining power reinforcing the asymmetry of the business climate. “Influential” firms, including many that have benefited from decades of import substitution policy, are more prone to lobby governments, including to preserve local market power. Larger firms also have rents to share between owners, employees, and public officials. Even apparently profitable larger firms will not grow rapidly in small markets and they may find it hard to surmount the “export productivity hurdle” because measures of their productivity are exaggerated by monopoly profits on domestic sales (van Biesebroeck 2005). On the side of governments, as explained below in many countries the business sector does not have strong natural political constituencies. Emery (2003) noted that the regulatory system is often used to control the productive sectors and is structured to ensure that most firms are in violation of at least some regulation. Nugent (1995) described the example of successive Ghanaian regimes that were open to foreign investment but significantly less enthusiastic about the creation of a broad-based, indigenous private sector because wealthy indigenous businessmen were viewed as potential political rivals. The government’s ambiguity about private sector development was also reflected in public opinion polls that showed Ghanaians to be enthusiastic about democracy but less positive on market-based reforms (Bratton et al. 2001). Business has thus been left more vulnerable to swings in public policy and dependent on maintaining close relationships with government, eroding the impact of already weak competition policy.22

7.4.3  Firm ownership and management capacity A number of factors that constrain the convergence of productivity operate at the firm level. We focus on the ownership of larger-scale manufacturing firms, including the role of entrepreneurship and business networks, and on management capacity.23 History has bequeathed Africa a distinctive legacy in these areas, although naturally there are variations across countries and over time. In some countries, the private sector has been seen as “alive, doing well, and owned by the government.”24 In many others, foreign investors and ethnic minorities of European, Asian, or Middle Eastern descent play a dominant role in the leading firms or sectors, a pattern of concentrated ownership with a long historical basis that in some cases predates the colonial period. Ethnically based business networks are of course not an exclusively African phenomenon; they are prominent in many countries, including in emerging industrial powers in Asia. Nevertheless, the dominance of minority-owned business in commodities trade was recognized early on by Bauer (1954).25 Several studies, including Himbara (1994) for Kenya 22  For an in-depth study of the state of competition and particular policies and cases, see Ellis and Singh (2010). 23  We do not discuss human capital as labor input in the production process and its impact on firm productivity or aggregate outcomes. 24  The words of a USAID mission analyzing Malawi’s private sector, cited in Harrigan (2001: 38). 25  In this context, we consider minority groups that are originating from outside the continent. In countries like Vietnam, in the Andean region, or in parts of Southern Africa, indigenous minorities are in fact highly disadvantaged.

Development as Diffusion    129 and Fafchamps (2001), address the question for industry, as do a number of enterprise surveys of the formal manufacturing sector. These show that a few minority-owned firms often account for a disproportionately large share of overall value added. Surveys for 14 countries showed that minority-owned firms produced 50 percent or more of total value added, and more than 80 percent in Guinea, Tanzania, and Kenya.26 Sutton’s enterprise maps for four countries show how modest the role of indigenous private owners is in the more advanced parts of industry. Only 51 of 200 leading firms started up as domestic privately owned firms, 57 evolved from trading enterprises, 63 were foreign-owned (in some cases with state participation), and 29 were state enterprises. Especially in East Africa, the domestic private business sector appears to be overwhelmingly dominated by non-African minorities, and while the picture is somewhat more balanced in Ethiopia27 and Ghana, on average only about 17 percent of firms appear to be owned by indigenous African entrepreneurs. Significant differences exist between minority-owned and indigenous firms. The latter are significantly smaller at start-up and grow more slowly than those owned by minority entrepreneurs.28 Minority entrepreneurs show an advantage over indigenous entrepreneurs in terms of education and work experience, as well as social background: surveys show that Asian entrepreneurs were between five and 10 times more likely to have parents in the same line of business as indigenous African entrepreneurs (Ramachandran et al. 2009). Minority entrepreneurs often belong to communities that have dominated external trade and commercial relationships for generations and that have built networks and credit relationships spanning countries and sometimes continents. Some operate within family-owned groups that have diversified across sectors, partly in response to the limits to growth created by small market size. Many of Africa’s firms have come about through the transformation of “merchant capital” into “industrial capital”. Africa, unlike China, has largely not benefited from diaspora-driven investment in technology and management expertise.29 Minority entrepreneurs play a positive role in the growth of the local private sector. They bring in skills, financial resources, networking channels, and knowledge of products and markets.30 They benefit from trust-based network relationships that can compensate for shortcomings in the business climate to provide finance and knowledge and substitute for weak contract enforcement. Like multinationals, family groups can diversify against country risk more easily than indigenous investors. These are powerful advantages—and as 26

  Indigenous firms are those that are black African-owned, including by black African majority shareholders or black Africans from other countries in Africa (Ramachandran et al. 2009). For Kenya, anecdotal evidence on top incomes supports the proposition that economic opportunities are unequally distributed; many of the richest Kenyans appear to be from ethnic minority groups or connected to the families of Kenya’s presidents. 27  Ethiopia is distinctive, with a higher proportion of indigenous ownership; nevertheless, the ownership structure of its industrial firms is highly unbalanced in terms of ethnicity (Mengistae 2001). Page (2013) notes the “history of deep distrust between [Ethiopian] business and government” (p. 29). 28  The picture is different for the few indigenous firms that reach the stage of being able to diversify into other African countries; they also grow more rapidly than local indigenous firms. 29  For China, Dinh et al. (2013) note the central role of Chinese expatriate investors in building the capacity of domestic firms (p. 472). 30  Bräutigam (2003), relying on anecdotal evidence, argues that Chinese business networks have generated positive spillovers in Mauritius.

130   Concepts emphasized by Hausmann et al. (2008), agents facing less binding constraints are more likely to survive and thrive. On the other hand, beyond some point minority ownership can have negative side-effects. An ownership structure dominated by a few industrial-trading groups can further reduce competition in small markets, engender public distrust, and stir populist policies that increase country risk and deter investment and entry. The absence of a natural political constituency for these investors also makes the emergence of a secure broad-based business coalition more difficult. Patterns of ownership and control can also impact on convergence through their effect on the management practices of firms. While Lucas (1978) modeled the distribution of firm size and productivity as reflecting differences in entrepreneurial skills, until recently it has not been possible to test the model because of the lack of an index of management capability that did not directly draw on measures of firm performance. Bloom and van Reenen (2007) addressed this gap with a survey tool that measures management practices in 18 dimensions covering operations, monitoring, targets, and incentives.31 They found that average management scores vary greatly across countries, and that they are correlated with income and aggregate productivity. Management scores also differ across firms within individual industries and countries and account for about 25 percent of productivity differences holding a number of other variables constant. Through a controlled experiment, firms that were helped to improve management scores achieved substantial productivity gains, suggesting a causal relationship (Bloom et al. 2013). Detailed Africa-specific results are not yet available, but average management scores for Ethiopia, Ghana, Kenya, Tanzania, and Zambia fall substantially below the score for India, which itself scores far lower than high-income countries. Several factors make it less likely that poorly managed firms will be forced out of business. Low levels of competition, measured economy-wide as well as reported by firm managers, are associated with poorer management practices. More restrictive labor market practices affect management quality by placing constraints on human resource management as well as by causing frictions in the hiring and firing of managers themselves. Government-owned firms are poorly managed, often being shielded from competitive pressures through subsidies, preferential regulatory treatment, or preferential access to value chains. Family, rather than professional, management also plays a role in reducing management quality and productivity, even for family-owned firms. Weak rule of law makes it less likely that managerial positions will be given to non-family members, effectively limiting the span of management control. This in turn constrains the expansion of productive firms and allows low-productivity firms to survive. These factors, as set out by Bloom and van Reenen (2007), are all relevant for most African economies.

7.5 Conclusion Just as Africa can learn from developing regions that have experienced more widespread structural transformation, the African development experience offers insights 31  The surveys initially covered firms in France, Germany, the United States, and the United Kingdom, but now include detailed data from about 15,000 firms in 30 countries including developing countries in Asia, Latin America, and sub-Saharan Africa. Background material and full survey data is available from worldmanagementsurvey.org.

Development as Diffusion    131 that may be valuable for other parts of the developing world and the discipline of Economics. Africa’s slow rate of productivity growth and structural transformation partly reflects slow productivity convergence both at sector and at firm level. Some of the research surveyed in this chapter is still in the process of being applied to Africa. The available evidence suggests that a number of factors are responsible, and that while every country has its particularities, there are common threads that characterize most experiences. They have roots in Africa’s geography and its distinctive history, including the legacy of its colonial period on state formation and market structure, as well as on the highly uneven distribution of human capital among its population. These factors have contributed to a political economy that has sustained a poor- and high-cost business climate which has both constrained the productivity of individual firms and slowed productivity convergence. None of these factors are immutable. Both history and geography are evolving in response to demographic, technological, and regulatory changes. Rapid population growth and urbanization are reducing land-to-labor ratios. That puts pressure on African governments to shift their economies towards manufacturing industry for better jobs and growth. The rapid spread of information and communication technology is breaking down distance barriers, at least in some dimensions. Trade reform and progress on regional economic integration, though slow, is helping to break down market barriers. The political economy of the private sector is also evolving, with growth in larger-scale African entrepreneurship, including a number of emerging trans-Africa businesses. This can help to complement foreign investment as well as strengthen the power of domestic business constituencies. Given the political power of established interests, it may be that a two-track policy along the lines of those implemented by Malaysia or Mauritius is more feasible for some African countries than across-the-board reforms. Any approach will need to open up opportunities for indigenous businesses as well as for foreign and minority firms, and be complemented by measures to strengthen the business climate and access to skills and management capacity.

Acknowledgments We are grateful to Célestin Monga, Nancy Birdsall, Michael Clemens, Louise Fox, Alvaro González, Leonardo Iacovone, Justin Yifu Lin, Margaret McMillan, Todd Moss, Gaël Raballand, Enrique Rueda-Sabater, Justin Sandefur, Finn Tarp, Gaiv Tata, and seminar participants at the Center for Global Development and the authors conference at the National School of Development at Peking University for thoughtful comments and suggestions, and to Sneha Raghavan for research assistance.

References Abramovitz, M. (1956). Resource and output trends in the United States since 1870. American Economic Review, 46(2):5–23. Acemoglu, D., Johnson, S., and Robinson, J.A. (2001). The colonial origins of comparative development: an empirical investigation. American Economic Review, 91(5):1369–1401.

132   Concepts Aghion, P., and Howitt, P. (1992). A model of growth through creative destruction. Econometrica, 60(2):323–351. Baily, M., and Solow, R.M. (2001). International productivity comparisons built from the firm level. Journal of Economic Perspectives, 15(3):151–172. Bartelsman, E., Haltiwanger, J., and Scarpetta, S. (2013). Cross-country differences in productivity: the role of allocation and selection. American Economic Review, 103(1):305–334. Bauer, P.T. (1954). West African A Study of Competition, Oligopoly and Monopoly in a Changing Economy. Cambridge: Cambridge University Press. Baumol, W.J., and Bowen, W.G. (1966). Performing Arts:  The Economic Dilemma; a Study of Problems Common to Theater, Opera, Music, and Dance. New  York:  The Twentieth Century Fund. Bernard, A.B., and Jones, C.I. (1996a). Productivity and convergence across U.S. states and industries. Empirical Economics, 21(1):113–135. Bernard, A.B., and Jones, C.I. (1996b). Comparing apples to oranges:  productivity convergence and measurement across industries and countries. American Economic Review, 86(5):1216–1238. Beuran, M., Mahihenni, M.H., Raballand, G., and Refas, S. (2012). The Impact of Demand on Cargo Dwell Time in Ports in SSA. Policy Research Working Paper 6014. Washington, DC: The World Bank. Block, S. (2010). The Decline and Rise of Agricultural Productivity in Sub-Saharan Africa Since 1961. NBER Working Paper 16481. Cambridge: National Bureau of Economic Research. Bloom, N., Eifert, B., Mahajan, A., McKenzie, D., and Roberts, J. (2013). Does management matter? Evidence from India. Quarterly Journal of Economics, 128(1). Bloom, N., and Van Reenen, J. (2007). Measuring and explaining management practices across firms and countries. Quarterly Journal of Economics, 122(4):1351–1408. Bratton, M., Lewis, P., and Gyimah-Boadi, E. (2001). Constituencies for Reform in Ghana. Journal of Modern African Studies, 39(2):231–259. Bräutigam, D. (2003). Close encounters: Chinese business networks as industrial catalysts in sub-Saharan Africa. African Affairs, 102(408):447–467. Bräutigam, D., Rakner, L., and Taylor, S. (2002). Business associations and growth coalitions in sub-Saharan Africa. Journal of Modern African Studies, 40(4):519–547. Caselli, F. (2005). Accounting for cross-country income differences, in P. Aghion and S. Durlauf (eds), Handbook of Economic Growth. Handbooks in Economics. Amsterdam: Elsevier, pp. 1A:679–741. Cattaneo, O. (2007). Tourism: unfulfilled promise, in A. Mattoo and L. Payton (eds), Services Trade and Development: The Experience of Zambia. New York, and Washington, DC: Palgrave Macmillan and The World Bank, pp. 213–258. Chenery, H.B. (1960). Patterns of industrial growth. American Economic Review, 50(4):624–654. Christie, I., Twining-Ward, L., Fernandes, E., and Messerli, H. (2013). Tourism in Africa: Harnessing Tourism for Growth and Improved Livelihoods. Washington, DC: The World Bank. Clark, C. (1940). The Conditions of Economic Progress. London: Macmillan. Cornwall, J. (1977). Modern Capitalism:  Its Growth and Transformation. New  York:  St. Martin’s Press. Dinh, H.T., Rawski, T.G., Zafar, A. et al. (2013). Tales from the Development Frontier: How China and Other Countries Harness Light Manufacturing to Create Jobs and Prosperity. With contributions by Xin Tong and Pengfei Lin. Washington, DC: The World Bank.

Development as Diffusion    133 Easterly, W., and Levine, R. (2001). It’s not factor accumulation: stylized facts and growth models. World Bank Economic Review, 15(2):177–219. Eifert, B., Alan, A., and Ramachandran, V. (2008). The cost of doing business in Africa: evidence from enterprise survey data. World Development, 36(9):1531–1546. Ellis, K., and Singh, R. (2010). Assessing the Economic Impact of Competition. ODI Research Report. London: Overseas Development Institute (ODI). Emery, J.J. (2003). Governance and private investment in Africa, in N. van de Walle, N. Ball, and V.Ramachandran (eds), Beyond Structural Adjustment: The Institutional Context of African Development. New York: Palgrave Macmillan. Fafchamps, M. (2001). The role of business networks in market development in sub-Saharan Africa, in M. Aoki and Y. Hayami (eds), Communities and Markets in Economic Development. Oxford: Oxford University Press. Faggio, G., Salvanes, K.G., and Van Reenen, J. (2010). The evolution of inequality in productivity and wages: panel data evidence. Industrial and Corporate Change, 19(6):1919–1951. Finn, A., Leibbrandt, M., and Wegner, E. (2012). Policies for reducing income inequality and poverty in South Africa, in J. Hofmeyr (ed.), Transformation Audit 2011: From Inequality to Inclusive Growth: South Africa’s Pursuit of Shared Prosperity in Extraordinary Times. Cape Town: Institute for Justice and Reconciliation, pp. 72–81. Fox, L., and Sekel Gaal, M. (2008). Working Out of Poverty:  Directions in Development. Washington, DC: The World Bank. Gelb, A., Meyer, C.J., and Ramachandran, V. (2013). Does Poor Mean Cheap? A Comparative Look at Africa’s Industrial Labor Costs. Working Paper 325. Washington, DC: Center for Global Development. Grossman, G.M., and Helpman, E. (1991). Quality ladders in the theory of growth. Review of Economic Studies, 58(1):43–61. Hall, R.E., and Jones, C.I. (1999). Why do some countries produce so much more output per worker than others? Quarterly Journal of Economics, 114(1):83–116. Hallward-Driemeier, M., Khun-Jush, G., and Pritchett, L. (2010). Deals versus Rules: Policy Implementation Uncertainty and Why Firms Hate It. NBER Working Paper 16001. Cambridge, MA: National Bureau of Economic Research. Harrigan, J. (2001). From Dictatorship to Democracy: Economic Policy in Malawi 1964-2000. Farnham: Ashgate. Harrison, A.E., Lin, J.Y., and Xu, L.C. (2012). Performance of formal manufacturing firms in Africa, in H.T. Dinh and G.R.G. Clarke (eds), Performance of Manufacturing Firms in Africa: An Empirical Analysis. Washington, DC: The World Bank, pp. 27–46. Hausmann, R., Hwang, J., and Rodrik, D. (2007). What you export matters. Journal of Economic Growth, 12(1):1–25. Hausmann, R., Klinger, B., and Wagner, R. (2008). Doing Growth Diagnostics in Practice: A ‘Mindbook’. CID Working Paper 117. Cambridge, MA: Center for International Development at Harvard University. Herbst, J. (2000). States and Power in Africa: Comparative Lessons in Authority and Control. Princeton, NJ: Princeton University Press. Herrendorf, B., Rogerson, R., and Valentinyi, A. (2013). Growth and Structural Transformation. NBER Working Paper 18996. National Bureau of Economic Research. Himbara, D. (1994). Kenyan Capitalists, the State and Development. Boulder, CO:  Lynne Rienner Publishers.

134   Concepts Hsieh, C.-T., and Klenow, P.J. (2009). Misallocation and Manufacturing TFP in China and India. Quarterly Journal of Economics, 124(4):1403–1448. Hsieh, C.-T., and Klenow, P.J. (2010). Development accounting. American Economic Journal: Macroeconomics, 2(1):207–223. Iacovone, L., Ramachandran, V., and Schmidt, M. (2013). Stunted Growth: Why Don’t African Firms Create More Jobs? Policy Research Working Paper 6727. Washington, DC:  The World Bank. Johnson, S., Ostry, J.D., and Subramanian, A. (2010). Prospects for sustained growth in Africa: Benchmarking the constraints. IMF Staff Papers, 57(1):119–171. Jones, C.I., and Romer, P.M. (2010). The new Kaldor facts: ideas, institutions, population, and human capital. American Economic Journal: Macroeconomics, 2(1):224–245. Kaldor, N. (1961). Capital accumulation and economic growth, in F.A. Lutz and D.C. Hague (eds), The Theory of Capital. New York: St. Martin’s Press, pp. 177–222. Kaldor, N. (1966). Causes of the Slow Rate of Economic Growth in the UK. London: Cambridge University Press. Klenow, P., and Rodríguez-Clare, A. (1997). The neoclassical revival in growth economics: has it gone too far? in B.S. Bernanke and J. Rotemberg (eds), NBER Macroeconomics Annual 1997. Cambridge: MIT Press, 12:73–114. Kuznets, S. (1966). Modern Economic Growth: Rate, Structure, and Spread. New Haven: Yale University Press. Lederman, D., and Maloney, W.F. (2012). Does What You Export Matter? In Search of Empirical Guidance for Industrial Policies. Washington, DC: The World Bank. Lewis, W.A. (1954). Economic development with unlimited supplies of labour. Manchester School, 22 (2):139–191. Lucas, R.E. (1978). On the size distribution of business firms. Bell Journal of Economics, 9(2):508–523. Mankiw, N.G., Romer, D., and Weil, D.N. (1992). A contribution to the empirics of economic growth. Quarterly Journal of Economics, 107(2):407–437. McMillan, M. (2013). The changing structure of Africa’s economies. Background paper African Economic Outlook 2013. Unpublished. McMillan, M., and Rodrik, D. (2011). Globalization, structural change and productivity growth, in M. Bacchetta and Jansen (eds) Making Globalization Socially Sustainable. Geneva: WTO Publications, pp. 49–84. Mengistae, T. (2001). Indigenous Ethnicity and Entrepreneurial Success: Some Evidence from Ethiopia. Policy Research Working Paper 2534. Washington, DC: The World Bank. Milanovic, B. (2013). All the Ginis Dataset (Updated June 2013). Available from http:// go.worldbank.org/9VCQW66LA0 (accessed January 14, 2014). Nugent, P. (1995). Big Men, Small Boys, and Politics in Ghana: Power, Ideology, and the Burden of History, 1982-1994. London and New York: Frances Pinter. Page, J. (2013). Industrial Policy in Practice:  Africa’s Presidential Investors’ Advisory Councils. Paper presented at UNU-WIDER Conference Learning to Compete: Industrial Development and Policy in Africa. Unpublished draft. Parente, S.L., and Prescott, E.C. (1994). Barriers to technology adoption and development. Journal of Political Economy, 102(2):298–321. Ramachandran, V., Gelb, A., and Kedia Shah, M. (2009). Africa’s Private Sector: What’s Wrong with the Business Environment and What to Do About It. Washington, DC: Center for Global Development.

Development as Diffusion    135 Restuccia, D., and Rogerson, R. (2008). Policy distortions and aggregate productivity with heterogeneous establishments. Review of Economic Dynamics, 11(4):707–720. Restuccia, D., and Rogerson, R. (2013). Misallocation and productivity. Review of Economic Dynamics, 16(1):1–10. Robinson, J.A. (2002). States and power in Africa by Jeffrey I. Herbst: a review essay. Journal of Economic Literature, 40(2):510–519. Rodrik, D. (2013). Unconditional convergence in manufacturing. Quarterly Journal of Economics, 128(1):165–204. Söderbom, M., and Teal, F. (2004). Size and efficiency in African manufacturing firms: evidence from firm-level panel data. Journal of Development Economics, 73(1):369–394. Sutton, J., and Kellow, N. (2010). An Enterprise Map of Ethiopia. London: International Growth Centre. Sutton, J., and Kpentey, B. (2012). An Enterprise Map of Ghana. London: International Growth Centre. Sutton, J., and Langmead, G. (2013). An Enterprise Map of Zambia. London:  International Growth Centre. Sutton, J, and Olomi, D. (2012). An Enterprise Map of Tanzania. London: International Growth Centre. Syverson, C. (2004). Product substitutability and productivity dispersion. Review of Economics and Statistics, 86(2):534–550. Syverson, C. (2011). What determines productivity? Journal of Economic Literature, 49(2):326–365. Te Velde, D.W. (ed.) (2010). Effective State-Business Relations, Industrial Policy and Economic Growth. London: Overseas Development Institute. Te Velde, D.W. (ed.) (2013). State-Business Relations and Industrial Policy. London: DFID-ESRC Growth Research Programme. Tilly, C. (1990). Coercion, Capital, and European States: AD 990-1990. Cambridge, MA: Basil Blackwell. Timmer, P, Chenery, H.B., and Nilakanta Srinivasan, T. (1988). The agricultural transformation, in Handbook of Development Economics. Handbooks in Economics 9. Amsterdam: Elsevier, 1:275–333. Van Biesebroeck, J. (2005). Exporting raises productivity in sub-Saharan African manufacturing firms. Journal of International Economics, 67(2):373–391. Woolard, I., and Leibbrandt, M. (2010). The Evolution and Impact of Unconditional Cash Transfers in South Africa. SALDRU Working Paper 51. Cape Town: Southern Africa Labour and Development Research Unit, University of Cape Town.

Chapter 8

Em pl oyme nt, Unempl oyme nt, a nd Underempl oyme nt i n Afri c a Stephen Golub and Faraz Hayat

8.1 Introduction Generation of “good” jobs and economic development are closely connected. Rising labor incomes are the primary means through which growth is translated into improved standards of living and lower poverty rates. Moreover, employment in “modern” sectors involving skill development and technological learning in turn can promote productivity growth, economic development, and demographic transitions with lower birth rates. The last 50 years have witnessed a virtuous cycle of rapid growth of export-led labor-intensive manufacturing, growth of employment, slowing population growth, and rising wages and living standards in a number of emerging countries, particularly in East Asia (e.g. Pack 1988; Radelet, Sachs, and Lee 1997; World Bank 1993), as labor has been absorbed into modern industry out of subsistence agriculture and urban informal activities. The most dramatic recent example is of course China, where 75 million private sector jobs have been created since China’s opening to the global economy, resulting in the largest poverty reduction program in world history (World Bank 2013: 58). What about Africa? Much has been made of the emerging “Cheetah” economies of Africa (e.g. The Economist 2011; Radelet 2010) and indeed African growth has picked up substantially since the mid-1990s in many countries and on the continent as a whole. Important strides have been made in health and education indicators. But, relative to other parts of the developing world and in absolute terms, African growth in per capita GDP has been limited and poverty reduction has been disappointing. African employment consequently remains overwhelmingly informal. This chapter documents and analyzes the predominance of informal employment and argues that lack of demand for labor is the main problem. Integration into the global economy and exports of labor-intensive products are vital to boosting the demand for labor in Africa. Africa has

Employment, Unemployment, and Underemployment    137 some potential to become competitive in light manufacturing, but the most promising avenue for export-led growth in many African countries is agriculture, including traditional cash crops such as cotton, coffee, cocoa, and groundnuts. Contrary to common perceptions, traditional cash crops, which are the source of livelihood for millions of Africans, have many of the features of manufacturing exports: high labor-intensity, potential for quality improvements through technological transfer, and lucrative but quality-sensitive markets in developed countries. The same obstacles inhibit traditional and non-traditional agricultural exports as manufacturing: inhospitable business climates characterized by corruption, high transactions costs, and deficient infrastructure. Section 2 presents the basic facts of pervasive under-employment and dualistic labor markets, section 3 makes the case that underemployment results primarily from lack of demand rather than worker characteristics, section 4 reviews relevant theoretical models, section 5 discusses policies for boosting employment and incomes through export-led growth, and section 6 concludes.

8.2  Patterns of Employment, Unemployment, and Underemployment in African Labor Markets 8.2.1  Employment and unemployment patterns Data on employment in Africa are sparse and not very up to date. The very concepts of labor force participation, employment, and unemployment used in developed economies are problematic in low-income Africa (Fox and Pimhidzai 2013; Fields 2012). Nevertheless, the available information paints a consistent pattern: African labor markets are marked by sharp dualism with very small formal employment. Agriculture and urban informal sectors1 feature pervasive underemployment rather than open unemployment. Labor force participation rates in sub-Saharan Africa (SSA) are not dramatically different from other developing regions. The historically unique aspect of African labor markets is the extent of informality (Roubaud and Torelli 2013). Table 8.1 shows the distribution of employment into government, formal private sector, and informal sector for selected countries, based on labor market surveys, around 2006. Informal employment is defined here as agricultural work, non-wage-employment, and part-time wage employment. For SSA low-income countries, informal employment defined in this way accounts for at least 80 percent of total employment, and often 90–95 percent.2 In half of the low-income SSA countries in Table 8.1, government employment exceeds formal private sector employment. In all these countries, however, both formal private and

1  As Benjamin and Mbaye (2012) note, definitions of the informal sector differ, with various studies using alternative criteria. The overwhelming share of informal employment, however, is not likely to be sensitive to the chosen definition. 2  Informal employment in Table 8.1 is slightly higher than wage employment in Fox et al. (2013) because informal employment includes some part-time and informal wage employment. See also Note 3.

138   Concepts Table 8.1  Distribution of employment by sector, selected African countries Year of survey

Public sector including stateowned enterprises

Formal private sector

Informal sector

2.6% 4.3% 4.9% 6.3% 3.9% 6.4% NA 9.0% 3.1% 8.0% 3.7% 1.8% 3.0% 2.8% 5.2%

2.1% 1.0% 4.7% 1.8% 6.2% 7.0% NA 11.5% 0.4% 0.3% 1.2% 6.1% 1.5% 14.2% 6.8%

95.3% 94.7% 90.4% 91.9% 89.9% 86.6% 86.5% 79.5% 96.5% 91.8% 95.1% 92.1% 95.5% 83.0% 88.0%

2006 2007

25.0% 16.0%

37.0% 45.6%

38.0% 38.4%

2005

30.0%

10.0%

61.0%

Sub-Sahara low-income Benin Burkina Faso Cameroon Congo Rep. Ethiopia Ghana Madagascar Malawi Mali Nigeria Rwanda Senegal Tanzania Uganda Zambia

2005 2005 2005 2005 2005 2010 2005 2004 2007 2004 2006 2001 2006 2006 2005

Sub-Sahara middle-income Botswana South Africa North Africa Egypt

Sources: Benin, Burkina Faso, Senegal: Benjamin and Mbaye (2012); Ethiopia, Mali, Malawi, Madagascar, Rwanda, South Africa, Uganda, Tanzania, Nigeria: Stampini et al. (2013); Botswana: Van Klaveren et al. (2009a); Egypt: As’ad (2009); Cameroon, Democratic Republic of Congo: Razafindrakoto et al. (2009); Zambia: Van Klaveren et al. (2009b); Ghana: Data Portal Ghana (2010).

government employment are under 10 percent, and often below 5 percent of the labor force. Informal employment in middle-income SSA economies Botswana and South Africa is lower, although still sizeable at 38 percent in both cases. Egypt, typical of North Africa, is an intermediate case, with 61 percent informal employment, with the bulk of the remainder employed in the public sector (30 percent). In a study of the urban informal sectors of ten francophone countries, Roubaud and Torelli (2013) confirm the dominance of informal employment even in the capital cities, finding that on average 77 percent of these cities’ labor forces is informally employed.3 Fox et al. (2013) provide a comprehensive analysis of African employment patterns, and find very low levels of wage employment in 2005, typically about 10–15  percent of the labor force. 3 

Informal employment in Roubaud and Torelli (2013) includes some workers employed in the formal sector but there is a high correlation between informal employment and workers in the informal sector—97 percent of informal sector jobs are informal while informal employment accounts for 41 percent of the much smaller number of jobs in the formal private sector.

Employment, Unemployment, and Underemployment    139 Open unemployment rates are generally very low in low-income SSA, often well below the levels in developed economies, for example, 0.7 percent in Benin, 2 percent in Uganda, 2.3 percent in Burkina Faso, and 2.6 percent in Madagascar. Unemployment is higher in middle-income SSA countries, particularly South Africa (Kingdon and Knight 2004). Also, unlike developed countries, in Africa recorded unemployment rates rise with the level of education, and university graduates tend to have the highest levels of unemployment (African Development Bank 2012). Unemployment is simply not an option for the poor and unskilled, who find refuge in subsistence agriculture and the urban informal sector (Fields 2012). The quality of the unemployment data is open to question, with unemployment and exit from the labor force difficult to distinguish. Nevertheless, it is clear that Africa has an employment rather than an unemployment problem (Fields 2012). Following independence, almost all African countries adopted highly interventionist import-substitution industrialization (ISI) policies characterized by growth of the public sector and protection of domestic industries. Widespread economic crises in the 1980s led to structural adjustment policies involving contraction of the public sector and reduced protection of formal import-competing industries. Public employment declined in absolute terms, and even more so as a share of the labor force, between the late 1970s and the mid-1990s (Goldsmith 1999). Structural adjustment programs initially also entailed declines in private sector industrial employment as inefficient import-substituting industries collapsed and non-traditional export growth was disappointing. Since about 1995, African growth has picked up, resulting in rising formal employment, but from a low base (African Development Bank 2012; Fox and Gaal 2008). Growth of wage employment has been insufficient to make much of a dent in underemployment (Kingdon, Sandefur, and Teal 2006; Haywood and Teal 2009; Fox and Gaal 2008; Fox et al. 2013). Private sector wage employment grew too slowly to offset declining public sector employment, or even to keep up with labor force growth in some countries. In recent years, self- and family- (largely urban) employment rose sharply as a share of the labor force. In Zambia, for example, wage employment declined from 25 percent of the labor force in the 1970s to less than 10 percent in 2005 (Fox and Gaal 2008). Wage employment is much lower for women than men.

8.2.2  Earnings: dualism and underemployment Remuneration differs sharply between the formal and informal sectors in African economies. Table 8.2 shows that dualism is much greater in low-income Africa than other developing countries, comparing gross domestic product (GDP) per capita to wages and productivity in manufacturing, for selected countries, based on data availability.4 Productivity and wages in manufacturing are measured as annual value added and labor compensation per employee, respectively, using United Nations Industrial Development Organization (UNIDO) data.5 Manufacturing productivity and wages are very high relative to per capita 4 

Countries use varying definitions of these concepts, and the findings in Table 8.2 should be viewed as general tendencies rather than precise estimates. Also, the formal manufacturing sector is very small in most African low-income countries. See Mbaye and Golub (2002) and Golub and Edwards (2004) for more discussion of international comparisons of labor costs and productivity. 5  UNIDO labor compensation data do not include employer contributions to social insurance funds and fringe benefits. UNIDO statistics cover only formal firms.

140   Concepts Table 8.2  Indicators of labor costs,a selected regions and countries Manufacturing wage/ manufacturing productivity

Minimum wage/GDP per capita

0.34 0.20 0.18 0.31 NA 0.31 0.35 0.23 0.09

0.43 1.93 1.31 1.14 1.05 4.36 0.44

1.0 2.5

0.43 0.38

0.17 0.29

1.8 2.6 1.7

0.26 0.42 0.41

0.41 1.18 0.58

2.9 1.1 1.1 1.4 1.1 0.7 2.0

0.33 0.16 0.15 0.22 0.32 0.18 0.28

0.77 0.51 0.54 0.49 0.34 0.54 0.68

India Nepal

2.4 1.8

0.19 0.19

0.68 1.21

Latin America Mexico

0.7

0.24

0.15

Manufacturing wage/GDP per capita Low-income sub-Saharan Africa Cameroon Ethiopia Ghana Kenya Lesotho Malawi Senegal Tanzania Uganda

5.0 6.1 4.9 4.7 2.5 3.3 9.4 4.5 8.6

Middle-income sub-Saharan Africa Mauritius South Africa North Africa Egypt Morocco Tunisia East Asia Cambodia China Indonesia Korea Malaysia Thailand Viet Nam South Asia

Source: UNIDO Industrial Statistics Database, World Bank World Development Indicators, US State Department, and authors’ calculations. a  Manufacturing data around 2005, minimum wages in 2012.

GDP in low-income African countries. The ratio of manufacturing wages to per capita GDP is often about 5 or higher in these countries, and above 2 in all cases. In middle-income SSA countries and North Africa, the differentials are much smaller, especially in Mauritius. In Asia, particularly East Asia, the ratio of manufacturing wages to per capita GDP is usually not far from parity. In Mexico the ratio is actually below 1. Gelb, Meyer, and Ramachandran (2013) and Clarke (2011) also find that African manufacturing wages are very high relative to per capita GDP, using firm-level data.

Employment, Unemployment, and Underemployment    141 Roubaud and Torelli (2013) provide further evidence of dualism. They find that earnings are generally much higher in the formal public and private sectors than in the informal sector, resulting in extremely high Gini coefficients for labor income. Public enterprise and to a lesser extent general government earnings are particularly high, well above formal private sector earnings, which in turn are typically double to triple informal sector earnings. The gap between formal and informal earnings is even more pronounced for women than men. Moreover, job tenure is quite long in the formal sector and even in the informal sectors, indicating limited mobility between sectors. Formal sector jobs are primarily held by older workers, which, together with the evidence of limited formal job creation previously noted, suggests that the prospects for young people are even dimmer than the overall statistics suggest (African Development Bank 2012). In addition, Roubaud and Torelli note that the informal sector itself is segmented, as also stressed by Benjamin and Mbaye (2012). Roubaud and Torelli (2013) also document the pervasiveness of urban underemployment, which they divide into “time-related” underemployment where workers are involuntarily working part-time, and “invisible” unemployment, defined as those workers who earn less than the minimum wage. In addition, the vast majority of the workforce in SSA does not receive any social security or other fringe benefits. The unsatisfactory nature of African employment opportunities is manifested in workers’ answers to questions about their aspirations. More than half of young workers (aged 15–24 years) surveyed aspire to formal employment in public or private sectors (Roubaud and Torelli 2013), despite the paucity of formal job creation. The African Development Bank (2012), using Gallop poll data, provides similar evidence of mismatch between aspirations of young people and the realities of the job market. In summary, since the era of structural adjustment, employment opportunities in the public sector have dwindled and the formal private sector has failed to grow sufficiently to absorb the large majority of the working population in agriculture and the urban informal sector, earning very low incomes and lacking access to social insurance programs. Lewis (1954) noted that much of Africa did not fully fit his model of unlimited supply of labor in subsistence activities. However, due to rapid population growth combined with limited development of the formal sector, Lewis’s framework now fits very well for much of low-income Africa, dominated by subsistence agriculture and small-scale informal family firms: What we have is not one island of expanding capitalist employment surrounded by a vast sea of subsistence workers, but rather a number of such tiny islands … We find a few industries highly capitalized such as mining or electric power side by side with most primitive techniques, a few high class shops surrounded by masses of old style traders, a few highly capitalized plantations, surrounded by a sea of peasants. (p. 147)

8.3  Causes of Dualism and Underemployment in Africa There are two main explanations for the large differentials in earnings and pervasive underemployment described above: (i) heterogeneous labor, with the preponderance of the labor force having low human capital and limited skills, and (ii) low demand for labor combined

142   Concepts with labor market segmentation. The labor heterogeneity argument claims that differences in human capital and other worker characteristics explain income differentials. The segmentation argument shifts the focus to the product markets with a shortage of “good” jobs, and rationing of these jobs. It is important to ascertain which of these two explanations is relatively more important in Africa, although of course both likely have some validity. If higher wages depend on raising human capital then the focus of poverty-reduction strategies on education and health in Africa is appropriate. If, however, low demand for labor originates in the product market, improved educational attainments and health outcomes may not be sufficient to boost formal employment, and the focus instead should be on the business climate. The observed patterns of wages and employment suggest that low demand for labor is the primary cause. In general, education explains only about 30  percent of variations in labor compensation (Mortensen 2005). Although this may reflect unobservable labor skills (Rosenzweig 1988), Mortensen (2005) finds that labor heterogeneity is robust to inclusion of numerous controls. Teal (2011) cites recent evidence showing that segmentation is common in African labor markets, particularly between firms of different sizes. Söderbaum, Teal, and Wambugu (2002) show that when observable and unobservable aspects of human capital are controlled for, wages are much higher in larger firms. Kingdon, Sandefur, and Teal (2006) conclude that non-competitive theories such as efficiency wages and bargaining models explain this effect better than human capital theory. In the formal manufacturing sector, Fox and Oviedo (2008) show that wage premiums do not reflect productivity differences. Enterprise surveys and poll data provide further evidence that low demand for labor rather than lack of education is the most binding constraint. Respondents to enterprise surveys in Africa tend not to rate lack of education of the labor force as one of the top constraints. The African Development Bank (2012) used Gallop Poll Surveys conducted in ten North African countries (African Development Bank 2012), finding that factors relating to insufficient labor demand (lack of jobs, insufficient government efforts, weak economy, jobs being given to people with connections, and corruption) together account for about two-thirds of the reasons provided. Secondary and tertiary education improves the chances of having wage employment, but even for workers with university education, under-employment and unemployment is the norm. The African Development Bank (2012) reports that only 30  percent of young people with some tertiary education hold wage employment, another 30 percent are in “vulnerable employment”, and the remaining 40 percent are unemployed, inactive or discouraged. Open unemployment actually rises with education, as previously noted. Several other factors contribute to low formal employment. (i) Education may be expanding, but fail to impart useful skills (Page 2012; African Development Bank 2012), creating a mismatch between worker skills and employer needs. (ii) Rapid population growth exacerbates the excess supply of labor in Africa (Fox et al. 2013), offsetting the effects of output growth. To some extent, however, Africa’s failure to experience a demographic transition reflects the lack of structural transformation so cause and effect are difficult to distinguish. Third, wages could be driven up by “Dutch Disease” effects in some natural-resource abundant countries, but this cannot explain why informal sector labor incomes are so low relative to formal-sector wages. Moreover, natural resource abundance does not necessarily preclude labor-intensive manufacturing exports, as Malaysia and Indonesia have shown (Fox et al. 2013).

Employment, Unemployment, and Underemployment    143

8.4  Models of Dualism and Underemployment 8.4.1  Dualism and structural transformation:  the Lewis model The Lewis (1954) model still provides the starting point for understanding African dualism as resulting from low demand for labor in the modern sector. The model features a large traditional sector with subsistence incomes and a small modern sector paying much higher wages. The process of economic development involves expansion of the modern (formal) sector through capital accumulation, gradually absorbing surplus labor from the subsistence (informal) sector. Figure 8.1 depicts the intersectoral allocation of labor in the Lewis model between rural (r) and modern (m) sectors6. L represents the total labor force, MPL is the marginal productivity of labor, and W the real wage. Due to a “surplus” of labor, MPLr is very low, with the modern sector consequently facing a perfectly elastic supply of labor. For reasons not specified in Lewis (1954), however, Wm is set exogenously well above the subsistence level Wr. Initially, as the modern sector invests, raising MPLm, its employment expands, for example, from L1 to L2, absorbing labor from the traditional sector without raising Wr. Eventually, the absorption of labor in the modern sector reaches L3, the Lewis turning point, and incomes begin to rise above subsistence levels in the traditional sector. The modern sector’s output may be modeled using the a Cobb–Douglas function (subscript m suppressed), Q = F ( A, K , L ) = AK a L1− a, where A is technology, K is capital, and L is labor. Labor market equilibrium implies Denoting L = dL / dt and likewise for other variables, it is easy to show that: L

 A + αK − W L = . α

That is, the rate of growth of modern-sector employment depends on technological progress, capital accumulation, and real wage moderation.

8.4.2  Urban unemployment and informal employment Harris and Todaro (1970) (HT) elaborated on Lewis’s dualistic labor market to include large-scale urban unemployment and underemployment, making migration endogenous. Surplus rural labor migrates to the higher-paid urban (modern) sector as long as the expected urban wage is higher than the rural wage. Equilibrium occurs when expected wages are equalized through adjustments in unemployment. That is, rural–urban migration

6  Figures 8.1 and 8.2 are based on Basu (1997). This version diverges from Lewis (1954) in assuming that wages equal the marginal productivity of labor in the traditional sector, rather than average product.

144   Concepts

MPLr MPLm2

Real wage

Real wage

The Lewis model of labor market dualism

MPLm1 N1

Wm

N2

N3

Wr Lr

Lm Om

L1

L2

L3

Labor

Or

L

Figure  8.1   The Lewis model of labor market dualism. continues until unemployment rises such that the probability of a finding a high-paying job falls enough to equalize expected urban and rural wages. Figure 8.2 depicts the HT model. HH represents a rectangular hyperbola on which the rural equilibrium wage and employment level lie, for a given modern sector wage and employment level. Equilibrium may be stated as: Wm

where L − Lr is the urban labor force and

Lm = Wr , L − Lr

Lm is the probability of finding a job in the modern L − Lr

sector. Fields (1975, 1990) presented a further important extension of the HT model, distinguishing unemployment and informal employment. The economy’s labor force is now composed of four groups: workers in the urban modern sector; workers in the urban informal sector; the urban unemployed; and subsistence agriculture. In Fields’ model, taking a low-paying urban informal job facilitates searching for a modern-sector job relative to remaining in the countryside, although unemployed workers face even lower search costs. Fields’ (1975) framework implies that urban informal earnings are below rural incomes, although superior to incomes of the unemployed. In reality, urban informal incomes are higher than in agriculture (Fox and Gaal 2008), but real incomes of the urban informal sector could still be lower, considering the higher pecuniary and non-pecuniary costs of urban living relative to village life.

Employment, Unemployment, and Underemployment    145

MPLr

Real wage

Real wage

The Harris-Todaro Model

MPLm N

Wm

Wr Lm Om

Lr Lm L–Lr

U

Lr

Labor

Or

L

Figure  8.2   The Harris–Todaro  model. Both the Lewis and HT models assume but do not explain the reasons for high and sticky modern sector wages. These high wages could be due to minimum wages, unions, or efficiency wage considerations. The labor turnover model (Stiglitz 1974) proposes that firms pay higher wages to reduce quit rates. Alternatively, the biological efficiency wage model (Stiglitz 1976) assumes that firms pay higher wages so that workers have enough nutrition to work productively and avoid illness. The HT model and Fields extension show how surplus labor is manifested in open unemployment and urban informal employment, in addition to subsistence agriculture. The central underlying problem giving rise to dualism, however, remains the scarcity of relatively high-paying modern sector jobs, as stressed by Lewis (1954).

8.4.3  The roles of globalization and modernization of the informal sector Two important extensions to the Lewis perspective are relevant to contemporary Africa: (i) the role of globalization in accelerating structural transformation, and (ii) modernization of informal practices, especially in agriculture, through technology transfer.

8.4.3.1 Globalization In the original Lewis (1954) model, the speed of economic development depends on domestic capital investment and technological change in the modern sector. Fei and Ranis (1964)

146   Concepts and others extended the Lewis model by refining the intersectoral linkages and the role of agriculture, while still assuming a closed economy. Contemporary globalization requires some important amendments through two channels: (i) Foreign capital, particularly foreign direct investment (FDI), provides an alternative to domestic savings and technological change. Moreover, FDI provides higher-paying jobs, increased competition, more training, and knowledge spillovers (Javorcik 2012). (ii) Outsourcing of labor-intensive manufactured products such as apparel by global supply chains (e.g. Gereffi 1999) also raises the demand for labor in developing countries. Murphy et al. (1989) is in the spirit of Lewis (1954), allowing for exports but with costly access to foreign markets. Thus, the central problem becomes alleviating bottlenecks to labor-intensive exports. Golub, Jones, and Kierzkowki (2007) point to the importance of domestic “service links”, that is, infrastructure and public services, in enabling developing countries to participate in the international fragmentation of production. Viewed from this perspective, accelerating growth of the modern sector requires improvement of the business climate in order to attract FDI and other “footloose” inputs that are critical to global competitiveness in manufacturing.

8.4.3.2  Modernization of the urban informal sector and agriculture Lewis (1954) focused on shifting out modern sector labor demand but another possibility is to raise productivity in agriculture and the informal sector, shifting the demand for labor in the rural sector. Lewis recognized that the distinction between traditional and modern activities did not coincide with rural and urban, as shown by the quote above. Much attention now focuses on raising productivity of the urban informal sector (e.g. Fox and Sohnesen 2012) but agriculture may be more promising. The products of the urban informal sector are predominantly non-tradable services or artisanal manufacturing, with minimal exporting. Exports of traditional and non-traditional agricultural cash crops, on the other hand, are a viable African alternative to manufacturing for labor-intensive export-led growth (Golub, O’Connell, and Du 2008; Brenton, Newfarmer, and Walkenhorst 2009). This issue is discussed in more detail in the next section.

8.5  Expanding Employment through Labor-intensive Non-traditional Exports in Africa Export-led growth is often identified with manufacturing, based on East Asia’s and to a lesser extent Latin America’s successes. Collier (2008) is pessimistic about Africa’s ability to compete with Asian manufacturers, given their head start and competitive advantages. Dinh et al. (2012) argue that Africa can compete in some light manufacturing industries, but that weaknesses in the business climate must be remedied. Alternates to manufacturing for labor-intensive exports are available. Africa has promising export industries in tourism, fishing and especially agriculture, including horticulture (fruits, vegetables, and cut flowers) and perhaps most significantly, traditional cash crops.

Employment, Unemployment, and Underemployment    147 Agricultural exports share many of the features of manufacturing, both in terms of their potential to spur growth and employment, and the institutional constraints they face in achieving this potential. Several critical aspects of manufacturing exports promoting development and poverty reduction apply to traditional and non-traditional agriculture: (i) high labor-intensity, (ii) possibilities for technological upgrading and consequently raising producer incomes, (iii) access to state-of-the-art foreign technology through FDI and outsour­ cing, and (iv) the necessity of attaining international competitiveness, and thus (v)  the critical roles of low-cost labor and a favorable climate for investment. For agriculture, especially, sanitary and phyto-sanitary norms in developed country markets are a major hurdle for successful exporting (Golub and McManus 2008) analogous to the demanding specifications of global buyers of apparel. The augmented Lewis model in section 4 suggests two main institutional impediments to labor-intensive exports (i) wages are set too high in the modern sector, and (ii) the adverse business climate deters investment and technological upgrading in labor-intensive tradable industries, both in manufacturing and agriculture.

8.5.1  High labor costs Minimum wages, unions, and labor market restrictions can raise urban labor costs, reducing employment in the formal sector as firms adopt more capital-intensive techniques or exit the country (equation 1 in section 4.1). As shown in section 2.2, manufacturing wages in Africa are very high relative to per capita income and informal sector incomes. This is partly due to relatively high minimum wages and labor market restrictions. Table 8.2 shows that minimum wages relative to GDP are much higher in most low-income African countries than in other regions, particularly East Asia. This is not the case for the middle-income African countries, although South Africa has rather high manufacturing wages, likely due in part to strong unions rather than minimum wages. Minimum wages may therefore be part of the explanation for high manufacturing wages relative to GDP in low-income African countries. Some studies (e.g. Rama 2000; Fox and Oviedo 2013) have found that minimum wages and labor market restrictions are not a major constraint in Africa, unlike in other regions, despite often highly restrictive statutory provisions, perhaps because of lack of enforcement or because other constraints are more important. This benign perspective on labor market regulations may be overstated, for several reasons. First, labor market conflicts may be of lesser importance than infrastructure or corruption, but could still matter. Second, domestic firms, particularly in the informal sector, may be able to routinely disregard labor market statutes. Formal firms, especially foreign investors, however, may feel compelled to abide by local laws, due to lesser recourse to authority in the host country as well as pressures from labor-rights activists at home, and thus may simply eschew investing in countries with such laws even if they are not much enforced. Both wages and productivity in manufacturing are high in Africa relative to per capita GDP. Unit labor costs, the ratio of wages to productivity, also tend to be higher in Africa than in other developing regions, adversely affecting international competitiveness (Mbaye and Golub 2002; Edwards and Golub 2004; Clarke 2011; Gelb, Meyer and Ramachandran 2013). One possible way of improving competitiveness, adopted in East Asia, is to promote

148   Concepts exchange-rate undervaluation, but this is precluded in countries in monetary unions, as in francophone West and Central Africa.

8.5.2  Business climate for investment More importantly than labor market restrictions, the African business climate remains very problematic in some areas, notably poor infrastructure and public services and burdensome restrictions and regulations (Ramachandran, Gelb, and Shah 2009; Eifert, Gelb, and Ramachandran 2008). Deficiencies in infrastructure, red tape and corruption raise indirect costs of production, input sourcing, and distribution substantially. Low-income African countries mostly rank at the bottom of standard measures of competitiveness and the business climate, such as the World Bank’s Doing Business indexes. As Lin and Monga (2010), Rodrik (2008), Golub, Berrnhardt, and Liu (2011), UNCTAD (2010), and others have stressed the private sector is the engine of job creation, but economic development requires a “developmental state” that assists the private sector to overcome market failures, such as external economies of scale, coordination failures, and knowledge externalities. Yet in African countries state failures are often even worse than market failures. These state failures take the form of both errors of omission (failure to invest in infrastructure and provide public services) and commission (excessive regulation and predation on private businesses) (Krueger 1990). These institutional dysfunctions raise transactions costs in all areas of economic activity, but are particularly damaging for export-oriented industries where quality control and timeliness of delivery are paramount. Collier (1998) and Golub, Jones, and Kierzkowski (2007) attribute Africa’s failure to attract investment in labor-intensive manufacturing to state failures raising transactions costs. It is less well known that traditional cash crop production is also undermined by severe disorganization resulting from state failures. Agricultural primary products such as cotton, coffee, cocoa, and groundnuts still dominate exports of many African countries, affecting the livelihoods of very large numbers of people, often smallholder farmers. These products involve complex value chains, including research and extension, provision of credit and inputs (seeds and fertilizer), storage, collection of the crop, transport, processing (e.g. shelling peanuts or ginning cotton), and marketing, in addition to planting and harvesting. Particularly for smallholders, arrangements for provision of credit and repayment of loans are major issues. Consequently, as Poulton et al. (2004) and Tschirley et al. (2009) have documented for cotton, there is a fundamental trade-off between competition and coordination in the organization of the value chain. Exports of specialty coffees, for example, in Rwanda, illustrate the potential gains from exports of agricultural commodities through technological transfer and product upgrading (Golub, O’Connell, and Du 2008). But more often, the same institutional obstacles to manufacturing competitiveness block progress in agriculture, as illustrated by the cases of cotton in Benin and groundnuts in The Gambia and Senegal. As in other countries, these cash crops were controlled by state marketing boards in the first decades of independence and privatized in the 1990s and 2000s. The integrated state-controlled system was evidently flawed, but reforms have had mixed success. Opening the market has often entailed opportunistic behavior rather than open competition. Provision of public goods has suffered.

Employment, Unemployment, and Underemployment    149

8.5.3  Cotton in Benin With assistance from the World Bank, in the 1990s Benin phased in a complex system of private organization, involving limited competition. The reforms succeeded in spurring the entry of domestic entrepreneurs. Some of these entrepreneurs have proved effective, while others have been incompetent and opportunistic, relying on political connections to remain in business. The government failed to enforce the rules and sanction cheaters. Instead, it sometimes intervened in support of special interests and disrupted the functioning of the system. Benin suffers in some respects from the worst of both worlds: limited coordination, due to the weakness of the institutions in enforcing compliance, and limited competition, resulting in depressed production and incomes. The problem is not so much the design of the reforms but the government’s inability to implement them effectively (Golub 2009).

8.5.4  Groundnuts in Senegal/The Gambia Groundnuts remain the two countries’ dominant cash crop (Golub and Mbaye 2002; Mbaye 2005; Integrated Framework 2007). Groundnuts can be sold in either edible form or processed into peanut oil and oil-cake. Contrary to the view that developing countries should strive for greater processing rather than selling products in raw form, edible groundnuts can fetch much higher prices in the European market than groundnuts pressed into oil. However, concerns about aflatoxins7 and pesticide residues have ratcheted up the quality standards in the edibles market, with higher qualities commanding increasing premiums. Processes for controlling aflotoxins are well-known and not difficult to implement in purely technical terms, requiring attention to moisture control and rapid shipment, in turn demanding investment in storage and transport infrastructure, and training of personnel in proper hand­ling (Mbaye 2005). Senegal and The Gambia have made little or no progress in these areas, so Senegalese and Gambian edible groundnuts have very elevated aflotoxin levels and have thus largely been shut out of the European market.8

8.6 Conclusions African economies have picked up but structural transformation remains limited. In this setting, employment opportunities are barely keeping up with rapidly growing labor forces. In low-income countries, this translates into large and sometimes growing underemployment rather than open unemployment, as people are simply too poor not to work. The vast majority of the work force remains in subsistence agriculture and, increasingly, the urban

7  Aflatoxins are a known cancer-causing substance that contaminates groundnuts when handling and storage are slow and the crop is exposed to inappropriate moisture and temperature. 8  Gambian raw groundnuts have been sold at discounted prices for birdseed, but even this limited market is threatened as animal rights activists protest against exposure of birds to high levels of aflotoxins!

150   Concepts informal sector, with very low and uncertain incomes and no access to social insurance programs. Public sector jobs have dwindled since the era of structural adjustment in the 1980s and 1990s, and private formal sector employment growth has been too small and started from too low a base to make a significant dent in underemployment. With its rapidly growing populations, small enclaves of relatively well-paying modern sectors, and vast informal economies, Africa resembles the situation described by Arthur Lewis (1954) as “unlimited supply of labor” more so today than at the time Lewis presented his classic analysis. Lewis’s (1954) depiction of development as the absorption of underemployed labor from subsistence activities into modern industry is still valid, with two amendments. First, developing countries can harness the forces of globalization to generate unprecedentedly rapid growth through labor-intensive exports, as successive waves of East Asian countries have been demonstrating for 50 years. Second, exports of traditional and non-traditional agricultural crops, tourism, and fishing are viable alternatives to manufacturing in Africa. Africa has opportunities to benefit from globalization, in manufacturing but even more in agriculture. When the government fails to provide public goods and harasses formal-sector firms, domestic enterprises will shut down or become informal, and foreign investors will look elsewhere. The work force pays the price in the form of fewer employment opportunities and lower incomes. Many African countries have made considerable progress in restoring macroeconomic stability and improving the business environment, but further efforts are needed to attain global competitiveness in labor-intensive industries to spur sustained employment growth and rising earnings.

Acknowledgment The authors thank Justin Lin, Célestin Monga, Aly Mbaye, Nancy Benjamin, Howard Pack, Louise Fox, Arne Bigsten, Morten Jerven and participants in the Beijing conference on the Oxford Handbook of Africa and Economics for comments and discussion.

References African Development Bank (2012). African Economic Outlook 2012:  Promoting Youth Employment. Tunis: African Development Bank. Asʻad, R. (2009). The Egyptian Labor Market Revisited. Cairo:  American University in Cairo Press. Basu, K. (1997). Analytical Development Economics:  the Less Developed Economy Revisited. Cambridge, MA: MIT Press. Benjamin, N., and Mbaye, A. (2012). The Informal Sector in Francophone Africa: Firm Size, Productivity and Institutions. Washington DC: World Bank. Brenton, P., Newfarmer, R., and Walkenhorst, P. (2009). Avenues for Export Diversification:  Issues for Low-Income Countries. Commission on Growth and Development Working Paper No. 47. Clarke, G. (2011). Wages and Productivity in Manufacturing in Africa: Some Stylized Facts. Washington DC: World Bank. Background paper V for Light Manufacturing in Africa.

Employment, Unemployment, and Underemployment    151 Collier, P. (1998). Globalization: Implications for Africa, in Z. Iqbal and M.S. Khan (eds), Trade Reform and Regional Integration in Africa. International Monetary Fund, pp. 147–181. Collier, P. (2008). Growth Strategies for Africa. Commission on Growth and Development Working Paper No. 9. Data Portal Ghana (2010). Data Portal Ghana. Retrieved July/August, 2013, from http://data. gov.gh/. Dinh, H.T., Palmade, V., Chandra, V., and Cossar F. (2012). Light Manufacturing in Africa:  Targeted Policies to Enhance Private Investment and Jobs. Washington DC: World Bank. Economist (2011). Africa’s Hopeful Economies:  The Sun Shines Bright, December 3.  http:// www.economist.com/node/21541008. Edwards, L., and Golub, S.S. (2004). South Africa’s International Cost Competitiveness and Exports in Manufacturing, World Development, 32(8):1323–1339. Eifert, B., Gelb, A., and Ramachandran, V. (2008). The cost of doing business in Africa: evidence from the investment climate data. World Development, 36(9):1531–1546. Fei, J.C.H., and Ranis, G. (1964). Development of the Labor and Surplus Economy: Theory and Policy. Homewood: Irwin. Fields, G.S. (1975). Rural-urban migration, urban unemployment and under-development, and job-search security in LDCs. Journal of Development Economics, 2(2):165–187. Fields, G.S. (1990). Labor market modelling and the urban informal sector: theory and evidence, in D. Turnham, B. Salomé, and A. Schwartz (eds). Paris: OECD, pp. 49–69. Fields, G.S. (2012). Working Hard, Working Poor. New York: Oxford. Fox, L., and Oviedo, A.M. (2008). Are Skills Rewarded in Sub-Saharan Africa. World Bank Policy Research Working Paper No. 4688. Fox, L., and Oviedo, A.M. (2013). Institutions and job growth in African manufacturing: does employment protection regulation matter? Journal of African Economies, 22(4):616–650. Fox, L., Haines, C., Muñoz, J.H., and Thomas, A. (2013). Africa’s Got Work to Do: Employment Prospects in the New Century. IMF Working Paper 13/201. Fox, L., and Sekkel Gaal, M. (2008). Working Out of Poverty: Job Creation and The Quality of Growth in Africa. Washington DC: World Bank. Fox, L., and Pimhidzai, O. (2013) Different Dreams, Same Bed: Collecting, Using, and Interpreting Employment Statistics in Sub-Saharan Africa—The Case of Uganda. World Bank Policy Research Working Paper No. 6184. Fox, L., and Sohnesen, T.P. (2012). Household Enterprises in Sub-Saharan Africa: Why They Matter for Growth, Jobs, and Livelihoods. World Bank Policy Research Working Paper No. 6184. Gelb, A., Meyer, C., and Ramachandran, V. (2013). Does Poor Mean Cheap? A Comparative Look at Africa’s Industrial Labor Costs. Center for Global Development, unpublished. Gereffi, G. (1999). International Trade and Industrial Upgrading in the Apparel Commodity Chain. Journal of International Economics, 48(1):37–70. Goldsmith, A.A. (1999). Africa’s Overgrown State Reconsidered: Bureaucracy and Economic Growth. World Politics, 51(4):520–546. Golub, S.S. (2009). Specific constraints in the cotton and transit trade sectors, in Benin Country Economic Memorandum. Washington, DC: World Bank, pp. 38–77. Golub, S.S., and Mbaye, A.A. (2002). Obstacles and Opportunities for Senegal’s International Competitiveness: Case Studies of the Peanut Oil, Fishing and Textile Industries. World Bank Africa Region Working Paper Series No. 37, September.

152   Concepts Golub, S.S., Bernhardt, A., and Liu, M. (2011). Development and Trade Strategies for Least Developed Countries. Paper prepared for UNCTAD, March. Golub, S.S., and McManus, J. (2008). Horticulture Exports and African Development. Paper prepared for UNCTAD. Golub, S.S., Jones, R., and Kierzkowski, H. (2007). Globalization and country-specific service links. Journal of Economic Policy Reform, 10(2):63–88. Golub, S.S., O’Connell, S.A., and Du, W. (2008). Export Competitiveness and Development in LDCs: Policies, Issues, and Priorities for Least Developed Countries. UNCTAD. Harris, J.R., and Todaro, M.P. (1970). Migration, unemployment and development: a two-sector analysis. American Economic Review, 60(1):126–142. Haywood, L., and Teal, F. (2009). Employment, Unemployment and Incomes in Nigeria: 1999-2006. Center for the Study of African Economies. Report prepared for the World Bank. Integrated Framework (2007). The Gambia:  From Entrepot to Exporter and Eco-Tourism. Diagnostic Trade Integration Study, July. Javorcik, B.S. (2012). Does FDI Bring Good Jobs to Host Countries? Background Paper for the World Bank World Development Report 2013. Kingdon, G., and Knight, J. (2004). Unemployment in South Africa: the nature of the beast. World Development, 32(4): 391–408. Kingdon, G., Sandefur, J., and Teal, F. (2006). Labor market flexibility, wages, and incomes in sub-Saharan Africa in the 1990s. African Development Review, 18(3):392–427. Krueger, A.O. (1990). Government failures in development. Journal of Economic Perspectives, 4(3):9–23. Lewis, W.A. (1954). Economic development with unlimited supplies of labour. The Manchester School, 22(2): 139–191. Lin, J.Y., and Monga, C. (2010). Growth Identification and Facilitation: The Role of the State in the Dynamics of Structural Change. World Bank Policy Research Working Paper No. 5313. Mbaye, A.A. (2005). Sanitary and Phytosanitary Requirements and Developing Country Agro Food exports: An Assessment of the Senegalese Groundnut Sector. World Bank Agriculture and Rural Development Discussion Paper. Mbaye, A.A., and Golub, S.S. (2002). Unit labor costs, international competitiveness, and exports: the case of Senegal. Journal of African Economies, 11(2):2003. Mortensen, D. (2005). Wage Dispersion:  Why Are Similar Workers Paid Differently. Cambridge: MIT Press. Murphy, K.M., Shleifer, A., and Vishny, R.W. (1989). Industrialization and the big push. Journal of Political Economy, 97(5): 1003–1026. Pack, H. (1988). Industrialization and trade, in H. Chenery and T.N. Srinivasan (eds), Handbook of Development Economics, Volume I. Amsterdam: North Holland, pp. 334–380. Page, J. (2012). Youth, Jobs, and Structural Change: Confronting Africa’s Employment Problem. African Development Bank Working Paper No. 155. Poulton, C., Gibbon, P., Hanyani-Mlambo, B. et al. (2004). Competition and coordination in liberalized African cotton market systems. World Development, 32(3):519–536. Radelet, S. (2010). Emerging Africa:  How 17 Countries Are Leading The Way. Washington, DC: Center for Global Development. Radelet, S., Sachs, J.D., and Lee, J.W. (1997). Economic Growth in Asia. HIID Development Discussion Paper No. 609, November. Rama, M. (2000). Wage misalignment in CFA countries: were labor market policies to blame? Journal of African Economies, 9(4): 475–511.

Employment, Unemployment, and Underemployment    153 Ramachandra, V., Gelb, A., and Shah, M.K. (2009). Africa’s Private Sector: What’s Wrong with the Business Environment and What to Do About It. Washington, DC: Center for Global Development. Razafindrakoto, M., Roubaud, F.F., and Torelli, C. (2009). Measuring the Informal Sector and Informal Employment:  the Experience Drawn from 1-2-3 surveys in African countries. Paris: Université de Paris, Dauphine. Rodrik, D. (2008). Normalizing Industrial Policy, Commission on Growth and Development. Working Paper No. 3. Rosenzweig, M.R. (1988). Labor markets in low-income countries, in H. Chenery and T.N. Srinivasan (eds), Handbook of Development Economics, Volume I. Amsterdam: North Holland, pp. 713–762. Roubaud, F., and Torelli C. (2013). Employment, unemployment and working conditions in urban labor markets of sub-Saharan Africa: main stylized facts, in P. de Vreyer and F. Roubaud (eds), Urban Labor Markets in Sub-Saharan Africa. Washington, DC: World Bank, pp. 37–80. Söderbom, M., Teal, F., and Wambugu, A. (2002). Does firm size really affect earnings? Oxford: Centre for the Study of African Economies, Department of Economies, University of Oxford. Stampini, M., Leung, R., Diarra, S.M., and Pla, L. (2013). How large is the private sector in Africa? Evidence from national accounts and labour markets. South African Journal of Economics, 81(1): 140–165. Stiglitz, J.E. (1974). Alternative theories of wage determination and unemployment in LDCs: the Labor turnover model. Quarterly Journal of Economics, 88(2):194–227. Stiglitz, J.E. (1976). The efficiency wage hypothesis, surplus labor, and the distribution of income in LDCs. Oxford Economic Papers, 28(2): 185–207. Teal, F. (2011). The price of labour and understanding the causes of poverty. Labour Economics, 18:S7–S15. Tschirley, D., Poulton, C., and Labaste, P. (2009). Organization and Performance of Cotton Sectors in Africa. Washington, DC: World Bank. UNCTAD (2010). The Least Developed Countries Report 2010: Towards a New International Development Architecture for LDCs. Geneva: UNCTAD. Van Klaveren, M., Tijdens, K., Hughie-Williams, M., and Martin, N.R. (2009a). An Overview of Women’s Work and Employment in Botswana. Amsterdam: AIAS Working Paper 09-81. Van Klaveren, M., Tijdens, K., Hughie-Williams, M., and Martin, N.R. (2009b). An Overview of Women’s Work and Employment in Zambia. Amsterdam: AIAS Working Paper 09-80. World Bank (1993). The East Asian Miracle. Washington DC: World Bank. World Bank (2013). World Development Report 2013: Jobs. Washington DC: World Bank.

Chapter 9

Inclu sive G row t h in Afri c a Mthuli Ncube

9.1 Introduction Africa has enjoyed a significant increase in economic growth over the past decade, growth that is spread across countries and sectors. This is certainly a welcome improvement over the previous two decades’ experience, yet many policymakers and analysts remain concerned that economic growth in and of itself is not sufficient. While the satisfaction with the growth acceleration is nearly universal, the concerns over its limits are quite varied. Some analysts simply feel that economic growth, while better, remains inadequate to generate rapid development, especially given Africa’s still high population growth rate. Others are more concerned with the unequal distribution of the benefits of growth, noting that increased inequality has limited growth’s impact on poverty reduction. Still others worry that the growth has not brought with it the structural transformations that are the hallmark of economic development, including growth in employment generating activities like labor-intensive manufacturing and services. And some argue that well-being broadly conceived to include non-income dimensions like health and education has not improved as much as incomes per se. Most recently, there is concern that growth is excluding important disadvantaged groups in society. Rapid economic growth in Africa, has failed in creating enough quality jobs for an estimated 36.3 million unemployed Africans. Poverty reduction has been short of expectations, with the proportion of people living on less than US$1.25 a day decreasing from 56.5 percent in 1990 to 47.5 percent in 2008. Africa’s record on achieving the Millennium Development Goals is also mixed, and several key targets are likely to be missed, including reducing child mortality, improving maternal health and achieving full and productive employment. The weak inclusion character of growth presents is equally of serious concern to many African governments. Non-inclusive growth is likely to result not only in rising and persistent inequalities in income and wealth, but also in health, education, participation in the political process etc. Leaving large sections of the population behind, is also likely to have repercussions on the sustainability of future economic growth, as human capital is

Inclusive Growth in Africa    155 left untapped and therefore stifling the full potential of aggregate demand and economic dynamism.

9.1.1  Inclusive growth, employment, and income distribution Rapid growth is a prerequisite for enabling the majority of the poor to lift themselves out of poverty. But for growth to be sustainable and inclusive, it is necessary that it impacts critical sectors that would offer employment country’s labor force. Inclusive growth also recognizes implicitly the linkages between the macro and the micro level, and the importance of structural transformation. Structural transformation is critical for job creation and eventual inclusive growth. Several African countries in the last ten years have achieved high growth rates and improved living conditions yet the long economic upswing has been accompanied by a tendency to growing in-country inequality and between-country inequality. Jobless growth is a reality that plagued many African countries recent growth history. In the coming four decades, the largest wave of young people ever will enter the labor market, most of them from Africa. With one half of Africa’s population under the age of 20 and a median age of 18, the continent is facing a “youth bulge.” Of the 2.4 billion people who are projected to be added to the world by 2050, 46 percent will be born in sub-Saharan Africa. The region will contribute 77 percent of the total increase in global population by 2100. Thirty-one countries out of the region’s 54 are projected to at least double their population by 2050. Economic opportunities for the young are scarce and although the young constitute around two fifths of the continent’s working age population, they make up three fifths of the total unemployed. Youth unemployment rates exceed those of adults, often by a ratio of two to one. In some countries (e.g. South Africa), one in two young people are unemployed. However, this youth bulge is a potential gold mine to be exploited. Some analysts have argued that the “economic miracle” of the East Asian Tigers can to a large extent be attributed to a “demographic dividend” which these countries were able to reap thanks to effective policies which led to the expansion of employment and labor force participation. Productive employment is a key dimension of inclusive growth.

9.1.2  Middle class growth and demographic dividend Africa’s middle class has been growing. The emerging middle class on the continent will continue to grow from 355 million (34 percent of Africa’s population) in 2010 to 1.1 billion (42 percent of the population) in 2060. Africa’s middle class which is strongest in countries that have robust and growing private sectors, is not only crucial for economic growth but it is also essential for the growth of democracy. This middle class will assume the traditional role of the US and European middle classes as major consumers, and will play a key role in rebalancing the African economy. Consumer spending in Africa, primarily by the middle class, has reached an estimated US$680 billion in annual expenditures in 2008 (based on per capita consumption of more than US$2)—or nearly a quarter of Africa’s GDP based on 2008

156   Concepts 100%

6%

6%

4%

3%

80%

28%

29%

30%

33%

67%

66%

66%

64%

1990

2000

60% 40% 20% 0%

1980 Rich class (+$20)

Middle class ($2–$20)

2010 Poor class (–$2)

Figure  9.1 Africa’s middle of the pyramid:  distribution of the African population by classes. Source:  Department of Statics, African Development Bank  Group.

purchasing power parity. By 2030 Africa will likely reach US$2.2 trillion in annual consumer expenditures and comprises about 3 percent of worldwide consumption (Figure 9.1). Compared with other parts of the developing world, Africa’s demographic transition is delayed. A positive aspect of the delayed demographic change is that Africa could benefit from what is called the “demographic dividend.” Driven by a delayed demographic transition, the share of youth (aged 15–24) in Africa—both north and south of the Sahara—has been rising over time and is now higher than in any other part of the world. The demographic dividend will also exhibit itself in increased importance and role of Africa’s middle class in socioeconomic development. All of these may lead to higher productivity and more rapid economic growth. Africa’s population is young and growing, and a rapidly expanding number of jobseekers must be incorporated into labor markets. The number of graduate students tripled in sub-Saharan Africa between 1999 and 2009, yet young people account for about 60 percent of the region’s unemployed. This demographic bulge offers the possibility of a growth dividend, if as in the case of East Asia, a rapidly growing work force can be combined with capital and technology. But it can also represent a major threat. Africa is not creating the number of jobs needed to absorb the 10–12 million young people entering its labor markets each year, and as recent events in North Africa have shown, lack of employment opportunities in the face of a rapidly growing, young labor force can undermine social cohesion and political stability. Each of these concerns suggests that policy makers should pay attention to other measures in addition to, or perhaps instead of, economic growth. Some such measures already exist. For those concerned about increasing inequality there are pro-poor growth measures and Datt-Ravallion decompositions.1 For those interested in non-income dimensions 1  See Ravallion (2004) on pro-poor growth and Datt and Ravallion (1992) and Kakwani (1997) on growth/inequality decompositions.

Inclusive Growth in Africa    157 of well-being there are a host of multidimensional measures of well-being, with the Human Development Index being the best known.2 For those concerned about structural transformation, there are labor market data on sector of employment. To date, however, there is no generally accepted measure of “inclusive growth.” Our aim is to lay out a framework for thinking about and choosing such an index. As we will see, this choice involves many decisions about what to measure and how to measure it, and each choice has advantages and disadvantages. It is important for policy makers to have a clear understanding of these choices before settling on a particular index. Indeed, it may be the case that no single index is adequate to address the concerns expressed above and that any attempt to force these many factors into one number would yield an index that has little meaning. On the other hand, the main argument in favor of a single number is that it helps to focus attention on a measure that reflects economic development more accurately than GDP per capita. In reality, GDP per capita is the default yardstick for measuring economic development. In part, this is because it is a useful measure: how much an economy produces per person matters. But it is also in part because GDP per capita is readily available in most countries. If an alternative and more appropriate measure of “inclusive development” were available, policy makers might focus on it instead of, or in addition to, GDP, to the benefit of development in their countries.

9.2  Approaches to Inclusive Growth Inclusive growth, shared growth, broad-based growth, and pro-poor growth are all conceptual responses to the concern that economic growth alone is not sufficient to generate economic development. As such, they are sometimes used interchangeably, though researchers try to distinguish them from each other.3 Not all authors’ definitions are the same for these concepts, and the definitions sometimes overlap. The definition of inclusive growth will what sub-groups matter in a population? What dimensions of well-being matter? Do opportunities or outcomes matter? And, does relative or absolute progress matter? These issues are discussed in Ncube, Shimeles, and Younger (2013).

9.2.1  GDP per capita, sub-groups and pro-poor growth GDP per capita is not a sufficient measure for progress on inclusive growth. There is a need to subdivide the population in such a way that the poor are targeted. Targeting the poor, results in pro-poor growth and shared growth. It this case, we have a grouping the population by “poor” and “non-poor” or by income/consumption quantile.4 The key for pro-poor or shared growth, then, is not whether the entire economy grows, but whether the incomes or 2  See Alkire and Foster (2011), Bourguignon and Chakravarty (2003), Tsui (2002), and Duclos, Sahn, and Younger (2006). 3  See Rauniyar and Kanbur (2010), Klasen (2010), Ianchovichina and Lundstrom (2009), and African Development Bank (2013), among others. 4  These are discrete groupings, though the continuity of the income distribution allows for a continuous representation, the growth incidence curve (Ravallion and Chen 2003).

158   Concepts consumption of key sub-populations—the poor, or the lower quantiles—grows. A key difference between inclusive growth and pro-poor growth is that the latter focuses only on a subset of the groups, the poorer ones, while at least some definitions of inclusive growth insist that all subgroups’ incomes grow. Inclusive growth is concerned with opportunities for the labor force in the poor and middle-class alike, unlike the pro-poor growth agenda which focuses mainly on the welfare of the poor. To qualify as “broad-based,” growth must occur in most or all sectors of the economy. Proponents of this approach note that countries tend to diversify as they grow, at least up to relatively high income levels (Imbs and Waczziarg 2003).

9.2.2 Well-being Growth in per capita incomes, pro-poor growth, broad-based growth, all take income or consumption to be the measure of well-being. Inclusive growth should include consideration of non-income dimensions of well-being, especially access to infrastructure and basic social services. Rauniyar and Kanbur (2010) argue that this should be considered as “inclusive development,” with “inclusive growth” only applying to the income component. For the non-income dimension, we can classify people by education status. Various definitions of “Inclusive Growth” all express the need for new approaches to address social inequalities especially in the developing world. These include inequalities in income; assets, both financial and human, education and health; economic opportunities and all spheres of life. The AfDB defines Inclusive Growth (IG) as “economic growth that results in a wider access to sustainable socioeconomic opportunities for a broader number of people, regions, or countries, while protecting the vulnerable, all being done in an environment of fairness, equal justice, and political plurality.” The last part of that definition seems to favor the development of capabilities that go well beyond economic opportunities or income growth. One dimension of well-being that does receive considerable attention, as well as most others writing about inclusive growth, is access to good quality employment for all. This echoes the writings of the World Bank (Ianchovichina and Lundstrom 2009) and the Asian Development Bank (Klasen 2010). Of course, employment and income/consumption are very closely related, but other characteristics of a job—security, dignity, voice—matter as well.

9.2.3  Opportunities or outcomes In some of the literature, there is an argument that opportunities, not outcomes, do matter in considerations of inclusive growth. In some parts of the literature, there is reference capabilities and, not functionings, as being important. However, it is more difficult to measure opportunities/capabilities, and much easier to measure outcomes/functionings. It is easy to observe a person’s income and consumption, but much more difficult to measure what they are capable of earning. Some of the literature distinguishes inequality in opportunities and inequality in outcomes in the income dimensions.5 Ferreira, Gignoux and Aran (2011) pursue this distinction empirically. Ianchovichina and Lundstrom (2009) and Klasen (2010) argue that a measure of 5 

See Roemer (1998) and also Ramos and van de Gaer (2012) for a good review.

Inclusive Growth in Africa    159 inclusive growth should include both outcomes (especially income) and opportunities (especially regarding employment, but also access to basic social services).

9.3  Inclusive Growth Components Many authors have criticized the weights used for the HDI which highlights the fact that reasonable people can and do disagree. A particular concern has been the implied marginal rate of substitution between the various components of the index.6 Similar disagreements are sure to plague any choice of weights for an inclusive growth index. One way to avoid this problem is to use a “dashboard” of indicators rather than a single index (Stiglitz et al. 2009; Ravallion 2010). A car’s dashboard has multiple indicators for key variables—oil pressure, engine temperature, fuel level, battery (dis)charge, etc.—which help you (or your mechanic) to assess the car’s status. Combining all of these indicators into a single index of your car’s well-being probably would not be helpful, except perhaps to tell you whether you should be driving it or not.7 Ravallion argues that rather than “mashing up” many disparate indicators into a single index, policy makers are better off seeing a dashboard that includes all of the indicators that went into the index. The argument has two main points. First, the indices used in practice rarely have weights derived from any sort of economic or ethical theory.8 They are chosen arbitrarily by the indices’ creators and can lead to unattractive marginal rates of substitution between the components (Ravallion 2010 lists several) Second, the trade-offs between the various dimensions of the index that the weights imply are often opaque. This makes it difficult for policy makers to understand exactly how much the index would change if the value of an included variable changes, and what the trade-off is between improving one variable at the cost of reducing another. Thus, even though the main motivation of most such indices, including a presumed inclusive growth index, is to focus policy makers’ attention on something that matters, this will not do much good if policy makers do not understand how their actions affect the value of the index. In fact, for most policy purposes, Ravallion argues that the individual components of any purported index are more useful. Even though the dashboard avoids the arbitrary weights needed to aggregate across the rows, it still must choose weights to aggregate down each column. The dashboard approach also has its limitations. As a technical matter, it is a “columns first” approach, it just does not take the second step of aggregating across the rows. As a consequence, the dashboard does not take into account the correlations among indicators. Practically, it may be the case that, presented with a wide range of indicators—a really comprehensive dashboard—policy makers, the press, and even analysts focus on only one or a few of them. The highest profile dashboard approach is the Millennium Development Goals, which 6 

See for example Ravallion (2010) who is particularly critical of the lack of transparency in many columns first type indices. 7  Such an “index of car health” would almost certainly not be an average of all these variables, but a maximum function: you should not drive if the worst of the indicators is below its acceptable level. 8  One exception is GDP, which is an index of disparate quantities of apples, oranges, cars, etc. produced and with weights equal to the prices of those items. For competitive economies, these price weights have a clear economic and welfare justification.

160   Concepts include eight goals and 21 targets.9 Since the goals are mash-ups of the targets, a proper MDG dashboard should include at least those 21 targets, and perhaps the 70 indicators. But even though almost all the targets are readily measured, some are followed closely (e.g. reducing the dollar-aday headcount by 50 percent, eliminating gender disparities in primary and secondary education), while others are barely noticed (reduce biodiversity loss; achieve decent employment for men, women, and young people). In practice, given the option to evaluate many indicators, we may choose to focus only on some. The dashboard allows one to assign an implicit weight of zero to some if its variables. An index at least forces their inclusion with some positive weight. Inclusive Growth (IG) as economic growth that results in a wider access to sustainable socio-economic opportunities for a broader number of people, regions or countries, while protecting the vulnerable, all being done in an environment of fairness, equal justice and political plurality.

Inclusive growth is important for ethical considerations of fairness and equity. Growth must be shared and should be inclusive across different segments of populations. In countries where there is growth with persistent inequalities, it may engender social peace, force poor and unemployed into criminal activities make women vulnerable to prostitution, force children to undesirable labor. Continued inequalities in outcome and access to opportunities may result in civil unrest and violent backlash from people who are continually deprived, derailing a sustainable growth process. This may create political unrest and disrupt the social fabric and national integration. Inclusive growth focuses on the rate and pattern of growth, which must be addressed together because they are interlinked. Long-term sustainable economic growth rates are necessary to reduce poverty and must be accompanied by growing productive employment to reduce inequality. Therefore, inclusive growth is about raising the pace of growth and enlarging the size of the economy, while leveling the playing field for investment and increasing productive employment opportunities as well as ensuring fair access to them. It allows every section of the society to participate in and contribute to the growth process equally irrespective of their circumstances. Inclusive growth indicators can be grouped into four categories, namely:

• • • •

economic inclusion; social inclusion; spatial inclusion; and political/institutional inclusion.

We discuss each of these below.

9.3.1  Economic inclusion Economic growth should create productive employment. As a necessary condition for inclusive growth, it provides the resources for investment in infrastructure, expanding the

9 

And about 70 specific indicators suggested to measure those targets.

Inclusive Growth in Africa    161 private sector, providing resources for social protection programs to protect the vulnerable, better access to health and education and gender equity, in dealing with inequality.

9.3.1.1  Financial inclusion Africa’s financial services sector has grown rapidly in response to its changing economic landscape as rapid urbanization, rising incomes, and technological advances bring more people-many of whom were formerly locked out of the formal financial system into contact with banks and other similar institutions. Financial inclusion is being driven largely by mobile telephony. The expansion of the financial sector not only creates new jobs and other economic opportunities, but it helps establish formal identities for millions of market participants and it provides greater security than the current cash-based transactions. The most visible case in Kenya is MPESA where active bank accounts have grown fourfold since 2007 aided by some 17 million M-PESA mobile money accounts, empowering them to move an estimated US$7 billion annually, an amount equivalent to 20 percent of the country’s GDP. The real breakthrough in the Kenyan market has been in people’s ability to send and receive money, with more than two-thirds doing so by phone. East Africa’s biggest success has been M-Pesa, a mobile-based money-transfer system pioneered by Safaricom, a leading Kenyan operator. It is simple interface, which works on any phone, has brought financial services to Kenya’s poor majority, enabling the movement of some US$8.6 billion in the first half of 2012. Safaricom also launched M-KESHO in March 2010, which allows for the movement of funds to and from an interest bearing account with Equity Bank. The success of Safaricom has compelled other M-money operators to enter the Kenyan competitive landscape. Drawing on Kenyan successful experiences, many low-income African countries have followed suit and adopted Mobile Network Operators (MNO)-led models for extending access of the unbanked population, in particular, to payment services though mobile phones and retail agents. The success of the (MNO)-led model is dependent on a large reliable network of agents and low risk management of electronic value for a cheaper but secure solution to financial inclusion in low-income African countries.

9.3.1.2 Inequality Inequality can be accounted for by adjusting GDP per capita using the Dalton-Atkinson framework which relies on the estimation of an inequality-aversion parameter. This approach was proposed by Ncube, Shimeles, and Younger (2013). Following Ncube, Shimeles, and Younger (2013), the adjustment is made by adapting Dalton-Atkinson inequality index, which use the social utility function with constant elasticity of marginal utility ε: y i1− ε − 1 ε ≠ 1, (1) 1− ε



U ( yi ) =



U ( yi ) = log( yi ) ε = 1, (2)

162   Concepts where yi is the income of the individual i, and ε is the elasticity of marginal utility also know as inequality aversion parameter. This utility function can be aggregated to have the social welfare function,

SWF ( y1 , ... , yn ) = ∑U ( y j ). j

The main purpose of the inequality index is to measure the gap between the actual social welfare and the social welfare compute on perfectly equally distributed incomes. The Dalton index is define by using U(ӯ) which is the social utility for each person in the community in case of perfectly equally distributed incomes, each person get the average income, and Ū which is the average of the individual social utility (see Dalton 1949),



Dε = 1 −

U =1− U( y)

1 n 1− ε ∑  y − 1 n j =1  j y 1− ε − 1

.

(3)

Instead of using social welfare gap, Atkinson index used the gap between the level of income which gives equally distributed social welfare and the actual average income. The Atkinson index is based on the concept of Equally Distributed Equivalent (EDE) income (see Atkinson 1970). Equally Distributed Equivalent is that level of income that, if obtained by every individual in the income distribution, would enable the society to reach the same level of welfare as actual incomes. The EDE is approximated by U−1(Ū) which is the individual income in case each person in the community have the same social utility Ū: 1



 1 n  y 1 − ε  1 − ε U −1 (U ) j Aε = 1 − = 1 −  ∑    . (4) y  n j =1  y    

Following Lambert et al. (2003), we computed the inequality aversion using alternative formula of Atkinson index for data partitioned into k equal-sized: 1



1 n 1− ε  1− ε Aε = 1 −  ∑ kq j     k j =1

ε ≠ 1 (5)

Aε = 1 − kqF ε = 1, (6) where qj is the share of aggregate income belonging to the group j, and qF is the geometric mean income share. This formal has been inversed to have the parameter value for a given value of Atkinson index. To adjust the per capita GDP, we used the Dalton inequality index formula where we approximated the average income by the per capita GDP and the average social welfare by the per capita GDP adjusted by inequality aversion parameter which is described as follows:

Inclusive Growth in Africa    163

pc _ GDP _ Dε =

pc _ GDP ε . (7) (( pc _ GDP 1− ε − 1) (1 − ε))

The attraction of the IGI based on the Dalton-Atkinson inequality index stems from the fact that the possibility it opens for collecting periodically, say on a yearly basis, subjective measure of inequality aversion by society to different opportunities, such as social services, natural and physical resources, political processes, etc. This allows for a consistent comparison of countries across time and space. In this illustration, the inequality-aversion parameter was simulated from quintile distribution data for each African country and applied to per capita GDP. By construction, the inequality aversion parameter computed for each country is highly correlated with the Gini coefficient but offers an opportunity to operationalize the Dalton–Atkinson measure of adjusting per capita GDP for inequality. The results also penalize hugely for high inequality demonstrating the strong preference to equity as compared to simple transformation of the Human Development Index or per capita GDP using just the Gini coefficient. The results are in Table 9.1.

9.3.2  Spatial inclusion Regional trade and integration will create larger and more attractive markets, turning landlocked countries into land-linked countries and integrate Africa to international markets and support intra African trade. Lack of regional infrastructure is a key impediment. Reducing regional inequality will help equalize per capita output, thus encouraging the spatial flow of capital and labor mobility and the diffusion of innovation across countries and regions leading to economic convergence. In order to track progress on spatial inclusion, where infrastructure development plays a big part, we can make use of the Africa Infrastructure Development Index (AIDI) developed by the African Development Bank (2013). The components of the index include access to transport infrastructure, energy infrastructure, ICT Infrastructure, and Water and sanitation infrastructure. Since the components of the AIDI are originally measured in different units, the observations are “standardized” or “normalized” to permit averaging, with the average regarded as a composite index. The normalization procedure used is the min–max formula applied to all observed values of each component during the period 2000–2010. This procedure adjusts the “normalized component” to take values between 0 and 100 over the indicated period. The composite index is calculated as a weighted average of indicators for each component that comprise more than one indicator. The weights are based on the inverse of the standard deviation of each normalized component:  y t = ( σ tot / σ x ) * x t ; where σtot is given by 1 / σ tot = Σ x (1 / σ x ) and σx is the standard deviation of the normalized component x. The rationale for this step is to reduce the impact of the most volatile components on the composite index and consequently the volatility of the rankings. The AIDI composite Index is computed using the sub-indexes of the four components and using the same method described above.

Table 9.1  Africa GDP per capita and GDP per capita adjusted for inequality (2006–2010) Period: 2006–2010

Country Egypt Botswana Mauritius Tunisia Algeria Gabon Comoros Zimbabwe Guinea South Africa Morocco Madagascar Namibia Congo, Rep. Rwanda Senegal Ghana Cameroon Gambia Tanzania Kenya Togo Guinea-Bissau Benin Angola Liberia Uganda Burundi Cape Verde Malawi Djibouti Zambia Mozambique Sierra Leone Mauritania Burkina Faso Congo, Dem. Rep. Côte d’Ivoire Ethiopia Nigeria Niger Mali Chad Central African Republic Sudan

Per capita GDP

Per capita GDP adjusted for inequality

Gini

Inequality aversion

5436.793 13440.54 12296.86 7874.435 7960.784 14192.92 1171.387 missing 1029.478 10007.61 4334.487 990.3096 6256.833 3987.188 1071.208 1792.537 1468.647 2165.558 1329.025 1224.974 1548.279 864.239 1036.866 1471.418 5297.671 384.3406 1144.933 382.8807 3380.736 754.7202 2082.813 1369.317 837.3094 761.8589 1956.71 1167.505 312.041 1694.006 867.8976 2108.136 681.3209 1108.58 1341.723 741.8057 2079.392

4.41E+08 7.050981 1053567 176235.8 1401891 1111304 81.1443 missing 3326.967 376.1365 401635.1 4936.06 891.3613 10466.02 13628.63 16221.69 41354.62 168795.6 673.1029 185813.5 2638.006 286931.7 1328.773 200495 689.8597 3896.628 14571.28 35877.86 2871.364 6726.824 40890.72 209.845 2558.519 11523.34 18195.23 22586.06 947.7646 52115.13 1746995 3923.782 33851.91 31196.32 36731.2 7.090085 189332.8

31.56 57.585 38.9 40.60333 37.76 41.45 64.3 50.1 42.84 60.846 39.872 44.68 69.115 47.32 46.08 43.264 38.596 40 48.755 35.34333 47.43 36.85 41.68 38.62 50.65 38.16 42.83857 36.33 50.52 44.41333 39.96 51.54 45.75333 38.935 42.00667 44.2375 44.43 40.48667 33.162 43.605 39.0175 40.645 39.78 53.73333 35.29

2.283514 1.0073 1.5387 1.439971 1.627829 1.524286 0.8792857 1.201386 1.339286 0.9248 1.601914 1.382743 0.8202 1.281514 1.475957 1.416786 1.543557 1.628571 1.183886 1.747457 1.270557 1.877543 1.257471 1.719443 0.8302286 1.510257 1.472457 1.801714 1.202614 1.455729 1.486686 1.116314 1.341214 1.5144 1.415771 1.5166 1.381586 1.544714 2.106557 1.269857 1.663414 1.5611 1.546257 1.0358 1.647843

Source: Ncube, Shimeles and Younger (2013).

Inclusive Growth in Africa    165 Seychelles South Africa Egypt, Arab Rep. Libya Mauritius Tunisia Morocco Algeria Cape Verde Botswana Namibia Gabon Sao Tome and Principe Zimbabwe Gambia, The Djibouti Swaziland Senegal Ghana Comoros Zambia Rwanda Kenya Uganda Cote d'Ivoire Nigeria Equatorial Guinea Cameroon Malawi Burkina Faso Lesotho Angola Burundi Benin Sudan Congo, Rep. Mauritania Guinea Liberia Togo Guinea-Bissau Central African Republic Mozambique Mali Tanzania Sierra Leone Eritrea Congo, Dem. Rep. Madagascar Chad Ethiopia Niger Somalia

27.83 25.90 24.75 24.72 24.71 23.45 22.30 21.66 21.11 20.95 20.13 18.65 18.43 17.88 17.75 17.58 17.30 16.79 16.45 15.33 15.11 14.99 14.57 13.72 13.21 13.12 12.53 12.43 11.18 10.80 10.61 10.52 10.45 10.29 10.21 7.55 7.25 6.81 6.60 5.46 5.37 5.04 2.79 -

10.00

20.00

30.00

33.50

40.00

51.81 47.78 44.11

50.00

59.51

60.00

67.01

70.00

71.37

78.97 77.67

80.00

84.41

90.00

100.00

Figure  9.2   Africa Infrastructure Development Index,  2010. Source:  The Africa Infrastructure Development Index, April 2013, Africa Development Bank  Group.

The results (scores and ranks) of the Africa Infrastructure Development Index and its components are presented in Figure 9.2 for 2010. The top ten ranked countries of the AIDI in 2010 were Seychelles, South Africa, Egypt, Libya, Mauritius, Tunisia, Morocco, Algeria, Cape Verde, and Botswana. Of these, five countries are in North Africa and three are Small Island countries where tourism constitutes an important sector of their economies. They have therefore traditionally focused on improving infrastructure to attract visitors. The bottom ten countries of AIDI in 2010 were Somalia, Niger, Ethiopia, Chad, Madagascar, Democratic Republic of Congo, Eritrea, Sierra Leone, Tanzania, and Mali. One shared characteristic of the bottom ten countries is that they are mostly fragile states, recently involved in some form of conflict.

9.3.3  Social inclusion Social inclusion ensures that all sections of the population including the disadvantaged due to their individual circumstances have equal opportunities. To ensure equal access to

166   Concepts opportunities, human capacities should be enhanced to bridge the gap that arise due to circumstances beyond the control of individuals, especially those from marginalized and disadvantaged section of the society including women. Service delivery in health such as basic health facilities and the provision of access to education, and the infrastructure to ensure access to these services, is key component of inclusive growth. Social protection is a key to social inclusion (see Ncube and Jones 2012). The role of social protection in preventing people entering into poverty, and in reducing the duration of poverty is well known. For some time social protection has been recognized as instrumental to achieve greater equality. More recently, experience has taught that when it is well-designed, social protection can both redistribute the gains from growth and, at the same time, contribute to higher growth. The recent food, fuel and financial crises have highlighted the importance of effective safety nets for reducing poverty and vulnerability. Growing evidence of the positive impacts of social protection safety net programs in both low and middle income countries as diverse as Brazil, Ethiopia, Mexico and Rwanda has helped placed safety nets firmly on the development agenda. The field of social protection has changed significantly in the last decade, and social protection programs now support more people than ever before. Social Protection has also gained importance and scope across Africa. Several African countries have started designing and implementing comprehensive social protection strategies. Global initiatives such as the Social Protection Floor10 are providing a far-reaching consensus and momentum behind developing and extending social protection beyond the lucky few. Social protection includes all initiatives, both formal and informal, that provide: • Social assistance to extremely poor individuals and households. • Social services to groups who need special care or who would otherwise be denied access to basic services. • Social insurance to protect people against the risks and consequences of livelihood shocks. • Social equity to protect people against social risks such as discrimination or abuse.11 Therefore social protection covers a wide array of instruments and objectives encompassing “risk reduction, risk mitigation, risk coping measures and transformative”. Within the context of a wider package of social protection, social transfers have a role in promoting growth by helping tackle risk and vulnerability. They can help address the high levels of inequality that can both reduce growth and the impact growth has on poverty giving poor households with some productive capacity greater confidence to undertake more risky activities, knowing they will have a minimum income to fall back on. When hit by crises, they have less need to sell their productive assets, thus are also more able to delay sales of produce, thereby obtaining a better price. Evidence is growing that beneficiaries of social transfers use them to invest in small-scale productive activities and assets, thereby setting in motion a potential multiplier effect. 10  11 

http://www.socialprotectionfloor-gateway.org/index.html. Devereux and Sabates-Wheeler (2004).

Inclusive Growth in Africa    167 Evidence shows that social protection is crucial for inclusive development by building assets to withstand future shock. Although it is becoming clear that economic growth is important and can be attained by different means, growth is unlikely to be sustainable in the long term unless all sections of society participate, derive benefits from and develop a stake in the growth process. Global shocks and crises, such as the food–fuel–financial (3F) crisis and more localized shocks (floods, droughts, hurricanes) impact society as a whole but tend to have a greater impact on the poor, particularly women, who have least resilience to cope with income fluctuations and little or no access to insurance and credit markets to help them maintain consumption (particularly food) when income falls or food prices rise. While the immediate effect on household level well-being may be visible through losses in income and increases in poverty rates, important long term effects some of which are irreversible, arise as households seek to cope. It is well documented for example that families may sell assets that they later struggle to recover resulting to borrowing food or cash, reducing meal frequency (eat less nutritious foods postpone or go without medical care and withdrawing children from school and encouraging them to work. The experience of two developing countries, Brazil and India show how social protection can aid in coping and recovery at the macro level. Brazil was one of the last economies in the world hit by the financial crisis of 2008 and also one of the first to recover. An important reason for this was an increase, since 2003, in the coverage and depth of Brazil’s social protection policies (Barbosa 2010). In India, the expansion of the National Rural Employment Guarantee Scheme (NREGS) was deemed to be a useful means to rapidly shore up aggregate demand while minimizing concerns about medium term fiscal sustainability (Chakroborty 2007). In Africa, social protection has been presented as an agenda that can strengthen the legitimacy of the state by allowing it to re-shoulder responsibilities for ensuring the basic survival of its citizens and so contribute to reducing political fragility and reducing the risk of a lapse into crisis. Social protection has established itself firmly on the policy agenda in most African countries and is increasingly being seen as an appropriate and affordable response to address long-term poverty and vulnerability. African governments are signatories to the African Union (AU) Social Protection Framework (SPF) enshrined within the Windhoek declaration of 2008.12 This agreement, aspires to provide a minimum package of social protection provision, comprising of grants for children, informal workers, the unemployed, older persons and the disabled, together with broader social policy provision, including basic health care, and an implied commitment to ongoing contributory pension schemes for civil servants. Social protection in sub-Saharan Africa takes several forms. Three of the most prevalent are social security, emergency relief, and social transfers.13

9.3.3.1  Social security Most countries in Africa have formal social security schemes for public sector workers and private sector employees. Social security typically includes unemployment insurance, 12 

13 

http://www.un.org/ageing/documents/SocialPolicyFrameworkforAfrica.pdf. Devereux and Cipryk IDS (2009).

168   Concepts disability provision and old age pensions. Government workers receive civil service pensions on retirement, either paid by the state or through employee contributions, while private sector workers have access to contributory private pensions. The main limitation of these schemes is their limited coverage. Formally employed workers in the public and private sectors, mostly living in urban centers, are covered, but this rarely extends beyond 10 percent of the population. The majority of citizens—rural smallholder farmers, informal sector workers, the self-employed—are not covered at all.

9.3.3.2 Emergency relief Much social protection in Africa has been delivered in the form of humanitarian relief in response to emergencies such as conflict, or following natural disasters such as drought. Chronically food insecure countries become chronically dependent on emergency food aid over several decades. The dominance of food aid in these humanitarian interventions, which usually target smallholder farming families, raises concerns about the disincentive effects on food production and local trade. The design of emergency relief interventions has provided the model on which many predictable social transfer programs in Africa are based.

9.3.3.3  Social transfers In African countries where financial and administrative capacity constraints make comprehensive social security systems unfeasible at this time, social transfer pilot projects are being introduced to provide social assistance to poor and vulnerable families. Social protection in Africa is dominated by ‘social cash transfers. They are targeted mostly at poor households with children and provide cash on condition that the children attend school and health clinics. Work programs, in which the state provides cash or food to the unemployed in exchange for work, combine elements of a social transfer with an insurance function, offering a safety net to those in the labor market. The main argument in favor of making social transfers conditional is that they provide strong incentives for families to invest in the health and education of their children. However, they require greater administrative capacity than simple unconditional cash transfers and depend on other services being in place. There are several positive examples of government-run social transfer programs. These include “social pensions”—more accurately, unconditional cash transfers targeted at older citizens—which were introduced in South Africa in 1928 and Namibia in 1973, but more recently have been adopted by Botswana (1996), Lesotho (2004), and Swaziland (2005). Other government programs include disability grants (South Africa and Namibia), poverty-targeted social assistance schemes (“Destitutes Support” in Botswana, “Food Subsidy Programme” in Mozambique, “Public Welfare Assistance Scheme” in Zambia), and child-focused schemes (“Child Support Grant” in South Africa), Ethiopia’s productive safety net program and Rwanda’s Vision 2020 Umerenge program.

Inclusive Growth in Africa    169

9.3.4  Political/institutional inclusion Inclusive growth requires supporting the voices and democratic accountability of the poorest and vulnerable groups, the democratization and representation at all strata of the population in all the economic and political spheres. Strengthening institutions of accountability and the rule of law can foster political stability. Inclusiveness also means public participation in the control and monitoring the management of public resources and that governments are held accountable for their economic and fiduciary responsibilities. Such system requires strong control, audit and judicial systems that can enforce the rule of law and the management of public affairs. Good governance, efficient government and strong institutions can promote inclusive growth. The lack of effective and accountable states in some parts of Africa hinders inclusion. The quality of a country’s institutions—including the quality of political representation and policy-making processes, the competence and integrity of the bureaucracy and the ability to enforce contracts and property rights—are a major factor in its overall economic performance and its ability to tackle poverty. Around the world, countries with better governance also tend to have lower inequality. Yet, governance remains Africa’s Achilles’ heel. The region has consistently performed poorly on standard governance indicators, scoring 30 percent lower than the Asian average and 60 percent lower than the average among industrialized countries. This poor governance performance significantly compromises Africa’s ability to lift its population out of poverty. Corruption costs Africa a quarter of its GDP annually, with the burden falling heavily on the poorest. It is estimated that, if Africa had the quality of institutions that most Asian countries achieved in the early phase of their industrialization, its collective GDP would be 80 percent higher than it is today.

9.4  Inclusive Growth Index In summary, the specific variables for the inclusive growth indicators, form the discussion above, are in Table 9.2. The indicators encompass economic, social, spatial, and political/ institutional inclusion. An inclusive growth index for Africa countries was created from the above variables, for the period 2006-2010. The results are in Table 9.3. Adjusting for inequality changes the inclusive growth standing of a country rather dramatically. For example Botswana drops dramatically to further down the rankings due to high inequality. South Africa is also in a similar position due to high inequality while have a high GDP per capita level. On the contrary Egypt, due to the impact of subsidies, and Ethiopia, their inclusive growth rankings, even adjusted for inequality, are favorable.

Table 9.2  Some indicators for inclusive growth Economic Inclusion Poverty/Inequality 1.  poverty headcount 2.  Gini coefficient 3. income share of the poorest 60 percent of the population 4. ratio of income/ consumption of top 20 percent to bottom 20 percent

Social Inclusion Social Infrastructure 23. net primary school enrollment ratio. 24. net secondary school enrollment ratio. 25. govt expenditure on education as % of total govt. expenditure. 26. under 5 mortality rate. 27. mortality rate for under age 40. 28. % of under 5 years who are underweight. 29. govt expenditure on health as % of total govt. expenditure. 30. % of population with access to safe water. 31. % of population with access to adequate sanitation.

Spatial Inclusion 32. intra African trade 33. regional labor mobility. 34. openness to trade. 35. spatial inequality. 36. Infrastructure access

Political/Institutional Inclusion 37. voice and accountability. 38. governance 39. the revenue/ GDP ratio. 40. public investment/ GDP ratio. 41. property rights

Productive Employment 5. share of employed or economically active in industry. 6. share of employed or economically active in manufacturing. 7. share of workers in non-agricultural paid employment. 8. own account and contributing to family workers Economic Infrastructure 9. average electric power consumption per capita. 10. proportion of population with access to electricity. 11. percentage of paved roads. 12. no of depositors per 1000 adults.

(continued)

Inclusive Growth in Africa    171 Table 9.2 Continued Economic Inclusion

Social Inclusion

Spatial Inclusion

Political/Institutional Inclusion

13. number of mobile phone subscribers per 100 people. 14. number of internet users per 100 people. Agriculture 1 5. fertilizer consumption 16. stable crop yield index 17. investment in irrigation 18. credit for inputs Gender Equity 19. gender parity in labor force. 20. share of women in nonagricultural wage employment. 21. ratio of literate females to literate males. 22. ratio of girls to boys in secondary/tertiary education.

9.5 Conclusion The analysis above has presented a discussion on inclusive growth in Africa. We have outlined Africa impressive growth in the last decade which has not created jobs and not reduced poverty fast enough. The growth has not been inclusive. The growth has been explained largely by a strong commodity price trend, while domestic demand factors have not been as strong relatively. We have presented what inclusive growth considerations are such as economic, social, spatial, and political/institutional inclusion. Under economic inclusion we presented an approach for dealing with inequality considerations. Finally, we presented an inclusive growth index for Africa, which allows us to compare countries. Questions still remain on issues such as:

• • • •

relative importance of inclusive growth factors; comparison of indices and measures that are aimed at capturing inclusive growth; policies and strategies for achieving inclusive growth; and customizing inclusive growth indicators to country specifics.

These would be issues for future research.

Table 9.3  Inclusive Growth Index (IGI) for Africa (2006–2010) Period: 2006–2010 Country

IGI without Inequality adjustment

IGI adjusted for Inequality

Rank of Country on IGI adjusted for Inequality

Egypt Botswana Mauritius Tunisia Algeria Gabon Comoros Zimbabwe Guinea South Africa Morocco Madagascar Namibia Congo, Rep. Rwanda Senegal Ghana Cameroon Gambia Tanzania Kenya Togo Guinea-Bissau Benin Angola Liberia Uganda Burundi Cape Verde Malawi Djibouti Zambia Mozambique Sierra Leone Mauritania Burkina Faso Congo, Dem. Rep. Côte d’Ivoire Ethiopia Nigeria Niger Mali Chad Central African Republic Sudan

0.5528308 0.706965 0.7047445 0.640328 0.6061221 0.7014509 0.5243774 0.5086617 0.5163358 0.585966 0.5328174 0.49342 0.5341125 0.5110457 0.4712551 0.4755806 0.470842 0.4671314 0.4503763 0.4446685 0.4397641 0.4337871 0.435576 0.4329841 0.4796888 0.4081629 0.4108735 0.4000683 0.4432636 0.394871 0.4154776 0.3980763 0.3897084 0.3754105 0.3844209 0.3745221 0.3594285 0.3642586 0.3372078 0.3393672 0.3177463 0.3194507 0.3150117 0.3046063 0.3114393

1.105662 0.7973578 1.05078 0.9115322 0.9406515 1.03896 0.5455507 0.6411654 0.6994971 0.6278517 0.8174527 0.6831058 0.5341125 0.6721544 0.6824394 0.6694724 0.7035923 0.7251865 0.5623108 0.7264412 0.5751081 0.7472274 0.5657359 0.6990635 0.4829763 0.6006409 0.5940161 0.668413 0.5591035 0.5663665 0.6047124 0.4786304 0.5284644 0.5535063 0.5408809 0.5527594 0.4973195 0.5446099 0.6336373 0.4436503 0.5008432 0.4811938 0.471312 0.349486 0.4875877

1 44 5 11 3 4 42 47 31 40 6 28 37 26 24 22 14 12 39 10 33 7 35 8 38 30 23 17 32 27 15 41 34 25 21 20 36 13 2 29 18 19 16 43 9

Source: Ncube, Shimeles and Younger (2013).

Inclusive Growth in Africa    173

References African Development Bank Group (2013). The African Inclusive Growth Index, April 2013, Tunis. Alkire, S., and Foster, J.E. (2011). Understandings and misunderstandings of multidimensional poverty measurement. Journal of Economic Inequality, 9(2):289–314. Atkinson, A. (1970). On the measurement of income inequality. Journal of Economic Theory, 2(3):244–263. Bourguignon, F., and Chakravarty, S., (2003). The measurement of multidimensional poverty. Journal of Economic Inequality, 1:25–49. Chakraborty, P. (2007) Implementation of the National Rural Employment Guarantee Act in India: Spatial Dimensions and Fiscal Implications, Working Paper no 505, National Institute of Public Finance and Policy, New Delhi, India. Dalton, H. (1949). Inequality of Income. London: Routledge and Kegan Paul. Datt, G., and Ravallion, M. (1992). Growth and redistribution components of changes in poverty measures: A decomposition with applications to Brazil and India in the 1980s. Journal of Development Economics, 38(20):275-295. Duclos, J.Y., Sahn, D.E., and Younger, S.D. (2006). Robust multidimensional poverty comparisons. Economic Journal, 116(514):943–968. Ferreira, F., Gignoux, J., and Aran, M. (2011). Measuring inequality of opportunity with imperfect data: the case of Turkey. Journal of Economic Inequality, 9(4):651–680. Imbs, J., and Wacziarg, R. (2003). Stages of diversification. American Economic Review, 93(1):63–86. Ianchovichina, E., and Lundstrom, S. (2009). Inclusive Growth Analytics:  Framework and Application. Economic Policy and Debt Department. Washington, DC: World Bank. Kakwani, N. (1997). Inequality, welfare and poverty: three interrelated phenomena. Discussion paper, School of Economics, The University of New South Wales, Sydney, Australia. Kakwani, N. (1997). On measuring growth and inequality components of poverty with application to Thailand. Discussion paper, School of Economics, The University of New South Wales, Sydney, Australia. Klasen, S. (2010). Measuring and Monitoring Inclusive Growth: Multiple Definitions, Open Questions, and Some Constructive Proposals. ADB Sustainable Development Working Paper Series, No. 12. Mandaluyong City, Philippines: Asian Development Bank. Lambert, P.L., Millimetb, D.L., and Slottjeb, D. (2003). Inequality aversion and the rate of subjective inequality. Journal of Public Economics, 87:1061–1090. Ncube, M., and Jones, B. (2012) Social protection and Inclusive Growth in Africa. Tunis: African Development Bank Group, Tunisia. Ncube, M., and Ondiege, P. (2012). Silicon Kenya: Harnessing ICT Innovations for Economic Development. African Development Bank Group. Ncube, M., Shimeles, A., and Younger, S. (2013). An Inclusive Growth Index for Africa. Tunis: African Development Bank Group. Ramos, X., and van de Gaer, D. (2012). Empirical Approaches to Inequality of Opportunity:  Principles, Measures, and Evidence, working paper, faculteit economie en bedrijfskunde. Rauniyar, G., and Kanbur, R. (2010). Inclusive Development: Two Papers on Conceptualization, Application, and the ADB Perspective. Mandaluyong City, Philippines:  Asian Development Bank.

174   Concepts Ravallion, M. (2004). Pro-Poor Growth: A Primer, World Bank Policy Research Working Paper 3242. Washington DC: World Bank. Ravallion, M. (2010). Mashup Indices of Development, Policy Research Working Paper 5432. Washington, DC: World Bank. Ravallion, M., and Shaohua, C. (2003). Measuring pro-poor growth. Economics Letters, 78(1):93–99. Roemer, J.E. (1998). Equality of Opportunity. Cambridge: Harvard University Press. Stiglitz, J., Sen, A., and Fitoussi, J.-P. (2009): Report by the Commission on the Measurement of Economic Performance and Social Progress. http://www.stiglitz-sen-fitoussi.fr/documents/ rapport_anglais.pdf. Tsui, K.Y. (2002). Multidimensional poverty indices. Social Choice and Welfare, 19:69–93.

Chapter 10

P overt y

 Shifting Fortunes and New Perspectives Abebe Shimeles

10.1 Introduction The first decade of the new millennium ended well for Africa compared with the previous “lost decades” where per capita incomes stagnated or declined in most countries. With the growth trajectory of the early decades reversed, Africa’s macroeconomic performance is celebrated globally. Leading media outlets such as The Economist (2011, 2013) punctuated Africa’s recent growth performance with “Africa Rising,” “Africa Emerging” in sharp contrast to “Hopeless Africa” it chronicled in early 2000. Others, like McKensie (2010), produced reports that heralded the beginning of new era in Africa. Indeed Africa has produced a contrasting growth narrative that clearly challenges the prevailing view and analysis on its economic prospects. Has this shift in the growth trajectory of Africa been translated into a sustained and comparable reduction in extreme poverty? The answer depends very much on how poverty is conceptualized and measured. The seminal work by Sen (1976, 1985) spurred research unrivalled in development economics that refined and deepened our understanding of poverty. Most of the empirical tools in applied poverty research have theoretical foundations in social choice theory that comply with widely accepted normative axioms. Refinements in methods of measuring poverty have also led to massive data collection efforts, particularly in developing countries in the form of household income and consumption surveys. Notable among this is the Living Standard Measurement surveys spearheaded by the World Bank in poor countries. Despite such advances, still major controversies abound on how poverty rates are computed across countries. There are two contending views on the path of poverty in Africa whose key difference lies in whether mean per capita income in a country should be drawn from national accounts or household surveys (Deaton 2005). The common approach promoted mainly by the World Bank is to draw distributional information and average welfare levels (per capita income or per capita consumption) from household budget surveys to compute income-based poverty. As a result, nearly all official poverty statistics reported

176   Concepts 80

Headcount

60 40 20 0 1980

1990

2000

2010

Year East Asia Latin America South Asia

Central Europe Middle East and NA SSA

Figure  10.1   Trends in extreme poverty by region. Source:  author’s computation from data provided in www.povcalnet.org.

by national governments in Africa are based on information drawn from household surveys. According to this approach poverty in Africa, defined as the percentage of the population earning an income level below 1.25 dollars a day per person, has been declining slowly (McKay 2013; Page and Shimeles 2013). Africa is considered “the last frontier” in the fight against extreme poverty in the world. In the early 1980s, sub-Saharan Africa had the lowest levels of extreme poverty compared to Latin America, East Asia, and South Asia. At the end of the 2000s, however, it had the highest rate of extreme poverty among these regions (Figure 10.1). On the other hand, Pinkovskiy and Sala-i-Martin (2013, 2014) argued that survey-based methods have overstated initial poverty and understated the pace at which it has declined over time. In their approach, which is based on mean income drawn from national accounts, initial poverty in sub-Saharan Africa in 1990 was around 34 percent and declined steadily in 2010 to around 21 percent, at a rate of almost 2 percent per annum. The true extent of poverty in Africa would probably continue to be a controversial issue. However, what is evident is that, extreme poverty in Africa, particularly sub-Saharan Africa, is still a major challenge that may have to be seen from the perspective of sustainability of the current growth spurt and transformation of sources of livelihoods in future. Much has been written in the last decade on persistence of poverty, household assets, and other indicators of well-being in Africa.1 The most widespread and contentious narration is whether African countries are confronted with initial conditions that seemingly make a sustained reduction in poverty an insurmountable task without meaningful positive exogenous shock, such as foreign aid, foreign direct investment, or other sources of development finance. The assertion began with the analysis of why

1 

For example, see McKay (2013), Booysen et al. (2008), and Sahn and Stifel (2000).

Poverty   177 Africa is growing slowly or not at all, with some attributing it to the hazards of bad climate and geography (e.g. Sachs and Warner 1997); anti-growth syndromes of different origins, such as bad policy, chronic corruption, etc. (Fosu 2009); artificial boundaries (Alesina et al. 2011); conflict (Collier 2004; Andrimihaja et al. 2012), and even system of slavery in pre-colonial periods (Nunn 2008). These studies implicitly or explicitly suggest that most African countries are too poorly endowed to grow and are locked in a low-income equilibrium trap (Sachs et al. 2007). The connection between growth and poverty is self-evident; without growth a sustained reduction in poverty and wealth creation is not feasible. There is enough evidence to suppose that the growth narrative has changed, and the poverty numbers seem to be improving over time. The question remains would African countries be able to sustain these gains? What steps should be taken to prevent growth collapse and rise in poverty? Has the current growth episode been accompanied by sufficient momentum in job creation? These are the issues most policymakers and development partners ponder in contemporary Africa. Thus, the quest for “inclusive” growth is now full steam in many countries. Against this background, this chapter attempts to provide perspective on the lingering issue of poverty traps (or reversal of fortunes) implied by most analytics and empirics, and addresses the potential for growth to affect poverty on a sustainable basis. The remainde of the chapter is organized as follows: section 2 reviews briefly the conceptual issues in the definition and measurement of poverty as it evolved in the literature in the last four decades, section 3 presents the theoretical basis for poverty traps in the context of Africa, with balanced review of the evidence. Section 4 discusses the link between growth and poverty, section 5 outlines the future policy shaped by the emerging reality, and section 6 concludes.

10.2  Measurement of Poverty: An Overview The seminal paper by Sen (1976) laid the foundation for the measurement of poverty. He argued that any measure of poverty should be able to provide information on two basic elements of poverty. Who are the poor? And how much poverty is experienced by them? The first is often associated with the prevalence of poverty in society and the second on the degree of deprivation experienced by the poor population. At the time, the headcount ratio (proportion of a population falling below a certain poverty line) and the income gap (the average income needed to lift one person out of poverty) were the popular measures of poverty. The Sen Index, introduced in Sen (1976) laid the foundation for an ethically consistent measure by combining the two basic aspects of poverty into a single index. More formally, consider a vector of income accruing to individual i in a given year in a country with ascending order: Y = (y1, y2, … , yn), so that y1< y2 … < yn. yi represents the income of individual i in a country. If z represents the poverty line, then, H (headcount ratio) can be written as:

H=

q

n

. (1)

178   Concepts Where, q represents the number of people with income level no higher than the poverty line z and n represents the total number of individuals in the country. In the same manner, we may express the income gap ratio as follows:

q

(z − yi ) . (2) n i =1

IG = ∑

Sen pointed out that H remains invariant to any change in the income of the poor (nothing happens to poverty until income of a poor person crosses the poverty line) and the IG is also remains invariant to any regressive transfer of income among the poor people. These observations led to the basic poverty equation formulated by Sen:

q

S(z , y ) = A(z , y )∑ (z − yi )vi (z , y ), (3) i =1

where S(z, y) is the aggregate income-gap of people whose income is no more than z, vi(z,y) is a non-negative weight given to the individual i, and A(z,y) is a normalizing factor. Subsequent developments in the measurement of poverty followed two approaches. Thon (1979, 1981), Takayama (1979), Kakwani (1980), Foster, Greer and Thorbecke (1984), and Foster and Shorrocks (1991) pursued Sen’s axiomatic approach to derive a poverty measure that satisfied certain desirable properties. Blackorby and Donaldson (1980), Clark et al. (1981), and Chakravarty (1983) used the notion of social welfare function and the underlying concept of “equally distributed income” to obtain an index of poverty along Atkinson’s (1970) inequality index. Currently applied research in poverty invariably uses the index by Foster, Greer, and Thorbecke (1984), which combines the headcount ratio, the income gap ratio and squared-poverty gap ration in three distinct measures. a



q  z − yi  p (z , y ) = 1 = ∑  , (4) n i =1  z 

where α ≥ 0. If α = 0, then p(z, y) reduces to H, whereas if α = 1 it reduces to the average income-gap. A higher value for α indicates increased concern for the poorest. Ravallion (1992) suggested that H measures the prevalence of poverty, I its intensity, and an indicator with α = 2 its severity. From equation (4) it is possible to note that the headcount ratio (α = 0) ranges between 0 (no one is poor) and 1 (everyone is poor). On the other hand, for the poverty gap (α = 1) and severity of poverty measures (α = 2), the values range between 0 and H, thus requiring rescaling to be meaningful and comparable across time and groups. The Foster–Greer–Thorbecke measures of poverty given in (4) have a number of desirable properties such as sub-group consistency and decomposability which are important for understanding poverty changes across sectors of employment, demographic characteristics and other desirable disaggregation. Equation (4) is estimated often using data on per capita income or per capita consumption collected through household surveys and it is known as money-metric measure of poverty. Sen (1976) expanded the definition of poverty and welfare by departing away from its utilitarian focus (happiness by having more of goods, etc., that underlie utility functions)

Poverty   179 to freedom of choice and the ability to function in society unfettered by circumstances. The capabilities approach to poverty has found practical and policy significance with the Human Development Index that is published by UNDP in its Human Development Report and recent research in multidimensional poverty formalized by Alkire and Foster (2011). For practical purposes, it is useful to combine different approaches to the measurement of poverty to gain robustness as well as clarity on the dimensions in which a country needs to focus to speed up improvements in overall well-being. The measurement of poverty in Africa has been problematic mainly for lack of reliable and comparable data on income, or consumption at the household level. The availability of household surveys in multiple waves in a number of countries in recent years has improved considerably our understanding of poverty in the continent. However, more work and concerted effort is needed to carefully construct cross-sectional as well as temporal trends in poverty. The sharp contrast between survey-based measures of poverty reported in Figure 10.1 and the one by Pinkovskiy and Sala-i-Martin (2013, 2014), which is based on national accounts, is further evidence that our knowledge of poverty in Africa is constrained by reliable data. While there are pros and cons in the use of either national accounts or surveys to track poverty it is reasonable to assume that significant variation between the two should be a cause for concern. Improvements in survey design and administration can bridge the gap between the two approaches substantially. For instance, replacing the collection of data on consumption expenditure from memories (recall data) by diaries could improve per capita consumption expenditure figures considerably (see Beegle et al. 2012). The use of other data sources such as the Demographic and Health Surveys could help to cross-reference the evolution of poverty using multidimensional indicators, such as disease burden, education, and access to clean water, electricity, and household assets. The availability of longitudinal data in recent years in selected countries has also provided insights into the persistence of poverty which is relevant for the discussion on potential for poor people to escape poverty, for the time it takes to end poverty spells, etc, and policy interventions that could assist in fighting chronic poverty.

10.3  Is Poverty Entrenched in Africa? Economic theory attributes self-reinforcing poverty due to either market failure or bad institutions (Azariadis and Stachurski 2005).2 Neo-classical growth theory predicts that if markets and institutions perform well, then, poor countries should be able to grow faster than richer countries due to diminishing returns to capital. That is, return to capital would be consistently higher in low-income than in high-income countries. However, Sachs et al. (2007) argued that in the case of African countries, particularly in sub-Saharan Africa (except South Africa) the assumption of high returns to capital is unrealistic in an environment where basic infrastructure (road, power, human capital) is nearly non-existent. There is minimum threshold of capital needed before self-reinforcing growth can be realized. This non-convexity in production functions generates two types of economies: one that 2  For an example of theoretical models that describe different mechanisms by which poverty traps may result, see for instance Lopeza et al. (2011), Goodhand (2003), and Ghiglino and Sorger (2002).

180   Concepts perpetually grows, and another that experiences growth collapse. Africa is in the latter category. The implications of such characterization of African economies as articulated in Sachs et al. is that a massive injection of capital in the form of aid is needed before these economies are ready for take-off. This is indeed a resurrection of the Big-Push approach that justified for development assistance in the 1950s. The assertion African countries are too poor to grow sparked research to investigate the empirical basis of its predictions and assumptions. Easterly (2006) undertook extensive documentation of growth performance of African countries between 1950 and 2001, finding no basis for zero per capita growth in the long term, which is the empirical implications of the poverty trap hypothesis. Similarly Kraay and Raddatz (2007) could not find evidence of the poverty trap using a canonical neo-classical model along the lines of Sachs et al. (2007) for African countries. On the other hand, Berthlemy (2006), based on semi-parametric estimate of growth dynamics for individual African countries, reported prevalence of poverty traps for most countries where growth episodes remained cyclical reverting to initial per capita levels.3 Table 10.1 updates Easterly (2006) and reports per capita gross domestic product (GDP) growth for countries who were in the bottom quintile in 1962 by setting them as dummies for three overlapping periods: 1962–2011; 1962–1995; and 1995–2011, covering 42 countries for which we have balanced data for the entire period. In the long term, if initially poor countries grow slower than the “rest,” then there is a “sign” that initially poor countries may be “stuck” in low-income equilibrium, with no potential of catching up with the relatively “richer” countries, which is the prediction of a self-reinforcing low-income trap or poverty trap. The table indicates that for all periods examined, countries that had started out as poor in 1962 have been growing at a faster rate than the rest of the “better-off ” group. This trend is unchanged by looking at structural breaks as well, where during 1960–1995 Africa on the whole experienced a downward trend in the growth episode and has recovered since then. This evidence poses a challenge to the idea of stagnant per capita GDP for the poorest countries and at least at the macro-level there is no visible poverty trap and the neo-classical predictions of conditional convergence seems to be at work. Poverty trap studies at the macro-level generally focused on dynamics of per capita GDP growth in a cross-country context, finding on balance that countries with a low initial level of per capita income grow faster than those with high per capita income. Even among African countries, nearly all growth regressions indicated the existence of conditional convergence in incomes. If this is true, then such process should also imply a convergence in poverty levels as growth necessarily leads to lower poverty. Recent studies have shown a contrary result. For instance, Lopez and Servén (2009) and Ravallion (2012) using a sample of developing countries reported that high initial poverty would hinder growth. This is a very important result that could have serious implications, mainly to most countries in Africa which have a high incidence of initial poverty. The mechanism by which this empirical regularity is explained is along the lines of the theory of the poverty trap alluded to previously. As new and reliable data sets become available, or sub-samples are used, the empirical results may change. For example, for the Africa sub-sample using Ravallion’s (2012) data set, Shimeles and Thorbecke (forthcoming) found that high initial poverty does not seem to affect growth. A much more disaggregated and large survey data are needed to unpack results of the cross-country narrative. 3 

See also Cazzavillan et al. (2013) for recent evidence on poverty trap using cross-country data.

Table 10.1  Regression of growth by quintiles of initial per capita GDP in each period for Africa

Dummy for poorest quintile std error t-stat Constant (per capita growth of upper 4 quintiles) std Error t-stat Observations F(1, N-2) Prob > F R-squared Root MSE Countries Reject stationary income for poorest fifth

Absolute poverty trap 1962–2011*

Relative poverty trap* 1962–2011

Absolute poverty trap 1962–1995

Relative poverty trap 1962–1995*

Absolute poverty trap 1995-2011

Relative poverty trap* 1995-2011

0.0117 0.0031 3.7275 0.0095

0.0117 0.0033 3.5885 −0.0002

0.0128 0.0042 3.0278 0.0045

0.0128 0.0044 2.9043 −0.0003

0.0090 0.0039 2.2907 0.0195

0.0090 0.0041 2.1886 0.0004

0.0016 6.0411 2058 13.89 0.0002 0.006 0.062 42 Yes

0.0017 −0.1078 2009 12.8800 0.0003 0.0057 0.0641 41

0.0020 2.2876 1386 9.17 0.0025 0.0066 0.0646 42 Yes

0.0021 −0.1631 1353 8.43 0.0037 0.0062 0.06729 41

0.0025 7.7850 714 5.25 0.0223 0.0044 0.05539 42 Yes

0.0026 0.1509 697 4.79 0.029 0.0042 0.05729 41

Source: author’s computation based on Penn World Tables. *Here per capita growth refers to growth of the ratio of per capita GDP of each country to that of South Africa.

182   Concepts For studies that used micro-data in Africa the evidence is suggestive of existence of poverty traps in line with Lopez and Servén (2009) and Ravallion (2012). One of the most common causes of poverty traps in Africa is a situation where credit or borrowing constraint coupled with income risk could make an initially poor household or an individual to remain poor for an extended period (Dercon 1998; Barrett and Swallow (2006); Barnett and Barrett 2008; Santos and Barrett 2011). This is not surprising. Subsistence farmers in rural areas in most African countries lead precarious livelihoods (exposed to income risk due to shocks), and have no access to financial services to invest in high-return activities. As a result, those who are already poor are unable and unwilling to undertake risky and costly investments that could have higher future returns. Dercon’s (1998) work in rural Tanzania showed that poor people would not engage in cattle rearing even though this particular activity had a high potential for wealth accumulation. The reason is that poor households had no access to credit to finance initial cost of acquiring cattle and due to income risk they could not save enough to self-finance as they would have to smooth consumption in time of shocks. Without external intervention, livelihood for initially poor households would propagate poverty. Similarly, major but short-lived shocks, such as natural disaster (drought, crop failure, illness, etc.), conflict, and political instability would lead to persistence of the shocks for a long time. Studies by Dercon (2006), Dercon and Christiansen (2011), Bigsten and Shimeles (2008) for Ethiopia, Giesbert and Schindler (2012) for Mozambique, Carter et al. (2007) for Ethiopia and Honduras, Radeny et al. (2012) for Kenya, and other studies showed the existence of path dependence in the incidence of poverty at the household level.4 The body of work based on micro-data seems to support the finding that some livelihood systems in Africa, particularly farming and informal activity are prone to self-reinforcing poverty.5 While it is possible for a country at a macro-level to experience faster growth over extended period of time, a large segment of its population thriving in farming, small-scale informal activities, and other labor-intensive activities could be mired in poverty traps. This is the reality most low-income African countries are confronted with. This partly may explain the apparent inconsistency between high economic growth and low pace of poverty reduction, which is the subject of the next section.

10.4  Growth, poverty reduction, and wealth creation It is established fact that economic growth has been high in most African countries in the past decade. The prevailing view that African countries are too poor to grow is increasingly

4 

See for instance Naschold (2012) for reported existence of poverty traps for households in India living in semi-arid areas. 5  The body of work that focused on investigating poverty traps at the household level is still evolving. There are studies that reported of finding no poverty traps in the African case as well (e.g. McKay and Perge 2011). However at least there is strong evidence from household panel surveys of multiple rounds that there is a ‘true’ state dependence in the evolution of poverty and poverty spells, which suggest the persistence of poverty following short-lived shocks.

Poverty   183 7.6

7.4 .02 7.2 .01

Per capita GDP

Real per capita GDP growth

.03

7

6.8

0 1960

1970

1980

1990

2000

2010

Year

Figure  10.2   Log per capita GDP and per capita GDP growth for Africa:  1960–2011. Source:  author’s computation based on data from Penn World Tables.

challenged by these recent trends. Some may argue that the recent growth spurt is nothing else but a recovery of the ground that has been lost in previous decades (e.g. Larke and Milanovic 2013). It is true that Africa experienced significant growth regression during the 1970s, 1980s, and early 1990s. However, the growth experienced in recent decades was more than a recovery and per capita GDP levels have been much higher than they were in early decades (Figure 10.2). In fact, there is evidence suggesting that recently widespread growth acceleration has taken place unprecedented for decades. Following the definition of growth acceleration by Hausmann et al. (2005),6 Figure 10.3 provides the proportion of countries that have completed growth acceleration in a space of five years since the 1960s. According to the Figure 10.3, during the early years of the 1960s there was no African country that registered a growth episode that could be labeled as a growth acceleration based on the definition adopted here. During 1966–1971, 15 percent of the 48 countries for which data were available had at least one growth acceleration. Since then, the proportion of countries having completed growth acceleration in a space of five years started to decrease at a rapid rate, reaching a bottom of 2 percent in the early 1990s. Then, things started to improve and in the early part of 2000, close to 23 percent of African countries from the same sample completed growth acceleration. It is also important to note that only eight countries completed multiple growth accelerations in the last 45 years, indicating the challenge Africa as a whole faces in sustaining rapid growth over an extended period. Still, the recent improvement may not be underestimated. 6 

A country is said to have experienced an episode of growth acceleration if the following three conditions are met: (i) per capita GDP has grown at a rate of at least 3.5 percent or more, (ii) growth acceleration (the rate of growth in per capita GDP growth during the same period) is at least 2 percent, and (iii) per capita GDP at initial period is higher than the last period in the growth episode.

184   Concepts

Proportion of countries with growth acceleration

.25 .2 .15 .1 .05

2002 to 2006

1997 to 2001

1992 to 1996

1987 to 1991

1982 to 1986

1977 to 1981

1972 to 1976

1967 to 1971

1962 to 1966

0

Period

Figure  10.3  Lowess estimate of proportion of African countries with at least one growth acceleration (Penn World Tables:  N  =  48). What is not very clear is whether poverty has been declining and wealth widely distributed corresponding to the high growth experienced by many countries. It is difficult to provide conclusive evidence to the link between growth and poverty. Most African countries undertake household surveys in intervals of three to five years, and often with no regard to comparability and consistency of survey designs (Deverajan 2013). Thus, one has no option but to patch up the evidence from fragments of individual surveys collected in different periods. The most widely used data by researchers is that provided by World Bank on its website www.povcalnet.org where “official” income distribution data are available for a large number of African countries for the period 1981–2010.7 The evidence from these data as shown in Figure 10.1 for sub-Saharan Africa does suggest that poverty has declined only by about 5 percentage points in the last decade or by about 1 percent per annum. When one compares with the per capita growth rate of close to 2.5 percent, the pace of poverty reduction is low. This evidence is consistent with other studies that used unit record data for selected African countries for two or more waves (e.g. Page and Shimeles 2013). It is important to point out that alternative approaches that rely on a combination of national accounts (to estimate mean income) and surveys (to estimate distribution of income) have reported a rapidly falling poverty in the last two decades (e.g. Pinkovskiy and Sala-i-Martin 2013, 2014). These estimates suggest a fall in poverty at a rate of 1.9 percent per annum, almost double to that obtained from household surveys. The other piece of evidence that may shade light on growth and poverty reduction can be obtained from the Demographic and Health Surveys (DHS) that document household

7  This poverty data is currently under revision using the results of the recent Purchasing Power Parity data from the International Comparison Program, which may significantly affect the trend reported here.

Poverty   185 14.00 12.00 10.00 6.00 4.00 2.00 0.00 Se ne ga Ke l ny G a ha N na am ib N ia ig e M ria or oc Ta co Cô nza te nia d'l vo ire Ch a G d ui ne a N ig U er M gan ad da a Bu gas rk car in af as o Be ni Rw n an d M a a Ca law m i e Zi roo m n ba bw e Eg yp t

Percentage (%)

8.00

–2.00 –4.00 –6.00 –8.00

Figure  10.4   Change in middle class status on the basis of household wealth/asset for selected African countries during 1990–2010. wealth or asset in great detail, and are comparable across a large number of countries in multiple waves. Based on this data set, Young (2012) reckons that per capita consumption on average has increased at a rate of 3.5 percent to 3.7 percent in the last two decades in Africa. This implies that on average most African countries might have grown at a rate of 7 percent in the last two decades. During this period, Ncube and Shimeles (2013) using DHS data reported that the size of the middle class increased in 21 of the 25 countries that had multiple waves (Figure 10.4). Some countries like Senegal, Ghana, and Kenya achieved rapid increase in the size of the middle class, but others have made slight improvements. The average change in the size of the middle class between the 1990s and the 2000s was about 3 percentage points or a rise from 7 percent to 10 percent. This is not comparable to the average expansion in African economies Young (2012) reported using similar data sets. In addition, if we examine the possibility of transiting into a middle class from being poor in terms of wealth, the picture we get is not encouraging. Table 10.2 presents transition matrix based on a synthetic panel constructed from a large number of African countries in multiple waves using time-invariant characteristics such as age and sex as a cohort. What emerges is that there was only a 5 percent probability for a household that was asset poor to transit into a middle class status and

Table 10.2  Transition matrix by wealth status for African countries 2005-2011 Pre-1995

Poor

Middle class

Upper class

Total

Poor Middle class Upper class Total

95.61 2.33 0 87.5

4.39 69.77 0 8.81

0 27.91 100 3.69

100 100 100 100

Source: computations by Ncube and Shimeles (2013) based on DHS data for 35 African countries.

186   Concepts vice versa. There is a clear persistence of class over time. Similarly, those that started out as a middle class in pre-1995 had a 70 percent probability of remaining middle class and a very small chance of slipping back into poverty. Actually, they had a better chance of moving up into the upper class over time. On the whole, the probability of being asset poor at any time was around 87 percent, which still is extremely high incidence when one takes into account housing conditions, household utilities, and other indicators of wealth. The DHS data could also be used to construct a multidimensional measure of asset-based poverty that may provide additional insight into the evolution of household welfare through access to a wide range of amenities and utilities. Table 10.3 reports the trend in asset-based poverty constructed on the basis of access to nine household amenities and utilities, including type of housing (roof-top and floor); clean water, electricity, and toilet; ownership of household durables such as radio and television. By our definition, households who live in houses with a grass roof-top and mud floor, no access to clean water, electricity, and toilet, and do not own a radio or television are classified as poor. Table 10.3 reports asset-based multidimensional poverty for a block of four periods, since very few countries had their surveys in the same year. The results bear some similarities with both national accounts and survey-based estimates of poverty. It is remarkably similar with estimates by Pinkovskiy and Sala-i-Martin (2013, 2014) in its range of poverty estimate between the end periods (pre-1995 and 2005–2011). On the other hand, the pace of poverty reduction stalled or declined very slowly after 1995, still echoing the pattern in the survey-based estimates of poverty. To examine the link between the asset-based poverty measures and (long-term) growth, Figure 10.5 provides a simple correlation between poverty and log of per capita GDP from World Penn Tables. The implied elasticity is approximately −0.94, that is, if per capita GDP rises by 1 percent then poverty might fall by about 0.94 percent, which is close to unity. This result is closer to earlier estimates on growth elasticity of poverty for Africa (UNECA 1999; Dollar and Kraay 2002). The corresponding elasticity implied by the poverty trend in Pinkovskiy and Sala-i-Martin (2013, 2014) is about −1.3. That is, a 1 percent growth in per capita GDP would lead to more than 1 percent decline in poverty, which is higher than most available estimates. Analytical derivations have shown that the growth elasticity of poverty is mainly driven by the level of initial development, initial inequality and the position of the poverty line in the distribution of income (Bourginon 2002; Bigsten and Shimeles 2006). Poorer countries with high initial inequality may find it harder to attain rapid reduction in poverty through growth alone.

Table 10.3  Multidimensional asset poverty for selected African countries

Period

Number of countries

Asset poverty Population (percent) coverage (Median, (percent) unweighted)

Asset poverty (percent) (Median, weighted)

Asset poverty (percent) (Mean, unweighted)

Asset poverty (percent) (Mean, weighted)

1990–1994 1995–1999 2000–2004 2004–2011

16 22 18 24

42.5 47.9 56.4 63.5

41.3 24.4 19.1 26.3

38.7 (15.7) 27.7 (17.9) 28.4 (20.4) 26.1 (15.4)

40.6 (14.0) 21.0 (18.2) 25.8 (27.2) 27.8 (20.4)

36.5 27.1 26.1 25.8

Source: author’s computations from 82 country-year matched Demographic and Health Survey waves.

Poverty   187 80

Proportion of asset poor

60

40

20

0 5

6

7 Log per capita GDP

8

9

Figure  10.5  Multidimensional asset-based poverty and per capita GDP for selected countries in Africa. Source:  author’s computations.

One of the reasons why growth might not lead to a significant reduction in poverty may have to do with the sectoral composition of growth and employment. From Table 10.4 it is easy to infer that in Africa close to 85 percent of poverty originates either in agriculture (54 percent) or services (31 percent), and the poverty impact of growth depends on what has happened to these sectors in the last decade. Furthermore, as extreme poverty rises, the gap in poverty between those employed in agriculture widens in comparison to those in industry or services (Figure 10.6). For some poor countries, bringing poverty levels in the agricultural sector down to those in services

Table 10.4  Decomposition of poverty by sector of employment in Africa Sector of employment

percent

Agriculture Industry Services Residual Total

54 12 31 3 100

Source: author’s computation based on recent 26 household surveys (2005 and latest) for 18 African countries.

188   Concepts

Headcount ratio

80 60 40 20 0 0

20

40

60

80

Average headcount ratio Agriculture Services

Industry

Figure  10.6   Poverty in agriculture, services and industry sectors by the level of aggregate poverty. Source:  author’s computations.

would take them a long way in dealing with extreme poverty. The picture we get for most African countries is a clearly dichotomous economy where on the one hand a higher level of poverty in agriculture coexists with relatively low poverty in the industrial sector. In the extreme case where poverty is rampant, it does not matter in which sector of the economy one is employed. Almost everyone is poor. The importance of structure of employment, rather than average growth in per capita GDP for impacting significantly on poverty, is illustrated in Table 10.5. It shows that for a sample of developing countries, extreme poverty responded much strongly to industrialization (or rise in employment in the industrial sector) than to average growth in per capita incomes. A 1 percent increase in the share of employment in industry could lead to a 0.7 percent decrease in poverty. On the other hand, a 1 percent increase in per capita GDP was associated with only a 0.3 percent reduction in poverty. The picture for the sample of African countries is different. Structure of employment in agriculture and services had a rather poverty increasing effect (though the associated coefficient was significant at 10 percent), and no statistically significant effect was observed for changes in employment in the industry sector. This result is not surprising. Few countries had rising industrial employment in Africa. The fact that the developing world has managed to significantly reduce extreme poverty in the last decade, and that this achievement is mainly driven by shifts in the structure of employment than in growth in average incomes sends a clear message to African countries that have still a long way to go in fighting poverty.

10.5  Implications to Development Policy Despite encouraging growth and falling poverty, Africa still harbors a significant share of its population living in extreme poverty. A typical African country may not be in a poverty trap at the macroeconomic level. Positive exogenous shocks such as improvements in terms

Poverty   189 Table 10.5  Two-step GMM estimate of the relationship between poverty and sectoral shares of employment Dependent variable (Log headcount ratio) Log share of labor in agriculture Log share of labor in industry Log share of labor services Log consumption Log gini coefficient Constant Sargan’s statistic (over- identification test Number of observations

All Developing Countries

Africa Sample

Coef.

z

Coef.

z

0.05 −0.79** −0.61 −0.30** 2.52*** −0.94 0.8492

0.19 −2.87 −0.58 −2.44 6.34 −0.34

1.23* -0.29 1.29* -1.12** 2.77*** -10.54 0.3464

1.95 -0.37 1.94 -2.14 4.61 -2.08

328

58

*** Significant at 1 percent; ** significant at 5 percent; * significant at 10 percent. Source: Page and Shimeles (2013).

of trade, particularly the steady rise in the prices of commodities, increase in the flows of FDI and remittances, and better macroeconomic management, all contributed to reviving growth and spreading it across many countries in Africa in the last decade. On the other hand, progress in poverty and wealth creation is not keeping pace with the growth story. It is changing very slowly and it is worrying, particularly in light of recent findings that high initial poverty may hinder growth. The obvious reason why poverty is not declining faster lies in the fact that fast-growing sectors do not employ many people, and sectors that do employ many people have not been growing faster. Until recently, Africa has not seen structural change as a driver of growth rather than an increase in productivity in the small but dynamic modern sector (McMillan 2013). The challenge for African countries and their development partners is to figure out how to set in motion structural change in their economies and what the available policy options are. One of the recognizable, but less emphasized, defining characteristics of most African economies is that they have a traditional sector that has been a mainstay of the bulk of the population for generations, and a modern sector that evolved in recent decades through urbanization and a process of natural resource extraction. These two economic systems are governed by a completely different set of technology, incentive structure, risk, access to resources, and infrastructure (Monga 2013).8 Yet, the overwhelming policy orientation in the past two decades that governs the whole economy, including the traditional sector, has been the neo-liberal approach that focuses on getting fundamentals right and everything else falls into place. Early development economists on the other hand made a point that to spur growth in a dual-economy setting, labor has to move away from the traditional to the modern sector. That is structural change is the driving force for economic growth.

8  Paternostro (1997) outlined the poverty trap underlying a dual economic system such as described here.

190   Concepts In this connection, Rodrik (2013) argues that both the neo-classical growth model and the structural change approach to development in a dual-economy setting complement each other, and he outlined a development framework that combines the two approaches. In his approach, countries will grow fast and sustain growth if they deepen reform in getting the fundamentals right and also promote structural change in their economy with sector specific programs, such as industrial policy. Focusing on one while neglecting the other would lead to a sub-optimal growth trajectory. In his typology, Rodrik outlines that countries that focus and invest less in economic fundamentals (improved governance, macroeconomic management, openness, rule of law, property rights, better investment climate, etc.), and also less in promoting structural transformation (industrial policy, subsidies of specific sectors, infrastructure and technology investment, rural transformation, etc.), will have no growth at all. Similarly, countries that invest in fundamentals but fail on structural transformation could only see episodic growth that is not sustainable. Those that do very well in improving economic fundamentals but are slow in promoting structural transformation grow only slowly, etc. In terms of priorities, improving the fundamentals of the economy is a necessary condition for achieving sustainable growth. The policy sequences and implementation strategy that echo Rordrik’s typology is described at great length in Li and Monga (2011). Poor countries intent on breaking the vicious circle of poverty on a sustainable basis will have to make a conscious effort to learn from the experiences of countries that have had a similar endowment structure, and take concrete actions to remove constraints that prevent existing firms with similar economic structure from upgrading their technologies; or if such firms do not exist in the domestic economy pursue ways and means of creating them; pay due attention to potential innovations by private enterprises for scaling up and replication, etc. In short, Li and Monga (2011) emphasize the role of the state in forging structural change in poor countries. Policymakers in Africa and their development partners may need to pay heed to this emerging new policy orientation. Recently, Page and Shimeles (2013) argued that for development aid to have stronger impact on employment and poverty, it should focus on activities that promote structural transformation. In their review of the trend of development assistance in the last two decades, there has been a clear shift away from “productive” sectors to financing “social or non-productive sectors.” In addition, most policymakers in low-income countries have withdrawn their effort to modernize their economy, leading in some instances to a process of deindustrialization, particularly during the era of Structural Adjustment Programs. There has been some effort in early 2000 by the World Bank to revive the making of industrial policy in Africa through government–private sector dialogue using the East Asia model. A synthesis of the case study of Ethiopia, Senegal, Tanzania, and Uganda by Page (2013) on a Presidential Advisory Council that was set up with the support of the World Bank revealed that the process has created a useful forum for the government to understand the key practical constraints investors face, and agree on specific measures. Except for Uganda, so far the influence of these councils in facilitating bold and experimental undertakings that push the agenda of industrialization is questionable. Currently there is a wide recognition on the need for industrial policy, but governments in Africa need a practical guide on how best to intervene to kick-start the process. In their extensive discussion, Stiglitz et al. (2013) outlined a useful framework to implement industrial policy with a view of initiating structural transformation where they focused on how to deal with issues of coordination failure, externalities and other “hard” and “soft” infrastructure needed for a process of unfettered industrialization.

Poverty   191 Structural transformation is not only about industrialization. It is also transforming the traditional sector through a set of policies that reduce income risk, provide access to modern technology, financial services, and markets in order to increase productivity in the sector. There are a number of programs in several African countries that are designed to transform the rural economy and small businesses in urban areas. The success of such programs reduces the risk of self-reinforcing poverty conditions that are reported in a number of countries. Finally, investment in health and education, apart from playing an important role in breaking the poverty trap, also assists in reducing wealth and income inequality. Figure 10.7 illustrates a strongly declining wealth inequality as a country’s mean level of education rises. In poor countries, differences in educational attainment are one of the key drivers of differences in observed inequality, particularly at higher levels of education. For instance, Ncube and Shimeles (2013) reported that the role of education in explaining the variance in wealth status of households exceeded 25 percent in 30 of the 83 regression decompositions they had undertaken. Not only that, the “returns” to education in terms of asset or wealth acquisition was always higher for those who completed either tertiary or secondary education in comparison to those individuals with no education across the entire spectrum of educational achievement. This is one of the neglected areas in most parts of Africa. The rise of educated unemployed speaks for the mismatch between demand and supply as well as the quality of education (African Economic Outlook 2012). The same also applies with respect to inequity in health (e.g. Moradi and Baten 2005).9

10.6 Conclusion Development economics has pushed the boundaries in the last four decades by providing the analytical foundations to understand poverty and its link with economic growth. Poverty statistics are now routinely reported by national governments and development agencies to monitor its incidence and severity. As the discussion in this chapter indicated, average growth in the economy may be necessary but not sufficient to affect the pace of poverty reduction significantly. The pattern and source of growth remains important. In particular, coexistence of persistent poverty among the self-employed in the rural traditional and urban informal sectors with a modernizing and rapidly growing modern sector puts structural transformation at the center stage of development policy for African countries. The evidence for a sample of developing countries suggests that extreme poverty can be dealt better by moving labor from low to high productivity sectors. The last decade has witnessed a shift in the fortunes of most African countries, where per capita incomes have been steadily rising for the first time in decades. The corresponding effect on poverty however has been less clear and controversial. This is worrisome for a continent that still harbors a very large segment of its population mired in abject poverty. 9 

See also Kalwij and Verschoor (2007) for a further discussion of vagaries of inequality.

192   Concepts

Gini coefficient for asset index

.8

.6

.4

.2

0 .1

.2

.3

.4

.5

Proportion of households with secondary or higher education

Figure  10.7  Gini index for asset (wealth) and educational achievements for selected African countries. Source:  computations based on unit record data from DHS for 82 country-year matched  data.

Most African countries indeed have seen a rise in the middle class and some reduction in poverty. On the other hand, poverty is also deeply entrenched. The urgency for rethinking and calibrating policy options cannot be overemphasized. The future if harnessed properly favors Africa. It is the only continent where its labor force is young and is growing, offering it an opportunity known as the “demographic dividend” at the time when the global labor market might face an excess demand. In addition, there is a natural resources boom in nearly 45 African countries whose potential revenue to the government coffers is in the order of 150 percent of current GDP of the continent. These two opportunities need not be wasted. This chapter brought into perspective the analytical and empirical literature that seemingly told inconsistent stories on Africa’s prospects. The empirical growth literature based on cross-country data generally could not establish the long-held view that Africa had adverse initial conditions that keep it locked in a poverty trap. While this narration may fit into the growth spurt that has been observed in the last decade, another strand of literature brought out a powerful result that high initial poverty is bad for growth, which also echoes the findings that high initial inequality is bad for growth (e.g. Fosu 2010). This link between poverty and growth is also consistent with the findings of micro-studies that investigated the existence of multiple equilibriums using longitudinal data for selected African countries. There are large segments of Africa’s population for whom staying poor for an extended period of time has become the fate of life. Such a phenomenon of low productivity in the subsistence and informal sector, coexisting with a dynamic, rapidly growing modern sector is responsible for the weak link between average growth and poverty reduction.

Poverty   193 The dual economic structure typical of most low-income African countries calls for a policy paradigm that improves both economic fundamentals at the macro-level and concerted efforts to speed up structural transformation at the sectoral and micro-level (e.g. Monga 2013; Rodrik 2013). There has been much improvement in Africa over the past years in areas of market-led economic reform that included privatization, liberalization, and stabilization. While important, these policy measures are not sufficient to spur structural transformation that allows people to move from the low- to the high-productivity sector, and thus reduce poverty significantly. The experiences of developing countries suggest that structural change is more powerful than the average growth in per capita GDP to bring about a meaningful impact on poverty. The range of policy measures needed to achieve such transformation varies with the specific context prevailing in each country, but one thing is clear. Without structural transformation, accompanied by ongoing reforms to improve economic fundamentals, current growth would eventually peter out or slow down significantly. Some African countries that have started but slowed down the path towards modernizing their economy through continuous dialogue with the private sector may have to reconsider resuming this practice and others to follow suit as these forums provide a practical guide on what is needed by investors and entrepreneurs to move the agenda forward. Commitment to reform the educational and healthcare systems, particularly for the poor and vulnerable, and facilitating financial and technical support to farmers and small businesses, need to be backed by meaningful action and experimenting with new ideas and innovations.

Acknowledgments I thank Célestin Monga, co-editor of the Handbook for extensive and insightful comments that considerably improved the initial draft. I express my gratitude to Tiguene Nabassaga, African Development Bank, for excellent research assistance. I remain responsible for all errors in the chapter. The views expressed in this chapter are that of the author, and not that of the African Development Bank Group and its Board of Directors or the countries they represent.

References African Economic Outlook (2012). Promoting Youth Employment. AfDB, OECD, UNECA, UNDP. Alkire, S., and Foster, J. (2011). Counting and multidimensional poverty measurement. Journal of Public Economics, 95(7–8):476–487. Alesina, A., Easterly, W., and Matuszeski, J. (2011). Artificial states. Journal of the European Economic Association, 9(2):246–277. Andrimihaja, N.A., Cinyabuguma, M., and Devarajan, S. (2012). Avoiding the fragility trap in Africa. Working Paper Series, 5884. Washington, DC: World Bank. Atkinson, A.B. (1970). On the measurement of inequality. Journal of Economic Theory, 2:244–263.

194   Concepts Azariadis, C., and Stachurski, J. (2005). Poverty Traps. Handbook of Economic Growth, Volume 1A. Amsterdam: Elsevier B.V. Barnett, J.B., and Barrett, B.C. (2008). Poverty traps and index-based risk transfer products. World Development, 36(10):1766–1785. Barrett, B.C., and Swallow, B.M. (2006). Fractal poverty traps. World Development, 34(1):1–15. Beegle, K., De Weerdt, J., Friedman, J., and, Gibson, J. (2012). Methods of household consumption measurement through surveys: experimental results from Tanzania. Journal of Development Economics, 98(2012):3–18. Berthelemy, J., (2006). Convergence and Development Traps: How did emerging economies escape the under-development trap? Growth and integration – Annual World Bank Conference on Development Economics. F. Bourguignon and B. Pleskovic (eds), pp. 127–157. Bigsten, A., and Shimeles, A, (2008). Poverty transition and persistence in Ethiopia. World Development, 36(9):1559–1584. Bigsten, A., and Shimeles, A. (2006). Can Africa reduce poverty by half by 2012? Development Policy Review, 25(2):147–166. Blackorby, C., and Donaldson D. (1980). Ethical indices for the measurement of poverty. Econometrica, 48:1053–1060. Booysen, F., Van der Berg, S., Burger, R., and Maltitz, V.M. (2008). Using an asset index to assess trends in poverty in seven sub-Saharan African countries. World Development, 36(6):1113–1130. Bourguignon, F. (2002). The Poverty-Growth-Inequality Triangle. Washington, DC: World Bank. Carter, R.M., Little, D.P., and Mogues, T. (2007). Poverty traps and natural disasters in Ethiopia and Honduras. World Development, 35(5):835–856. Cazzavillan, G., Donadelli, M., and Persha, I. (2013). Economic growth and poverty traps in sub-Saharan Africa: The role of education and TFP shocks. Research in Economics, 67: 226–242. Chakravarty, S.R. (1983). Ethically flexible measures of poverty. Canadian Journal of Economics, 16:74–85. Clark, S., Hemming, R., and Ulph, D. (1981). On indices on the measurement of poverty. Economic Journal, 91:525–526. Collier, P. (2004). Conflict and Development. Oxford: University Center for the Study of African Economies. Deaton, A. (2005). Measuring poverty in a growing world (or measuring growth in a poor world), NBER Working Paper Series, WP 9822. Dercon, S. (1998). Wealth, risk and activity choice:  cattle in Western Tanzania. Journal of Development Economics, 55:1–42. Dercon, S. (2006). Economic reform, growth and the poor:  Evidence from rural Ethiopia. Journal of Development Economics, 81:1–24. Dercon, S., and Christiaensen, L. (2011). Consumption risk, technology adoption and poverty traps: evidence from Ethiopia. Journal of Development Economics, 96(2011):159–173. Dollar, D., and Kraay, A. (2002). Growth is good for the poor. Journal of Economic Growth, 7(3):195–225. Deverajan, S. (2013). Africa’s statistical tragedy. Review of Income and Wealth, 59: S9-S15. doi: 10.111/roiw.12013 Easterly, W. (2006). Reliving the 50s: the big push, poverty traps, and takeoffs in economic development. Journal of Economic Growth, Springer, vol. 11(4), pp. 289–318, December.

Poverty   195 Foster, J., Greer, J., and Thorbecke, E. (1984). A  class of decomposable poverty measures. Econometrica, 52(3):761–766. Foster, J., and Shorrocks, A.F. (1991). Subgroup consistent poverty indices. Econometrica, 59(3):687–709. Fosu, A. (2010). Does inequality constrain poverty reduction programs? Evidence from Africa. Journal of Policy Modeling, 32:818–827. Fosu, A. (2009). Impacts of Policy Syndromes and Governance on Productivity and Growth of African Economies. Seminar presentation at Center for the Study of African Economies, the University of Oxford. Ghiglino, C., and Sorger, G. (2002). Poverty traps, indeterminacy, and the wealth distribution. Journal of Economic Theory, 105:120–139. Giesbert, L., and Schindler, K. (2012). Assets, shocks, and poverty traps in rural Mozambique. World Development, 40(8):1594–1609. Goodhand, J. (2003). Enduring disorder and persistent poverty:  a review of the linkages between war and chronic poverty. World Development, 31(3):629–646. Hausmann, R., Prichett, L., and Rodrik, D. (2005). Growth Accelerations. Cambridge, MA: Harvard University. Kakwani, N. (1980). Income Inequality and Poverty:  Methods of Estimation and Policy Applications. Oxford: Oxford University Press. Kalwij, A., and Verschoor, A. (2007). Not by growth alone:  The role of the distribution of income in regional diversity in poverty reduction. European Economic Review, 51:805–829. Kraay, A., and Raddatz, C. (2007). Poverty traps, aid, and growth. Journal of Development Economics, 82:315–347. McKay, A. (2013). Growth and Poverty Reduction in Africa:  Evidence from AERC Growth-Poverty Project And Beyond. Nairobi: AERC. McKay, A., and Perge, E. (2011). How strong is the evidence for the existence of poverty traps? A  multi country assessment. Economics Department Working Paper Series:  25–2011, University of Sussex. McKinsie (2010). Lion’s on the move. www.mckinsey.com/insights/africa/lions_on_the_move. McMillan, M. (2013). Structural Change in Africa: Harnessing Natural Resources for Inclusive Growth. Tunisia: African Development Bank. Monga, C. (2013). The Mechanics of Job Creation: Seizing the New Dividends of Globalization. World Bank Working Paper Series, WP 6661. Moradi, A., and Baten, J. (2005). Inequality in sub-Saharan Africa: new data and new insights from anthropometric estimates. World Development, 33(8):1233–1265. Naschold, F. (2012). The poor stay poor: household asset poverty traps in rural semi-arid India. World Development, 40(10):2033–2043. Ncube, M., and Shimeles, A. (2013). The making of middle class in Africa: evidence from DHS data, IZA Discussion Papers, 7352, IZA. Nunn, N. (2008). The long-term effects of Africa’s slave trades. The Quarterly Journal of Economics, 123(1):139–176. Larke, C. and Milanovic, B. (2013). Global Income Distribution: From the fall of the Berlin Wall to the Great Recession, World Bank Working Paper Series, 6719. Li, J., and Monga, C. (2011). Growth identification and facilitation: the role of the state in the dynamics of structural change. Development Policy Review, 2011:264–290. Lopez, H., and Servén, L. (2009). Too Poor to Grow. Policy Research Working Paper 5012. Washington DC: The World Bank.

196   Concepts Lopeza, M.S., Molerob, J., and Santos-Arteaga, J.F. (2011). Poverty traps in a frictionless world: the effects of learning and technology assimilation. Structural Change and Economic Dynamics, 22:106–115. Page, J. (2013). Industrial Policy in Practice in Africa. Tunisia: African Development Bank. Page, J., and Shimeles, A. (2013). Aid, Employment and Poverty in Africa. Tunisia: African Development Bank. Paternostro, S. (1997). The poverty trap: the dual externality model and its policy implications. World Development, 25(12):207 I–208 I. Pinkovskiy, M., and Sala-i-Martin, X. (2014). Lights, Camera, Income: estimating poverty using national accounts, survey means and lights, NBER Working Paper Series, 19831. Pinkovskiy, M., and Sala-i-Martin, X. (2013). Africa Is on Time, Processed. New York: Colombia University. Radeny, M., Van Den Berg, M., and Schipper, B. (2012). Rural poverty dynamics in Kenya: structural declines and stochastic escapes. World Development, 40(8):1577–1593. Ravallion, M. (2012). Why don’t we see poverty convergence? American Economic Review, 102(1):504–523. Ravallion, M. (1992). A Guide to Concepts and Methods, Working Paper 88.World Bank Living Standard Measurement Series. Washington. Rodrik, D. (2013). Structural Change, Fundamentals, and Growth:  An Overview. Princeton, NJ: Princeton University. Sachs, J.D., and Warner, A. (1997). Sources of slow growth in African economies. Journal of African Economies, 6(3):335–376. Sachs, J.D., Mcarthur, J.W., Schmidt-Traub, G. et al. (2007). Ending Africa’s Poverty Trap. Brookings Papers on Economic Activity. Sahn, E.D., and Stifel, C.D. (2000). Poverty comparisons over time and across countries in Africa. World Development, 28(12):2123–2155. Santos, P., and Barrett, C. (2011). Persistent poverty and informal credit. Journal of Development Economics, 96:337–347. Sen, A.K. (1976). Poverty: an ordinal approach to measurement. Econometrica, 44:219–231. Sen, A.K. (1985). Commodities and Capabilities. Oxford: Elsevier Science Publishers. Shimeles, A., and Thorbecke, E. (forthcoming). Does initial poverty bad for growth in Africa? A Revisit. Tunisia: African Development Bank. Stiglitz, J., Li, J.Y., and Monga, C. (2013). The Rejuvenation of Industrial Policy. World Bank Working Paper Series, 6628, World Bank, Washington DC. Takayama, N. (1979). Poverty, income inequality and their measures: Professor Sen’s axiomatic approach reconsidered. Econometrica, 47:747–759. The Economist (2013). Emerging Africa: the hopeful continent. March. The Economist (2011). The hopeful continent: Africa rising. December. The Economist (2001). Hopeless Africa. March. Thon, D. (1979). On measuring poverty. Review of Income and Wealth, 35(4):429–440. Thon, D. (1981). Income inequality and poverty: some problems. Review of Income and Wealth, 25:429–440. Vaughan, R.N. (1987). Welfare approaches to the measurement of poverty. Economic Journal, 97(suppl):160–169. UNECA (1999). African Economic Report. Addis Ababa, Ethiopia: UNECA. Young, A. (2012). The African Growth Miracle. NBER Working Paper No. w18490.

Chapter 11

Dimensions of African Inequa l i t y Arne Bigsten

The chapter discusses dimensions of inequality in sub-Saharan Africa (hereafter referred to as Africa) and their causes. We first review empirical evidence about inequality during the colonial period as well as the post-independence era. Then we focus on the forces that determine inequality change, focusing on factor accumulation and structural change. Next we discuss the relationship between inequality and growth, the role of agriculture in the development process, the relationship between ethnicity and social stratification and governance, and external influences on inequality. Finally, we consider what policy interventions can do to reduce inequality.

11.1  The Historical Legacy There is not much quantitative evidence on the extent of inequality in pre-colonial times in Africa. The meager evidence there is on economic growth suggests that per capita incomes were virtually unchanged between years 1000 and 1820 (Maddison 2003). During this period there was external economic involvement in slave trade and raw material extraction, but by 1820 there were only about 50,000 people of European descent in Africa (half of them in South Africa). This share increased relatively fast once colonial powers took control, and by 1913 the number had increased to 2.5 million (Maddison 2005). Although it is hard to discern any trend in per capita incomes until the nineteenth century, there were early growth accelerations—typically related to the expansion of raw material exports—in some countries followed by busts (Jerven 2010). The international slave trade, which lasted until 1865, had profound consequences for particularly West Africa. Its end meant that there was scope for the expansion of smallholder cash-crop production with positive income effects. Colonialism led to increased diversification of these economies and increased inequality. In pre-colonial African societies there had been kingdoms with established elites, but otherwise few people had per capita incomes beyond subsistence levels. This suggests that

198   Concepts inequality was fairly low at this time. The inequality increases which followed were related to the arrival of European colonizers. This set off a process of differentiation along the lines outlined in Lewis (1954) dual-economy model. It meant the introduction of modern enclaves in mostly traditional agricultural economies. Most African countries were colonized by European powers in the late nineteenth century. At that time most parts of sub-Saharan Africa had limited contacts with the outside world, while North Africa and South Africa had closer ties. There was trade along the coast, also some long-distance trade, but the most of the inland was little integrated with the international economy. The British colonization of Kenya can serve as an example of the impact of colonialism on economic inequality. In Kenya the inland households were pastoralists, settled farmers, small craftsmen, or traders, but there was limited specialization. At the time of the arrival of the British to East Africa, households had generally access to enough land to ensure a standard of living roughly comparable to that of the other members of the community (Bigsten 1986). Differences in incomes or welfare levels were therefore relatively modest. With the building of the railway to Uganda, the Kenyan inland was opened up to trade and white settlement. The railway construction also brought in Indian workers, and some of those stayed and set up small stores or sought employment in industry or government. A three-layered society emerged, with the white settlers in control. Some Africans became engaged in settler agriculture either as squatters or as contract labor, while others became traders or businessmen (Kitching 1980). By restricting the scope for development on African farms and by hut and poll taxes, the British sought to maintain a cheap supply of labor for settler agriculture. Still, there was a gradual expansion of cash crop production on African farms, so inequality among African smallholders increased. Urban real wages increased significantly after World War II, partly due to increased minimum wages. Rural–urban differences in living standards grew, and rural–urban migration accelerated. Independence in Kenya in 1963 implied a dramatic change in the inter-racial distribution of both political power and incomes. The average income in the post-independence period was still the highest for Europeans and the lowest for Africans, with the Asians in between, but the overlap increased a lot. The evolution of inequality of Kenya during the colonial era is typical of colonized sub-Saharan Africa, although the share of white settlers was higher in Kenya than in most other countries excluding South Africa, Rhodesia, and the Portuguese colonies. The contribution of inter-racial inequality to overall inequality was higher in those countries than most colonies. In Kenya overall inequality increased rapidly until 1950, then fluctuated. The income gaps among the racial groups in the case of Kenya at this time were huge. The Asians had 27 times the African average income and the Europeans 95 times (Bigsten 1986). The pattern of change of the income Gini coefficient for Kenya 1914–1976 shows that when the modern–traditional income gap increased, the Gini coefficient went up and vice versa. Structural change was then as now a key determinant of inequality change. The evolution of inequality during the colonial rule in Africa was similar in other African countries, where elites of European descent took control and earned very high incomes by African standards. Inequality was highest in the countries where the white minority was strong and tried to maintain white control. In 1980 Rhodesia became Zimbabwe and white control came to an end and in 1994 South Africa followed, but the racial income gaps remained large.

Dimensions of African Inequality    199 The evidence for Africa thus suggests that there was a rapid increase in inequality when colonization started at the end of the nineteenth century. The inequality increase was eventually halted, but inequality has remained very high in Africa.

11.2  Post-colonial Inequality The first decade after independence in the early 1960s African countries grew quite fast, but most of them stagnated from the mid-1970s to the mid-1990s. Then growth recovered, but it was not until 2006 that sub-Saharan Africa as a whole achieved per capita incomes higher than the peak in the 1970s. Since then growth has continued with a temporary setback during the financial crisis 2008–2009. A challenge that one encounters when trying to measure growth or income distribution in African economies is that the database is weak. Gross domestic product (GDP) estimates vary a lot between series with different base years (Jerven 2013), and it is difficult to measure incomes by households or individuals in a systematic fashion. Much of the inequality analysis therefore is based on household consumption data. We report here what the available evidence has to say about recent changes in inequality. Arndt (2012) summarizes changes in poverty in 22 African countries from 1996 to the most recent estimates, using national poverty rates and data from World Development Indicators supplemented by data from McKay (2013), who report results for ten AERC case studies. In as many as 19 out of the 22 cases there was a reduction in the poverty rate. Arndt also looks at the evolution of poverty using the World Bank’s one dollar a day poverty line. For sub-Saharan Africa he notes that poverty rates increased until the mid-1990s (alongside falling per capita incomes), but in 1996 the poverty rate started to decline. When growth is positive poverty tends to fall and vice versa. Young (2012) constructed an alternative measure of real consumption based upon data from Demographic and Health Surveys. It includes the ownership of durable goods, the quality of housing, the health and mortality of children, the education of youth and the allocation of female time in the household. These estimates suggest that living standards in sub-Saharan Africa have been growing at about 3.4 to 3.7 percent per year during the two last decades! He paints a more optimistic picture of African development than what consumption estimates suggest. Less effort has been devoted to the analysis of the evolution of inequality than to the change in poverty. Available estimates of inequality are typically from the household surveys that are used to measure poverty. It is a concern that the poorest and the richest households are poorly covered in the surveys, at the same time as these groups are particularly important for inequality measurement. Ravallion and Chen (2012) have anyway computed estimates of inequality on the basis of 850 household consumption surveys for 1979–2008 covering 125 developing countries. They show that sub-Saharan Africa has higher inequality than other regions with the exception of Latin America and the Caribbean. They also find that there is no clear trend over time in SSA inequality. It showed some increase for 1999–2005 followed by a decline for 2005–2008. Overall the level of inequality has not changed much in sub-Saharan Africa since the first of those estimates for 1981. Arndt (2012:  14)  found no significant correlation between growth and inequality change. At

200   Concepts present we seem to have a distributional-neutral growth pattern in sub-Saharan Africa. However, this is based on data on consumption, which typically miss out on the highest and the lowest incomes, so we cannot be sure that the conclusion holds also for income inequality. Although most analysts have looked at income inequality and poverty on the basis of household budget surveys, there are also some alternative estimates. Pinkovskiy and Sala-i-Martin (2010) computed inequality by disaggregating total GDP with the help of distribution functions. Their estimates suggest that inequality for the whole of sub-Saharan Africa increased from around 1970, peaked in the 1980s, and was back at the 1970s level in 2006. Within country inequality was more stable, but showed a modest decline from 1990 to 2006. One can consider the extent of intergenerational mobility as an indicator of inequality of opportunity. Cogneau et al. (2006) find that two countries with relatively low cross-sectional income inequalities, Ghana and Uganda, also display relatively high intergenerational mobility and low inequality of opportunity. Sahn and Stifel (2003) showed that urban–rural gaps in living standards are high and show now tendency to decline in a the African countries included in their study. The picture based on income and consumption estimates can be complemented by evidence on other important dimensions of human welfare. The UN’s Human Development Report provides statistics about changes in the inequality of attributes like health and education. For both these indictors sub-Saharan Africa shows consistently declining inequality, like other regions of the world. Many countries in Africa also show improvements in the Gender Inequality Index 2000–2012. So a broader measure of multidimensional inequality would indicate that inequality in SSA was declining between 1990 and 2010, in spite of income inequality remaining essentially unchanged (UN 2013:  32). The indicators here relates to fairly basic needs. The poorer segments of society thus seem to have improved their access to these basic services relatively to the richer segments.

11.3  Factors of Production, Structural Change, and Inequality The distribution of income and its change is strongly related to the distribution and evolution of asset ownership. Factors to consider are capital and land including natural resources (very important in Africa) as well as labor and human capital. The impact of education and human capital on inequality is strongly related to wage and employment determination. Kuznets’ (1955) classical paper suggested that inequality first increases and then decreases as per capita incomes go up. He suggested that early inequality increases are more or less inevitable and that it might be detrimental to growth to try to counteract this. Inequality increases when labor is transferred from a large traditional or agricultural sector to a more productive modern industrial sector. This effect was important in the early stages of modernization in sub-Saharan Africa, but the effect on the Gini coefficient declined when the modern income share increased. Still, structural change remains a major determinant of the evolution of income distribution.

Dimensions of African Inequality    201 African countries are generally poorly integrated, which means that inequalities between regions as well as urban and rural areas are pronounced. The classical Harris–Todaro (1968) model of rural–urban migration was inspired by the experience of the migration to Nairobi from the rural hinterland in the face of moderate expansion of formal urban jobs. The explanation they provided in their model was that formal sector wages had been pushed up by minimum-wage legislation, and that people moved to Nairobi to have a chance to get these jobs. In the model those who fail to get a job are assumed to be unemployment without income, but in reality most of them are absorbed in some fashion into the informal sector at modest wages. To explain changes in inequality we need to look at changes in factor proportions and structural change. Africa is a vast continent with unused land in some areas, but in many parts of the continent there is a rapidly increasing pressure on land due to the rapid growth of the labor force. Typically, models of economic development assume that there is a process of capital deepening, but this process has been slow in Africa. This has had implications for the pattern of structural change. Since capital-to-labor ratios have been stagnant for decades, people that have been pushed out of agriculture have generally not been absorbed by the modern sector. They have instead been absorbed by the informal sector, where incomes are often not higher than in smallholder agriculture. This has meant that the shrinking of the agricultural share has not led to a decline in overall inequality. Capital-poor countries tend to start from similar patterns of specialization and low wages. Leamer (1987) has developed a useful model of development that helps us understand how the pattern of specialization and then employment changes over time. His model includes three factors of production, namely labor, capital, and land, and the focus of the analysis is on the impact of capital accumulation on employment structure and wages and other factor rewards. Economies move between patterns of specialization as factor abundance changes. The typical African economy may broadly be divided into four sectors. The non-agricultural part of the economy, where we assume that only capital and labor are used, typically consists of a less capital-intensive or informal sector and a more capital-intensive or formal sector. Then we may distinguish between two agricultural sectors, a smallholder sector using only land and labor, and a modern agricultural sector using all factors, that is, capital, labor, and land. The relative size of the sectors will be determined by the relative availability of factors. When factor endowments change, the pattern of specialization changes as does the factor price structure. Inequality levels will depend on the relative size of different activities and the relative rewards in those. We can discuss the process of structural change in four steps. First, we may ask what changes in factor endowments do to the production structure. When capital is accumulated at a rate sufficient to increase the capital/labor ratio, capital-intensive goods will increase their share of production. But if labor grows faster than capital there will instead be a shift towards a more labor-intensive product mix. The outcome will be moderated depending on what happens to the supply of land, but in many parts of Africa it is hard to increase the area under cultivation. An increase in the capital/labor ratio would therefore tend to lead to a more capital-intensive mix of production and higher wages. When land cannot grow there will also be an increase in land rentals. What happens to the rate of capital formation is therefore crucial for what happens to structural change and inequality in African countries.

202   Concepts A second key aspect concerns how the relationship with the world market via goods prices affect factor rewards and the economic structure. Prices can change autonomously or the government can intervene to change the domestic price structure. For example, after independence African countries typically followed an import-substitution strategy with high tariff protection of the industrial sector. This meant that the wage gap between urban workers and rural labor and urban–rural inequality were pushed up. A third force that changes the economic structure is technical progress. The effects depend on the character of the technical progress and in which sector it occurs. If technological change is labor-augmenting, it can help increase wages and to some extent compensate for a slow rate of capital accumulation. Fourthly, there may also be factor market distortions that affect factor rewards. The factor prices will also depend on how the domestic factor markets are integrated with international factor markets. There are barriers to mobility between the informal and the formal labor market, which implies that labor with similar skills will be paid differently in the two markets (Bigsten et al. 2013). Within the formal market there exist minimum wage legislation and trade union contracts that may generate wages that are different from equilibrium wages. In the discussion of structural change labor was treated as a homogeneous category, but there are of course considerable variation of skills and incomes within the group. Particularly within the urban economy earnings vary considerably. One important determinant of the evolution of inequality is the expansion of education, which has an impact on overall inequality by affecting the share of incomes that accrues to labor and the dispersion of wages. The latter effect depends on what happens to the structure of wages and the number of employees in the various labor categories. Typically there would be a narrowing of the wage gaps across categories of labor when the relative supply of higher education increases, which has happened in Africa. Knight and Sabot (1990) investigated how the expansion of secondary education in Kenya and Tanzania affected the inequality of pay in those countries and find that there is a compression effect. Educational policy can thus be a tool for inequality reduction, but it is not self-evident that this will be done. Education is typically skewed in favor of the rich, and it could possibly function as a tool of exclusion (Gradstein 2003). What we have discussed in this section is how gross incomes of individuals are determined. But if we want to explain inequality in consumption or net income we need to take various redistribution mechanisms into account. This includes taxation and redistribution by the government, but in Africa there is also redistribution going on within the extended family system. The pension system does not redistribute significant amounts of money in most of sub-Saharan Africa, although it is an important equalizer in South Africa.

11.4  Inequality, Poverty Traps, and Growth The traditional view of the effect of inequality on growth is that higher inequality makes higher savings possible and provides incentives and therefore is growth enhancing. However, there is more recent literature, which suggests that inequalities affect growth negatively (Persson and Tabellini 1994; Alesina and Rodrik 1994). The explanation most discussed in this context relates to the credit markets. There are credit market failures, which affect the

Dimensions of African Inequality    203 access of the poor to credit negatively, and this means that large parts of the population are unable to realize their potential (Galor and Zeira 1993). Inequality may also lead to instability and lower investments, rent-seeking, higher transaction costs and more insecure property rights with negative effects on growth. In Africa, rent-seeking is particularly problematic in resource-rich countries, which at least until the last decade had mediocre growth outcomes. There is also extensive rent-seeking and corruption relating to public procurement, which fuels inequality at the same time as it hampers growth. Social protests may create uncertainty about the enforceability of contracts, increase transaction costs for businesses, and force a diversion of public expenditure to security areas from growth enhancing investments. High levels of inequality may also lead to high crime rates and possibly be detrimental to health. Inequality may also reduce possibilities for broad-based participation in the political process, which in turn may lead to political and social instability and reduce the government’s ability to pursue efficient policies. So lower inequality is not only a social target, but it may also have an instrumental value for growth. The effectiveness of growth in reducing poverty is strongly linked to inequality, since the elasticity of poverty reduction with regard to growth falls with the degree of inequality (Ravallion 2001; Bigsten et  al. 2003; Bigsten and Shimeles 2007; McKay and Perge 2009; Fosu 2009). Duclos and O’Connell (2009) argued that a development trap exists when low incomes hold back growth for an extended period of time. Low levels of human capital could explain poor returns to investments in situations with low incomes. Poverty-alleviation policies may then also have a functional justification in so far as they improve overall economic performance.

11.5  Transformation of Agriculture and Inequality Structural change is an integral part of economic development. Typically the agricultural sector’s share in output and employment shrinks as incomes increase, while the shares of industry and services expand. Gollin (2012) contrasted two extreme views on the role of agriculture in this transformation. One says that agriculture is a source of labor only, while the other says that it must generate growth to be able to release labor. Most SSA countries have two-thirds or more of its labor force in agriculture. The gap between agriculture and non-agriculture in labor productivity is according to Gollin, 7.8:1. The gap in productivity between agriculture and the rest of the economy is larger in sub-Saharan Africa than in other regions of the world with strong implications for inequality. The reasons for this huge gap include low skills, poor management, and poor technology in agriculture. Africa has over recent decades seen very slow growth in agricultural productivity. A key question is then why so many are stuck in subsistence agriculture? Schultz (1953) argued that an agricultural surplus was necessary to start the transformation process indicating that one should seek policies that boost agricultural productivity. However, it has been hard to sort out causality between agricultural growth and overall growth. Non-agriculture has to grow faster than agriculture for the structural transformation to come about. The role of agriculture is less crucial once the economy is opened up (Dercon

204   Concepts 2009), since the crucial demand linkage to agriculture is removed. Labor-intensive growth in other sectors can reduce poverty effectively if it absorbs labor. So the gradual integration of Africa into the world economy would make growth less dependent on the development of agriculture. Still, agriculture supports a large share of the population in Africa and it is particularly important in the land-locked resource poor economies. So far we have discussed the change in economic structure from a macro-perspective, but structural change also takes place at the household level. Smallholders in Africa were originally almost exclusively farmers, but over time they have shifted into production for the market and to non-agricultural activities as well. Income diversification is a result of households’ allocation of its assets across income-generating activities. Households seek to achieve an optimal balance between expected returns and risks in different activities given the constraints they face (Barrett et al. 2006). Income structures vary between households according to endowments and constraints. Since African markets often are poorly integrated, different households have access to different sets of income opportunities. There is also a large variation in transaction costs and market prices, and households differ in terms of property rights, labor availability, and access to credit. Barrett et al. (2006) analyzed how income sources and diversification vary in Kenya, Cote d’Ivoire, and Rwanda. They note that households that do not possess sufficient human and financial resources do not have access to potentially lucrative activities, and are forced to choose low-return activities. The endowments are of course a key determinant of smallholders’ activity choices, in a similar fashion as the national endowment structure determines the sector structure of the whole economy. The labor/land ratio of the household is a key determinant of movement into off-farm activities. The human capital of the household is also a key factor determining activity choices. Then it is clearly easier to diversify out of agriculture if the household has good access to a thriving off-farm sector, which normally would be the case closer to urban areas. Thus, the main factors behind allocation choices and the resulting inequality are differences in endowments, differences in access to markets, and access to finance. Reardon (1997) concludes from a review of the income diversification literature that non-farm income is generally regressively distributed. This means that households with the highest farm income also have the highest incomes from non-farm activities and that diversification generally is a way up the income scale. Bigsten and Tengstam (2011) show that smallholder diversification is associated with higher incomes in the case of Zambia, and that the scope for diversification depends on endowments and access to markets and finance. However, there are also instances where you see distress diversification, that is households having to take bad jobs just to survive (Barrett 1998).

11.6  Inequality and Inclusive Governance In recent years there has been an intense debate on the relationship between governance and institutions and economic development. The basic hypothesis in Acemoglu’s and Robinson’s (2012) theory of development is that economic and political institutions shape the incentives of business, individuals, and politicians. Economic institutions provide incentives to become educated, to save and invest, to innovate and adopt new technologies. But it is the political processes that determine what economic institutions people live under and how the processes

Dimensions of African Inequality    205 work. Furthermore, it is the distribution of power in society which shapes both institutions and the outcomes of the political process. So while economic institutions determine whether countries are poor or rich, it is politics and political institutions that determine what economic institutions a country has. They also note that there is high institutional persistence because of the way that political and economic institutions interact. Extractive institutions tend to persist because it is in the interest to those in power. Isaksson (2011) argued that social divisions have a negative effect on perceived institutional inclusiveness, which in turn should depress institutional payoffs. Empirical estimations confirm a weaker association between property rights and economic performance in societies marked by social divisions. Since the political institutions are a strong influence on the economic institutions that generate development, their development is clearly crucial. The desired institutional set-up is a system of governance that distributes power broadly in society and subjects it to constraints. This means that political power should rest with a broad coalition or a plurality of groups. Outcomes depend on which group wins in the political process, which in turn depends on the distribution of political power. And this distribution of power is strongly related to inequality. There is a strong presumption that inequality is a crucial determinant of the inclusiveness of governance and political institutions, and inequality is of course affected by governance processes. The question is how one can get into a virtuous circle of improved governance and reduced inequality. Many countries in Africa have been in a vicious circle, which Rothstein (2011) refers to as a low trust—corruption—inequality tap. Lower inequality would increase the prospect of broader coalitions getting together in collective actions to build inclusive governance. But we should note that there have been considerable improvements in governance in Africa, which probably have helped reduce inequality in some broader indicators of development such as health, education, information, security and political influence. And these factors in turn support the growth acceleration we have observed. Johnson, Ostry, and Subramanian (2007) discussed whether Africa can achieve sustained growth with the current institutions. The find that there are many shortcomings, but they also find that Africa is not much worse than East Asia was in the 1960s. Therefore it should be possible also for Africa to achieve economic improvements and to reduce the institutional restrictions. So what constrains the implementation of “best practice” policies and institutions, which can reduce transaction costs and improve access to the world market? One concern is the vested interests of policymakers, which opens up for rent-seeking behavior. There is also a lack of skills, which hampers reforms even when the will is there. There is a challenge of formulating policy, but there is an even greater challenge in Africa relating to policy implementation. For policy reforms to be credible and sustainable they should be grounded in a democratic process. The causal link between democracy and growth is somewhat unclear, though.

11.7  Ethnic Inequality and Social Stratification It is easier to undertake reforms in a socially cohesive society, where citizens feel they are part of the same community, face the same challenges, and reap similar societal benefits (Easterly et al. 2006). In Africa the picture is often very different from this.

206   Concepts Africa stands out as a region where ethnic divisions play the most important role in how the governance of society works. Kimenyi (2006) explained how the most common form of corruption entails the distribution of rewards, jobs, contracts, and promotions on the basis of ethnicity. This then leads to ethnic and/or regional inequality that affects the level of overall inequality. There is also significant gender inequality in Africa, but since income generally is shared within families it is hard to get good measures of gender differences in consumption standards. There is evidence on other aspects of inequality relating to work and influence though. Policies are often heavily influenced by ethnic loyalties, and this may help explaining the under-provision of public goods and the prevalence of patronage goods (Wantchekon 2003; Kimenyi 2006; Habyarimana et al. 2007; Vicente 2008; Baldwin and Huber 2010). And it also explains why we often see ethnically and regionally biased allocation of public goods. The extent of ethnic diversity in access varies across types of public goods. Jackson (2013) finds that the supply of community type of goods like electricity and water a more equally distributed than education, where the locally dominating group has better access. Jackson relates this to effects on the demand side, where the dominant group has a higher demand for education because they have better labor market opportunities once educated. Alesina, Michalopoulos, and Papaioannou (2012) showed that ethnic inequality is inversely related to per capita income, and that differences in geographic endowments across ethnic homelands explain much of ethnic inequality. These imbalances thus have a long history. They also show that individuals from the same ethnic group are worse off when they reside in districts with a high degree of ethnic inequality. Kyriacou (2013) found in a cross-section of countries that governance is worse where ethnic group inequality is large. Alesina and Zhuravskaya (2011) showed that ethnically and linguistically segregated countries, i.e. countries where these groups live more separated, have lower quality of government. South Africa is an extreme case in Africa because of its apartheid history. Post-apartheid growth in South Africa has been sluggish, but there has at least been a modest reduction in poverty (Leibbrandt, Finn, and Woolard 2012). The main driving force of inequality change in South Africa post apartheid was that the share going the top decile increased. Social grants became much more important as sources of income in the lower deciles, but overall it is the labor market which is the main driver of aggregate inequality. Inequality within each racial group has increased, while the contribution of between-race inequality has decreased. Growth in Africa tends to be concentrated in small geographical areas, so spatial inequality is large (McKay and Perge 2009). In sub-Saharan Africa this is often related to the location of natural resources. There are clearly advantages of agglomeration such as economies of scale, lower transport and transaction costs, and forward and backward linkages matter. Spatial imbalances are particularly serious in Africa because of its high ethno-linguistic fractionalization. High spatial inequality can be bad for growth by creating conflicts and tensions and lead to demands for redistributive measures. Successful countries are characterized by greater density, shorter distances, and fewer divisions (World Bank 2009). The World Development Report 2009 concludes that urbanization and concentration of production is unambiguously good for growth and thus poverty reduction in the long run. The African story does not seem to be one of consistently increasing spatial inequality.

Dimensions of African Inequality    207

11.8  External Influences on Inequality Africa has for a long time had high and increasing trade intensity, but still it supplied a smaller and smaller share of global exports (until the last decade). Sub-Saharan Africa has been specialized in resource exports and it has been unable to diversify into sectors with larger spillovers and dynamic externalities. Nissanke (2009) refers to this as a commodity-dependence trap. African countries have been suffering from the consequences of swings in commodity prices, and the international community has not provided any efficient contingency financing. To be able to follow the diversification pattern of Asia, sub-Saharan Africa needs to have a level of education that is so high that it can benefit from the dynamic forces of globalization, mainly within intra-firm trade in parts. Sub-Saharan Africa needs a strategy for upgrading its comparative advantages and climb to higher value-added activities. This requires a capable nation state. This process would be easier if trade partners such as the EU had more generous rules relating to African exports. The concern today for poor African countries is not primarily the level of tariffs on final goods, but instead other barriers like rules of origin which make it hard to become involved in global value chains. One external influence that has been and still is important in sub-Saharan Africa is foreign assistance. The effect of aid on inequality depends on how it is allocated across regions and groups. Donors are concerned about the distributional implications of aid, and in recent years the focus has been on poverty reduction. The question is how the aid relationship should be designed to lead to an equitable allocation of the aid. Donors have attempted to use policy conditionality to achieve desired outcomes, but the effectiveness of this mechanism has been weak. Because of this donors have sought other ways to ensure that aid is effectively used, summarized in the Paris Agenda. One key dimension there is the emphasis on ownership, that is, higher recipient control of aid use. The question then is what measures other than policy conditions that can help keep biased allocations in check. It would to a higher degree have to depend on domestic checks and balances. Many countries in sub-Saharan Africa have political systems that deliver poor governance, but donors wanting to reduce poverty there still need to channel money into these countries. When the government lacks capacity to handle aid effectively it may be advisable to seek channels outside the government. Donors could for example set up their own non-governmental investment institutions that support private investment directly, rather than trying to do so indirectly via support for the government infrastructure. If this is successful it would also strengthening civil society, which could possibly put pressure on the political and institutional system to enhance governance.

11.9  Policy for Equity The development strategy of a country affects poverty via its impacts on growth and income distribution (Bourguignon 2004; Thorbecke 2013). The development policies pursued in Africa since independence have followed the trends in the international development debate closely. In the immediate post-independence period the main strategy was one of import-substitution industrialization. When concerns emerged about the

208   Concepts distributional consequences of this trickle-downs strategy, the focus shifted to policies such as Redistribution with growth and Basic needs. In the late 1970s many countries faced large economic imbalances, which meant that the strategy focus shifted to macro-stabilization and structural adjustment. This was pursued in the 1980s and 1990s, with moderate short-term success. Around the turn of the century there was again a shift towards a poverty focus and parallel to this a shift in focus from macro-stabilization policies to one focusing on governance issues. These shifts in policy have been closely related to perceptions about how inequality and poverty have evolved. To reduce inequality and poverty Africa needs a strategy that can help absorb the surplus of unskilled labor, and this could be an export strategy based on labor-intensive manufacturing. But also agricultural and rural development, with encouragement of new technologies, must play a role. Investment in physical infrastructure and human capital are crucial. To back this up one needs efficient institutions that provide the right set of incentives to farmers and entrepreneurs. Social policies to promote health, education, and social capital need to be discussed, as well as the scope for safety nets to protect the poor. One needs to address the issue of governance, which is crucial for the development of African economies. To be able to formulate a policy of redistribution (or poverty reduction) it is important to understand the causes of inequality. It can relate to human and physical capital, land or public goods. The geographical and sectorial pattern of growth also matters. The most effective policy of redistribution would relate to assets rather than incomes, but asset redistributions are hard to undertake except under exceptional circumstances—often related to political violence. It is easier to redistribute incomes with the help of taxes and transfers, but these may have detrimental effects on growth incentives. By reducing returns on human and physical capital income taxation reduces incentives to save and invest. If one assumes that it is primarily the rich who have the possibility to save, redistribution away from them in favor of the poor would reduce savings. Poverty can limit growth in the long run via negative effects on productivity. In such a situation income transfers to the poor groups could be more positive for growth to the extent that they make it possible for them to invest in human and physical capital. Transfers also have an insurance dimension and protect households against negative shocks and thereby make it possible for them to avoid negative savings or having to take their children out of school. Large income gaps also increase the risk of macroeconomic instability and makes it harder for governments to undertake reforms, which require collaboration and trust among people or groups (Alesina and Perotti 1996). It seems clear that a reasonably even distribution is good for growth. The effect will, however, depend on how one gets to a relatively even distribution given that you start form a very uneven one. Most types of redistribution policy are controversial, and to be able to undertake it there must be support from influential groups. It could be argued that it may be in the interest of the elite to see a strong middle-class emerge, which might mean that they for example may be willing to support a broad push for education. This would be good for the growth in the long run, which would be to its benefit, but it could also undermine the power of the elite. Still, the growth of a middle class would tend to reduce social tensions and reduce the risk of future confiscation of its assets. To be able to reduce poverty countries need long-term growth. The growth acceleration we have seen in Africa in recent years is primarily due to two factors. First, there have

Dimensions of African Inequality    209 been considerable improvements in the policy environment with regard to macroeconomic policy and extensive structural reforms reducing market distortions have been undertaken. Secondly, the natural resource boom has increased incomes in resource-rich countries. The number of armed conflicts has been reduced, there were democratic advances, increased political stability, and many countries have received considerable debt write-offs. All these factors have contributed to faster growth. To reduce poverty and to achieve a more equal income distribution one must build up the resources of the poor such as human capital, but it also requires a growth process that generates demand for the resources the poor. To get a process that generates jobs is vital for Africa. The developments over the last couple of decades have mainly generated low-paid jobs in the informal sector, and this is not a transformation process that reduces inequality significantly. A policy for formalization should seek to make the formal sector attractive enough by making rules and regulations simple and transparent (Aryeetey 2009). There is a risk of policy errors if the policy process focuses too much on policies that have short-term poverty-reducing effects. The optimal development path from a poverty reduction perspective would probably best be defined as one that minimizes the discounted sum of future poverty. A policy package that achieves this would be different from one that minimizes poverty in the short term. There are many policies that increase consumption today at the expense of consumption tomorrow. At the same time there are policies aimed at financing investments in infrastructure (e.g. taxation) that generate growth and poverty reduction in the longer term, while they may have negligible or even negative effects on the consumption of the poor today. Redistribution from the future to the present and from the currently non-poor to the poor can reduce poverty in the short term, but the question that needs to be addressed is how it affects future poverty. The future of inequality in Africa hinges on what happens to structural transformation. To lower inequality and poverty one needs growth that generates labor demand outside traditional agriculture and the natural resource sector. In Asia the successful poverty reduction was achieved by having a rapid increase in the demand for unskilled labor in the manufacturing sector. This change was moreover often preceded by a green revolution in agriculture, which increased productivity and incomes in that sector. This both created demand for manufactured products and released resources for the expanding sector. We have not seen such a breakthrough in African agriculture yet.

Acknowledgment I am grateful for comments from Ann-Sofie Isaksson, Sven Tengstam, Hippolyte Fofack, and participants in the book contributors’ workshop in Beijing, December 2013.

References Acemoglu, D., and Robinson, J.A. (2012). Why Nations Fail: The Origins of Power, Prosperity and Poverty. London: Profile Books Ltd.

210   Concepts Alesina, A., Michalopoulos, S., and Papaioannou, E. (2012). Ethnic Inequality, NBER Working Paper no 18512. Alesina, A., and Perotti, R. (1996). Income distribution, political instability, and investment. European Economic Review, 40(6):1203–1228. Alesina, Q., and Rodrik, D. (1994). “Distributive Politics and Economic Growth”. Quarterly Journal of Economics, 109:465-490. Alesina, A., and Zhuravskaya, E. (2011). Segregation and the quality of government in a cross section of countries. American Economic Review, 101:1872–1911. Arndt, C. (2012). Chapter 2: Green Growth and Africa in the 21st Century, for the African Development Report 2012, mimeo. Tunis: African Development Bank. Aryeetey, E. (2009). The Informal Economy, Economic Growth and Poverty in Sub-Saharan Africa, mimeo. Nairobi: AERC. Baldwin, K., and Huber, J.D. (2010). Economic versus cultural differences:  forms of ethnic diversity and public goods provision. American Political Science Review, 104(4):644–662. Barrett, C.B. (1998). Immiserized growth in liberalized agriculture. World Development, 26(5):743–753. Barrett, D.B. et al. (2006). Welfare dynamics in rural Kenya and Madagascar. Journal of Development Studies, 42(2):248–277. Bigsten, A. (1986). Welfare and economic growth in Kenya 1914-1976. World Development, 14(9):1151-1160. Bigsten, A., Kebede, B., Shimeles, A., and Taddesse, M. (2003). Growth and poverty reduction in Ethiopia:  evidence from Household Panel Surveys. World Development, 31(1):87–106. Bigsten, A., Mengistae, T., and Shimeles, A. (2013). Mobility and Earnings in Ethiopia’s Urban Labor Markets, 1994-2004. Economic Development and Cultural Change, 61(4):889–931. Bigsten, A., and Shimeles, A. (2007). Can Africa reduce poverty by half by 2015? Development Policy Review, 25(2):147–166. Bigsten, A., and Tengstam, S. (2011). Smallholder diversification and income growth in Zambia. Journal of African Economies, 20(5):781–822. Bourguignon, F. (2004) The Poverty Growth Inequality Triangle. Working paper 125. Indian Council for Research on International Economic Relations. Cogneau, D., Bossuroy, T., de Vreyer, P. et al. (2006). Inequalities and equity in Africa. Document de travail DT/2006-11, DIAL, Paris. Dercon, S. (2009). Agriculture, growth and rural poverty reduction in Africa: fallacies, contexts and priorities for research, mimeo. Nairobi: AERC. Duclos, J.-Y., and O’Connell, S. (2009). Is poverty a binding constraint in sub-Saharan Africa? mimeo. Nairobi: AERC. Easterly, W., Ritzen, J., and Woolcock, M. (2006).Social cohesion, institutions, and growth. Economics and Politics, 18(2):103–120. Fosu, A. K. (2009). Inequality and the impact of growth on poverty: comparative evidence for sub-Saharan Africa. Journal of Development Studies, 45(5):726–745. Galor, O., and Zeira, J. (1993). Income distribution and macroeconomics. Review of Economic Studies, 60:35–52. Gollin, D. (2009). Agriculture as an Engine of Growth and Poverty Reduction: What We Know and What We Need to Know, mimeo. Nairobi: AERC. Gradstein, M. (2003). The Political Economy of Public Spending on Education, Inequality, and Growth, World Bank Policy Research Working Paper 3162, Washington DC.

Dimensions of African Inequality    211 Habyarimana, J., Humphreys, M., Posner, D.N., and Weinstein, J.M. (2007). Why does ethnic diversity undermine public goods provision? American Political Science Review, 101(4):709–725. Harris, J.R., and Todaro, M.P. (1970). Migration, unemployment and development: a two-sector analysis. American Economic Review, 60(1):126–142. Isaksson, A. (2011). Social divisions and institutions: assessing institutional parameter variation. Public Choice, 147(3):331–357. Jackson, K. (2013). Diversity and the distribution of public goods in sub-Saharan Africa. Journal of African Economies, 22(3):427–462. Jerven, M. (2010). African growth recurring:  an economic history perspective on African growth episodes, 1690–2010. Economic History of Developing Regions, 25(2):127–154. Jerven, M. (2013). Poor Numbers: How We Are Misled by African Development Statistics and What to Do about It. Ithaca, NY: Cornell University. Johnson, S., Ostry, J.D., and Subramanian, A. (2007). The Prospects for Sustained Growth in Africa: Benchmarking the Constraints, NBER Working Paper 13120, Cambridge, MA. Kimenyi, M. (2006). Ethnicity, governance and the provision of public goods. Journal of African Economies, 15(1):62–99. Kitching, G. (1980). Class and Economic Change in Kenya. The Making of an African Petite-Bourgeoisie. New Haven and London: Yale University Press. Knight, J.B., and Sabot, R.H. (1990). Education, Productivity, and Inequality: The East African natural experiment. Oxford: Oxford University Press. Kuznets, S. (1955). Economic growth and income inequality. American Economic Review, 45(1):1–28. Kyriacou, A.P. (2013). Ethnic group inequalities and governance: evidence from developing countries. Kyklos, 66(1):78–101. Leamer, E.E. (1987). Patterns of development in the j-factor n-good general equilibrium model. Journal of Political Economy, 95(5):961–999. Leibbrandt, M., Finn, A., and Woolard, I. (2012). Describing and decomposing post-apartheid income inequality in South Africa. Development Southern Africa, 29(1):19–34. Lewis, W.A. (1954). Economic development with unlimited supplies of labour. Manchester School of Economic and Social Studies, 22(2):139–191. Maddison, A. (2003). The World Economy: Historical Statistics. Paris: OECD. Maddison, A. (2005). Growth and Interactions in the World Economy. The Roots of Modernity. Washington, DC: The AEI Press. McKay, A. (2013). Growth and poverty reduction in Africa in the last two decades:  evidence from an AERC growth-poverty project and beyond. Journal of African Economies, 22(Supp. 1):i49–i76. McKay, A., and Perge, E. (2009). Spatial Inequality and its Implications for Growth—Poverty Reduction Relations. Draft Framework Paper. Nairobi: AERC. Nissanke, M. (2009). Linking Economic Growth to Poverty Reduction under Globalization: A Case for Harnessing Globalisation for the Poor in Sub-Saharan Africa, mimeo. Nairobi: AERC. Persson, T., and Tabellini, G. (1994). Is inequality harmful for growth? American Economic Review, 84:600-621. Pinkovskiy, M., and Sala-i-Martin, X. (2010). African Poverty is Falling … Much Faster than You Think, NBER Working Paper 15775. Cambridge, MA: National Bureau for Economic Research.

212   Concepts Ravallion, M. (2001). Growth, inequality and poverty: looking beyond averages. World Development, 29(11):1803–1815. Ravallion, M., and Chen, S. (2012). Monitoring Inequality. Mimeo. Available at: https://blogs. worldbank.org/developmenttalk/files/developmenttalk/monitoring_inequality_table_1_. pdf. Reardon, T. (1997). Using evidence on household income diversification to inform study of the rural non-farm labor market in Africa. World Development, 25(5):735–748. Rothstein, B. (2011). The Quality of Government. Chicago and London:  University of Chicago Press. Sahn, D.E., and Stifel, D.C. (2003). Urban-rural inequality in living standards in Africa. Journal of African Economies, 12(4):564–597. Schultz, T.W. (1953). The Economic Organization of Agriculture. McGraw-Hill. Thorbecke, E. (2013). The interrelationship linking growth, inequality and poverty in sub-Saharan Africa. Journal of African Economies, 1(22, Suppl):i15–48. UN (2013). Human Development Report 2013. New York: UN. Vicente, P. (2008). Is vote buying effective? Evidence from a field experiment in West Africa, mimeo. Wantchekon, L. (2003). Clientelism and voting behaviour: Evidence from a field experiment in Benin. World Politics, 55(3):399–422. World Bank (2009). World Development Report 2009. Washington, DC: World Bank. Young, A. (2012). The African Growth Miracle. Journal of Political Economy, 120(4):696–739.

Chapter 12

Inclu sive G row t h and Devel opme nta l Governa nc e  The Next African Frontiers Richard Joseph

12.1 Introduction Africa has undergone significant changes in the past quarter-century. Economic and political liberalization have facilitated an era of unprecedented growth. This growth, however, has not been accompanied by structural transformation. Poverty levels, though declining, are still high; increases in formal sector jobs are inadequate; agricultural yields are still low, and so are labor productivity and global competitiveness. Two crucial frontiers remain to be breached: patterns of growth that do not disproportionately benefit elites while the material condition of the greater majority stagnates; and acquiring modes of governance that are developmental in the sense of steadily optimizing performance and output.1 The 1980s were framed by two pivotal World Bank reports: Accelerated Development in Sub-Saharan Africa (1981), which declared the need for economic liberalization and structural adjustment policies; and Sub-Saharan Africa: From Crisis to Sustainable Growth (1989), which emphasized governance and institutional reforms. Subsequently, many reports by international agencies and policy action in Africa have been anchored to the arguments and perspectives in these studies. Taken together, a paradigm for the development community, in Africa and abroad, was established and its consequences have been extensive.2 Today, this paradigm is under challenge. There is now a revisionist paradigm regarding its main provisions. Economic liberalization, pro-market reforms, and a reduction in the state’s

1 

This exercise continues, and complements, the analysis and arguments in Richard Joseph (2013c). An excellent introduction to the changes in policy and action can be found in Todd J. Moss (2013). His chapter, “Africa’s growth puzzle,” pp. 91–104, provides a useful background to the issues to be discussed here. 2 

214   Concepts role, often grouped under the “Washington Consensus,” are challenged by those advocating industrial policies which emphasize the crucial role of the state in promoting growth and development. Allied to this critique is a questioning of the “good governance agenda,” which is viewed as mimicking the political principles and practices of Western democracies while overlooking the pathways followed by Asian countries in their rapid transition from poverty to prosperity. A decade after the pursuit of pro-market reforms advocated by Western aid agencies and international organizations, a wave of democracy swept over Africa. When it subsided, a good governance agenda persisted of the rule of law, accountability, transparency, and human rights. A third of the states of sub-Saharan Africa are today substantially democratic while the rest consists of quasi-democratic, electoral authoritarian, autocratic, and failed states. As will be shown, a number of experts who have studied the economic upswing, and the obstacles still to be overcome, acknowledge the pivotal role that the political abertura has played in Africa’s economic revival and its evolution.3 Having drifted from Africa during the quarter-century of economic stagnation and contraction that began in the 1970s, leading economists have returned to the study of the continent. Drawing on the experiences of Asian economies, some are challenging prevailing paradigms that regard governance and institutional failures as the greatest impediments to sustainable and transformative growth in Africa. Social scientists in other disciplines often use a different conceptual language. But bridges are being built between these schools, especially as they focus on common goals: expanding growth and eliminating mass poverty. Economist Roger Myerson expresses what is the consensus view of many political and other social scientists: “The great central question is what can anyone do to try to improve the quality of governance [in Africa]. How can development aid have any hope of actually doing good for poor people in poor nations if we don’t understand what kinds of assistance policies can encourage positive political development?”4 I have written that African countries “cannot achieve sustained and transformative growth, involving the shift from largely subsistence agricultural economies to globally competitive ones, and the generation of well-paying jobs for their urbanizing populations, without significantly improving ways of managing their collective affairs through appropriate institutions” (Joseph 2013c: 308). After reviewing the different approaches to this task, it was stated that we must seek to harmonize two contrasting but equally important perspectives: “one claims that growth-enhancing governance capabilities can be acquired through concrete economic endeavors; the other contends that bad governance, and deficient governments, will obstruct and corrupt even the suggested growth-enhancing process” (Joseph 2013c: 301). These arguments were summarized as follows: The Pretoria Roundtable and Good Growth and Governance brought to my attention an important wave of work by political economists concerned with how Africa could exit the labyrinth of low productivity and multiple social ills … I find the proposal that governance capabilities will be “discovered” in the growth process as insufficient as the one being rejected,

3 

I earlier used the Portuguese term, abertura, meaning opening to refer to the political and social transitions experienced worldwide, starting in the late 1980s. See Richard Joseph (1998). I now see the importance of the economic and social, as well as political, dimensions of the abertura. 4  Personal Communication, June 19, 2012.

Inclusive Growth and Developmental Governance    215 namely, that improvements in governance will open the door to accelerated growth. Neither perspective, as it concerns contemporary Africa, is compelling on its own. I suggest instead the need to harmonize them, in theory and practice.5 (Joseph 2012c: 310)

While preparing to write the sequel to that paper, a book appeared on the same topic: David Booth and David Cammack, Governance for Development in Africa: Solving Collective Action Problems (2013). This study synthesizes the findings of a large body of field research conducted under the auspices of the Africa Power and Politics Programme (APPP) in Britain.6 Unlike studies promoting industrial policy conducted by economists, the work of the APPP has been undertaken principally by scholars in other social sciences. The latter also have substantial Africa experience and draw on recent field research projects. Both communities of scholars, however, are advocates of the revisionist paradigm. This paper will examine, and respond to, this capstone study. While identifying its insights and limitations, a case will be made for combining two projects:  democracy-building and developmental governance. The revisionist paradigm downgrades the former.

12.2  The Case for Developmental Patrimonialism From the outset, APPP set about rethinking governance for development or what is referred to here as developmental governance. The writings of Tim Kelsall have been influential in this exercise. Kelsall called for “going with the grain” in African development, a notion that emphasized avoiding projecting onto Africa externally derived templates and ideologies. To this notion was added others such as “the right fit” and “building on what works” which also suggest that empirical diversities in Africa should determine which policies are introduced, where, when, and how. Two key theoretical notions have flowed from these researches. The first is the concept of developmental patrimonialism which is exemplified by the political economy strategy in Rwanda under the direction of its president, Paul Kagame, and the Rwanda Patriotic Front (RPF).7 The second is the insistence, repeatedly stated in Governance for Development in Africa, that a collective action framework must replace the dominant principal action framework in development policy thinking.

5  The Roundtable in Pretoria, July 3–4, 2012, on the topic, “New Thinking on Industrial Policy: Implications for Africa,” was led by Joseph E. Stiglitz, Justin Yifu Lin, and Célestin Monga. An earlier conference involving several of the same participants led to the publication: Akbar Noman et al. (2012). 6  APPP was funded by the UK Department for International Development and Irish Aid along with other donors. A subsequent paper will take up the arguments in the important APPP book by Tim Kelsall, Business, Politics and the State in Africa: Challenging the Orthodoxies of Growth and Transformation (Zed Books, 2013). 7  See D. Booth and F. Golooba-Mutebi (2012) and others papers posted on the APPP website, http:// www.institutions-africa.org/.

216   Concepts Developmental patrimonialism, it is claimed, has enabled Rwanda to avoid the ills of most of sub-Saharan Africa. The dominant pattern in the continent is seen as rent-seeking by officials and politicians in the conduct of public affairs. Such practices are “widespread and uncontrolled, and associated with political and administrative corruption” (Booth and Golooba-Mutebi 2012: 385). In Rwanda, however, while economic rents are generated, they are managed by a political elite and a techno-bureaucracy to serve national development, not just enrich big men and their cronies. In Rwanda, as in other African countries, there are “politically generated opportunities” to garner profits [rents], but they “are comprehensively institutionalized and centralized.” The key mechanism for accomplishing these goals is the creation of holding companies under the control of the RPF. They include Tristar Investments and Crystal Ventures Ltd. Other politically controlled entities include the Rwanda Investment Group, which raises funds for large projects. Among the many benefits of developmental patrimonialism in Rwanda is the opportunity for economic rents to be deployed with a long–term perspective. Another is the control of corruption. The authors contend that “there is no evidence of direct profit taking by individual politicians or military leaders” (Booth and Golooba-Mutebi 2012: 390). Moreover, they claim, “corruption is quite uncommon in Rwanda’s public service at any level and corruption with impunity is largely absent” (Booth and Golooba-Mutebi 2012: 392). A third positive feature is the building of a capable civil service in which “recruitment and promotion” is based “to a large extent on merit and effectiveness” (Booth and Golooba-Mutebi 2012: 392). The outcome of this political economy has been rapid advances in key social sectors. Before turning to Governance for Development in Africa, which follows this analysis, a few observations can be made. First, the developmental model of RPF-ruled Rwanda might be exceptional in nature, as has been the more democratic examples of Mauritius and Botswana, but they are difficult to replicate in the rest of the continent. Second, Rwanda, as Israel, has a post-liberation government deriving from, and resting upon, strong military capacities as well as being post-genocide entities. These features facilitate their political autonomy strategies. While Paul Kagame may be one of the most capable African leaders, he is also one of the most autocratic (Gettleman 2013). The regime has no democratic agenda beyond presentability (Joseph 1998). Its corporatist system is difficult to imagine in many other African countries in which the business interests of the head of state and members of the ruling elite are often managed in a semi-clandestine way. David Booth and Diana Cammack present the well-wrought theoretical aspects of the revisionist paradigm and highlight its empirical dimensions based on a small number of country case studies: Rwanda, Malawi, Niger, and Uganda. Their book represents an important fruition of the APPP project. Instead of opening analytical doors, however, it culminates in a cul-desac. The authors express pessimism about the likely impact of their study which appears predetermined by several factors. First is the use of Rwanda as the exemplar of developmental governance. Second is the dismissive attitude toward democratization in Africa. And third is the small number of cases, one a dynamic country with a regional influence, Uganda, and two countries that are impoverished and not regionally influential, Malawi and Niger. Finally, they anchor their analysis to an unpersuasive distinction between collective action and principal action frameworks. Although they emphasize the importance of local solutions, they recognize the significance of downward-imposed disciplines and even contend that “the masses are not significant players on their own behalf ” (2013: 4). As I and others argue, however, the

Inclusive Growth and Developmental Governance    217 activation and empowerment of the African demos—the large and usually poor populace—are key dimensions of tackling obstacles to inclusive growth and development. The authors ask key questions regarding developmental governance. How can blockages in the provision of public goods in Africa be reduced? (2013: 4). “What works, and what does not, in the provision of public goods?” (Booth and Cammack 2013: 6). “How are economic rents managed and distributed?” Virtually repeating the question raised by the Roger Myerson quotation cited earlier, they inquire about “the kinds of political processes and regime types that provide a more enabling environment for addressing typical blockages” (Booth and Cammack 2013: 7). In place of the good governance agenda emphasizing constitutionalism, the rule of law, human rights, transparent government, and anti-corruption efforts, they insistently ask: “what are optimal in given countries” (2013: 10). “Rather than donor ideals,” it is best” they say, “to start from country reality” (Booth and Cammack 2013). Much attention is devoted by Booth and Cammack to distinguishing the principal agent approach from the collective action one they recommend. Much of the work of development agencies, they claim, reflects efforts to overcome principal action impediments through improved information flows. Collective action, however, recognizes that the blockage is not about aims but “expectations about the behavior of others” (2013: 18). The challenge is to facilitate the collective action needed to solve development problems by citizens, officials, and others in specific local contexts. They do not examine how this contention applies to autocratic as contrasted with democratic systems. The fact that collective action can be facilitated to different degrees in different political systems is not considered. Democracy in Africa does not only reinforce clientelistic patterns of resource acquisition as they contend. The processes are far more complex.8 The chief principal actor in Rwanda, President Kagame and his regime, secure compliance from government officials through a variety of means, and they in turn from the populace. So principal agent strategies are strongly in evidence. Collective action is complicated in Rwanda as a dominant explanatory framework by the importance of top-down disciplines. The authors discuss in great detail the sub-optimal provision of public goods in their other focus countries, especially regarding maternal health, and contrast it with the optimizing approach in Rwanda. The information and insights provided are valuable. In their conclusion, as throughout the book, however, they downplay the broad economic and social advances taking place in much of Africa. Theirs is definitely not a perspective about a “Rising” or “Emerging Africa”: Countries of low-income Africa are once again experiencing sustained economic growth, and several indicators of human well-being are steadily improving too. But African development is not yet on a safe upward path. Current growth is not bringing widespread productivity gains or leading to structural economic change. Smallholder agriculture, on which the majority in most countries still depends, remains untransformed, and partly as a consequence, new sources of gainful employment are not being generated … When it comes to routine public services and regulation in the areas where ordinary people live, conditions remain abysmal. (2013: 122)

8 

These complexities are demonstrated in several chapters in Adebanwi and Obadare (2013).

218   Concepts Although their work has been financed by development agencies, the authors state that they are pessimistic “about the possibilities for getting real uptake of our perspectives on how to work more effectively on African governance among official donor agencies” (Booth and Cammack 2013: 138). Citing the experiences of Malawi and Niger, they contend that there is a conspiracy among “line-ministries, senior politicians and the donors to suppress any real sub-national experimentation if it appears to threaten their control over policy-making” (Booth and Cammack 2013: 129). It can be contended that, if there has not been “a turnaround in practice, among either aid donors or African reformers,” it is not surprising given the unremarkable nature of the advice they proffer: “finding new ways of addressing collective action problems”; “a robust commitment to “convening and brokering” as opposed to ‘delivering stuff ’; and “greater use of arm’s length assistance—funding of organizations that can do a better job of facilitating governance for development along the lines outlined” (2013: 139). 9 They do not provide African policymakers, and their external funders, as Justin Yifu Lin and Célestin Monga demanded in another context, “actionable prescriptions on how to design policies to achieve their economic and governance goals” (2012: 3).

12.3  Developmental Governance:  The Need for a Broader Aperture Instead of kicking away the ladder of externally prescribed development strategies, I contend that African countries should strengthen the ladders that have been painfully constructed since the 1980s. The East Asian model of authoritarian capitalism is important in understanding how growth and development could be accelerated and sustained, but it is not likely to be the model most African countries will adopt. They will arrive at their own variants and hybrid systems. Inspired by Richard Sklar’s notion of Africa as “a workshop of democracy” (1983), it is pertinent to consider many of the 48 states of sub-Saharan Africa as workshops of developmental governance. We can learn from examining the variety of models being devised in situ. The approach I recommend emphasizes learning from the realities (plural) of Africa, but not examining them from a single theoretical aperture, or simply based on the experiences of countries in a quite different cultural area of the world. Since the 1980s, sub-Saharan Africa has experienced profound transformations. APPP underplays the significance of the most sustained period of economic growth in Africa (mid-1990s to the present) since the independence era, and the political dimensions of the abertura that has led to dozens of liberalized, if not fully democratized, polities. Inclusive growth and developmental governance, the next frontiers, will require analysis and action on a broader canvas. Six key dimensions of the evolving transformations: economic growth, democracy, state and security, infrastructure, discordant development, and developmental governance will be considered.10 The focus will

9 

The work of myriad external agencies in Africa over the past three decades demonstrates how much these strategies have been attempted. 10  Steven Radelet (2010) identifies five areas of transformation in Africa, some of which coincide with the six identified here.

Inclusive Growth and Developmental Governance    219 be on the last five since economic growth is extensively discussed in the available literature and will be fully covered by contributors to the Handbook.

12.4  Democratic Transitions It is easy to criticize the quality of democratization in Africa, but APPP overdoes the criticism. When Booth and Cammack refer to democracy, it is usually to show how unprepared Africa is for it, and how little it has delivered in development. They make no reference to the Afrobarometer surveys that have tracked the course of democracy-building, and the responses of many societies, for over a decade. Nor do they draw on the growing literature comparing democratic outcomes in Africa.11 Contrary to the arguments of APPP, democratization has been a significant transformative force. At the end of the 1980s, only four sub-Saharan countries could be classified as being democratic to a certain degree: Botswana, Gambia, Mauritius, and Senegal. In 1991, Africa Demos included among the democratizing countries four additional ones: Benin, Cape Verde, Namibia, and São Tomé & Principe.12 According to the consensus scholarly view, the number of liberal democracies in Africa has remained around eight for over a decade, while there has been a substantial increase in the number of competitive authoritarian regimes and electoral democracies.13 Crawford Young has compiled an insightful table that draws on several indices of democracy, human development, transparency, and state capacity in continental Africa (2012: 360–361). What is noticeable is that nine of the ten sub-Saharan countries at the top of the list are among the most democratic in Africa:  Mauritius, Seychelles, Cape Verde, Botswana, South Africa, Namibia, Ghana, São Tomé, Benin, and Senegal.14 The last 15 countries on the list, all in sub-Saharan Africa, have autocratic governments. The Crawford Young Index also matches the Mo Ibrahim Index, which combines a large number of variables including “sustainable economic opportunities, human development, safety and rule of law, political participation, and human rights” (Young 2012: 358). The more democratic countries lead the list based on 57 variables while the non-democratic ones are found at the bottom. The point that these tables suggest, and it can be illustrated in other ways, is that not only is democracy—and even political liberalization—positively correlated with growth, security, and development in Africa, it can often prove essential. Michael Bratton and Carolyn Logan

11  There are now several sources that track democratization in Africa. They include the long-established index of Freedom House and more recent ones such as http://www.democracyinafrica. co.uk/ directed by Oxford scholar, Nicholas Cheeseman. 12  The full set of Africa Demos bulletins, 1990–1995, is now available online: http://books. northwestern.edu/viewer.html?id=inu:inu-mntb-0006443104-bk. The early categorization of African political systems in transition appeared in Richard Joseph (1991). 13  In Larry Diamond’s typology of liberal democracies of 2009, he removes Gambia and Senegal from the Africa Demos 1991 list, and substitutes Ghana and South Africa. I agree with these changes. “Introduction.” Joel Barkan (2009) arrived at the same set of liberal democracies. 14  Although Senegal follows Lesotho on the list, it will have climbed a few spots since the removal of Abdoulaye Wade as its president in the 2012 elections.

220   Concepts have closely explored the democracy and development nexus using their important concept, “Claiming Democracy,” and drawing on the abundant survey data of Afrobarometer. Bratton and Logan acknowledge:  … the full potential of democracy—including the promise of accountable governance—has yet to be fulfilled. Economic growth is still elusive, corruption remains widespread, and aid dependency continues to frustrate recipients and donors alike. As global leaders contemplate the massive increases in international assistance to the continent, questions have therefore been asked about indigenous capacity to absorb an influx of new funds without exacerbating old problems. A smart aid approach would seem to require that African political leaders are held accountable—not only to donor agencies but, more importantly, to their own people—for sound policy choices and the effective use of resources. (2009: 181) [italics added]

The authors then ask: “why has democratization so far failed to secure better governance?”; “why is policy performance so poor?”; “Why haven’t competitively elected governments in Africa demonstrated a significantly greater degree of accountability to their publics than the authoritarian systems that they replaced?” (Bratton and Logan 2009: 181). These questions overlap with the APPP research agenda, but the responses differ. While “the road to accountable governance may be a long one,” Bratton and Logan state, transforming African governance depends on Africans grasping “their political rights as “citizens,” notably to regularly claim accountability from leaders” (2009: 182). They provide a chart that traces how Africans view the accountability of elected leaders to their constituents. It shows great variation, from 3 percent in the case of Benin to just under 40 percent in Namibia. Bratton and Logan do not regard this dimension of political behavior as static. “Despite the residual influence of historical legacies,” they contend, “there remains considerable room for popular political learning” (Bratton and Logan 2009: 202). They then suggest ways in which donors can “foster mechanisms of vertical accountability” that both differ from, and overlap, with APPP proposals: • Start locally … Transfer authority over local affairs not only to locally elected officials but to citizens themselves • Promote transparency and access to information … Support processes of public budgeting, which can engage the public while reducing the diversion of public funds. • Give voice to citizens … insist on public popular input into public policy processes. • Encourage representatives to engage with their constituents … Consider providing incentives for greater engagement … publicize the constituency service of every elected representative. • Fight corruption and legal inequality … Encourage citizen engagement in the fight against corruption. • Build on elections … use elections as vehicles for publicizing the broader rights and responsibilities of citizenship. • Above all, do no harm … Purge aid programs of requirements that cause governments to concentrate more on accountability to foreign agencies than to their own citizens. What we see in Bratton and Logan is a strong affirmation of collection action strategies as facilitated by democratization as well as one of the principal action strategies

Inclusive Growth and Developmental Governance    221 derided by Booth and Cammack: “Only internal, vertical accountability imposed by an active and lasting democracy,” they contend, “can provide a truly secure foundation for an effective and lasting democracy, and for a sustainable program of social and economic development” (2009: 201) [italics added]. Four years later, drawing on the most recent round of Afrobarometer surveys, Bratton and Logan ask: “Do people in African societies remain subject to authoritarian legacies that left little room for questioning leaders or developing expectations of government, much less demanding accountability?” (Logan and Bratton 2013). They respond that the percentage of voters who declare that leaders should be chosen via “regular, open, and honest elections” has remained consistently high at 82 percent in the countries surveyed. According to Bratton and Logan, political learning is steadily taking place. “Two-thirds (66 percent) of Kenyans now assert that it is voters who must hold MPs to account, compared to 45 percent in 2005” (Logan and Bratton 2013). APPP had targeted how well governments perform, that is, in their provision of public goods and services. Afrobarometer surveys show that Africans are not just concerned about service delivery but also the democratic character of their governments. When asked if it was important for governments to get things done or be able to hold it accountable, even if it slows down decision-making, 55 percent in 15 countries surveyed chose the latter. While there is a widening gap between the demand and supply of accountability, it has not resulted in a decline in the demand for democratic governance. Bratton and Logan see a two-stage process that is still unfolding two decades following the abertura: “it was no easy matter for individuals to transit from “subjects” under authoritarian rule to “voters” in electoral democracies—and thereafter to rights- and accountability-demanding citizens” (Logan and Bratton 2013). The Afrobarometer data suggest that learning and attitude changes are occurring, and that the demands of citizens for political accountability are strengthening in several parts of the continent. APPP contend that it is external donors, and international NGOs, that promote the agenda for democratic advance, civil society strengthening, and information flows that can empower African citizens to hold governments accountable. This is not the whole truth and underplays the considerable amount of African-led action in all these realms. As Afrobarometer shows, the support of African people for democratic government has remained constant at over 80 percent. These findings suggest the importance of the unfulfilled demand for more accountable government as a means of pursuing more socioeconomic progress. The evidence shows that African citizens want both club goods and to enjoy the supreme public good of a democratic polity.15 Even where economic advances and social sector gains have been substantial, as in Ethiopia and Rwanda, agitation for democratic progress has not abated. Booth and Cammack and Bratton and Logan present diametrically different approaches to the same key questions about governance for development in Africa. Transcending, or harmonizing them, is an important objective of this exercise.16

15 

This is an argument Amartya Sen (1999) has cogently made. The exercise referred to include my three sequential papers on the “revisionist paradigm”—referred to by APPP as unorthodox strategies—and a collaborative effort that is being launched in 2014 involving a network of researchers. 16 

222   Concepts

12.5  State and Security Dilemmas The revisionist paradigm does not take sufficiently into consideration the evolving nature of African state systems. From the seminal works by Robert Jackson and Carl Rosberg (1982) on the survival of weak states, through Jeffrey Herbst’s path-breaking study of the projection of state power (2010), to Crawford Young’s summation of decades of study of the post-colonial state (2012), generations of scholars have traced the shaping and reshaping of territories crafted by Europeans and transferred to Africans. It is important to bring the still-evolving nature of African statehood into considerations of economic and social progress. Moreover, as Tom Carothers appropriately remarked, overlooked in the euphoria over the political transitions in Africa in the 1990s was the fact that these were taking place in “devastatingly weak states” (2002: 24). A major transformation in the African state system has been the ending of large-scale wars (Straus 2012). In the 1970s and 1980s, but petering out in the subsequent two decades, Africa was the scene of major wars, most internal to particular countries. Angola, Mozambique, Namibia, South Africa, Ethiopia, Liberia, Sierra Leone, Sudan, Congo, and Somalia have all been wracked by armed struggle. With the exception of Congo, Somalia, Sudan, and South Sudan, these have largely ended. As Paul Collier has persuasively argued (2007), one of the greatest obstacles to growth and development is armed warfare. Every African country that escapes armed struggle is another than can embark on long-term growth. Post-liberation regimes in Africa are disproportionately capable of driving accelerated growth. They include Angola, Ethiopia, Rwanda, Uganda, and Namibia. As Booth and Cammack show in the case of Rwanda, these regimes can enforce, when they choose, more centralized allocation, or access to, economic rents. They can restrict the extent of decentralized corruption that usually drains development resources and adopt long-term horizons, the latter being a key APPP recommendation. Other states that made socioeconomic advances under the auspices of militarized regimes are Burkina Faso and Ghana. Ghana is today an important model of democratic development as it made important shifts: from a quasi-socialist to a market system; from an authoritarian to a democratic political order; and from a fragmented and weak state system to an increasingly capable one. As Guillermo O’Donnell (2001) argued, “democracy should not be analyzed only at the level of the political regime … it also must be studied in relation to the state—especially the state qua legal system.” It can further be said, no capable state, no sustained economic growth and development. Insecurity as a consequence of excessive violence takes many forms in Africa. In the case of Nigeria, the prolonged conflicts in the Niger Delta, and their extension to the hinterland, affected a wide geographical area. The north-east of the country has been trapped for five years in brutal combat conducted by the militant jihadist group, Boko Haram, and matched by the brutalities of state security forces. Frequent atrocities in the Middle Belt area around the city of Jos have severely harmed the economies of these areas. The loss of economic output of Nigeria’s northern region, and the high expenditures for state security operations, constitute a profound drain on the resources of the Nigerian state. On the other hand, there are positive dimensions of state-building that can enhance Africa’s economic prospects. Achille Mbembe’s notion of the “re-territorializing” of Africa

Inclusive Growth and Developmental Governance    223 (2001), caused by the movement of peoples and goods, is reflected in the emergence of new geographical spaces of economic activity that transcend provincial and national boundaries. The case of Lagos State in Nigeria is a case in point. It is the fulcrum of an economic zone in West Africa that now extends into neighboring states (as well as peripheral Nigerian ones) as entrepreneurs operate over widening areas just as small traders, money-changers, and smugglers have done for decades.17 Restructuring is also taking place at the level of formal institutions. Among the most dynamic is the East Africa Community, which is playing an increasing role in financial, trading, and investment activities. Already, elements of a transnational state order can be perceived. Another dimension of transnational institution-building is the emergence of a judicial system to adjudicate matters across borders.18 There are advances in West Africa, often conducted without fanfare, in creating a judicial infrastructure that can facilitate the conduct of economic activity over a large region by both African and non-African entrepreneurs. APPP researchers recognize the importance of operational innovations at sub-national levels, but they have been too insistent on forcing developments into a specific theoretical grid. There is a lot to learn about how economic activity is actually conducted at local and provincial levels. In the case of Nigeria, the extraordinary achievements under the auspices of a democratically elected government of Lagos state have shown that state capacity can be built in a transparent way alongside extensive social and infrastructure investments.19 What Booth and Cammack call “pockets of effectiveness” can become “nodes” of effective governance whose elements can be replicated internally and externally.20 And all this can be accomplished under democratic auspices. Perhaps the greatest security dilemma in Africa today concerns the spread of violent militias, especially in “a band of insecurity and instability” that extends from Mauritania in the southwest to Djibouti in the north-east (Joseph 2012a). The anti-terrorism priorities of Western governments, and the increasing presence of overt and covert security forces in at-risk nations, can replicate the Cold War experience. It will take great vigilance on the part of legislatures, media entities, and local and international non-governmental organizations to ensure that the financial and manpower resources used to bolster anti-terrorist capacities are not diverted to non-developmental activities.

12.6 Infrastructure AAAP has focused on the provision of public goods and examined specific social sectors. Of equal importance is the building and maintaining of core infrastructures. Africa is almost universally under-supplied with adequate roads, railways, and other public transport; 17 

See Howard French (2013). This important development is the focus of ongoing research by Karen Alter of Northwestern University and her associates. 19  The Lagos State developments can be compared with similar advances in Jakarta, Indonesia. 20  In Nigeria, government officials from other states regularly visit Lagos to study the “Fashola” model of developmental governance. 18 

224   Concepts water and sanitation systems; electric power generation, transmission, and consumption; and health and educational units at all levels. Indeed, Africa today is a frontier for the rapid installation and improvement of these core infrastructures. Led by China, contractors are now present all over Africa seeking to fill the infrastructure gap. The USA, appropriately, devised the Millennium Challenge Account to enable recipient countries to decide how their sizeable grants should be invested. Many have chosen to tackle core infrastructure needs. Africa’s infrastructures must be rapidly improved to match its economic advances. Some countries with an autocratic government and ample financial resources, such as Gabon, can contract with external firms, such as the Bechtel Corporation, to undertake far-reaching infrastructure projects. Chinese, Japanese, and Indian firms, along with large Western industrial companies such as GE, are taking part in long-term infrastructure projects in the continent. In keeping with the governance-enhancing growth perspective, this is an obvious case in which “learning by doing” will be put to the test. The training of large numbers of Africans in these projects, and providing them with the framework to sustain the norms, attitudes, and skills these projects require, should be an important concern for all involved. Japan, electing to scale up its engagement in Africa in response to China’s, has made training and capacity building a central focus. Many African countries, such as Ghana and Nigeria, once possessed railways that were gradually eroded.21 The same is true of universities, hospitals, power-generating systems, oil refineries, agricultural extension stations, and road networks. A revolution in institution-building and–maintaining behaviors must accompany the upsurge in infrastructure investments (Joseph 2002).

12.7  Discordant Development Discordant development is another concept introduced to the study of contemporary Africa (Joseph 2013a). It applies more widely as is apparent from contemporary debates about the disproportionate amount of wealth that flows to the more affluent members of American society and those of other countries. Africa’s growth expansion is taking place in contexts in which neo-patrimonial and clientelistic systems persist. The “political settlements,” a concept favored by new paradigm advocates, have shown surprising continuities from the past despite the wide adoption of multiparty politics and regular elections. The great disparity in socioeconomic gains is unmistakable. When the masses of the population, as is currently the case in South Africa, see that political transitions have produced new dominant classes who disproportionately benefit from state-mandated economic changes, while their material conditions stagnate or decline, it is inevitable that political unrest and instability will follow. Mali is a textbook case of discordant development (Whitehouse 2013). A government that was widely praised for its democratic advances, and lavishly supported by external donors, facilitated the consolidation of a 21  Todd Moss cites the case of the Kenya Railways, and Daron Acemoglu and James Robinson that of Sierra Leone, to show how these systems suffered from disruptive politicized decision-making.

Inclusive Growth and Developmental Governance    225 self-serving political class while the economy eroded and minimal material resources flowed downwards. The natural resource expansion in Africa through major petroleum, gas, and coal discoveries means that the flow of external rents into government coffers will significantly expand. There are many countries, for example Nigeria, Congo, and Guinea, in which the population has generally failed to prosper from their nations’ mineral wealth. Some authors propose that resource wealth should flow more directly to the people themselves rather than to government officials in the hope that they will be used for development purposes (Diamond and Mosbacher 2013). There are few cases to point to in Africa, with the exception of Botswana, in which resource flows have yielded improved institutions and sustained material gains for the population. While the old paradigm has not worked to counter these tendencies, the new paradigm takes the authoritarian regimes of Ethiopia and Rwanda as their models. In the current growth and security era, external donors will experience constraints in their capacity to influence the behavior of African governments. There is a need for strategies based on analyses of leadership, institutions, political culture, and resources (Joseph 2013c). There must also be a further strengthening of both governmental institutions, and civic organizations to hold them accountable, and also a deliberate promotion of non-elite empowerment. Advances in democratic developmental governance are as imperative as the economic and political liberalization that have introduced important but incomplete changes to Africa’s political systems and political economy. Non-democratic developmental governance, now called developmental patrimonialism, is not the preferred option for Africa. As Botswana and Lagos State demonstrate, African political systems can be both democratic and developmental. There should now be an affirmation, in theory and practice, of democratic development that includes political, social, and economic dimensions.

12.8  Developmental Governance:  Divergent Pathways Steven Radelet (2010) has provided a robust defense of the consensus in development thinking on governance and development that came to dominate policy-thinking regarding Africa after 1981. He identifies five fundamental changes that have been responsible for the economic turnaround. The first is the rise of more democratic and accountable governments. The others are more sensible government policies, the end of the debt crisis, the spread of new technologies, and the emergence of a new generation of policymakers, activists, and business leaders. Radelet’s synthesis has a central theme: the significant political and social changes in Africa from the late 1980s that accompanied and facilitated economic liberalization and eventually sustained economic growth. And, as Crawford Young cogently contends, “the political opening produced by the 1990s democratization surge is far from being erased” (2010: 335). The end of the debt crisis involved major changes in Africa’s relations with the international financial and donor community. After decades of trying to negotiate reforms with African governments whose officials had a limited, and often ideologically

226   Concepts distorted, understanding of economic issues, the abertura facilitated the emergence of economically competent officials who were empowered to take charge of negotiations with external agencies and foreign and domestic investors. The “spread of new technologies,” Radelet points out, were “creating new opportunities for business and political accountability” (2010:  20). Finally, “the emergence of a new generation” of actors across government, business and civil society created a significantly altered political and social environment from the one that prevailed during Africa’s decades of stalled growth (Radelet 2010). As mentioned earlier, APPP emphasizes Africa’s neo-patrimonial political economies while minimizing the new socioeconomic trends. We can speak of two political economies today, one modernizing, the other stubbornly prebendal. In the case of Rwanda and Ethiopia, prebendal practices can be constrained by a modernizing authoritarian state. In the majority of African countries, however, both can be simultaneously in play as the case of Ghana exemplifies. Radelet’s summary of the Ghanaian turnaround merits citing: The Ghanaian economy today bears little resemblance to the basket case of the early 1980s. Economic growth has averaged 5 percent for 25 years, translating into a 70 percent increase in the income of the average Ghanaian. The poverty rate … fell dramatically to 30 percent by 2005 … Investment, which had dropped to just 7 percent of GDP between 1978 and 1984, has risen to more than 25 percent of a much larger GDP … The economy is more diversified, with the over-reliance on cocoa supplanted by larger contributions from minerals, timber, manufacturing and a growing range of nontraditional exports … (2010: 72–73)

Radelet contends that there is a causal link between democratization, even flawed, and economic expansion in much of Africa. “The shift towards more democratic governance,” he states, “appears to have (imperfectly) increased accountability and significantly improved economic management relative to authoritarian governments. Policies are not solely determined by patronage politics but are responses to the need for sensible economic policies for a broader segment of the population” (2010: 88) [italics added]. As asserted earlier, there exists in Africa a variety of developmental models and they are steadily evolving. Instead of advocating one model over others, this is a time to take account of Todd Moss’ contention that we are far from understanding which “combination of factors are most relevant and in which cases” (2010: 103). We should not, Moss cautions, “expect to find a precise combination that applied across multiple countries” (2010: 103). What this exercises points to is the need for collaborative studies of growth and developmental governance in Africa, pursued with Moss’ injunction in mind: “a precise combination that applied across multiple countries” is not the “holy grail” of African development policies. “The quest for the answer to Africa’s growth puzzle is unlikely to find a universal answer” (2010: 103). This is the time to widen the theoretical aperture and seek to understand what is actually taking place in Africa’s workshops of developmental governance.22

22  This is, indeed, how I arrived at the theory of prebendalism to explain key features of the dynamics of Nigerian politics and political economy (Joseph, 1987/2014; 2013b).

Inclusive Growth and Developmental Governance    227 The AAAP researchers arrived at a cul-de-sac because their analyses not only do not correspond with the proclivities of donor agencies, as they admit, but also do not correspond with an Africa that has seen such significant changes in just two decades. For all we know, these transitions may only be at early stage. And it is an Africa that will enjoy substantial resource inputs from mineral exports, the talents of a large overseas diaspora, investment flows from an ever-increasing group of foreign companies, and the targeted interest of external nations such as Brazil, Japan, and especially China. We should not presume to know how these processes will evolve.

12.9  Frontiers in Theory and Practice We have entered a new era in Africa following the growth acceleration in the mid-1990s. There are major hurdles still to be surmounted in the building of adequate infrastructures, the provision of basic services to the populace, human capital development to provide the skills needed to build globally competitive economies, and structural changes that would make commodity price rises and earnings from mineral extraction serve as inputs for more productive economies. There are two contrasting assessments of Africa which can be called the disaster and progress narratives (Joseph 2012b, c). Crawford Young captures them by citing two authors: “by and large, the states of sub-Saharan Africa are failures”; and “there is no reason to downplay the progress that African countries have made in the past two decades under very difficult circumstances” (2012: 334). The first quote, from Pierre Englebert, is an exaggeration. The second, from Goran Hyden, is undeniable. They can be blended: African states have made substantial progress in a quarter-century but are far from realizing their potential to optimize growth, security, and development. In “Industrial Policies and Contemporary Africa” (Joseph 2013c), the growth-enhancing ideas of economist Mushtaq Khan were summarized in eight provisions. The eighth of these reads: “Ultimately, societies have to devise their own political compromises and government institutions that can pragmatically address their growth challenges given their historical and political constraints.” [italics added] Experts who subscribe to the two prevailing paradigms would agree with this observation. A decade after the political abertura in Africa, it was acknowledged that a variety of political experiments had been launched. “Identifying the ones,” it was argued, “that are likely to combine state building, democracy, and economic growth is a vitally important project that requires similar studies of the actual exercise of power and the building of institutions in a range of countries”(1999: 13). Today, Crawford Young recognizes the same challenge: “the striking diversity of state itineraries since 1990 produced a variety of outcomes in terms of political form, developmental performance, and quality of governance …” (2012: 335). Such a comparative project can be anchored to Dani Rodrik’s contention (2007) that high-quality institutions are required for countries to expand beyond the growth acceleration stage. And also to his view of democracy as a meta-institution that encourages and facilitate the spawning of such institutions. It is through institutions that governance takes place. This is a basic but crucial observation that should not be overlooked. The essence of developmental governance is the building of institutions, and their management, to optimize the production of public goods and services. The pace of population growth, urbanization,

228   Concepts climate change, and the coming on stream of new petroleum, gas and other mineral resources—together with the search for structural economic transformation—will require more effective, law-based, and accountable governments. Acquiring governments capable of “guiding the economy and overcoming obstacles to the process of continued economic upgrading” should not be minimized; nor should the importance of simultaneously improving the quality of growth. The paradigm the revisionists are seeking to supplant is itself not that old. A quarter-century ago, it was the new paradigm seeking to displace complacency regarding repressive, autocratic, and corrupt forms of governance that had undermined economies and livelihoods. Today, it is being shown the door by the revisionists (who prefer the rubric, “unorthodox”). But the essence of the situation is that both the new and no-longer-new frameworks are now contending for primacy. The former cannot fully take over and the latter is not willingly departing. Several scholars are helpful in pointing the way to their harmonization. Steven Radelet contends that the economic expansion in Africa is attributable both to the economic reforms introduced in the 1980s and the political transitions that occurred from the early 1990s. He contends that the widening economic space in Africa, and pressures to deliver benefits to formerly marginalized groups such as farmers, is undeniably connected to the political openings that gave voice to the voiceless:  … the shift toward democracy, greater transparency, and increased accountability has fundamentally changed the dynamic for the formation of public policies. The press is freer and livelier in many countries, and there are a growing number of NGOs, think tanks, business groups, and other organizations that push for more sensible economic and social policies. A new generation has come of age that expects and demands competent economic management. (2010: 88)

Shantayan Devarajan and Walter Fengler echo Radelet in arguing that political changes opened up spaces that enhanced the formulation, and acceptance, of better economic policies. Political liberalization was part of a broader process that involved freer media, more vigorous civil societies, and also greater access to information. They also contend that the persistent problems which inhibit structural transformation also “reflect government failures that will be very difficult to overcome.” “The obstacles to durable growth in the region are largely political” (2013: 81), they claim, echoing Myerson. According to Devarajan and Fengler, “basic public services have been stolen by or diverted to political elites,” a striking assertion by senior World Bank officials (2013: 79). Such practices represent an alarming addition to the phenomena of state capture and prebendalism. Despite the huge sums expended on health and education, Devarajan and Fengler claim, echoing Booth and Cammack, that there is “little to show for it.” Access to water by urban households, they contend, “has declined in almost every urban area of Africa” (2013:  77). The economic reforms and political abertura have contributed to the emergence of politico-business elites who not only benefit from new forms of accumulation but also “thwart the delivery of public services” (2013:  79). Clientelism, patronage, and cronyism determine how basic public services are distributed. While democracy is publicly applauded, access to wealth and even basic public goods are not being democratized.

Inclusive Growth and Developmental Governance    229 What responses are there to these dilemmas? Diamond and Mosbacher, while addressing the substantial expansion of exploitable oil and gas resources in Africa, and the possible emergence of a dozen more major mineral exporters, propose ensuring that the masses of African people are not excluded once again from their nations’ bounties. The potential earnings from the new resources, they claim, could “revolutionize social wellbeing” and transform politics and economies if properly utilized. To achieve such an outcome, a radical change is needed. Diamond and Mosbacher caution that the new revenue flows could drag these countries “down to the miserable governance level of the current exporters” (2013: 94).23 While the new programs to enhance transparency about mineral revenues such as Publish What You Pay, Extractive Industries Transparency Initiative, and other anti-corruption programs are helpful, a more radical break from past systems of revenue allocation is needed. They recommend that a portion of the new oil revenues should be distributed directly to a country’s citizens “as taxable income.” Under this “oil-to-cash” program, these allocations would go directly to the bank accounts of citizens. Other cases of cash transfers have shown that such funds are mostly used for “food, education, health care, and business investments.” The Diamond/Mosbacher suggestion, mooted in other contexts, would represent a real “paradigm shift.” It would demonstrate that “a country’s natural resources belong not to the state but to its people”; and it would “change the relationship between citizens and the state” as the people are incentivized to monitor how “their” oil and gas wealth was managed (2013: 98). There is a fundamental difference between the claims of Booth and Cammack that “the masses are not significant players on their own behalf ” (2013: 4), and the perspectives of Bratton/Logan and Diamond/Mosbacher who agree but aim to change a relationship they see as flawed. As long as African peoples are subjects and not citizens, their nations’ wealth will not be used for their benefit. Diamond and Mosbacher are adamant on this point: “… the argument that poor people don’t understand their best interests as well as bureaucrats and public servants do is a paternalist myth” (2013:  96). By providing cash reserves directly to them, in other words, reversing the pyramid of resource allocation, “a broad constituency of citizens” can be constituted “in place of the often passive populations of corrupt resource-cursed states” (2013: 95). Shifting power to populations through endowing them with cash resources will alter some of the power imbalance. But institutional dilemmas will persist. Devarajan and Fengler contend that a program to provide vouchers to farmers for a fertilizer subsidy in Tanzania ended up with most of the benefits captured by officials (2013: 75). Clientelist patterns can therefore override and warp policy initiatives. I have therefore argued the need for a rupture of the macro-institutional political framework in Africa. Effective and transparent institutions in Africa cannot be achieved without the exercise of reformist political power, such as has been moderately accomplished in Ghana and significantly thus far in Lagos State. Optimizing institutional performance in Africa is a major challenge yet to be overcome. This is the essence of developmental governance. Moreover, inclusive growth depends on the aims of those who direct these institutions, who participate in designing and monitoring them, and whose interests are actually served. Devarajan and Fengler concur with Booth 23  Radelet (2010) avoided this dilemma by excluding oil exporters from the countries included in “Emerging Africa.”

230   Concepts and Cammack regarding the low quality of services available for the masses of the population. The former see this state of affairs as a direct outcome of political forces and practices. Who gets what, when and how, the political science axiom, also determines in Africa who does not. The latter set of scholars seeks the answer in the bypassing of these political processes and the substituting of collective action by officials and citizens in local contexts to solve local problems, strongly guided by an autocratic head of state, and a bureaucracy incentivized to carry our instructions and pursue established goals.

12.10 Conclusion In conclusion, we can return to the notion of “rupture” that was made in “Industrial Policies and Contemporary Africa” (Joseph 2013c). There is need for a rupture in the macro-institutional framework of African economies and societies as occurred in the Nordic countries in the last decades of the nineteenth century and in East Asia in the second half of the twentieth. This rupture can take place under authoritarian, democratic or hybrid political auspices. Distilling and contrasting these emerging models, and not steering analyses unduly in a particular direction, is what is needed during this period of progress and uncertainty. Booth and Cammack see the need for rupture but center it on promoting collective action in local contexts without necessarily requiring greater democratization. Bratton and Logan disagree and so do Diamond and Mosbacher. The latter advocate a profound rupture in how rents from mineral exports are allocated. Bratton and Logan envisage a gradual strengthening of processes underway, building on changing attitudes regarding the role of citizens in holding governments accountable. What should be evident from this discussion is that no one has as yet the answer to the key governance and development quandary in Africa enunciated by Myerson. There is likely to be different paths to a common goal (Ohno and Ohno 2012). All the experts cited here recognize the basic facts of the situation: Africa has experienced significant political and economic changes in a quarter-century but is still hampered by defective systems of governance. They all recognize that discordant development and non-transformative growth have been the consequences. The answers will be found, to use a no-longer-fashionable word, in praxis: simultaneous explorations in theory and practice. The conversations now taking place within communities of economic and other social science experts must be carried over into the public sphere; and policy initiatives in the latter must be distilled for the knowledge emerging from the practical solving of development problems. Twenty years after the ending of apartheid, South Africans wonder how they will escape the perverse impact of the proclivities of a new politico-business elite. The answers will come less from abroad than the near-abroad because many African states are engaged in the third stage of a multiplex transition: seeking to advance from acquiring freer economies and political systems and transiting to more developmental ones. Africa’s bounty is not only its natural and human resources, but also the many workshops of growth and developmental governance. What is essential is to make use of the revolution in information technology to distill the advances being made on the ground. This exercise suggests the urgent need for a comparative project on inclusive growth and developmental governance. Such a project would

Inclusive Growth and Developmental Governance    231 build on the analytical approaches discussed here, and ongoing efforts in Africa’s national, sub-national, and transnational workshops, with the aim of designing more democratic and developmental systems.

References Adebanwi, W., and Obadare, E. (2013). Democracy and Prebendalism in Nigeria:  Critical Interpretations. London: Palgrave. Barkan, J. (2009). Advancing democratization in Africa, Table 5.2, in J.S. Morrison and J.G. Cooke (eds), U.S. Africa Policy beyond the Bush Years. Washington, DC: Center for Strategic & International Studies, p. 96. Booth, D., and Golooba-Mutebi, F. (2012). Developmental patrimonialism? The case of Rwanda. African Affairs, 111(444):379–403. Booth, D., and Cammack, D. (2013). Governance for Development in Africa: Solving Collective Action Problems. London: Zed Books. Bratton, M., and Logan, C. (2009). Voters but not yet citizens: democratization and development aid, in R. Joseph and A. Gillies (eds), Smart Aid for African Development. Boulder and London: Lynne Rienner, pp. 181–206. Carothers, T. (2002). The end of the transition paradigm. Journal of Democracy, 13(1):5–21. Collier, P. (2007). The Bottom Billion: Why The Poorest Countries Are Failing And What Can Be Done About It. Oxford: Oxford University Press. Devarajan, S., and Fengler, W. (2013). Africa’s economic boom: why the pessimists and the optimists are both right. Foreign Affairs, 92(3):68–81. Diamond, L., and Mosbacher, J. (2013). Petroleum to the people: Africa’s coming resource curse—and how to avoid it. Foreign Affairs, 92(5):86–98. Gettleman, J. (2013). The global elite’s favorite strongman. The New York Times, September 4. Herbst, J. (2000). States and Power in Africa: Comparative Lessons in Authority and Control. Princeton: Princeton University Press. French, H.W. (2013). How Africa’s New Urban Centers Are Shifting Its Old Colonial Boundaries. http://www.theatlantic.com/international/archive/2013/07/how-africas-new-u rban-centers-are-shifting-its-old-colonial-boundaries/277425/. Jackson, R.H., and Rosberg, C.G. (1982). Why Africa’s weak states persist: the empirical and juridical in statehood. World Politics, 35(1):1–24. Joseph, R. (1987/2014). Democracy and Prebendal Politics in Nigeria: The Rise and Fall of the Second Republic. Cambridge: Cambridge University Press. Joseph, R. (1991). Africa: the rebirth of political freedom. Journal of Democracy, 2(4):11–24. Joseph, R. (1998). Africa 1990–1997: from Abertura to closure. Journal of Democracy, 9(2):3–17. Joseph, R. (1999). State, conflict and democracy in Africa, in R. Joseph (ed.), State, Conflict, and Democracy in Africa. Boulder and London: Lynne Rienner, pp. 3–14. Joseph, R. (2002). Smart Partnerships for African Development: A New Strategic Framework. Special Report. Washington, DC: United States Institute of Peace. Joseph, R. (2012a). Insecurity and counter-insurgency in Africa. Foresight Africa 2012. Washington, DC: The Brookings Institution. Joseph, R. (2012b). Strategic Priorities in Contemporary Africa, Part I. http://2012summits. org/commentaries/detail/joseph_1.

232   Concepts Joseph, R. (2012c). Strategic Priorities in Contemporary Africa, Part II. http://2012summits. org/commentaries/detail/joseph_2. Joseph, R. (2013a). Discordant Development and Insecurity in Africa, Foresight Africa 2013. Washington, DC: The Brookings Institution. Joseph, R. (2013b). The logic and legacy of prebendalism in Nigeria, in W. Adebanwi and E. Obadare (eds), Democracy and Prebendalism in Nigeria: Critical Interpretations. London: Palgrave, pp. 261–280. Joseph, R. (2013c). Industrial policies and contemporary Africa: the transition from prebendal to developmental governance, in J.E. Stiglitz et al. (eds), The Industrial Revolution II: Africa in the 21st Century. New York: Palgrave, pp. 293–318. Kelsall, T. (2013). Business, Politics and the State in Africa: Challenging the Orthodoxies on Growth and Transformation. London: Zed Books. Lin, J.Y., and Monga, C. (2012). Solving the mystery of African governance. New Political Economy, 17(5):659–666. Logan, C., and Bratton, M. (2013). Claiming Democracy: Are Voters Becoming Citizens in Africa? http://africaplus.wordpress.com/2013/05/14/claiming-democracy-are-votersbecoming-citizens-in-africa/. Moss, T.J. (2013). African Development:  Making Sense of the Issues and Actors, 2nd edition. Boulder and London: Lynne Rienner. Mbembe, A. (2001). On the Postcolony. Berkeley: University of California Press. Noman, A., Botchwey, K., Stein, H., and Stiglitz, J.E. (2012). Good Growth and Governance in Africa: Rethinking Development Strategies. New York: Oxford University Press. O’Donnell, G.A. (2001). Democracy, law, and comparative politics. Studies in Comparative International Development, 36(1):7–36. Radelet, S. (2010). Emerging Africa:  How 17 Countries Are Leading The Way. Washington, DC: Center for Global Development/Brookings Institution. Rodrik, D. (2007). One Economics, Many Recipes: Globalization, Institutions, and Economic Growth Princeton: Princeton University Press. Sen, A. (1999). Development as Freedom. New York: Alfred A. Knopf. Sklar, R.L. (1983). Democracy in Africa. African Studies Review, 26(3/4):11–24. Stiglitz, J.E., Lin, J.Y., and Patel, E. (2013). The Industrial Revolution II: Africa in the 21st Century. New York: Palgrave. Strauss, S. (2012). Wars do end! Changing patterns of political violence in sub-Saharan Africa. African Affairs, 111(443):179–201. Whitehouse, B. (2013). The Power is in the Street: The Context of State Failure in Mali. http://africaplus.wordpress.com/2013/04/19/the-power-is-in-the-street-the-context-ofstate-failure-in-mali/. World Bank. (1981). Accelerated Development in Sub-Saharan Africa: Agenda for Action [Elliott Berg Report]. Washington, DC: World Bank. World Bank. (1989). Sub-Saharan Africa: From Crisis to Sustainable Growth—A Long Term Perspective Study. Washington, DC: World Bank. Young, C. (2012). The Postcolonial State in Africa:  Fifty Years of Independence, 1960–2010. Madison: University of Wisconsin Press.

Chapter 13

E c on omics and t h e St u dy of C orrup tion i n A fri c a M. A. Thomas

Corruption is frequently raised as an important concern in discussions of Africa by both international and domestic stakeholders. It is a recurring theme in the media, in dialogue between foreign aid donors and recipient governments, and in African domestic politics. How has economics approached the study of corruption in Africa, and how does the study of corruption in Africa shape the economic study of corruption? The question is complex, turning on the definition of corruption, the boundaries between academic disciplines, and the utility of Africa as an analytic category for the study of corruption. The work of political economists focused on national political economy features—much of it focused on Africa and in line with an older literature in political science—challenges older economic models of corruption and even the assumptions that underlie the definition of corruption itself. The attention given to corruption in Africa may reflect the increased attention to corruption as a topic. There has been an explosion of academic literature on corruption in the last decade. In addition to more traditional work based on survey data, the availability of indicators of corruption and perceptions of corruption provided by entities such as the World Bank, Transparency International, and commercial political risk firms has led to thousands of studies seeking to establish the relationship between corruption and other variables of interest, such as foreign aid, foreign trade, inequality, conflict, and economic growth. New experimental work in corruption, including laboratory, field, and natural experiments, has explored the relationships between the structure of incentives and individual decision making in corrupt transactions. It is not always clear what economists have learned from this work on corruption. As Olken and Pande (2012: 3) put it in a survey of the economic literature on corruption in developing countries if we were asked by a politician seeking to make his or her country eligible for Millennium Challenge aid or the head of an anti-corruption agency what guidance the economic literature could give them about how to tackle a problem, we realized, beyond a few core economic principles, we had more questions to pose than concrete answers.

234   Concepts Lively methodological debates have been valuable in themselves, but also make it more difficult to identify where knowledge has advanced. It is even more difficult to identify what economists have learned about corruption in Africa. There is a long history of treating Africa as if it were sui generis, but is there such a thing as African corruption, deserving of special consideration and treatment? If Africa is not a special case, then the question of what economists have learned about corruption in Africa is indistinguishable from the question of what economists have learned about corruption generally. At the same time, more targeted studies of corruption raise questions of external validity, making it hard to say whether studies conducted outside of Africa might nevertheless be relevant to the study of corruption in some African context, or whether studies conducted in Africa can be generalized to other African contexts, or even the same place at a different time. If the study of corruption in Africa is to be distinguished from the study of corruption more generally the distinction must lie in interrelated features of political economy that, for historical reasons, are common in African countries (although neither unique to Africa nor universal in Africa). These include national poverty with its consequent low government revenue and capacity, governance by means of the distribution of private goods to elites, dependence on revenue from natural resources or foreign aid, and conflict. The study of the relationship between these factors and corruption—much of which has focused on African countries—implicitly challenges the assumptions of much of the corruption literature. Section 1 explains the common paradigms in economics and political economy used to model corruption, how corruption is commonly defined, and why the assumptions on which the definition relies can be limiting. Section 2 describes commonly encountered features of African political economy and their relation to corruption. Section 3 concludes by suggesting that economists can further contribute to the study of corruption by reexamining governance assumptions implicit in models, studies, and policy advice. A more contextualized approach means that Africa is not a useful analytic category for the study of corruption.

13.1  Corruption: Definition and Paradigms The most commonly used definition of corruption is “use of public office for private gain,” a simplified version of a definition first advanced by Nye (1967). This definition encompasses many possible behaviors such as sale of influence, nepotism, embezzlement, and theft, and many types of private gain including personal gain and gain for one’s family, friends, or political party. It does not depend on the legality of the action, and some types of corruption may be legal, as was congressional insider trading in the USA until 2012.1 Within economics and political economy the study of corruption has focused primarily on bribe exchange. The earliest and simplest microeconomic models of bribe transactions

1  In 2012, Congress passed the STOCK Act that prohibited this type of insider trading. See The Stop Trading on Congressional Knowledge Act (“STOCK” Act, Pub.L. 112–105, S. 2038, 126 Stat. 291, enacted April 4, 2012).

Economics and the Study of Corruption in Africa    235 assume that there is a probability that a corrupt act will be detected and punished, and that a rational actor maximizes his expected utility by weighing the bribe against the probability and severity of punishment (see, e.g., Klitgaard 1988; Cadot 1987). Principal-agent models of corruption posit a corrupt agent imperfectly monitored by an honest principal. In some versions of the model, the agent is a lower level bureaucrat and the principal is the supervisor. In other versions, the agent is a corrupt politician and the honest principal is the citizenry. Both the definition of corruption and the economic models that rely upon it rest on a number of assumptions that limit their utility. First, they focus exclusively on the behavior of those holding public office, ignoring the role of private parties. Many corrupt transactions include both a government actor and a private party. Although private parties may be victims of extortion, they may also be fully complicit or instigators of corrupt transactions. Where governments are weak, private actors may even be in the position to coerce government actors into participation in corrupt acts. The exclusive focus on the role of government actors also means, perversely, that corruption can be eliminated by definition by eliminating the role of government: by privatizing, outsourcing, or abandoning government functions. For the customer who must still pay a bribe for an electrical hookup, however, it matters little whether the bribe must be paid to an employee of a public or a private utility company. Alternate definitions of corruption have been proposed in an attempt to capture private sector activity. Second, the definition rests on implicit assumptions about the nature of governance. “Corruption” is defined as a deviant act in a broader political context in which there is a clear distinction between public and private spheres and there exist public offices whose powers and assets are entrusted to an office holder solely for use to benefit the public. This is not an empirical description of government, but a normative aspiration, and it is not universally shared. The distinction between public and private government office and resources would not apply, for example, in absolute monarchies. It can also be questioned in other contexts where the ordinary operation of the government does not seem to respect such distinctions. For example, President Bongo of Gabon did not appear to embrace such distinctions when he argued that discretionary funds allocated to presidents by the state could be used for public, political, or personal purposes (“to pay for the liberation of hostages under the table, to buy out political opponents, to finance their political parties, to buy jewels, or to give the money to dogs”) (Bongo 2001: 277). He similarly pointed to his private accounts in Citibank and argued that they were used to serve public purposes such as redistribution to the Gabonese, support of universities, hosting of diplomatic summits, or the purchase of ambassadorial residences (Bongo 2001: 291–292). When challenged by a journalist, Bongo asked whether the Palace of Versailles was built with King Louis XIV’s money or the money of France, implying that there was no distinction. His willingness to discuss this openly with a journalist suggests that he would face no consequences in Gabon for failure to observe the distinction. Finally, the word “corruption” is a normative and pejorative umbrella term. A synonym of “corruption” in English is “depravity” (Merriam-Webster). The normative load risks tangling social science efforts to study behaviors with the policy and advocacy efforts to reduce or eliminate them. At the same time, the breadth of the term obscures the differences between the causes, consequences, and frequencies of different behaviors such as bribe exchange (whether consensual or extorted), nepotism, or embezzlement.

236   Concepts The limitations of the definition are of special concern in the study of corruption in Africa. Africa is home to a number of weak governments, as well as governments that, like Bongo’s Gabon, may not fully subscribe to the implicit governance assumptions of the definition. This is true notwithstanding the fact that these distinctions are made in domestic law.

13.2  African Political Economy Factors With 55 recognized states that include low-income and high-income economies, with parliamentary democracies and military dictatorships that are coastal and landlocked, desert and rain forest, in conflict and in peace, and home to more than 2000 languages, there are few statements that could be true of all of the African continent. However, the twin processes of colonization and decolonization by which Africa was folded into the modern state system left broad imprints. Much of Africa was claimed by European colonial powers in the “scramble for Africa” during the late nineteenth and early twentieth centuries. The scramble was a competitive rush to establish exclusive spheres of influence in order to gain control of African resources, markets, and land. European colonial claims encompassed both areas of colonial settlement and areas that colonial powers had not even explored and over which their control was never more than nominal. They claimed both areas that were well resourced and populated and barren areas in which few people lived. Some dependencies were poor and never profitable enough to pay for their own administration. They created two-tier political systems that distinguished between colonial citizens, to whom they had some degree of accountability, and colonial subjects, to whom they did not. At the beginning of the colonial era, there was little consideration of the obligation to invest in the welfare of colonial subjects and no intention of moving towards independence for most colonies. By the 1940s, colonial powers began to feel a greater sense of responsibility to colonial subjects, in conjunction with changing ideas of the role of government at home. However, the same shifts in norms that catalyzed this sense of responsibility also made colonialism indefensible and untenable, and contributed to bringing the colonial era to a close. Most of Africa gained independence by the 1960s; the last African state to gain independence was Zimbabwe in 1980. Colonial administrative boundaries became international boundaries, spawning a number of new states, some of which were landlocked. At the same time, to facilitate decolonization, the international community waived the requirement of international law that states have effective governments, recognizing new African states regardless. As a consequence, some African states were created with governments that, like the colonial administrations they succeeded, had only nominal control over territories that had never previously been governed as states. For them, state-building was a post-independence task. For some, it has faltered. Civil conflict broke out in a number of African states. Military regimes replaced elected governments by coup, often citing government corruption as their motivation, until a wave of democratization began in the 1990s. While some countries saw successful peaceful transitions of power through democratic elections, for most democratic institutions remain fragile.

Economics and the Study of Corruption in Africa    237 The new African states also inherited the formal institutions and laws of their former colonizers. However, the rules themselves did not necessarily reflect local norms and African governments often lacked the revenue and capacity to implement and enforce them. Many African polities were characterized by “big man” or neopatrimonial politics, in which a leader maintained power by doling out positions and resources to political supporters. Democracy, when it arrived, was characterized by clientelism, in which politicians dole out private goods to a small base of supporters. Notwithstanding, the process of transference of laws and institutions from the West by imitation or under pressure continues. Most African countries are parties to the United Nations Convention Against Corruption, which requires parties to pass domestic laws criminalizing corruption, and most corrupt acts are criminal under domestic law.2 The consequence of this historical legacy was the creation of a number of African states that are poor, with governments with little revenue and capacity, governments that have weak mechanisms of popular accountability, governments that depend on redistributing private benefits informally to a limited elite, governments whose revenue derives from natural resources or foreign aid, governments whose control of their territories is challenged by conflict, and governments that have high levels of corruption. Because many African countries have more than one of these interrelated characteristics, it can be difficult to tease apart the impact of these factors, making cross-country studies susceptible to problems of endogeneity.

13.2.1  Poverty, low government revenue, and low capacity Although there are methodological debates about measuring corruption (see, e.g., Thomas 2010), and existing measures of corruption are heterogenous (Olken and Pande 2012) “[w]‌ith few exceptions, the most corrupt countries have low income levels” (Svensson 2005: 24). This is an important finding for the study of corruption in Africa because Africa is home to most of the world’s poorest countries. Studies have focused on the ways in which high corruption levels could impede economic growth and contribute to poverty; however, there are also channels by which poverty could influence corruption levels. Poor countries have poor governments. The lack of government resources and capacity undercuts the government’s ability to make, disseminate, implement, and enforce laws, including anticorruption and criminal laws. Poor governments also have poorly and irregularly paid civil servants, tacitly relying upon them to pay their own wages by monetizing their offices. The limited opportunities for employment and wealth accumulation in the private sector may lead people to seek government employment in search of financial rewards. The type of government envisioned by the definition of corruption—namely, a government that governs by means of the legitimacy gained from providing public goods and services impersonally to all citizens under the rule of law—is expensive, becoming available to a handful of wealthy industrialized countries only in the last century (Thomas 2015). Accordingly, there is a threshold effect. Although increased revenue does not guarantee a shift to this type of government, governments that lack the revenue to provide universal public goods and services must of necessity rely on older, cheaper strategies of governance, 2 

United Nations Office on Drugs and Crime. “UNCAC Signature and Ratification Status as of 27 September 2013.” http://www.unodc.org/unodc/en/treaties/CAC/signatories.html (accessed 10/22/2013).

238   Concepts such as patronage. The higher levels of corruption in poor countries are therefore a consequence of both low capacity and strategies of governance. That these practices are in conflict with domestic and international law further weakens the law as a mechanism of coordination for both government and society.

13.2.2 Governance Two related lines of argument point to the way in which governments hold power as critical to the understanding of corruption levels. The first focuses on the strength of mechanisms of popular accountability, which are thought to allow citizens to check government corruption; the second, on governments that govern through the distribution of private benefits to elites. Many African governments are thought to have weak popular accountability and to depend on distribution of private benefits to elites to govern.

13.2.2.1 Accountability One approach to the study of corruption is through the lens of principal-agent theory in which the government is cast as the agent tempted to corruption and the citizenry as the honest principal who imperfectly monitors and sanctions the agent. Drury, Krieckhaus, and Lusztig (2006), for example, find that corruption does not impede economic growth in democracies, as corruption annoys voters who can then remove politicians (see also Adserà, Boix, and Payne 2003). In Africa, however, democratic accountability is historically weak. Corruption is vocally condemned, yet there is evidence that the leading concerns of African citizens are poverty and unemployment, rather than corruption in government.3 Popular tolerance of corrupt practices varies substantially at the national and subnational level and depends on the exact behavior under discussion. The ability of the public to monitor and its willingness and ability to sanction are also likely to be highly contextual. One mediator is thought to be government transparency, although Kolstad and Wiig (2009) point out that transparency alone is insufficient to control corruption where citizens lack effective means for exerting control over politicians; on the contrary, in some circumstances greater transparency could lead to greater corruption Tests of the impact on corruption levels of improvements to transparency have yielded mixed results. The Public Expenditure Tracking Survey (PETS) was pioneered by the World Bank in Uganda in 1996 to trace funds disbursed from the Ministry of Finance through intermediate levels of government to the final spending unit, typically a health clinic or school (Reinikka and Svensson 2004). The original Ugandan PETS found that most schools received nothing of what was budgeted to them for nonwage expenditures. Anecdotal evidence suggested that the funds were diverted for use for political patronage. Reinikka and Svensson (2005) examined the impact of informing parents of the school budget to see whether parents empowered with 3  See Afrobarometer Round 5 (2010–2012) in which survey respondents were asked to name the most important problems facing the country that the government should address. www.afrobarometer.org (accessed 10/22/2013).

Economics and the Study of Corruption in Africa    239 information would then monitor and correct leakages of funds. They found that following the information campaign the flow of funds to schools was much improved. However, it is difficult to tease out the impact of this intervention from the larger context of donor scrutiny of the education sector and the related conditionality that followed the alarming result of the PETS (Hubbard 2007). Humphreys and Weinstein (2012:  36)  disseminated information about the performance of Ugandan members of parliament to voters, but found that these efforts had no impact on the reelection of MPs. “One preliminary conclusion from this experience is that the popular hypothesis that transparency alone leads to improvements in performance is overly optimistic” (Humphreys and Weinstein 2012:  36). They conclude that either a stronger dissemination campaign might have been necessary or that perhaps reelection depends on factors other than performance in office, such as personalistic ties or financing. Another mechanism of popular control that has been suggested because of its role in British history is the financial dependence of government on domestic taxes. Some argue that the payment of direct taxes such as income tax creates both popular demand for and the means of control of government (see, e.g., Moore 2004). However, African governments rely heavily on other financing sources, such as indirect taxes, taxes on trade, foreign aid, and revenue from natural resources.

13.2.2.2  Extractive institutions Principal-agent arguments that cast the public as the principal assume that the power to govern depends on the public, whether because a democratic government requires voter support or because it depends on the public for revenue. A second line of literature considers governments that are not as dependent on popular support to govern. It describes governments that are the product of a redistributive bargain among elites in which government authority, jobs, contracts, or monopolies are awarded not for the purpose of serving the public, as is assumed in the definition of corruption, but as private benefits that recipients are free to monetize. Political scientists and Africanists studying the newly independent African states named this type of governance “neopatrimonial,” building on the work of sociologist Max Weber (1947). Although the political science literature is outside the scope of this article, there is an extensive political science literature on neopatrimonial regimes in Africa and researchers interested in corruption should not hesitate at disciplinary boundaries (see, e.g., Roth 1968; Lemarchand 1972; Eisenstadt 1973; Le Vine 1980; Médard 1991; Bayart 1993; Bayart, Ellis, and Hibou 1999; Van de Walle 2001; Chabal and Daloz 1999; Herbst 2000). Economists have paid increasing attention to this type of governance and its implications for economic growth, distribution, technological innovation, and policy implementation. A canonical piece by Olson considered the incentives of governments that provide a modicum of security to citizens in return for the extraction of productive surplus for the benefit of the ruling “stationary bandit” (Olson 1993). North, Wallis, and Weingast (2009) described “limited access orders” as one of three types of social orders. Under limited access orders, elites provide social stability in return for privileged control over resources or activities from which they earn rents. They simultaneously restrict both the number of and access

240   Concepts to economic, political, religious, educational, and military organizations as well as entry into trade, because expanding the pool of elites diminishes rents. Violence is a regular form of political and economic competition and elites are aligned with military specialists. The uncertainty and distortion of market and price signals in limited access orders impedes economic growth as well as the collection of tax revenue. Acemoglu and Robinson (2012) focused more narrowly on political institutions as determinative of the type of economic institutions that can be created and adopted. Their framework defined “inclusive” and “extractive” political and economic institutions. Extractive political institutions are those in which political centralization may be lacking and power is held by a narrow elite who use their political power to maintain economic institutions designed to enrich them at the expense of non-elites. Extractive political institutions usually support and maintain extractive economic institutions. Elites resist centralization and act to suppress technological innovation and social change that might result in a loss of their privilege. As a consequence growth comes only from the reallocation of existing resources and so is self-limiting and not sustainable. Growth also invites elite fighting over the division of rents and is potentially destabilizing. North, Wallis, and Weingast (2009) described limited access orders as the default organization of human society. However, Acemoglu and Robinson (2012) pointed to the legacy of colonialism to explain variation in the nature of African political institutions. They argued that where geographic conditions made European settlement more attractive, Europeans settled and established more inclusive institutions; where settlement did not happen, they established extractive institutions (Acemoglu, Johnson, and Robinson 2001; Acemoglu and Robinson 2012). They argued that extractive institutions explain African poverty, conflict, and failed states, while the preservation or construction of inclusive institutions explains African successes. Although these governments are sometimes described as systemically corrupt, corruption is a deviation from an assumed norm of public office for public benefit. In neopatrimonial governments, the political systems of limited access orders, or governments with extractive institutions there is no “public” office whose purpose is to be used for the “public” benefit, or at the very least, this norm is weakly established. The private use of government office and authority is not a deviant and opportunistic act but the routine operation of government and the glue that holds the government together and makes it possible. Scholars have described many governments, including some African governments, as the product of such redistributive bargains.

13.2.3 Resource curse Another line of literature considers the relationship between corruption and government dependence on natural resource revenues. Africa is gifted with rich reserves of oil, natural gas, minerals, and revenues from natural resource extraction are an important source of government finance for a number of governments. However, studies have shown that in some countries resource extraction that has provided the government with revenue has failed to

Economics and the Study of Corruption in Africa    241 translate into higher standards of living and instead has been accompanied by lower growth rates, higher levels of corruption, and sometimes conflict (Van der Ploeg 2011). Literature on the “resource curse” examines both the circumstances under which resources are a curse and the nature of the curse. Perhaps unsurprisingly, the consensus is that the impact of resource revenues is mediated by the quality of governance (Morrison 2010). Governments with stronger rule of law and more accountable institutions are more likely to manage resource revenues in ways that benefit the public. For example, Botswana’s management of diamonds has been held up as an African success story (see, e.g., Robinson, Acemoglu, and Johnson 2003). In countries with weak rule of law where governance is characterized by redistribution to elites, increased government revenues lead to higher levels of corruption and in some cases, increased competition for government rents that can lead to conflict. Vicente (2010) considered the impact of oil wealth in São Tomé and Príncipe, finding that anticipated revenues increased corruption levels, reflected in increased vote buying, state jobs, and corruption in the education and health sectors. Arezki and Gylfason (2013) considered the interaction between resource rents and corruption for 29 sub-Saharan countries from 1985 to 2007, finding that higher resource rents increase corruption particularly in nondemocratic countries. Voors, Bulte, and Damania (2011) similarly found that in Africa positive income shocks increase corruption, especially in countries that already have high levels of corruption. The governments of low-income African countries remain heavily dependent on foreign aid. Another literature considers whether foreign aid has a similar impact, given that it also provides non-tax revenue to governments (see, e.g., Morrison 2010; Svensson 2000). Collier (2006) argued that aid is subject to donor oversight and conditionality, unlike natural resource revenues. Okada and Samreth (2012) similarly concluded that foreign aid lessens corruption, particularly in countries that have low levels of corruption to begin with.

13.2.4 Conflict Most of the world’s conflicts today are in Africa and Asia (Buhaug, Gates, Hegre, and Strand). Conflict can drive corruption. It can lead government to deprioritize corruption control in favor of security issues, reduce the resources available for law enforcement, or lead to the bypassing of processes and organizations intended to ensure financial control even as government makes urgent defense expenditures. A line of literature, however, examines the impact of corruption on conflict, particularly in the context of conflicts over natural resources. The literature is divided on whether higher resource rents lead to violent competition for access, or may be stabilizing as they give the government more patronage to disburse and offer the possibility of ending conflict by purchasing peace (see Le Billon 2003). Arezki and Gylfason (2013) found that in less democratic countries in sub-Saharan Africa, higher resource rents reduce conflict through redistribution to the public.

242   Concepts

13.2.5  High levels of corruption Whether because of poverty, governance strategies, dependence on resource revenues, or conflict, many African countries are considered to have high levels of corruption. High corruption levels shape the expectations, opportunities, and incentives of government actors, making high levels of corruption self-reinforcing and persistent. In other words, corruption can be both cause and consequence. In high corruption equilibria, the likelihood of punishment for corrupt acts is reduced and finding partners for corrupt transactions is easier. At the extreme, a refusal to participate in corrupt arrangements is unrewarded or even costly (Andvig 1991). Multiple equilibrium models have been used to explain how high levels of corruption can exist and persist, or to explore corruption dynamics (see, e.g., Andvig 1991; Lui 1986). Persson, Rothstein, and Teorell (2010) argued that persistent corruption is a collective action problem. Khan (2008) advanced a model of corruption in which people take their behavioral cues from others, positing a bandwagon effect in corruption. There is some evidence that citizens from countries with high levels of corruption may be more likely to participate in corrupt transactions even in countries with lower levels of corruption. Fisman and Miguel (2007), for example, use a change in the enforcement of parking laws for diplomats as a natural experiment in corruption. Until 2002, New York City did not have a mechanism for punishing diplomats with parking violations; starting in 2002, the city began to confiscate the diplomatic plates of offenders. The study finds that diplomats from countries perceived to be corrupt according to Transparency International’s Corruption Perception Index (CPI) were more likely to accumulate unpaid parking violations in the absence of enforcement. The authors hypothesized that diplomats carried national norms to their new environment. Barr and Serra (2010) attempted to replicate this result in a laboratory experiment in the United Kingdom with students from 34 countries. They found that while the CPI of the country of origin of the student was a statistically significant predictor of the propensity to engage in bribe exchange for undergraduate students, it was not for graduate students. They suggested that graduate students may have assimilated and lost their home country values, or that the selection process for undergraduate and graduate students may be different.

13.3 Conclusion The study of corruption in Africa exposes the fragility and specificity of the assumptions of governance that underlie both the definition of corruption and many economic models and studies. Where the government governs by sharing out government offices, authority, and resources as private benefits to powerful elites, corruption is much more than an individual act of bribe taking. It is a political system that is not described by distinct public and private spheres in which public offices, authority, and resources are to be used for the public benefit. These systems frame expectations and set the broader incentives for private use of government office. Both North, Wallis, and Weingast (2009) and Acemoglu and Robinson (2012) see such systems as persistent and difficult to change. If this political system is both persistent and the most common—the default system of human social organization—it makes the study of corruption parochial, and suggests that

Economics and the Study of Corruption in Africa    243 the research questions about corruption should be inverted. The question is not why corruption happens, but why it does not. Punishment for corrupt acts should be studied, not assumed. Studies of the efficiency costs of corruption usually assume a counterfactual of no corruption; perhaps they should assume a counterfactual of no government. Similarly, economists should revisit policy recommendations to ensure that they are robust to change in their underlying assumptions about both levels of corruption and punishment. For example, in the abstract, one-stop shopping for government permits and processes might be recommended on efficiency grounds, but this also creates a more valuable monopoly in government authority that can be exploited for rents. Some have suggested that creating multiple dispersed service providers would create competition that would result in rents being bid down to zero (see, e.g., Rose-Ackerman 1999), although this assumes perfect competition. If economists are to consider the broader incentive for corrupt acts in studies of corruption, they need more specific information about the incentive structures for a specific behavior in a specific place. Binary classifications such as “limited” versus “open” or “inclusive” versus “extractive” may not sufficiently capture either the complexity of national institutions or the variation among subnational ones. A classification system that puts most governments in the world in a single basket is of limited analytic utility; indeed, this is the complaint that has long been advanced about the term “neopatrimonial.” More work remains to be done to identify relevant incentive structures to ensure that macro-level studies of corruption do not suffer from endogeneity problems. Here it would be useful for economists to become more familiar with the political science literature on regime types that is both more extensive and more nuanced; and with the literature on patronage in anthropology, which considers patronage as a social system. Some economists do seek a more nuanced understanding of the local incentives of both government and private actors. An example of this kind of work is Sequeira and Djankov (2010), which studied the payment of bribes paid by shippers to clear the ports in Maputo, Mozambique, and Durban, South Africa. The study considered the amounts of bribes, but also the differing circumstances under which they were paid, the different incentives of custom officials on each country, and the different efficiency consequences for shippers between corruption that is “coercive” and corruption that is “collusive.” Another example is Delavallade (2012), which analyzes a 2004–2005 survey of North African firms to explore the relationship between tax evasion and bribe payments, concluding that when the risk of tax fraud detection is low, firms that hide sales pay more bribes; when it is higher, firms that hide sales are less likely to pay bribes. However, these results, like those of other targeted studies, cannot be generalized beyond their specific contexts. Just as the term “corruption” requires unpacking, there is sufficient variation to call into question the utility of regional consideration of corruption. “Africa” is not a useful analytic category for the study of corruption.

References Acemoglu, D., Johnson, S., and Robinson, J.A. (2001). The colonial origins of comparative development: an empirical investigation. The American Economic Review, 91(5):1369–1401.

244   Concepts Robinson, J.A., Acemoglu, D., and Johnson, S. (2003). “An African Success Story: Botswana”, in D. Rodrik (ed.), In Search of Prosperity: Analytic Narratives on Economic Growth. Princeton: Princeton University Pres, pp. 80–119. Acemoglu, D., and Robinson, J.A. (2012). Why Nations Fail: The Origins of Power, Prosperity, and Poverty. New York: Crown Publishers. Adserà, A., Boix, C., and Payne, M. (2003). Are you being served? Political accountability and quality of government. Journal of Law, Economics, & Organization, 19(2):445–490. Andvig, J.C. (1991). The economics of corruption: a survey. Studi Economici, 43:57–94. Arezki, R., and Gylfason, T. (2013). Resource rents, democracy, corruption and conflict; evidence from sub-Saharan Africa. Journal of African Economies, 22(4):552–569. Barr, A., and Serra, D. (2010). Corruption and culture: an experimental analysis. Journal of Public Economics, 94(11–12):862–869. Bayart, J.-F. (1993). The State in Africa: The Politics of the Belly. London: Longman. Bayart, J.-F., Ellis, S., and Hibou, B. (1999). The Criminalization of the State in Africa. Oxford, Bloomington & Indianapolis: James Currey & Indiana University Press. Bongo, O. (2001). Blanc Comme Negre. Paris: Grasset et Fasquelle. Buhaug, H., Gates, S., Hegre, H., and Strand, H. Global Trends in Armed Conflict. Oslo: Centre for the Study of Civil War, International Peace Research Institute, Oslo. http://www.regjeringen.no/en/dep/ud/kampanjer/refleks/innspill/engasjement/prio.html?id=492941 (accessed 10/22/2013). Cadot, O. (1987). Corruption as a gamble. Journal of Public Economics, 33(2):223–244. Chabal, P., and Daloz, J.-P. (1999). Africa Works:  Disorder as Political Instrument (African Issues). Bloomington: Indiana University Press. Collier, P. (2006). Is aid oil? An analysis of whether Africa can absorb more aid. World Development, 34(9):1482–1497. Delavallade, C. (2012). What drives corruption? Evidence from North African firms. Journal of African Economies, 21(4):499–547. Drury, A.C., Krieckhaus J., and Lusztig, M. (2006). Corruption, democracy, and economic growth. International Political Science Review, 27(2):121–136. Eisenstadt, S.N. (1973). Traditional Patrimonialism and Modern Neopatrimonialism. London: Sage Publications. Fisman, R., and Miguel, E. (2007). Corruption, norms, and legal enforcement: evidence from diplomatic parking tickets. Journal of Political Economy, 115(6):1020–1048. Herbst, J. (2000). States and Power in Africa: Comparative Lessons in Authority and Control. Princeton: Princeton University Press. Hubbard, P. (2007). Putting the Power of Transparency in Context:  Information’s Role in Reducing Corrupt ion in Uganda’s Education Sector. Center for Global Development Working Paper No. 136. http://www.cgdev.org/publication/putting-power-transparency-co ntext-informations-role-reducing-corruption-ugandas (accessed 10/8/2013). Humphreys, M., and Weinstein, J. (2012). Policing politicians: citizen empowerment and political accountability in Uganda: preliminary analysis. Typescript. Khan, F. (2008). Understanding the spread of systemic corruption in the third world. American Review of Political Economy, 6(2):16–39. Klitgaard, R. (1988). Controlling corruption. Berkeley, CA: University of California. Kolstad, I., and Wiig, A. (2009). Is transparency the key to reducing corruption in resource-rich countries? World Development, 37(3):521–532.

Economics and the Study of Corruption in Africa    245 Le Vine, V.T. (1980). African patrimonial regimes in comparative perspective. Journal of Modern African Studies, 18(4):657–673. Lemarchand, R. (1972). Political clientelism and ethnicity in tropical Africa: competing solidarities in nation-building. American Political Science Review, 66(1):68–90. Le Billon, P. (2003). Buying peace or fueling war: the role of corruption in armed conflicts. Journal of International Development, 15:413–426. Lui, F. (1986). A  dynamic model of corruption deterrence. Journal of Public Economics, 31(2):215–236. Médard, J.F. (1991). L’état néo-patrimonial en Afrique noire, in J.F. Médard (ed.), États d’Afrique Noire. Formations, Mécanismes et Crise. Paris: Karthala, pp. 323-354. Merriam-Webster. Merriam-Webster Dictionary. http://www.merriam-webster.com/dictionary/corruption (accessed 10/22/2013). Moore, M. (2004). Revenues, state formation, and the quality of governance in developing countries. International Political Science Review, 25(3):297–319. Morrison, K. (2010). What can we learn about the ‘resource curse’ from foreign aid? The World Bank Research Observer, 27(1):52–73. North, D.C., Wallis, J.J., and Weingast, B.R. (2009). Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Cambridge:  Cambridge University Press. Nye, J.S. (1967). Corruption and political development. American Political Science Review, 61(2):417–427. Okada, K., and Samreth, S. (2012). The effect of foreign aid on corruption: a quantile regression approach. Economics Letters, 115(2):240–243. Olken, B., and Pande, R. (2012). “Corruption in Developing Countries.” Annual Review of Economics, 4:479–509. Olson, M. (1993). Dictatorship, democracy, and development. American Political Science Review, 87(3):567–576. Persson, A., Rothstein, B., and Teorell, J. (2010). The Failure of Anti-Corruption Policies:  A  Theoretical Mischaracterization of the Problem. QoG Working Paper Series 19. Reinikka, R., and Svensson, J. (2004). Local capture:  evidence from a central government transfer program in Uganda. Quarterly Journal of Economics, 119(2):679–705. Reinikka, R., and Svensson, J. (2005). Fighting corruption to improve schooling:  Evidence from a newspaper campaign in Uganda. Journal of the European Economic Association, 3(2–3):259–267. Rose-Acerkman, S. (1999). Corruption and Government: Causes, Consequences, and Reform. Cambridge: Cambridge University Press. Roth, G. (1968). Personal rulership, patrimonialism and empire building in the new states. World Politics, 20(2):194–206. Sequeira, S., and Djankov, S. (2010). An empirical study of corruption in ports. http://mpra. ub.uni-muenchen.de/21791/ (accessed 10/8/2013). Shaxson, N. (2007). Oil, corruption and the resource curse. International Affairs, 83(6):1123–1140. Svensson, J. (2000). Foreign aid and rent-seeking. Journal of International Economics, 51(2):437–61. Svensson, J. (2005). Eight questions about corruption. Journal of Economic Perspectives, 19(3):19–42.

246   Concepts Thomas, M.A. (2010). What do the worldwide governance indicators measures? European Journal of Development Research, 22:31–54. Thomas, M.A. (2015). Govern Like Us: U.S. Expectations of Poor Countries. New York: Columbia University Press. Van de Walle, N. (2001). African Economies and the Politics of Permanent Crisis, 1979–1999. Cambridge: Cambridge University Press. Van der Ploeg, F. (2011). Natural resources: curse or blessing? Journal of Economic Literature, 49(2):366–420. Vicente, P. (2010). Does oil corrupt? Evidence from a natural experiment in West Africa. Journal of Development Economics, 92:28–38. Voors, M.J., Bulte, E.H., and Damania, R. (2011). Income shocks and corruption in Africa: does a virtuous cycle exist? Journal Of African Economies, 20(3):395–418. Weber, Max. (1947). The Theory of Social and Economic Organization. New  York:  Oxford University Press.

Chapter 14

Thoughts on Devel op me nt  The African Experience

François Bourguignon

14.1 Introduction If development is to be judged by the ability of developing countries to catch up with developed countries and reduce poverty, the performances of the last few decades can be considered mixed. Some countries, particularly in Asia and notably China, have met with undeniable success. On the other hand, living standards in Latin America have remained at about the same level in relation to the world average, while the initial backwardness of a number of countries of sub-Saharan Africa has become more acute. Do these results mean that economics and development practice have been merely a series of attempts, some of which have actually led to successes—at times resounding ones—but others, perhaps the great majority, have had only little impact? Should one go so far as to think, as some do, that development economics is purely and simply a failure and that little has been learned from the national development experiences as they have been taking place in real time over the past 50 years? Can nothing better be done today than simply to come to the aid of the world’s poorest people, bringing them additional purchasing power, or guaranteeing the education and good health of their children, while abandoning the attempt to find an engine enabling the creation of productive jobs thanks to which individuals and families will be able to improve their living conditions and freely exercise their talents, to take up Sen’s fine analogy between development and freedom (1999), in an environment of economic growth? Most fortunately, we have not come to that. A “corpus of knowledge” about the mechanisms of development has in fact accumulated, and does indeed show an extraordinary variability in these mechanisms in space and time, in the constraints under which they work, and in the policies to be implemented to foster development. The object of this chapter is to try to evaluate this knowledge by revisiting the great debates of development economics in

248   Concepts the light of the theoretical and empirical elements at our disposal today and very much with reference to the case of sub-Saharan Africa. It is divided into three sections. The first outlines the way this knowledge has progressed over the course of time, as influenced by favorable and unfavorable economic circumstances. The second deals with the role that the international development community and particularly the developed countries can play in the development of the poorest countries. The last section discusses the remaining challenges of development in sub-Saharan Africa, where it seems that world poverty will increasingly be concentrated in the coming decades, and the way in which it is hoped these challenges can be met.

14.2  Brief History of a Half-century of Economic Development and of Development Economics Figure 14.1 summarizes the performances of the developing world over the last 50 years by showing the evolution of the gross domestic product (GDP) per capita expressed in the purchasing power of the American dollar in 2005 prices. With the exception of Africa, the picture seems, all in all, fairly favorable. The world economy has grown at an annual rate of about 2% since 1961, Latin America at about the same rate, the Indian sub-continent at 2.9% and East Asia, led by China, at the record rate of 5.7% Insofar as it is possible to compare living standards over such long periods of time, the standard of living of that part of the world has increased almost 20-fold over the past half-century. The evolution has not been nearly as favorable in sub-Saharan Africa. Since these countries have become independent, the average annual growth rate of GDP per capita has been only 0.9%, with a large part of this result in fact acquired over the past 10 or 15 years after a long recession in the 1980s and the first half of the 1990s. The assessment changes markedly when, instead of looking at the absolute level of living standards, the gap separating the developing countries from the developed countries is examined. Figure 14.2 thus shows the evolution of per capita GDP in the developing regions in comparison with the per capita GDP of the United States (the result would be the similar if Europe were taken as reference). This time it can be seen that the curves are sharply downward sloping for Africa and Latin America, slightly upward sloping for South Asia and strongly upward sloping for East Asia. If one considers that catching up or convergence with the richest countries is a major objective of economic development, then the assessment is less encouraging. Asia seems to have begun such a catching-up process whereas in Latin America and especially in Africa the average lag behind the developed countries has increased considerably. Examining the performances of the past 30 years as concerns poverty also leads to a cautious diagnosis. World poverty has diminished considerably and the United Nations millennium goal to reduce by half the 1990 poverty rate by the year 2015 should be reached on the world level. Figure 14.3 shows, however, that this result is due above all to Asian performances, linked to their large populations. As in the previous figures, Latin America

Thoughts on Development    249 12000

ppp 2005 US dollars

10000

8000

Latin America World

6000

East Asia

Middle-East

4000

South Asia Sub-Saharan Africa

2000

0

1960

1970

1980

Year

1990

2000

2010

Figure 14.1   GDP per capita in developing regions: 1960–2012 (2005 ppp dollars). 35 30

Percents

25

Latin America

Monde

20

Middle East

15 10

East Asia

Sub-Saharan Africa

South Asia

5 0

1960

1970

1980

1990

2000

2010

Year

Figure  14.2  GDP per capita in developing regions as a proportion of that of the USA:  1960–2012 (USA = 100, GDP in ppp 2005 US dollars). and Africa do not have nearly the dynamism of the Asian continent. The situation is less serious in Latin America, for poverty is limited there. It is more serious in Africa, where almost 50% of the population is today still living below the subsistence level of $1.25 per day at international 2005 purchasing power, the poverty threshold set by the United Nations. Furthermore, the decrease of this percentage in the course of the last decade, an encouraging

250   Concepts 90 80 70

Percents

60 50

Africa

40 30

South Asia

20

World

10

East Asia

0

Latin America 1980

1985

1990

1995

Year

2000

2005

2010

2015

Figure 14.3  Poverty in the world and developing regions: 1980–2010. Proportion of population below 1.25 ppp 2005 USD per person and per day. prospect, should not conceal the fact that the number of poor people is still increasing in that part of the world due to very rapid demographic growth. Four hundred million Africans were concerned in 2010. All in all, the balance sheet of world development over the past half-century is thus mixed. An examination of the series discussed earlier, however, shows a certain heterogeneity not only between regions but also between periods. In fact, the past 50 years fairly naturally divide into three rather distinct phases of world development, each of which, interestingly enough, corresponds to a different approach to the problematique of economic development.

14.2.1  1960–1982: from decolonization to the debt crisis— voluntarist development policies If one word must be used to describe the approach to development in the 1960s and 1970s, it is without any doubt the term “planning” that comes to mind. Developing economies, many of which had just attained independence, were then following the path of a good number of European countries that had just come out of the post-war reconstruction period. The other example was obviously that of the Soviet economy whose planned growth was, in certain domains, notably the military, spectacular. In most of the developing countries, voluntarist policies to increase agricultural productivity and to industrialize were implemented, most often shielded by sizeable protective tariffs under the generic name of “import-substitution.” At first these policies, which had already proven themselves in Latin America during the Great Depression and the Second World War, produced, on the average, satisfactory results. Back in the 1970s, people thus spoke of the Brazilian miracle in Latin America or the Ivory Coast miracle in Africa. On the other side of the world, South Korea and the other “tigers” had similar, if not more rapid, performances. On the other hand, and despite

Thoughts on Development    251 robust planning, the Indian subcontinent was stagnating. As for China, it was blocked by the Cultural Revolution after the disastrous results of the “Great Leap Forward.” The theoretical model behind these development strategies was simple. Development is seen as the process by which production equipment, infrastructure, human capital can be accumulated in key sectors capable of pulling along the rest of the economy and absorbing the labor force employed in traditional, weakly productive activities. Did this model run out of steam at the end of the 1970s, whereas it had worked rather well up to that point? Had it reached its limits? Was import-substitution turning out to be increasingly inefficient? Had the proliferation of state interventions of all kinds sapped the economy’s fundamental dynamism? It is difficult to say, since the end of this first phase of contemporary development history coincides with the beginning of a major world economic crisis, the first of the post-war era. This crisis itself unfolded in two stages. The first corresponds to the oil crisis of October 1973, and more generally to the soaring prices of several commodities, whether agricultural products or mineral ores, during the same period. The effect of this boom on the developing countries was substantial, all the more so because it came after several decades of decline compared with the prices of manufactured products imported from industrialized countries. Some leaders of the time even spoke of a “change in the world economic order” and actually considered this change as definitive and permanent. All in all, the developing countries were therefore not much affected by the first oil boom. In effect, a recession in developed countries, lower interest rates due to the recycling of oil incomes, and bright expectations in most developing countries led them to heavy borrowing so as to accelerate the accumulation process. This would lead to a second stage where this debt proved to be unsustainable.

14.2.2  The 1980s and 1990s: “structural adjustment” and the change of “paradigm” The second oil boom in 1979 produced a new recession in the industrialized countries. The fall in world demand that followed led to a strong drop in the prices of commodities other than oil on the international markets. The rise in interest rates imposed by the American authorities in 1982 to curb the inflation caused by the oil booms then triggered a major balance of payments crisis, first in Latin America, then in Africa and some Asian countries. The accrued debt of the Latin American countries at that time amounted to about 50% of their GDP, a figure that would not be at all alarming today. However, as a large part of this debt was short-term, international interest rates suddenly hovering around 20%, the simultaneous appreciation of the American dollar, the fall in export earnings and the impossibility of obtaining additional credit from international bankers who had become wary made it totally impossible to service this debt. Mexico was the first to find itself insolvent. During the summer of 1982, it declared a moratorium on its debt, unleashing panic in the world banking system, reducing its lending ability a bit more and spreading the crisis to a great many other developing countries. One year later, no fewer than 27 countries were restructuring their debt with their creditors, a good half of them in Latin America. Brazil, Argentina, Venezuela, though an oil-producing country, but also Ivory Coast, Egypt, the Philippines, and Pakistan

252   Concepts joined Mexico. The international banks themselves were threatened with failure due to the inability of their debtors to honor their debts. The crisis was now literally worldwide. How it was resolved will not concern us here. From the point of view of real development, it is more important to examine its consequences on economic growth, the reduction of poverty and the implementation of new development policies. In the short and mid-term, these consequences were disastrous. In fact, it was to take Latin America and sub-Saharan Africa 10–15 years to get back to a growth rate similar to that preceding the crisis. This led to the often-heard description of the 1980s as the “lost decade of development.” A balance of payments crisis is typically resolved through the intervention of the International Monetary Fund (IMF), the lender of last resort, but a lender “with conditions.” The essential condition is the restoration of macroeconomic balance of the countries in difficulty by an adjustment of their fiscal and monetary policies and a devaluation of their currency. The scale of the crisis of the 1980s was nonetheless such that a special intervention program involving the Fund and the World Bank was created under the label of “structural adjustment program,” combining standard macroeconomic adjustment measures and structural free market reforms designed to prioritize the initiatives of private economic agents and market mechanisms rather than state interventions. From planning accumulation and productivity growth, the approach to development shifted to policies aimed at incentivizing private agents to accumulate and improve productivity, while recognizing that economic growth was essentially endogenous. These programs thus amounted to a paradigmatic change in the way of conceiving and analyzing development, whose basic components made up what was later to be called the “Washington Consensus.” In the case of the Latin American and African economies, this policy reorientation was not completely unjustified. In a great many of them, it is true that disorganized state intervention, often guided solely by the interests of the elite or of populist governments had reduced their capacity for growth and their resilience in front of an unprecedented crisis. Did these neoliberal reforms have results? The prolonged slowing of growth in Latin America and the inability of a great many African countries to resume growth led several analysts to express doubt about it, all the more so because, at the same time, development accelerated in Asia. Was the paradigmatic change the actual cause of this very long delay in recovery and the major social costs resulting from it? A before/after comparison has led some to answer affirmatively. But it is important to realize that in the absence of a more or less trustworthy economic model of the effect of diverse economic policies, no counterfactual scenario is available. What would have happened without this adjustment and a simple return to initial policies once the rescue had been made thanks to loans from the IMF? Unless the entire debt had been cancelled, would the same difficulties and sufferings not have become apparent? On the other hand, one must be careful about generalizations regarding the effectiveness of this or that policy based on the example of a few countries. An argument often advanced in favor of trade liberalization in structural adjustment programs was the example of the Asian economies strongly open to the rest of the world and their stellar rate of growth. Is it not true that the spectacular growth of the four tigers in the 1970s and then the takeoff of China, Thailand, and Indonesia beginning in the mid-80s were due above all to the liberalization of their trade? Here too, correlation is not causation. The case of city-states like Hong Kong and Singapore is too particular for them to be taken as models. And as the World Bank’s report “The East Asian Miracle” (1993) shows, the opening of countries like Korea and Taiwan

Thoughts on Development    253 took place only progressively and with sometimes very intensive state interventions and a mixture of measures promoting exports and aid to national producers. In the same way, the rapid expansion of Chinese manufactured exports is due to a strictly state-controlled strategy of joint investments with foreign firms. We now realize that a policy that has had good results in a given country at a certain time will not necessarily be as successful in another and/or at another time. The Asian miracle cannot necessarily be transferred to Africa, and it is illusory to think that going back to the import-substitution strategies of the 1960s and 1970s in Brazil would bring back the growth of those years. In the same way, it has not been established, either by theory or empirical analysis, that the principles of the Washington Consensus are a sort of development panacea. In effect, from the viewpoint of the experience and analysis of development, the period of the years 1980–1990 closed with growing doubts as to the existence of a more or less universal recipe leading to development. Africa was recovering too slowly from the debt crisis of the early 1980s and now Asia, home of the best development “performers” also went into crisis in 1997!

14.2.3  The prosperity of the 2000s and the new focus of development economics Asia recovered fairly quickly from a financial crisis, called the Asian crisis, one of whose causes appears a posteriori to be linked to the financial liberalization advocated by the Washington Consensus. They were helped by an increasingly favorable global situation. The developed economies were on a path of rapid growth and, briefly affected by the crisis of its neighbors, the Chinese economy took off. The impact of the events of September 11, 2001, followed by the burst of the so-called “dot.com” bubble on stock markets were rapidly overcome, thanks in part to a particularly accommodating American monetary policy. From 2003 to 2007, the world economy on the whole experienced several years of exceptional growth. There was a new boom in commodities after an almost continual decline since the beginning of the 1980s. All the developing regions, from Latin America to Africa, the Mid-East and Asia, experienced accelerated growth. For the first time a process of convergence, or catch-up, began between the developed and the developing countries. Everyone knows what followed. One of the engines of this global runaway enthusiasm was, once again, very low interest rates. Combined with faulty regulation of the financial sector and of its innovations, they led to the 2008 subprime crisis in the developed countries and to the worst economic recession since the Great Depression. The remarkable fact for the developing world is the resilience it has shown in the course of these past years. Affected by the crisis, it nevertheless avoided recession, and growth rather quickly returned to nearly its initial rate. In fact, several questions arise in view of this ability of the developing world to maintain its growth difference with respect to the developed world during the crisis and to recover much faster than it. Does this imply a growing autonomy of the developing countries, progressively freeing themselves from the demand coming from the developed countries, particularly due to the development of South–South trade? Is this resistance to the crisis of the developed

254   Concepts countries a sign that developing countries have, each in its own way, learned from their past mistakes and are today applying development policies and strategies better adapted to their context and the international environment? Or does it mean that today’s favorable situation combines certain countries’ autonomous development preceding the crisis with the effects of commodity prices that are still very favorable on the international markets and thus of still very advantageous terms of trade for the developing countries that export these products? It will be seen below that this last factor is probably crucial for African economies, but the others should not be ignored either. Regarding the economic approach to development, the paradigm seems again to have evolved in two important directions in the course of the last decade. In the first place, more emphasis is placed than previously on certain notorious inadequacies of the markets, most particularly in the financial sector, justifying state intervention—for example, with respect to capital flows. Secondly, and in the strictly opposite direction, more attention has been given to the political economy of development, notably the role of institutions and governance. The reflection is very much academic—see for instance Acemoglu and Robinson (2012)—but it has growing implications for the management of public development aid in several countries, including some in Africa. Finally, the relative failure to identify the combination of macroeconomic policies likely to give rise to durable development in a variety of contexts has led to a much stronger emphasis on microeconomic behaviors and the conditions for effective direct interventions as regards poverty, education, and healthcare. In particular, evaluation of the impact of these interventions by randomized control trials is seen as important progress (Banerjee and Duflo 2011). Some will find this rapid survey of development economics rather disappointing. It would certainly be ill-advised not to admit that we presently have no recipe available enabling today’s least developed economies to take off and to quickly reduce their poverty. But it would also not be good to underestimate the knowledge accumulated about development. On the macroeconomic level and in a given economy and context, it is often possible to identify the specific short- and medium-term constraints that are blocking the process of development, the fields of investment and of public spending to prioritize and the large structural reforms to implement in the long term. On the microeconomic level, progress is being made in fine-tuning direct interventions that actually improve the lot of the most disadvantaged and reduce poverty, provided that the requirements for the success of these interventions are met on the macroeconomic level. The experiences of developing countries, history and economic reflection suggest, however, that this strictly economic knowledge can be used profitably only in certain institutional contexts and certain initial conditions. Some institutions are favorable and evolve along with the development process in a sort of virtuous circle. Other institutions and some initial conditions are unfavorable and, on the contrary, lock economies into vicious circles of poverty. For example, a ruling elite may have only a limited interest in promoting the development of society as a whole beyond the minimum that will guarantee social peace, especially if this could threaten its power. Such behavior is obvious at certain times and in some countries, while economic knowledge suggests that more rapid development that is truly “inclusive,” that is, affecting the whole of society, is possible. The essential problem is getting out of such a vicious circle of poverty. It is, all in all, political rather than economic and its solution is likely to come from outside the countries enclosed in these kinds of poverty traps.

Thoughts on Development    255 From this point of view, the external context in which the developing countries evolve has changed considerably in the course of the last 50 years, notably in favor of the current acceleration of the globalization of national economies. Can the developing countries find in globalization an engine of development and structural change? Furthermore, what support can they expect from outside in the framework of international cooperation?

14.3  The Role of the External Context on Development: Globalization and International Cooperation Development’s external economic environment seems to have changed considerably with the currently ongoing process of globalization. National economies now have access to world markets of almost unlimited size in proportion to that of a given country. But they are also in competition with a growing number of actors, both on international markets and on their own. The process of globalization, however, goes beyond the mere expansion of international trade. It also involves population flows through migration, capital flows—whether foreign direct investments, that is, setting up foreign firms on national territory, or flows that are purely financial—transfers of knowledge, or access to other cultures via the development of the media. Another dimension of the external context is the evolution of international relations. From a development standpoint, the evolution of relations between developed economies in the North and less developed economies in the South, and most particularly of international cooperation regarding development and flows of financial aid, is particularly important.

14.3.1  North–South relations and globalization World exports accounted for a bit less than 20% of the world’s GDP at the beginning of the 1990s; they were nearly 30% in 2012. They have more than tripled in volume, whereas the world GDP has only doubled. This rapid expansion is largely due to the dynamism of the countries of the South and a massive relocation of manufacturing production from the North to the South. The South represented 25% of world trade in 1990, a percentage that had almost doubled by 2012. It must be emphasized, however, that these figures cover disparate evolutions. In particular, the South’s trade dynamism is largely due to Asia and more specifically to China. The share of the world’s other regions has in fact varied little. Regarding other developing regions, their share of world trade depends very much on the relative price of commodities. They themselves have been the victims of Asian competition, which reduced the share of their manufactured exports. Figures 14.4 and 14.5 show that for oil, as well as for non-oil products, the very long-term trend seems presently to be ascending.1 Even if this is the case, however, the oscillations around this trend are strong and make 1  These figures confirm Singer and Prebish’s thesis on the on-going deterioration of the developing world’s terms of trade beginning with the inter-war years. According to Figure 14.4, for non-oil products

256   Concepts 4.5 4.0 3.5 3.0 2.5 2.0 1.5

1875

1900

1925

1950

1975

2000

Figure  14.4   Real price of oil:  1875–2010 (logarithmic scale). the economies whose exports depend on these prices particularly vulnerable to a drop in their terms of trade. Are there reforms available to the international community that would enable a significant improvement in the external context of developing countries, particularly those exporting commodities? The World Trade Organization’s Doha negotiation cycle, known as the “development agenda,” was supposed to be working in this direction. Begun more than ten years ago, negotiations are at a standstill. The goal of this cycle was to favor the developing countries by obtaining enhanced access for their agricultural and industrial products on the markets of developed countries, in exchange for greater access to their own markets. Negotiations came up against the refusal of developed countries to improve access of developing countries to their agricultural markets, still protected by producer subsidies, and the resistance of emerging countries—particularly Brazil, China, India, South Korea, and South Africa—to agree to substantially reduce protection of their manufacturing sector. Another area where international cooperation could have beneficial effects on development concerns South–North migrations of unskilled labor. These flows, which have been favorable to the migrants themselves and to their communities of origin, through remittances to members of their families who stayed back home, are limited by the will, or perhaps the ability, of the developed countries to take in these migrants. From a strictly economic point of view, the world community cannot fail to benefit from the transfer of labor from the turnaround may not have taken place until around 1995 and some time is still needed before it can be confirmed – see Erten and Ocampo (2012).

Thoughts on Development    257 5.4 5.2 5.0 4.8 4.6 4.4 4.2 4.0

1875

1900

1925

1950

1975

2000

Figure  14.5   Real price of non-oil commodities:  1865–2010 (logarithmic scale). countries of weak productivity to countries where it is strong. This transfer has obvious social costs, but there is little doubt that the current volume of migration is far below what would be optimal on a world level, notably in view of the demographic aging and slowdown of a great number of rich countries. There is not much doubt either that these countries’ selection bias favoring skilled migrants runs counter to the development of their countries of origin. As in the case of trade, regulation imposed unilaterally by rich countries compromises world development. The subject is sensitive but, here too, international cooperation could greatly improve things.2 Another dimension of globalization and another source of constraint on development resides in imperfect capital mobility. Direct North–South foreign investments are not really involved. They are relatively free and over the recent period put money into most of the developing regions, more or less in proportion to their economic weight. Furthermore, a rapid increase in South–South investments by firms of large emerging countries has been noted. The situation is more problematical for access to international credit. If, in normal times, the risk premiums that increase the cost of credit have been greatly reduced for emerging countries, they are high and in fact prohibitive for the poorest countries. This is an area where international cooperation should take the place of the globalization of financial markets, which marginalizes part of the developing world. Official development assistance in its various forms, one of them being long-term subsidized loans, has long been the principal response to this state of affairs. But this comes with its own problems, which are discussed later. 2 

See, for example, Pritchett (2006).

258   Concepts Quite a few other areas involved in globalization call for international cooperation to minimize the burden that developing countries may have to bear. Space considerations prevent their being discussed here. But one cannot avoid mentioning environmental questions, particularly global warming whose first victims might well be developing countries in the tropics, or also the problem of epidemiological risks or the matter of intellectual property in areas as important as health—the sensational case that opposed the international pharmaceutical industry and the government of South Africa on the subject of producing generic medications against AIDS comes to mind. This world undergoing globalization provides numerous opportunities for international coordination and cooperation that would help correct factors unfavorable to the development of the poorest countries. Global political economy, however, makes their implementation difficult, blocking action in several areas. Moreover, things are not always simple either in areas where such cooperation exists in a permanent way, as seen in the current debate on the effectiveness of development aid.

14.3.2  Does development aid really aid development? From the beginning of the process of decolonization, international cooperation has been considered a major instrument for a community concerned, for humanitarian and other more geopolitical reasons, not to leave a large part of the world’s population in absolute destitution—think of what the current poverty threshold of $1.25 per day per person represents—or even in relative destitution, that is, a few percentage points of the standard of living of developed countries. International cooperation has taken different forms over the course of time and continues to evolve. What lessons should be drawn from these experiences? In particular, what should be thought of the current controversy on the effectiveness of official development assistance? Fairly rapidly the possibility that development aid might have “perverse” effects making it not very effective was put forward to explain poor reactivity of the beneficiary countries. Was not external aid being substituted, at least in part, for national savings? Was not the entry of foreign currency contributing to the real appreciation of national currency with negative effects on the exporting capacity of the countries? Was not aid a way for developed countries to buy the cooperation of the leaders of beneficiary countries in the international geopolitical interplay, and thus without great effect on the populations? Compelled to be paid to sovereign governments rather than directly to poor populations or to those in charge of particular investment projects, did it not encourage corruption and, at any rate, a lack of transparency in public spending with respect to the populations? Of course, a possible answer to these risks of diversion of public development aid was for the donors to impose conditions on its disbursement, to control directly the implementation of the projects, programs, and reforms that it was supposedly financing. But such control is difficult when it must be exercised on a sovereign country asserting its autonomy in the matter of development policies and projects. Thus began a heated controversy on the effectiveness of aid with, on one hand, those who asked that it be increased to allow the beneficiary countries to get out of the poverty trap (for instance Sachs 2005) and on the other, those who wanted it abolished, considering that, all in all, it was more a cause than a remedy for the development lag (Easterly 2006; Deaton 2013).

Thoughts on Development    259 A veritable industry also developed around the econometric analysis of the potential effects of development aid on growth, but without leading to decisive results. There are several reasons for this. Causality works in both directions: the growth of a country can perhaps benefit from aid, but aid itself tends to increase when the country does not grow much and poverty touches a larger part of the population. The impact of aid logically depends on policies implemented by the country and also on the quality of its institutions and, in particular, on the amount of corruption. The effect of aid also depends on the motivation of the donors: not much good can be expected of it when it is given for basically geopolitical reasons and openly and publicly misappropriated by the beneficiary countries’ leaders as was the case during the cold war. Finally, the favorable effects of a portion of the aid on economic growth sometimes appear only long afterwards as, for example, aid financing the educational sector. Controlling for all these factors proves difficult. For aid as for other policies likely to accelerate development, the difficulty of econometric analysis comparing the growth and policy experiences of different countries is the diversity of the individual situations and the impossibility of summarizing them in a few observable variables. Case study itself is not without difficulty. There are examples of countries whose development accelerated while they were receiving a large amount of aid. There is even mention of donors’ “pets”: Ghana, Mozambique, Rwanda, and Vietnam, among others. Without a doubt, these countries used the aid put at their disposal effectively. But at the same time, might they not have been capable of accelerating their development without this aid, provided they could have borrowed on the international markets? Conversely, there are also countries whose governance was far less satisfactory, which received aid that was possibly considerable, but whose growth did not accelerate. Should it be concluded that aid there was ineffective? Not necessarily. Growth might have been still slower and poverty more pronounced without aid. It is possible that, even if part of the aid was misappropriated or used badly, another part attained the set objectives. But it is also possible that the aid benefited only the governing elite and even strengthened its position and finally harmed the country’s development. Mobutu’s Zaire, Abacha’s Nigeria, Sassou Nguesso’s Congo or Teodoro Obiang’s Equatorial Guinea before the take-off of the oil sector are obvious examples. In the last case, external aid represented up to 40% of the GDP at a time when foreign companies had not yet begun to extract the oil and gas that are the riches of this country. Without a very detailed analysis relying, in one way or another, on the appropriate data and the elaboration of counterfactual scenarios, it is difficult to evaluate the effects of development aid. Unfortunately, the data enabling one to follow the use made of the aid, if only approximately, by an attentive monitoring of public spending, are rarely available due to a very partial information being available on budget implementation.

14.3.3  Allocation of aid: trade-off between  effectiveness and needs The growing doubts regarding the effectiveness of aid and the theorization of the relations between donors and beneficiary countries have had an effect on the management of aid and its allocation between countries and sectors.

260   Concepts From the standpoint of economic theory, the situation is simple. The donors are faced with several potential beneficiaries that differ according to their needs—that is, their degree of poverty—and their effectiveness at actually transferring the aid to the most destitute—that is, the quality of their governance or the degree of corruption of their leaders. If the donors’ goal is to maximize the reduction of poverty in the world, they will thus allocate the aid they wish to distribute according to both the quality of the beneficiary countries’ governance and their poverty level. If the poorest countries are at the same time the worst governed, then they will receive less aid than countries that are better governed but less poor. Donors thus make a trade-off between effectiveness—reduce poverty in the world where this is the easiest—and need—reduce poverty where it is greatest. Furthermore, the less effective the aid is, the fewer resources the donors will be ready to disburse as development aid.3 The trade-off discussed would be modified if the donors could exercise control over the use made of the aid. In practice, these means are limited, however. On one hand, it is difficult to prove that part of the aid was misappropriated. On the other hand, it is difficult to sanction the leaders without simultaneously affecting the poor population, so that any threat of retaliation in case of established fraud has little credibility. For if the donor threatens to suspend the aid if it is proved that too great a part of it was diverted from its objectives, acting on the threat will deprive the population of the part of the aid that benefitted it. This is the so-called Samaritan’s dilemma. Rational leaders will anticipate that an altruistic donor will not carry out such a sanction, and will ignore the threat of sanction.4 Another way to mitigate the trade-off between effectiveness and needs is for the donors to limit their aid to sectors where its use can be more easily observed, thus reducing the incentive to misappropriation by the leaders of the beneficiary country. But of course such a strategy can only partially modify the tradeoff since limiting oneself to certain sectors reduces the economic effectiveness of the aid granted. Practice is in line with these simple principles of selectivity. Thus, the International Development Association, the arm of the World Bank responsible for managing aid funds given it by donor countries, allocates its resources among countries with poor incomes on the basis of a mathematical formula dependent on three terms: the quality of policies and institutions (that is, of governance), the per capita income, and the population of the beneficiary countries.5 Among these factors, however, the first plays a disproportionate role. Inasmuch as the bilateral aid authorities in developed countries allocate their aid in a similar way, this is how on one hand there are the “pets” of aid—those whose governance is judged satisfactory—and on the other, the “orphans.” As regards the use of aid, moreover, a distinct evolution has taken place in the course of the last 15 or 20 years in favor of sectors considered “social”—health, education and social protection. This is because the misappropriation of funds by the leaders of beneficiary countries is said to be more difficult in these sectors than in the construction of infrastructure where contracts with construction firms more easily hide the payment of large bribes. 3  For a recent formal discussion of these questions, see Bourguignon and Platteau (2013, 2014a) and the references mentioned there. 4  On the “Samaritan’s dilemma” and related issues on the aid donor-recipient relationship, see the brief survey in Bourguignon and Platteau (2014b). 5  For a presentation and a discussion of that formula, see International Development Association (2010).

Thoughts on Development    261 The aid received by a country sometimes includes a purely humanitarian component, granted in cases of natural disaster (e.g. the Haitian earthquake) or to countries coming out of civil or military conflicts, countries most often considered as “fragile.” In both cases, the aid is distributed in a different way and controlled much more directly by the donors. This direct control is perhaps the only way to improve the effectiveness of aid in countries where governance is deficient. It can take the form of a return to aid granted in the form of investment projects supervised by the donors, or entrusted to local or international NGOs that have demonstrated their management ability and their transparency. After all, this second solution resembles China’s practice of delivering part of its aid in the form of infrastructure built by Chinese firms—and often Chinese workers—and handed to the beneficiary countries ready to use, often in exchange for contracts supplying them with commodities. Short-circuiting the ruling elites in this way could well be counterproductive, however, in countries where the principal progress to achieve is precisely to strengthen governance and the ability of the governments to manage their economies effectively. Furthermore, handing them resources that could maintain and even increase the degree of corruption in the economy is also counterproductive. At that level too, there is a Samaritan’s dilemma. All in all, optimizing development aid consists in identifying the most satisfactory balance between effective reduction of poverty and an inevitable loss of transferred resources because of poor management or corruption. It is understandable that in these conditions its effectiveness at the level of aggregate growth is limited. International cooperation on development seems to concentrate on the matter of aid. But the analysis mentioned above shows that the constraints the developed world at times imposes on the developing world through the mechanisms of globalization cannot all be resolved merely by the financial flows of aid. Whether it is a matter of trade, migratory policies, the environment or the transfer of knowledge, the developed countries and, increasingly the emerging countries are limiting the growth potential of the poorest countries.

14.4 Sub-Saharan Africa, Major Development Stake in the Coming Decades Africa south of the Sahara offers a perfect illustration of a certain number of points discussed on the pages above concerning the weals and woes of development economics and policies, and also of development aid. As it is at the same time a key region from the standpoint of the struggle against world poverty, it represents a major stake for understanding the development process and our ability to influence it. First, Africa presents three singular characteristics that make it a key player in the evolution of world poverty in the future decades: (i) most African countries are at the bottom of the world scale of per capita income; (ii) their average growth over the last 35 or 40 years is weaker than in all the other regions of the world; (iii) demographic growth is more rapid there than anywhere else. Secondly, most of the countries are faced with a difficult context: an economic autonomy often limited because of a strong specialization in the production and export of natural

262   Concepts resources, a small size making them very dependent on the outside and an often deficient governance, marked in some cases by great political instability. The question that arises in these conditions is the interpretation that should be given to the acceleration of growth seen since 2000 and, for some countries, since the middle of the 1990s. Is this an effect of the noted improvement in the price of mineral or vegetal commodities, emphasized above? Or, as some think, of an acknowledged improvement of their governance and the development policies they are implementing? In one case as in the other, however, there is the question of the long-term growth engine of sub-Saharan African countries. Could it be the rents from commodities or must this engine be found elsewhere? If this is the case, how can such an engine be started up and can the countries of the region do it without the aid of the international community? Such are the questions that will be quickly examined below.

14.4.1  The national and international context of  African development Among the world’s 30 poorest countries, 25 are African and the annual average growth rate of their per capita GDP since their independence is only 0.6%, most of this growth in fact having taken place in the last 15 years. In comparison, the average annual growth rate over the same period was 1.5% in Latin America, 2.8% in South Asia and more than 5% in East Asia. And today, some African countries particularly affected by civil and military conflicts are poorer than they were at the time of independence. This is the case, for instance, of the Democratic Republic of Congo, Niger, Madagascar, or even Ivory Coast. As to demographic growth, it has averaged 2.7% per year since 1980, that is, about double that of the world population. At this rate, the African population will reach more than 2 billion by 2050, representing a bit less than 20% of the world’s population. If poverty continues to diminish at the rate seen since the acceleration of 1995, a perhaps optimistic hypothesis, the number of poor in Africa would continue to progress and will then represent a large share of world poverty. Regarding the context in which African economic development is taking place, the following points should be emphasized. In the first place, the comparative advantage of the region’s countries clearly resides in their mineral or vegetal natural resources. Exports of manufactured products represent less than 5% of the total exports south of the Sahara. Some countries strongly specialize in one or two commodities whereas others are more diversified. For example, a country like Kenya exports tea, its product from the colonial era, but also horticultural, oil, and fishery products. The more diversified countries are obviously less sensitive to large price fluctuations in commodities on world markets. Nevertheless, most African countries have seen their terms of trade (price of exports/price of imports) evolve parallel (Figure 14.6) to a nearly continuous deterioration until the end of the 1990s and a vigorous appreciation since the beginning of the twenty-first century, hardly interrupted by the crisis of 2009. Unsurprisingly, this evolution reflects that of the real price of commodities seen above. A second contextual element that should be mentioned is the increased importance of trade with China, the rising power of the emerging world. Practically nonexistent in 1980, trade with China now represents about 20% of African foreign trade, chiefly

Thoughts on Development    263 350 300 250 Nigeria 200

Ghana Cote d'Ivoire RDC Ethiopia Uganda

150 100 Arithmetic mean 50 0 1980

1985

1990

1995

Year

2000

2005

2010

2015

Figure  14.6   Terms of trade of African countries:  1980–2012 (2000 = 100). with commodities as exports and manufactured products as imports. The rise is actually fairly recent. In 2000 foreign trade with China still represented only a bit more than 5% of African trade. A third important contextual element is the weakness of agricultural productivity in sub-Saharan Africa. This is a serious constraint on its development, but at the same time a considerable opportunity for progress. Today, Africa still has a greater area of arable land per capita than the totality of the developing regions except Latin America. On the other hand, the yield per acre is the lowest in the world, only about one-third of the world average. There are, of course, reasons for this state of affairs, notably the difficulty of irrigating the available land, but it is also remarkable that agricultural productivity has grown much more slowly in Africa than elsewhere in the developing world. The gaps are much greater today than they were 40 years ago. This suggests that there is indeed a potential that is untapped, in particular due to lack of necessary investments in water management and transportation infrastructure. Moreover, the link between weak agricultural productivity and poverty is obvious in a region where a large portion of poverty is essentially rural. The last contextual element is crucial. It is the quality of governance and institutions in sub-Saharan Africa. Several indexes for measuring this quality are available, focusing on one or another aspect of governance, among them the CPIA index6 used by the World Bank in allocating its aid funds among countries. Established on the basis of the subjective assessment of observers and analysts, all these indexes show a certain diversity. Researchers have compiled a great number of sources and worked out synthetic indicators7 that summarize these assessments according to five dimensions:  accountability, political stability, 6 

7 

Country Performance and Institutional Assessment. Kaufman, Kraay, and Mastruzzi (2009).

264   Concepts effectiveness of government, quality of regulation, rule of law, and control of corruption. In this database, almost half of the sub-Saharan African countries appear in the bottom quartile of the world hierarchy. Only a quarter of them should be ranked at that level if they had a level of governance comparable to that of other regions. For many countries, the reason can be found in civil or military conflicts that have seriously damaged the existing institutions, but these conflicts themselves often result from the weakness of the institutions.

14.4.2  The acceleration of African growth The mixed results of African growth over the very long-term hide, as has been seen, an underlying asymmetry. Stagnation and even recession for a great number of countries after a promising start at their independence, then a resurgence of growth at the turn of the twenty-first century and for some a bit earlier. This resurgence inclines some to optimism and there are quite a few reports and studies that see in it the manifestation of a veritable take-off for the region, based above all on a notable improvement of institutions and the implementation of effective development policies.8 For others, this rebound, certainly welcome from the standpoint of the welfare of African populations, reflects above all the end of a long period of deterioration of terms of trade and of difficult macroeconomic adjustments, facilitated several years ago by the cancellation of a part of the debts of African countries to private creditors or international financial organisms. This uncertainty regarding the causes of the acceleration of African growth raises the question of its “sustainability” and, more generally, of strategies to be implemented for a truly durable development in this region of the world. It is undeniable that economic policy has improved in sub-Saharan Africa, whether it is a matter of monetary policy, judged by the control of inflation, or public spending policy, judged by fiscal balance. In this regard, the way in which many countries of the region were able to come through the 2009 crisis without a drastic slowdown and rebound almost instantly is most remarkable. Furthermore, it is also obvious that, from a more structural point of view, efforts are being made in a great many countries to improve the “investment climate,” that is, to offer firms access to the infrastructure required for production, a skilled workforce, and an administration that is more transparent than it has been. Partly, the results shown by African economies in the course of the past decade are in fact due to improved policies and, perhaps, but it will take time to verify this, to improved institutions. But for all that, this does not mean that the rise in the terms of trade, which concerns extremely diverse countries, as can be seen in Figure 14.6, as well as the announcement of the discovery of new oil and gas fields, notably in East Africa, are not playing a major role in the recent acceleration of African growth. But they seem to be doing it chiefly by the demand they are generating from those who benefit directly from this rise in income, whether governments or large agricultural producers, rather than by the appearance of new autonomous production lines. Consequently, if the terms of trade were to deteriorate again, economic activity may well slow down in consequence. 8 

See McKinsey (2010).

Thoughts on Development    265 In support of this argument, it should be noted that the sectors producing non-tradable goods, that is services, construction, transportation and communication, and energy distribution, are those that benefit the most from the expansion of the GDP. These sectors basically produce goods and services in response to national demand. On the other hand, the GDP share of tradable goods excluding commodities—that is, agriculture and manufacturing industry—is stagnating or declining on average for the region and, in fact, in most, but not all of the countries (Figure 14.7). The African growth of recent years thus seems to be more “pulled” by the demand for goods coming from the income surplus due to the rise in terms of trade rather than “pushed” by autonomous expansion of the production system. If this is indeed the case, then there is in fact a risk that all the economies are fated to slow if the commodities cycle should reverse. It is also possible that spending by the governments on the basis of royalties and taxes levied on the export of commodities could improve the environment of the national productive sectors in terms of providing electricity and transportation and communication infrastructure and are actually preparing the way for a diversification of the economy in the production of tradable goods. There has also been a distinct acceleration in investment that may be going in the same direction, but that may also be intended for another type of activity. It is difficult to say what the situation actually is. After some ten years of improvement of the terms of trade, there is hardly a sign of diversification in production activity. There is no sign either of many African countries accumulating assets abroad that would, at a later stage, provide some revenue and therefore some autonomy with respect to commodity export and prices. Oil producing countries like Angola or Nigeria are indeed involved in such a strategy. Other countries are running a current account deficit and in several cases the Gross National Income share of their external debt is increasing,

35.000 30.000

Agriculture

Percents

25.000

Services

20.000

Extractive industry Trade

15.000 Manufacturing 10.000

Transport and communication

5.000 0.000 1970

Construction

1975

1980

1985

1990

1995

2000

2005

2010

2015

Year

Figure 14.7   Sectoral structure of GDP in sub-Saharan Africa: 1970–2012 (GDP shares at constant 2005 prices; arithmetic mean of national shares).

266   Concepts understandably from the very low level reached at the completion of the Heavily Indebted Poor Country debt relief program. A fortunate consequence of growth acceleration is the more rapid decline of poverty in the region. However, discrepancies can be noted in some countries. In Nigeria, for example, the per capita GDP has practically doubled, at inflation-adjusted prices, between 1996 and 2009. Nonetheless, poverty there has hardly changed. The problem may be that of the quality of household surveys that are the source of the data used to gauge poverty, but it may also mean that an important share of the increase in purchasing power resulting from the rise in oil prices has not, or has hardly benefitted households, particularly the poorest of them. Similar observations can be made in several other countries. In Mauritania, Kenya, or Zambia, renewed growth does not seem to have contributed to a meaningful increase in the average household income.9

14.4.3  Can natural resources be a development engine for Africa? The distrust of an African development strategy that would be centered on commodities may be exaggerated. On one hand, it is clear that a rise in the price of commodities and the discovery of new fields do not encourage national economic agents to do anything but respond to the increased demand for goods and services that they generate. On the other hand, why would development centered on exporting commodities not be possible? After all, it is the path taken by several oil-producing countries elsewhere in the world and they are not necessarily doing badly because of it. There are, however, several reasons for thinking that this scenario does not suit Africa very well. The chief reason is its demographic growth. Development on the basis of natural resources requires that the rents it generates increase in real terms at least as quickly as the population. Is this conceivable? On one hand, these resources are finite and it is important to anticipate what will happen when they near exhaustion. Of course, it is possible that new discoveries will continue to be made. After all, as Paul Collier has observed, Africa is the region of the world with the least known amount of natural resources per square mile.10 If this is because there has been less prospecting in the past, then the future in this regard is promising. If this is not the case, it must be assumed that real prices—that is, in purchasing power of goods imported from the rest of the world—will continue to rise. Here too, it seems that there are limits, if only in the area of fossil fuel, the inevitable brake that sooner or later will be applied to the emission of greenhouse gasses. But perhaps it could be imagined that the time horizon at which these events will take place is far enough in the future to be ignored at first, and that the development engine is only the demand of non-tradable goods resulting from the rents of natural resources. The problem then could be the growth of employment. Because this would be limited by the growth of real rents, from which the gains in productivity must also be subtracted. Without 9  Conclusions based on a comparison between per capita GDP in volume figures from the National Accounts and the percentage of poverty as described in the World Bank’s Povcal database. 10  Collier (2010).

Thoughts on Development    267 diversification of economic activity in the field of tradable goods, by import-substitution or the development of new exports, the growth of a rent economy is limited by the growth of these rents, regardless of their size in relation to the GDP.11 If the African population is to double in fifty years and if the real per capita income is to grow by 2% or 3% per year, then the rents from commodities must increase by 4% or 5%, that is, a multiplication by a factor of ten of the real price of commodities in 50 years. Such a rise is not likely, even over so long a period. As for the possibility of an equivalent flow of discoveries of new fields, that also seems very problematical. Investing part of the rents from natural resources abroad may be another way of accumulating and growing at a high rate. Such a strategy would be preferable to the diversification of the economy if the rate of return on this sovereign fund is higher than the potential return on physical investment in the production of tradables, including spillovers into the rest of the economy. It would also require institutions solid enough for the sovereign fund’s management to be fully transparent and with no risk of misappropriation by unscrupulous rulers. The long-term development of the African countries thus requires a diversification of their economies, including the sovereign fund strategy. In the mid-term, this diversification is all the more necessary because these countries are for now not much protected against a possible reversal, even temporary, of the current cycle of commodity prices. This diversification is also necessary to maintain the progress made in governance and to avoid the aspect of the “natural resource curse” that leads to opacity in public decision-making, corruption and open or hidden conflicts over the appropriation of rents.

14.4.4  Diversify the African production apparatus and increase the development of Africa Apart from the sovereign fund option, economic diversification can go in several directions. Increasing agricultural productivity is a first avenue. It concerns traditional agriculture rather than export agriculture and requires large investments in infrastructure, transportation and water management. In many countries, provided there is transportation infrastructure, the internal urban market often supplied today by imports offers important outlets to increased agricultural production, and the export of non-traditional agricultural products could be developed. National governments and foreign or international development agencies alike have too long neglected the potential represented by the agricultural sector.12 Things are more complicated in the manufacturing sector insofar as Asian competition makes it difficult to enter foreign markets and even to develop internal markets, often too small to allow the exploitation of economies of scale. There are several ways to remedy this state of affairs. Regional integration is one, at least if it takes the form of true customs unions rather than that of the many partial trade agreements reached until now, which often hinder rather than help the development of trade. A customs union within which trade is free and that offers the same protection to all its members over against a third country would 11  For an argumentation on the need for a structural change accompanying and leading African development and its link to employment, see Monga (2013) and Yifu Lin and Monga (2014). 12  See World Bank (2008).

268   Concepts get around the constraint of national markets that are too narrow and, possibly with the help of foreign firms, reach the critical size allowing the development of competitive industries. Regional integration in Africa is one of the main lines of the Economic Partnership Agreements currently being proposed by the European Union to several groups of African countries, although its protectionist aspect seems insufficient. The preceding avenue belongs to the practice of import-substitution, which has known limits. Africa, however, seems presently not to have reached these limits. Another path, which requires a certain amount of international cooperation, is that of export promotion. The model here is that of agreements reached between the large developed countries and the least advanced African countries, offering the latter unlimited access to the formers’ markets. The African Growth and Opportunity Act (AGOA) is the United States’ framework for all the products of the clothing and shoe industries. The Everything but Arms (EBA) agreement opens the European markets to the manufactured products of the less advanced African countries if they satisfy the “rules of origin” requiring that a large share of the value of the imported products be contributed by the country of origin, rather than imported from a third country. Neither of these two models has as yet had much success, the AGOA because it restricts too narrowly the range of products admitted to the United States without paying a tariff and the EBA because the rules of origin are too restrictive. Relaxing these agreements would not really cost the importing countries anything since the corresponding national production lines have already been outsourced. For African countries, on the other hand, this could form a strategic nucleus of industrialization. The example of Ethiopia that, thanks to the availability of cheap labor and quality leather, is perhaps developing, within the framework of the agreements above, a dynamic export sector in shoes, supported by Chinese firms and those of other African countries, should be considered. This model is undoubtedly applicable in other countries and to other product lines. The difficulty is obviously that it requires that “trade preferences” be granted by the advanced countries to African countries, preferences that would be contested by their non-African suppliers. Along with these avenues intended to diversify the African economies the more traditional role of international cooperation must not be forgotten. Even if their effectiveness needs improvement, the instruments of development aid can also participate in this diversification: by favoring infrastructure investments that facilitate internal and external trade, by contributing to the training of the workforce, and by setting up effective institutions. To close this rapid overview of the prospects for development in Africa, it is imperative to go back to that famous “natural resource curse.” This hypothesis is actually commonly used to explain the difficulties of a certain number of countries, in Africa as on the other continents. It is true that an economy’s dependence on a small number of products whose price fluctuates strongly on the international markets creates economic difficulties, and the resulting concentration of wealth is often responsible for deficient institutions and political instability. That being the case, many countries whose initial comparative advantage resided in a good endowment of natural resources have been able to develop in a satisfactory and diversified fashion. Indonesia and Malaysia are the examples that come to mind for Asia. In Latin America, the experiences of Chile over the past quarter century or of Brazil and Uruguay over longer periods are instructive. In Africa, the example of Mauritius, which was

Thoughts on Development    269 able to diversify its economy from its initial specialization in sugar exports and has been able to grow more rapidly that all the other countries of the region, should be considered.

14.5 Conclusion Several important points that the international community should bear in mind in the coming years and decades emerge from this all-too-brief overview of development and development economics and general remarks on African development. The first concerns the heterogeneity, not to say asymmetry of the developing world. On one hand there is Asia, and more precisely East Asia and China. It has made spectacular progress and has the ability to continue this progress. Things are a bit less clear for the Indian sub-continent, although the size of this region guarantees it a certain autonomy in development. On the other hand, sub-Saharan Africa has in recent years shown a renewed dynamism, but its lag has grown with regard to the other regions. The conditions of its development are probably less favorable as far as institutions are concerned and its international specialization makes it extraordinarily dependent on the price of commodities, and thus on their volatility. It is not sure that the traditional path of development, passing through industrialization, is open to this part of the world. If that is the case, a new development model must be invented for it. The second important point, also linked to these all in all rather average performances of the poor countries is urgency. Certainly poverty has diminished in the world when it is measured at an absolute level using as gauge $1.25 per person per day. But the diagnosis would be worse if a relative perspective were adopted, as in Europe, for example. By placing the poverty threshold at 50% of the median income of the world population, as is done in Europe, the proportion of poor people is still increasing in the world and most particularly in Africa. This is basically due to the fact that the asymmetry of performances in the developing world increases inequality there. The third point: there is no room nor is there reason for passivity and a wait-and-see policy on the matter of development. Reducing poverty demands ambitious interventions within the developing countries themselves, but also within the international community. Of course, there is no universal recipe that guarantees development and catching up with the rich countries. But we have in the course of time accumulated a capital of knowledge and experiences that can be implemented as long as the political, national and international will to actually reduce poverty exists. We have also learned that market mechanisms do not automatically function in favor of development. A certain amount of voluntarism is necessary, which in several respects concerns international cooperation. This is all the more true in that several aspects of world development economics have not been mentioned in this presentation focused on the countries’ development. I mean, of course, global public goods, and in particular the issues of climate change and water. It is to be hoped that the international community will quickly take the measures necessary to meet these challenges. In elaborating these measures, however, the question of development gaps should play a central role, as should the fact that, on the average, the developing countries are more vulnerable to climate change and water scarcity.

270   Concepts

References Acemoglu, D., and Robinson, J.A. (2012). Why Nationals Fail, the Origins of Power, Prosperity and Poverty. London: Profile Books. Banerjee, A., and Duflo, E. (2011). Poor Economics, A Radical Rethinking of the Way to Fight Global Poverty. New York: PublicAffairs. Bourguignon, F., and J. P. Platteau (2013). Optimal Discipline in Donor-Recipient Relationships Reframing the Aid Effectiveness Debate, Working Paper 34, G-Mond, Paris School of Economics. Bourguignon, F., and J.P. Platteau (2014a). Aid effectiveness revisited: the trade-off between needs and governance, Working Paper, G-Mond, Paris School of Economics. Bourguignon, F., and J.P. Platteau (2014b). The Hard Challenge of Aid Coordination, Forthcoming, World Development. Collier, P. (2010). The Plundered Planet. Oxford: Oxford University Press. Deaton, A. (2013). The Great Escape, Health, Wealth, and the Origins of Inequality. Princeton, NJ: Princeton University Press. Easterly, W. (2006). The White Man’s Burden. Oxford: Oxford University Press. Erten, B., and Ocampo, J.-A. (2012). Super-cycles of commodity prices since the mid-nineteenth century. Desa, Working Paper 110, Columbia University. International Development Association (2010). IDA’s Performance Based Allocation System: Review of the Current System and Key Issues for IDA16, IDA Resource Mobilization Department (CFPIR), Washington, DC. Kaufmann, D., Kraay, A., and Mastruzzi, M. (2009). Governance matters VIII: aggregate and individual governance indicators 1996–2008. Policy Research Working Paper Series 4978, World Bank. Lin Yifu, J., and Monga, C. (2014). The evolving paradigms of structural change, in Currie-Adler, Kanbur, Malon and Mendhora (eds), International Development: Ideas, Experience and Prospects. Oxford: Oxford University Press. McKinsey Global Institute (2010). Lions on the Move: The Progress and Potential of African Economies. The McKinsey Global Institute, Company. Monga, C. (2013). Winning the jackpot: jobs dividends in a multipolar world, in Stiglitz, Lin Yifu, and Patel (eds), The Industrial Policy Revolution II. Basongstoke: Palgrave MacMillan. Pritchett, L. (2006). Let Their People Come:  Breaking The Gridlock on International Labor Mobility. Washington, DC: Brookings Institution Press. Sachs, J.D. (2005). The End of Poverty. London: Penguin Books. Sen, A. (1999). Development as Freedom. Oxford: Oxford University Press. World Bank (2008). Agriculture for Development. World Development Report. Washington DC.

Chapter 15

T he Idea of E c onomi c Devel op me nt  Views from Africa

Hippolyte Fofack

15.1 Introduction Development may be defined as an endogenous, multifaceted, and continuous path-dependent process whereby a country’s aspirational goals are constantly refined by all the different actors (public and private) in a given society, taking into account the social fabric, the evolving stock of historical, cultural, and institutional knowledge, as well as scientific and technological infrastructure. Optimally, in a dynamic setting of constant feedback and interplay between the collectively defined aspirational goals and available instruments, the success on the development path in an open economy context will depend on the ability of a given society or country to timely harness local and global knowledge to move towards its aspirational goals. Taken as aspirational goals, the definition of development adopted in this chapter is much broader than the traditional single-dimensional approach which has emerged over the last few decades (Seers 1969; Harriss 2014). Reflecting the mainstreaming of rational behavior into existing models, that single dimensional approach has implicitly or explicitly focused on individual well-being and particularly on monetary variables. However, in addition to individual well-being, the collectively defined “aspirational goals” approach adopted in this chapter encompasses large groups and even the population universe in any given country and not just material goods, but other parameters such as human development and know­ ledge accumulation. In this regard, it is much closer to the definitions proposed by Sen (1988), Stiglitz (1998), Basu (2001), and Stiglitz et al. (2011). Viewed from this vantage standpoint, the most successful societies will emerge as the ones which are consistently approaching their aspirational goals or alternatively societies with low aspirational gaps; conversely, the least successful will be the ones that are either not approaching their goals or even worse, the ones without any goals to aspire to.1 Sadly 1 

Under the former scenario, countries may not be able to attend their collectively established aspirational goals either because available instruments have not been properly calibrated to support the

272   Concepts enough the majority of countries in sub-Saharan Africa have fallen under the latter category, as models underpinning the development process during most of the post-colonial period in the region have been extroverted—not endogenously developed, with the development thinking emanating not indigenously from within Africa but paradoxically, from the global industry of development experts largely based in donor countries. The extroverted development model has perversely suppressed national aspirations, and in the process, reduced the local development actors (policymakers, private sector operators, intellectuals, and civil society organizations) to bystanders of a process that they should have been leading.2 In addition to ignoring and crowding out local expertise and indigenous knowledge, the global industry of development experts has identified a monolithic and unified development goal for Africa as a whole, notwithstanding the continental diversity and heterogeneity of societies and economies. That goal has been seen as mainly increasing the national income of countries, measured in terms of changes in gross domestic product (GDP). In essence, national income growth has acted as a sufficient statistics invariably considered as the suitable indicator to capture the complexity and the whole comprehensive development process, even though measures of economic growth are not known to be particularly good correlates of other features of development. Indeed, studies have highlighted the negative correlation between economic growth and development, with growth directly associated with increased unemployment rates and income inequality (Lewis 1955).3 In addition to distributional issues, other problems associated with the aggregate view of income include the presence of externalities, non-marketability, and value-heterogeneity (Sen 1988). This reductionist approach of development reflects the golden era of economics as a discipline that trumps all the other fields of social sciences. Under that economic dominance, development is simply viewed as identical to economic development, which in turn is equiva­lent to economic growth. Still, another salient feature of development in post-colonial Africa is the implicit assumption that African countries should aspire to become like their former colonial powers, which have since mutated into the status of donors in the development industry. Having taken the present state of donor countries as a desirable objective, there was really no need for any country in the region to undertake the needed exercise of collectively defining its aspirational goals and development trajectory. This chapter reviews the possible implications and long-term costs of the prevalent extroverted development model that fails to take into account the particular local, historical or institutional contexts (“path dependency”) of sub-Saharan Africa (hereafter Africa). It proposes the contours of an endogenous development framework that is more reflexive and

collectively defined objective or because the aspirational goals are simply unrealistic. In either case, the failure to develop is a collective responsibility and not just the burden of the elites, rulers, or development partners. 2  After independence, a few leaders tried to establish reflexive and endogenous development models, which may still have been full of internal shortcomings, but their own ones nevertheless. These leaders faced uphill implementation challenges during the Cold War. 3  Sir Arthur Lewis highlighted the risk of confusing economic growth and development in Theory of Economic Growth in 1955 when he said “It is possible that output may be growing, and yet that the mass of the people may be becoming poorer” (Lewis 1955: 9).

The Idea of Economic Development    273 comprehensive and rooted in Africa’s institutional and traditional realities. Notwithstanding the heterogeneity of views, Africans4 are collectively aspiring to self-reliance. Economic transformation that is measured in terms of generalized balanced growth and structural changes is just one of the means to close aspirational gaps on the development path. Still, structural transformation advocated under the proposed framework also has the potential of mitigating the risk of reversibility on the development trajectory.5 The remainder of the paper is organized as follows. The next section provides an overview of the leading ideas of economic development that dominated development thoughts in post-independence Africa, with an emphasis on the dichotomy between state and market. Section 3 provides a contextual analysis of asymmetric forces that led the overwhelming majority of African countries to end up at the receiving end of the development thinking process. Section 4 provides an overview of what Africans would consider a fully reflexive and contextualized development model that is endogenous. Section 5 sketches out a simple analytical framework underpinning the emerging African development narrative. It is shown that the risks of reversibility on the development trajectory under such a framework are lower and development outcomes are hypothetically superior to the ones achieved under the prevailing model. The last section concludes.

15.2  History of Development in Africa: Overview of Big Ideas Although the history of development in post-colonial Africa has been singularly dominated by the lone goal of invariably increasing per capita GDP under “growth fundamentalism” models, the means, instruments, and leading ideas drawn upon to strive for sustained output expansion have been subject to changes over time (World Bank 2005a; Harriss 2014). In particular, over most of the post-independence era, the leading ideas driving the growth process in the region have alternated between advocacy for a stronger role for the state when the economic thought was dominated by structuralist theories and increasing reliance on markets when the neoliberal dogma became the dominant economic ideology (Stiglitz 2002; Currie-Adler et al. 2013; Stiglitz et al. 2013a). The tension between these two approaches has characterized much of the global policy debates in development. This twin characterization notwithstanding, the ideological landscape did not exactly follow a binomial distribution, with the prevailing ideology uniformly oscillating between either a model of full-state control or one that is exclusively the fact of free play of markets. Across the two opposite ends of the ideological spectrum there were many permutations. However, regardless of the weight assigned to each one of the two competing alternatives, the coexistence of state interventions and markets was often the norm. Even at the height 4  Hereafter, the word Africans is used in reference to both continental African residents but also to the growing African Diaspora predominantly in Europe and North America. 5  Risks of reversibility on the development trajectory are highly likely when development is reduced to an increase in national income. For one the commodity-driven growth often turned out to be merely a strong upswing in a boom-bust cycle in the region.

274   Concepts of the Cold War when a handful of African countries embraced the Soviet-style “dirigiste dogma,” few if any ever instituted a generalized collectivism and central planning system, with the state controlling all the means of production. Instead, markets continue to play a key role at strengthening incentives for productivity growth and aggregate output expansion. Nonetheless, despite the range of possible permutations on the ideological spectrum, the history of development thought in Africa can be divided into two broad periods. Taking a long-term view covering the five decades of independence, African governments played a more activist role in the immediate aftermath of independence up to the 1970s. During that period, the leading idea is structuralism and the majority of growth models focused on accelerating the rate of capital accumulation. In particular, capital accumulation is the cornerstone of Lewis’ model of unlimited supply of labor, of Nurkse’s balanced growth theory, and of the Harrod–Domar model, which posited a linear relationship between investments and economic growth. In practice, the establishment of state-owned enterprises (SOEs), a variety of price controls, and state interventions were the cornerstone of a policy mix that various governments used at one point or another to channel resources to sectors thought to be strategic and growth enhancing. Following the mantra of development economics of the time, there was a strong belief in the power of state-led industrialization. In the short term, addressing rural poverty and achieving food security were top among the development priorities of young nations emerging from colonialism. Through agricultural subsidies, governments supported policies that raised the productivity and agricultural output, for both food for local consumption and export crops, with the latter receiving more government support by virtue of its status as the main source of foreign reserves needed to import capital goods. In addition to expanding employment opportunities for rural households, the policies increasing agricultural productivity had the potential of stemming rural-to-urban migration. In the short term, most governments confronted with the perennial risk of adverse terms of trade (TOT) shocks and foreign reserve shortage understood that the opportunity cost of shifting labor out of agriculture was not necessarily equal to zero. Over time, the generalized push for the implementation of programs in support of rural development in the policy arena highlighted one of the main limitations of Lewis’ model of “economic development with unlimited supplies of labour” (Lewis 1954: 402), though that model was internally consistent with the primacy of capital underpinning most growth models developed after the Second World War. However, and unlike other regions of the world, the activist role played by African governments in the first decades of their respective independence, which may also be viewed as the era of “capital fundamentalism,” is also dictated by the imperatives of nation-building and the specific challenges facing the relatively young nations, also known as third world countries in reference to their chronic deficit of human capital and absence of basic physical infrastructure, including higher learning institutions which were conspicuously missing in most countries at independence (Gerschenkron 1962). State interventions tried to compensate for inadequate supplies of capital and physical infrastructures—critical drivers of competitiveness and productivity growth—with varying degrees of success in a context where industrialization was the path to economic development and capital was the binding constraint to closing technological and infrastructure gaps.

The Idea of Economic Development    275 Moreover, in countries where the colonial legacy of resource extraction and rents was the institutionalized policy, the scant infrastructure inherited was not dictated by the desire to promote development or by any logic of local and regional economic integration. Instead, the interior-tocoast transport infrastructure layout inherited after independence essentially served as conduit to extraction and transfer of natural resources from the “periphery” of the erstwhile colonies to the colonial empire to meet the growing demands for primary commodities and natural resources in support of industrial output expansion in Europe (Bonfatti and Poelhekke 2013; Fokam 2013; Fofack 2013). The interior-to-coast transport infrastructure layout is still the dominant pattern in most countries, and has been singled out as one of the leading causes of low intra-regional trade and greater economic integration with Europe (Bahadur et al. 2004; UNECA 2013).6 The second phase of African development history which started in the early 1980s and has been dominated by a decreasing role of the state and ascendency of liberalism is probably one of the most consequential in terms of development outcomes, transformation of the policy landscape, and institutional changes. Motivated by some of the failures of state-dominated approaches, the new paradigm saw markets as panaceas for African development. For example, among the few institutions which were affected by the unfolding neoliberal ideology, the monetary institutions discovered a new mandate, namely inflation target, and the reforms of labor markets gave birth to flexibility—implying empowering the clearing function of the market. In earnest, this era of free market dominance, which also marked the death of “capital fundamentalism” actually started in the late 1970s (Figure 15.1) and its power intensified over time, culminating with the demise of the Soviet Union in the late 1980s (Harvey 2005). Since then liberalism has, in effect, been the dominant economic ideology throughout Africa. More than in any other region of the world, it permeated the public policy arena in a profound way, resulting in a major institutional transformation with significant economic and social consequences. The reduction of the role of the state advocated under that ideology led to the adoption of reforms giving markets free rein and allowing foreign capital unprecedented access, via liberalization of capital account. Implementing austerity measures, getting prices right and achieving macroeconomic stability—meeting low inflation targets and reducing deficits to attain equilibrium in the domestic and external sector—became the mantra of the day, one that trumps the poverty reduction and unemployment objectives as it fails to take into account the inflation and growth trade-offs (Fofack and Ndikumana 2014). In Africa, the package put forward to advance the neoliberal agenda and also respond to macroeconomic imbalances and adverse TOT shocks which afflicted the continent included a range of policies: public sector downsizing, public expenditure cuts and switching, privatization of SOEs, deregulation, trade and financial liberalization, dismantling of state marketing boards, eliminating agricultural subsidies, tightening monetary policy, as well as unification and increased competitiveness of exchange rates (Williamson 1990; Adedeji 1999; Rodrik 2006).7 6 

Reflecting on the asymmetric market integration effect of the current transport infrastructure layout in the region, Bahadur et al. (2004: 182) maintain that: “Not only does Africa have extremely low per capita densities of rail and road transport, but indeed existing transport systems were largely designed under the colonial rule to transport natural resources from the interior to the nearest port. As a result cross-country transport connections within Africa tend to be extremely poor and are in urgent need of extension, to reduce intraregional transport costs and promote cross-border trade.” 7  Available estimates show that more than 50 percent of SOEs were divested in the 1990s (Nellis 2003).

276   Concepts Liberal Institutional Pluralism Focus on institutional diversity; Institutional conditions for successful growth; Institutions for effective public service delivery. Structural Approach

Washington Consensus

Focus on Market Failures: Import Substitution Strategy

Focus on Government Failures: Privatization, Marketization & Liberalization

Miserable results

Lost decades

1960s

1970s

1980s

1990s

New Structural Economics Focus on structural change; Industrial upgrading; Facilitating state providing soft and hard infrastructure and competitive markets.

2008

Figure  15.1   The evolving path of development theory. Source:  Author.

So deep and entrenched was the “market fundamentalism” ideology that it actually enriched the development economics’ lexicon: “Structural Adjustment Programmes” in reference to the package of policy reforms conditioning access to structural adjustment loans and “Washington Consensus” in reference to the fact that the Bretton Woods institutions (World Bank and IMF) which are both Washington-based became the seat of the neoliberal orthodoxy in the world of development policy. More than a development bank, the World Bank, in concert with the IMF, became the institutions enforcing the implementation of the neoliberal agenda in countries which entered a Bank or Fund-supported program as a result of balance of payment crisis. Compliance with conditionalities embedded in the Washington Consensus became a pre-condition for accessing foreign reserves for countries in need of external assistance. Yet, the Bretton Woods institutions played a less active role in the policy space in the immediate post-independence years when structuralism was the predominant ideology and the concept of development state was in full motion. As a development institution, the World Bank essentially provided the resources to expand public infrastructures to the newly independent nations of Africa, hence consolidating the rise of capital intensity under the “capital fundamentalism” model. For instance, the first World Bank loan ever extended to a country in Africa went to Ghana—the first independent nation in the region—to support the construction of the Akosomo Dam, which still powers Ghana to this day. The loan was approved by the Board of the World Bank in 1962 at a time when the concept of development state under President Nkrumah was in full swing.8 Over time, the Bretton Woods institutions expanded their mandate, in part as a result of the rise of the neoliberal ideology but also following a number of historical events which 8  After its emergence as the first independent nation of Africa in March 1957, Ghana became a member of the World Bank in September 1957 and the loan was approved on 8 February 1962 for Volta River Hydroelectric Project (Loan 0310).

The Idea of Economic Development    277 fundamentally changed the international financial architecture and development landscape. The most significant of these events included the unilateral decision taken by the Nixon administration in 1971 to cancel the direct convertibility of US$ to gold, a decision that essentially opened the era of freely floating currencies and balance of payments crises; the 1973 oil crisis and the dramatic interest rates hikes adopted by the US Federal Reserve Bank in response pushed many countries into default and resulted in international debt crisis. A new era of balance of payments crisis began. In Africa where most countries were natural-resource dependent, the costs of these two events were exacerbated by the deterioration of TOT and vulnerability to adverse shocks which became only more frequent over time. Despite the inflation of critics who contrast the promise of liberalism with results on the ground (see for instance Adedeji 1999; Stiglitz 2002; Fine et al. 2003), the costs of the neoliberal experiment for the region are not yet fully assessed. Still, in a number of areas the cost–benefit analysis points to a large negative balance sheet, though a number of authors are entertaining the view that the macroeconomic stability and growth enjoyed by countries in the region over the last few years could be attributed to decades of structural reforms advocated under the Washington Consensus (Devarajan and Shetty 2010). However, it has also been argued that debt relief extended to countries in the region under the Highly Indebted Poor Country Initiative (HIPC) expanded the fiscal space and reduced the risks of monetization of fiscal deficits which had been the main driver of inflation and macroeconomic instability (Fofack 2014). In other areas, where empirical evidence is clear cut, the costs far outweigh the benefits. According to Joseph Stiglitz, one of the leading critics of the Washington Consensus, the privatization which resulted in a gradual elimination of state monopolies instituted private monopolies as alternative in the areas of infrastructures without necessarily improving the quality and efficiency in the provision of public services (Stiglitz 2002; Fokam 2013). The costs associated with the rise of private monopolies are particularly obvious in the areas of energy production and distribution where delayed investments after the privatization of SOEs have resulted in a chronic deficit of power. Meanwhile, the short- and medium-term costs of the going neoliberal experiment are even broader and significant. In addition to worsening income inequality and poverty, the region achieved one of the most sluggish growth performances during that experiment. Despite the implementation of macroeconomic reforms, real GDP per capita fell below levels enjoyed by the region during the era of structuralism, with most countries recording negative growth in real per capita GDP (Adedeji 1999; Artadi and Sala-i-Martin 2003). In the words of Adedeji who headed the UN Economic Commission for Africa as UN Under-Secretary General at the height of the structural adjustment era, “The tragedy of the adjustment effort is that even for the narrow economic objective of growth in real per capita GDP, the record of the SAP over the two decades has been quite disappointing” (Adedeji 1999: 522) (see Table 15.1). More than three decades into the neoliberal experiment, Africa is the only region of the developing world that will miss the first Millennium Development Goal of halving poverty by 2015 (World Bank 2011). Its contribution to world trade has fallen below 1.5 percent, from more than 3.8 percent in the 1950s. Artadi and Sala-i-Martin (2003) have labeled the dismal growth performance achieved by the region during the neoliberal experiment as the economic tragedy of the twentieth century. Others have characterized the adjustment era as the

278   Concepts Table 15.1  Annual growth rates of GDP and per capita GDP by decade (in percentages) 1961–70 Average annual GDP growth Sub-Saharan Africa East Asia Average annual per capita growth Sub-Saharan Africa East Asia

1971–80

4.5 7.06

2.4 7.8

1.8 4.28

–0.4 5.47

1981–90 1.4 6.8 –2.6 4.8

1991–2000

2001–10

2.8 6.2

4.82 5.71

–0.4 4.44

2.26 4.48

Sources: UN Intellectual History Project (2009) and World Bank, World Development Indicators 2013.

lost decades (Mkandawire 2004). Remarkably, in Asia where the concept of development state remained in effect during the globalization of the neoliberal dogma growth was sustained and resulted in a significant improvement of living standards and reduction of poverty (Wade 1990; Chang 2002; Stiglitz et al. 2013b). Yet Africa’s growth prospects were seen as superior to those of overpopulated Asia in the 1960s—a view captured in Asian Drama (Myrdal 1972). The unexpected dismal growth performance achieved by the region is partly attributed to a dramatic decline in public investment and de-industrialization during the neoliberal experiment (Norman and Stiglitz 2012). Indeed, the transition from “capital fundamentalism” to “market fundamentalism” was particularly costly for the former. Aggregate investment fell from 15 percent of GDP in 1975 to 7.5 percent in the mid-1990s, significantly below the average of 30 percent enjoyed by East Asian emerging markets which surfed on high domestic savings over the same period (Artadi and Sala-i-Martin 2003; UNECA 2013). Paradoxically, the declining rate of investment during the neoliberal experiment was accompanied by massive capital flight from the region, with average capital flight rising from about US$21 billion in the 1970 decade to US$46 billion in the last decade (2000–2010) (Boyce and Ndikumana 2012; Fofack 2012). Overlooked by international financial institutions, capital flight was pernicious to long-term growth in Africa (Fofack and Ndikumana 2010). To the extent that capital flight may undermine domestic savings and investments, the positive correlation between capital account liberalization and capital flight suggests that the reforms advocated under the Washington Consensus may also have undermined the process of capital accumulation (Hermes and Lensink 2014). Even the most ardent ideologues of liberalism now concede that the push to impose the values of “market fundamentalism” on developing countries failed to meet expectations on most development counts. In a retrospective review of lessons learned from that experiment, the World Bank had the following conclusion “Despite good policy reforms, debt relief, continued high levels of official assistance, promising development on governance and a relatively supportive external climate, no take-off has ensued” (World Bank 2005a: 8).9 An IMF review of the adjustment era also had a concordant assessment, concluding that “Growth has 9  Commenting on that report, Rodrik (2006: 974) had the following words: “Not only were success stories in Africa few and far between, but the market-oriented reforms of the 1990s proved ill-suited to deal with the growing public health emergency in which the continent became embroiled.”

The Idea of Economic Development    279 been disappointing.” Summing up that experiment, Anne Krueger, one of the leading architects of adjustment, had the following words in 2004: “Meant well, tried little, failed much” (Krueger 2004: 1). In practice this impotence has led to the shift away from structural adjustment programs and the introduction of Poverty Reduction Strategy Papers and the birth of new lending instruments—the Poverty Reduction and Support Credit in the World Bank and the Poverty Reduction and Growth Facility in the Fund. In Africa where failed state interventions (as a result of misallocation of limited resources to ill-conceived government projects or poor governance) during the structuralism era exacerbated fiscal deficits and undermined macroeconomic stability, the rise of liberalism informed by the efficient market hypothesis was promoted as a viable growth and development alternative. However, in addition to its failure to meet expectations on growth, the extreme version of “market fundamentalism” advocated under the Washington Consensus failed to recognize that liberalism was not immune from market failures either, though instances of market failures producing outcomes that are not Pareto optimal abound. At the same time, there is a large body of work that attributes the rise of Asian emerging markets to a successful application of the “development state” model (Amsden 1989; Wade 1990; Chang 2002; Stiglitz et al. 2013a). The successful rise of “cohesive capitalist states” in Asia suggests that structuralism theories can indeed work when institutions limiting the risks of government failures are operational and effective (Kohli 2004; Lin and Monga 2011). The evidence of successful state-led development models in some parts of the world and the failure of the “market fundamentalism” ideology, made even more obvious by the 2008 Great Recession which highlighted the limits of self-regulated markets, brought renewed attention to the role of the state in economics.10 Increasingly, the emphasis on free markets as a viable alternative to state interventions is being viewed as a false dichotomy (Stiglitz et al. 2013b).11 The state has a role to play and may even reduce the adverse effects of negative externalities, especially in young nations where weak institutions of checks and balances could exacerbate the costs of market failures. Departing from decades of economic orthodoxy, a search process is ongoing to produce an evidence-based and useful new development framework that is more organic and takes into account the institutional and structural realities of countries at every stage of the development process. Although still in the burgeoning stage, two new development paradigms are leading the thought process: the “liberal institutional pluralism” and the “new structural economics.” The former which emphasizes institution—the rules and norms constraining human behavior—is advocated by North (1990), Acemoglu et al. (2005), Rodrik (2008), and Brett (2009). At the macro-level, the new synthesis advocated by these authors focuses on institutional conditions for economic growth and political transformation. This synthesis assumes that the economic performance of societies depends on the effectiveness of their underlying institutions at containing the predation by individuals and at resolving agency problems. This institutional approach to development received a strong boost when Easterly 10  The crisis also casted some doubt in the foundation of the neoliberal orthodoxy and led the IMF to reconsider its development framework (see IMF Staff Position Note on ‘Rethinking Macroeconomic Policy’) (Blanchard et al. 2010). 11  This view is not completely strange to the World Bank. Its 1997 World Development Report on “The State in a Changing World” concluded: “Development without an effective state is impossible … an effective state—not a minimal one” (World Bank 1997: 18).

280   Concepts and Levine (2003) showed that policies advocated under the Washington Consensus did not exert any independent effect on long-term economic performance, once the quality of domestic institutions is accounted for. In practice, Acemoglu et al. (2005) found that “Economic institutions encouraging growth emerge when political institutions allocate power to groups with interest in broad-based property rights enforcement, when they create effective constraints on power-holders, and when there are relatively few rents to be captured by the power-holders” (Acemoglu et al. 2005: 387). However, one emerging challenge associated with this line of research is the inability to establish a causal link between any particular institutional design feature and growth, reflecting the potential effects of extraneous factors on the efficacy of institutions. Indeed, as pointed out by Rodrik (2008), different institutions may have similar outcomes while the same institutions may give rise to different outcomes in different contexts. The difference between failure and success on the development path may therefore lie on singling out what makes institutions function effectively. The “New Structural Economics” put forth by leading proponents such as Chang (2002), Lin and Monga (2011), Stiglitz and Lin (2013) advocates a synthesis of structuralism and liberalism ideology, without necessarily striving for a unified theory of economic development. Their proposed framework is the application of a neoclassical economic approach to understand the drivers of economic structure and its evolution in development. It emphasizes the need to take into account structural features in the analysis of economic development and the role of the state as the agent driving the modernization of economic infrastructure for growth and integration of local and regional markets. The new approach recognizes the presence of both state failure and market failure. Although the market remains the basic mechanism for effectively allocating resources under this new framework, it is just one of the instruments in the development toolbox. More broadly, development is seen as a dynamic process that requires constant industrial and technological upgrading and corresponding improvement in hard and soft infrastructure. And the potentially large externalities to firms’ transaction costs and returns to capital investment associated with the need to constantly upgrade economic infrastructure calls for government intervention. The common thrust to the two emerging development frameworks is the departure from the one-size-fits all approach to take into account the specific institutional and structural features of each country at different stages of the development process. These approaches look at the realities in each country and adapt policy tools to local conditions. They could be appealing in Africa where poor governance and the deficit of infrastructure remain major constraints to economic development. Furthermore, the two approaches may also be highly complementary in the African context where widespread poverty is associated with massive capital flight. Strengthening the institutions of good governance could raise the prospects for domestic resource mobilization in support for infrastructure development.

15.3  Explaining Africa’s Position in  the Global Development Bargain Even reducing the concept of development to the single one-dimensional income vector, Africa consistently exhibited a poorer record compared to other regions of the developing world, both during the structuralist ascendancy and neoliberal hegemony. This

The Idea of Economic Development    281 time-invariant outcome suggests that other factors besides ideology have been at play. The economic literature has condensed these factors to a combination of a number of effects, including poor governance and institutions, macroeconomic instability, conflicts, and resource curse (Rodrik 2008; Sachs and Warner 2001; Easterly and Levine 1998). Regarding the latter, three channels of causation from natural resource abundance to poor growth outcomes are often put forward: rent seeking through the voracity effect, TOT shocks through the commodity prices volatility channel and Dutch Disease (Isham et al. 2005; Fofack 2010). Not much attention has been given to historical and philosophical determinants of development in the region, however. Yet, these additional drivers are extremely pertinent and may even trump economic factors in a zero-sum game globalization mindset where the gains of any given country from trade are increasingly viewed as resulting from losses of utility for other participants in the global trading game. Recently, the rise of these mercantilist policies has been reflected in the push for competitive devaluations and concerns about global macroeconomic imbalances. The case may even be stronger for historical and philosophical determinants, especially for young nations coming out of the colonial experience which subverted traditional institutions to serve the interest of imperial powers (OAU 2001; Akyeampong and Fofack 2014). For instance, to the extent that ideology may shape economic structure, with direct impact on patterns of trade, by focusing on TOT, one may be looking at the consequences of a phenomenon and overlooking the fundamental causes of underdevelopment. This session reviews the historical processes and asymmetric forces which have locked Africa in the short end of the global development bargain. A focus on ideological drivers suggests that the leading ideas shaping the development process in Africa over the last few decades have been largely determined by the dominant ideology at the global level (Adedeji 2002). In effect, until recently (rise of the Asian emerging markets) the development wisdom across Africa was the monopoly of industrialized countries in the north, even though the philosophy underpinning that wisdom did not necessarily reflect the social and historical contexts or Africans’ development aspirations. In part, this bias was the consequence of a mindset which assumes that the intellectual wisdom informing the development process could not possibly emanate from backward nations which were for the most part at the bottom of the development ladder. The asymmetric distribution of power and knowledge which confined Africa to the receiving end of the development wisdom affected development outcomes. The policy recommendations emanating from such models were not always grounded in Africa’s historical and contextual realities and often failed to lead to development takeoff. The push for privatization of SOEs in the region in a context of limited depth of capital and financial markets is one of the most glaring examples. Even though Africa provided the fertile ground for that experiment the reform dividend did not follow. In contrast, Asia, which was more refractory to it, produced more successful entrepreneurs and achieved economic diversification for a better integration into the world economy (Norman and Stiglitz 2012; Stiglitz et al. 2013b). However, the costs and implications of the much stronger bias of African countries for the globally dominant ideology have been much broader. Probably in no other area have these costs been more significant and consequential than the overemphasis on economic growth (i.e. rise in per capita income) under the neoliberal ideology and less on “growth fundamentals.” In particular, the exclusive monitoring of the one-dimensional income vector and concurrent benign neglect of “growth fundamentals” has perpetuated the colonial

282   Concepts production model of excessive reliance on primary commodities as the main connecting link to the global economy (Fofack 2013). As long as per capita income growth was secured over time, there was no need to be concerned with sources of growth and distributional issues. However, even under the best case scenario, the strategy did not lead to improvement of living standards, owing in part to the fact that the export of primary commodities entails huge forgone income through lack of value addition, the export of jobs to countries adding value, and exposure to high risks due to dependence on exhaustible commodities and fluctuations in commodity prices and demands. The consequences of this myopic approach to development and insularity in thinking are even broader. The continued benign neglect of “growth fundamentals” in the post-colonial period has reinforced the dependency syndrome whereby African economies essentially serve as feedstock in the global economy. However, the nature of the relationship linking African economies to the rest of the world has historical roots going back to the colonial era. It was part of the colonial construct which confined production in the colonies to primary commodities and natural resources, a pattern that has persisted in the post-colonial period in the region. Albert Sarraut (1872–1962), Colonial Minister for France from 1920 to 1924 and 1932 to 1933 best captured the structure and intent of the colonial economy. Economically, a colonial possession means to the home country simply a privileged market whence it will draw the raw materials it needs, dumping its own manufactures in return. Economic policy is reduced to rudimentary procedures of gathering crops and bartering them. Moreover, by strictly imposing on its colonial “dependency” the exclusive consumption of its manufactured products, the metropolis prevents any efforts to use or manufacture local raw materials on the spot, and any contact with the rest of the world. The colony is forbidden to establish any industry, to improve itself by economic progress, to rise above the stage of producing raw materials, or to do business with the neighboring territories for its own enrichment across the customs barriers erected by the metropolitan power (Fetter 1979: 109). The persistent bias for exogenous development models has shaped the structure of production and may be largely responsible for Africa’s commodity dependency trap. In particular, by specializing in the export of primary products and becoming dependent on the world economy for imports of manufacturing goods, African countries have fallen in a condition of underdevelopment (Rodney 1982). The idea that a foundation of growth solely resting on natural resource export may stifle the development of a country entered mainstream economics through the Dependency Theory in the early 1950s. This thesis took its origins in the seminal work on developing countries’ TOT by Prebisch (1950) and Singer (1950). It became very influential in the 1960s and 1970s as an antithesis to modernization theories (Amin 1972). In addition to its effects on economic structure, the costs of uncritically embracing exogen­ous development models for developing nations which are price takers is the rise of conformity to the norms and development paradigms imposed by imperial powers and surrendering of national aspirations (consciously or unconsciously). In essence, the logic goes as follows: if the more advanced and industrialized countries are successful development models to which the least developed and backward nations should aspire, then embracing the ideology which has taken these countries to the pinnacle of success should be the sure path to development for late-comers. This idea which assumes that the ideology sustaining development in more advanced economies has not been subject to changes over time is

The Idea of Economic Development    283 rooted in the grand Marxian generalization of history of development.12 It gained more traction with Rostow’s (1959) Stages of Economic Growth model. Arguably, the transcending power of exogenous development models, which are informed by the globally dominant ideology on backward nations grew significantly under the Washington Consensus when the conformity with the global trend of thought was no longer done on a voluntary basis, but became a precondition for accessing resources for countries facing recurrent balance of payments crises. Still, in the African context, the erosion in the sovereignty of nations was further exacerbated by debt overhang which consolidated the power of creditor nations. However, whether the process of surrendering national aspirations on the development path is voluntary or involuntary, a cross-section and trend analysis of the history of development suggests that development is fundamentally an endogenous process whereby the “spirit” and driving “ideology” are fully informed by the country’s political, historical and institutional contexts, which are themselves subject to changes over time (Gerschenkron 1962; Currie-Adler et al. 2013). This historical finding is supported by the contrasting nature of development models adopted by European countries, most notably France and Germany, after the industrial revolution in Great Britain. These countries did not exactly follow the path beaten by Great Britain decades earlier. Instead, their development process was informed by their historical and institutional settings as documented by Gerschenkron in Economic Backwardness in Historical Perspective. In effect, Gerschenkron’s historical analysis of development patterns in Europe after the industrial revolution suggests that the development of a backward economy might differ considerably from that of the now more advanced economies. In a context of rapidly changing technology, late industrializers might be able to even leap-frog into more technologically advanced sectors, through imitation and learning from pioneers. This historical analysis has highlighted two of the most critical development pitfalls which might have stifled the development process in Africa: the acquiescence to the globally dominant ideology which is neither immune from geopolitical considerations nor rooted in Africa’s historical and institutional realities, and the costs of development mimicry (or aspiring to become like developed countries in the north, even though models informing development in these countries have been in a constant state of mutation). Interestingly, other countries in the developing world, especially in Asia, successfully avoided these two development pitfalls through the emergence of development models of their own. The next sections discuss the meanings of development from an African perspective and sketch out the contours of an African development model.

15.4  An Emerging African View and Perspective of Development The pre-eminence of exogenous development models over the African development and policy space, especially in the post-structuralist era, is not necessarily a reflection of a larger 12  According to Marx, “The industrially more developed country presents to the less developed country a structure of the latter’s future” (Marx: 1867, Preface).

284   Concepts deficit of endogenous models rooted in African institutional and traditional realities. In effect, several attempts have been made to craft development models which are informed by indigenous development paradigms in the region and which address local realities. Hence, the analysis undertaken in this section is based on desk research and informed by a review of existing African-led conceptual development frameworks and representative surveys of opinion. These approaches, derived from evidence gleaned from several decades of development experience, should provide a good representation of Africans’ vision of their future. This is not the place for a comprehensive review of all African-led development frameworks.13 However, it is worth mentioning some of the most prominent ones. In particular, a few that fall under that category include the Lagos Plan of Action (LPA) for Economic Development of Africa (1980–2000), the Africa’s Priority Programme for Economic Recovery 1986-90 (APPER), the African Alternative Framework to Structural Adjustment Programme for Socioeconomic Recovery and Transformation (AAF-SAP) developed in 1989, the African Charter for Popular Participation for Development (1990) and the widely publicized New Partnership for African Development (NEPAD) released in 2001.14 The common thrust of these African-led development frameworks is the emphasis on self-reliance encompassing socioeconomic transformation accompanied by a holistic human development and democratization of the development and governance process (Adedeji 2002) (see Figure 15.2). Self-reliance does not, however, mean that Africa should embrace autarky as a development strategy. Instead, it implies that external support should not be the mainstay of African development, but rather a boost to existing endogenous efforts. After all, economic growth, which is necessary though not sufficient for development, depends on trade and investment, both domestic and foreign direct investment. Nevertheless, the emphasis on self-reliance in these development frameworks rhymes with the decolonization of political economy and affirmation of independence, especially for newly independent nations emerging from decades of colonialism. In particular, the quest for self-reliance reflects the growing aspiration of Africans to assume the full responsibility of their destiny by exiting the dependency trap as perpetual recipients of foreign aid. Most African countries became even more dependent on foreign aid during the implementation of adjustment programs, in part because the external response to balance of payments crises that was supposed to be short-term bailouts took a life of their own and became permanent operations. In practice, as the ultimate development objective, self-reliance cannot be achieved when economic decisions undertaken by countries aspiring to it are largely predetermined by foreign governments and international organizations acting as donors, though not always altruistic ones (Easterly 2006, 2013; Moyo 2009). In effect, bilateral and multilateral agencies providing assistance to African countries have influenced policy choices and the design of development programs through the implementation of conditionalities. To take one prominent example, the privatization of public enterprises which gave birth to the emergence of private monopolies largely owned by foreign firms and concessions was part of that effort.

13 

For a comprehensive review of these plans, see OAU (1980, 2001), Adedeji (2002), and Bujra (2004). NEPAD is thought to have been less organic and endogenous in its conception than the earlier development plans, particularly in light of its top-down approach and excessive reliance on external financing (Adedeji 2002). 14 

The Idea of Economic Development    285 At the same time, the choice of self-reliance as the long-term development goal fundamentally reflects the core values in effect in countries and across the region. Although society’s values tend to differ and may even be subject to changes over time as a result of shifting social norms and even religious beliefs (as convincingly established by Max Weber (1905) in the Protestant Ethic and the Spirit of Capitalism), a review of successive development strategies and analysis of surveys of opinion undertaken in the region single out trust, freedom and justice, self-respect and equalization of opportunity as core values (Justesen and Bjornskov 2012; Richmond and Alpin 2013). At the heart of these core values is the notion of equity, probably reflecting the focus on human capital as both a means and end objective in a context of corruption-fuelled growing income inequality and poverty. Perhaps the democratization of the development and governance process articulated in various African development plans has been seen as a possible solution to poor governance. Nonetheless, while the drive for good governance is rooted in the notion of social justice and therefore may enhance the quest for trust and freedom, inclusive economic growth is the path to sustenance and self-respect. By uniformly expanding the access to opportunities, a more balanced type of accumulation is likely to ensure that the benefits of growth are not regressive in their distribution. Still, the core values and development aspirations are very much intertwined. The development priorities are actually shaped by the core values which often reveal the priorities of societies. For one, the pursuit of self-reliance as the collectively shared aspirational goal can only be achieved at the national level if the core value of self-respect is uniformly shared and accepted by most if not all individuals. In this regard and reflecting the invisible hand metaphor the core value of self-respect which is intrinsically incompatible with the idea of being chronically dependent on the state or alternatively on a generous social capital in the absence of effective social welfare systems will eventually accelerate the convergence towards the ultimate development goal of self-reliance at the national and regional level. Arguably, this interaction between the societal core values and the derivation of development aspirations, whereby the latter is regularly refined according to prevailing social norms and cultural values, is also likely to affect the sequencing and choice of instruments. In particular, departing from the old framework (underpinned by the two-gap model) where development did not envisage structural transformation and instead implies extrapolating past trends, choosing investment patterns that would produce an acceptable increase in national income over time, the proposed framework starts with a vision of an optimal development path consistent with the collectively shared aspirations, and then works backward to align the development of human and physical capital to support the realization of that end-development objectives—self-reliance. In addition to affecting the quality of human resource development and content of education, the core values are also likely to condition the parameters defining the role of the state and the market. This is particularly the case in the field of education where markets have historically played a less prominent role during early stages of development when government policies have promoted equalization of opportunities and greater access to address chronic shortage of skills in a context of structurally low-income. However, the role of the state has decreased with economic expansion and per capita income growth. In this regard, the quest for development is clearly conceived as a medium to long-term process under the proposed development framework. It may not even follow a linear path, in the sense that the process of structural transformation expected in the medium and long run may involve short-term costs in terms of growth and welfare.

286   Concepts Now turning to instruments and mechanics of development, the quest for broad-based and inclusive growth is one means to achieve the ultimate goal of self-reliance (Figure 15.2). Taking into account the triangulation between growth, income inequality and poverty, these frameworks recognize that growth alone may not be sufficient to achieve sustainable development and poverty reduction, especially in a context of rising income inequality. In effect, in a region where the recurrent cycle of commodities price boom-and-bust has invariably been accompanied by the persistence of widespread poverty as the experience of petroleum economies show—(i.e. Gabon is often referred to as a middle-income country with low-income country development outcomes)15 it is not at all surprising that distributional issues are at the heart of development. In practice, the proposed African development frameworks have advocated policies in favor of inclusive growth, underpinned by expansion of labor-intensive industries and progressive eradication of unemployment. In other words and unlike Washington Consensus models, growth is not an end, but a means to expand employment opportunities, reduce income inequality and poverty. To a certain extent, the parameters underlying the Africans’ vision of development are consistent with models behind the transformation of Asian emerging markets where growth was not only sustained, but also more inclusive, resulting in a concurrent reduction of income inequality and poverty as a result of expanding labor-intensive employment opportunities in the manufacturing sector (Nelson and Pack 1999; Lin and Monga 2011; Stiglitz et al. 2013a). In other words, growth becomes development enhancing under these frameworks only when it leads to a concurrent decline in unemployment, inequality and poverty. Conversely, if there is deterioration in any of the three development outcomes, especially if all three worsened during growth spurts, then economic growth can hardly be associated with development, even in the face of sustained increase in per capita income. Equally important is the sharp contrast between development models advocated under the Washington Consensus and the vision Africans have of their production patterns and possibility frontiers. Although there is heterogeneity of views, the emerging consensus is that aspiring Africans do not see their production patterns as static, but see the modernization and diversification of their economies as a path to a more balanced growth and insurance against recurrent risks of adverse TOT shocks which for decades have made sustainable growth and poverty reduction elusive to the region (Stiglitz et  al. 2013b). Hence, departing from the highly orthodox era of adjustment which accepted the existing structures of African economies as given and set to adjust them within the prevailing production paths, African-led development models advocate a stronger role for the state in the economy and structural transformation—reallocation of economic activity across the three broad sectors (agriculture, manufacturing, and services) and greater value-addition nationally, as well as increased competition. Structural transformation entails a gradual reduction of the craft economy underpinned by the colonial mode of production (largely dominated by mono-culture, natural resources and increasingly informal activities), and transition to a more efficient and technology-based economy largely dominated by manufacturing industries. Over time, it is expected that the rise of successful entrepreneurs and technology adoption through 15 

See World Bank (2005b).

The Idea of Economic Development    287 knowledge transfers will tilt the balance of the production structure towards more technologically advanced and efficient methods of production and industries operating on frontier technologies. In the process, this transition will enhance the competitiveness of African economies and their integration into the world economy that is increasingly dominated by manufactured goods.16 Embracing the structural diversification path may also strengthen the quest for self-reliance by decolonizing the African political economy. This view is well articulated in the LPA. In particular, Article 52 of that Plan calls upon member states to give “a major role to industrialization, in view of its impact on meeting the basic needs of the population, ensuring the integration of the economy and the modernization of society. To this end, and in order for Africa to achieve a greater share of world industrial production and attain an adequate degree of collective self-reliance rapidly, Member States proclaimed the years 1980 to 1990: Industrial Development Decade in Africa.” However, these calls were not followed by actions at the policy level, in part due to the lack of commitment of African leaders, but also because the implementation of fiscal austerity during the adjustment era curtailed the expansion of public investment, with pernicious effects on growth. The emphasis on science and technology and call for a holistic human resource development is another constant in the various conceptual frameworks put forward by Africans (OAU 1980, 2001). The centrality of science and technology for progress and development has been consistent over the years and throughout the world. In fact, existing empirical evidence suggests that technological convergence between formerly backward nations and advanced industrialized countries has always preceded income convergence (Gerschenkron 1962; Fofack 2008). More recently, a review of various models that could produce a generalized balanced growth and structural transformation as simultaneous outcomes singled out technological progress as one of the primary drivers (Herrendorf et al. 2013). In this respect, the promotion of science and technology, which has become the main driver of productivity in the knowledge economy, may emerge as a key instrument to achieving the ultimate objective of self-reliance and structural transformation. In particular, technological absorption and application will enable resource-rich countries in the region to go beyond the simple production processes and export of primary products to move up the value chain. It will also boost agricultural productivity and accelerate progress toward the green revolution which is a pre-requisite to structural transformation, as it may reduce the risks associated with shifting labor away from that sector when the assumption of unlimited supply of labor embedded in Lewis’ model is relaxed. However, the success of countries in their attempt to move from factor-driven to efficiency-driven economies on the competitiveness ladder will depend on their ability to absorb existing technologies and global knowledge and on the speed with which they adopt new ones in order to constantly operate on global frontier technology. At the same time, it also depends on other efficiency enhancers such as the quality of higher learning institutions, adequate economic and scientific infrastructures, the pools of well-educated and skilled workers, market size, and efficiency of labor and goods markets (in terms of alignment of supply and demand).

16  Manufacturing products now account for more than 85 percent of developing country exports, from less than 15 percent in the 1960s (see World Bank 2002).

288   Concepts

Development Outcomes

National Statistical Systems

Human Development

Economy

Markets Regional Integration

Physical Capital

Human Capital

Realm of aspirations setting (self-reliance), Public policy decisions and regulations

Core Value System

Figure  15.2   A  framework for sustainable development in sub-Saharan Africa. Source:  Author.

Regional integration and intra-African trade expansion which constitute the mainstay of African development strategies provide the path to grow market size beyond the confinements of national boundary. In addition to encouraging greater exploitation of increasing returns to scale, economic integration may also reduce the recurrent risks of asymmetric shocks and exchange risks, especially in regional common currency zones. For one, regional integration may provide a much smoother path for firms to gradually grow into more competitive global value chains. Presently, the acute shortage of technical skills, the structural mismatch between supply and demand of labor as well as the deficit of adequate scientific and economic infrastructures is pervasive and highlights coordination and market failures. Another symptom of market failures specific to African economies is the absence of markets for the production

The Idea of Economic Development    289 and transfer of knowledge or the imperfection of these markets wherever they exist (Stiglitz et al. 2013b). In order to close Africa’s scientific and knowledge gap with the rest of the world and enhance the valuation of traditional knowledge, human resource development is set as a continuous process under the various African development plans. However, in the long run, the sustainability and success of that human resource strategy will depend on incentives put in place to continuously raise the rate of returns to education. In this regard, providing the right incentives for the growth of entrepreneurship could create a virtuous cycle under the complementarity of public and private capital accumulation argument. Meanwhile, more than just addressing market failures, the less dogmatic approach underpinning African development strategies has carved an even greater role for states. In addition to providing the right infrastructure for economic development, states are expected to play a more active role in the area of human resource development and utilization, industrialization, including the promotion of import substitution industries (ISI) which were central to economic diversification in Asia and are increasingly advocated by scholars of African development (Adedeji 2002; Lin and Monga 2011; Fokam 2013; Stiglitz et al. 2013b).17 In the African context, ISI and policies could be an important means to correct capital and sectoral misallocations and to address the pervasive discrepancies between private gains and social costs which skyrocketed after the privatization of public infrastructure in the region. Still, they could also ensure an adequate transfer of resources from low- to high-productivity sectors, including internal migration of Africa’s unskilled rural labor to unskilled labor-intensive industries as recently argued by Stiglitz et  al. (2013b). The new model foresees the rise of developmental states, more growth-oriented with greater sensitivity to prices and markets, than their predecessors in the 1960s and 1970s. Over the years, development models advocated by Africans have been intrinsically self-reliant and underpinned by development policies that take into account socio-cultural values. Human development is at the heart of these models and the various tools and instruments available to sustainably improve the living standards of the population include economic growth, science and technology, human resource development and utilization, massive increase in capital formation including through infrastructure development and regional integration to expand opportunities, both in terms of increasing returns to scale and employment. Human resource development is a means to promote economic growth through the productivity channel; but it is also an end in itself. Raising life expectancy and education levels will improve living standards and contribute to the construction of a more harmonious society where the core values of equity and fairness are enhanced (see Figure 15.2). Over time, the effectiveness of the proposed framework will be assessed on the basis of impacts on development outcomes, which include both monetary and non-monetary indicators to monitor progress towards self-reliance and account for emphasis on human development. In practice, the challenge associated with the operationalization of these development plans lies in the capacity of countries to establish adequate monitoring systems to assess development impacts on a regular basis, not only on the three aforementioned

17  A recent study assessing the development impact of ISI in Latin America also concluded on the efficacy and net positive returns of that experiment, with state-wide benefits largely exceeding budgetary costs (Ocampo and Ros 2013).

290   Concepts development outcomes (unemployment, inequality and poverty), but also vis-à-vis the declining rate of foreign aid or alternatively a diversification of sources of foreign aid and reduced dependence on a single donor. The ability of a country to sustainably draw on strong capital formation to wean itself off aid-dependency should be a good indicator of progress towards self-reliance in the medium and long term.18 Equally central to this new paradigm is the view that the agent of development is the state and the instrument for optimizing the allocation of public resources is the market. The state is expected to play an activist role in the sphere of modernization of the economy, including in the form of import substitution industries and increasing substitution of domestic factor inputs for external factor inputs. But it is also meant to support nascent industries while fostering competition by improving the business environment and regulatory framework for a better integration into the global economy. These policies were particularly successful in East Asia where the pervasive and catalytic role of governments and leadership commitment ensured their successful implementation. However, the success of Asian emerging markets over the last few decades has equally rested on the strength of national institutions and governance which insulated the meritocratic and competent bureaucracies from political capture. In particular, these institutional features and designs enabled civil servants to enforce strict performance criteria for industries that received support from governments. In the process, the ability of these able bureaucrats to systematically optimize the allocation of public spending and public investments largely contributed to the rise of homegrown multinational corporations in that part of the world. Hence, strengthening the governance framework will be essential to the creation of effective development states that will set African countries on the path of economic transformation. In addition to governance and institutional strengthening, another challenge which has stifled the implementation of national development strategies is the resource constraint. The implementation of fiscal austerity and the systematic implementation of a balanced budget rule in the face of macroeconomic imbalances constrained the policy space, significantly curtailing public investments. More recently, a combination of debt relief and commodities price boom has increased the fiscal space across the region. In the short term, countries, especially the resource rich, could further increase their fiscal space by improving the regulatory framework to raise additional revenues from mining, oil and gas concessions in support of infrastructure development, making the most of Africa’s commodities. At the same time, raising the marginal propensities to save and invest to levels enjoyed by Asian emerging market economies during their golden years is essential and should be a priority in the post-HIPC commodity price boom. This intermediate objective should be facilitated by increased additionality from improved regulatory framework in the management of natural resources. More generally, increasing prospects for domestic resource mobilization should be facilitated by the improvement of governance and institutions to systematically apply the golden rule in the rationalization of public investment. Financing sound projects under the ISI should also be facilitated by the adoption of a macroeconomic framework where deficit target setting takes into account inter-temporal fiscal sustainability. 18 

Although, the various plans have called for such systems, the production of timely statistics on poverty, income inequality and unemployment remains an exception rather than the rule in the region. South Africa is one of the few countries in the region which produces reliable statistics on poverty and unemployment on a regular basis.

The Idea of Economic Development    291

15.5  Analytical Framework for Sustainable Development in Africa We outline a simple framework that captures important elements of the African transitional development path. Central to this framework is generalized balanced growth (sustained over time) and structural transformation. Equally important is the underlying hypothesis of sustained accumulation of physical and human capital which is essential for a continuous increase in the level of output per worker. A two-sector growth model that incorporates balanced growth and structural transformation as desirable outcomes is used to represent that transitional development path. It closely resembles that in Nelson and Pack (1999), Acemoglu and Guerrieri (2008), and Herrendorf et al. (2013). Under the framework, development is measured by the transition from the traditional economy inherited from the colonial era to the modern and more technologically advanced one. Over time, structural transformation is measured by growth in the size of the modern sector. This is illustrated by a gradual shifting of resources from the traditional sector (mono-culture, natural resources and increasingly informal production) to the more technologically advanced sector dominated by manufacturing. Against this background, let’s assume that production functions in both sectors (traditional and modern) are Cobb–Douglas, with complementarity between factors. If we denote these production functions by r (for rudimentary) and by m (for modern), then the two-sector production function can be represented by:

Yi ,t = Ai ,t K iα,ti L1i −,tαi

i ∈{r , m}, (1)

where α with (0 < α 0 ( g > 1). The higher profitability of the modern sector provides an incentive for structural transformation. The reallocation of resources from the traditional sector to the rapidly growing one is motivated by that profitability. The factors behind the higher profitability in the modern sector include multiple drivers of total factor productivity, including quality of labor and capital, skilled labor, infrastructure, research and development, and entrepreneurial methods. Provided that the education premium associated with the increasing demand for skilled labor never becomes too large, the modern sector will continue to enjoy the productivity and technological advantage. From equations (2), (3), and (4) we know that as capital and labor shifts to the modern sector, K / L and Y / L will increase. As long as the amount of educated labor is responsive to demand, human capital will continue to expand. However, the relatively higher productivity in the modern sector that is driving the reallocation of factors also holds when the assumption of constant price across sectors is relaxed. This can be verified by looking at the implications of differences in sectoral capital intensities for structural transformation. In particular, assuming that capital intensities differ across sectors, then the first-order conditions for the stand-in firm in sector i ∈{r , m} can be derived from equation (1) as follows: K  Rt = Pit α i At  it   Lit 

α i −1

(6)

αi



K  Wt = Pit (1 − α i ) At  i ,t  , (7)  Li ,t 

where Rt is the rental rate of capital and Wt is the wage rate. Dividing these equations by each other gives:

1 − α i K it 1 − α j K jt . = . α i Lit α j L j ,t

(8)

294   Concepts Equation (8) implies that sectors with larger capital shares have larger capital–labor ratios. The capital–labor ratio grows at the same rate in all sectors. Substituting (8) into (7) produces the relative prices as follows:

K  Pit = Ωij  i ,t  Pjt  Lit 

α j − αi

i, j ∈{r , m}, (9)

where Ωij is a constant that depends on the capital share. Since the capital–labor ratio grows at the same rate in the two sectors, equation (9) implies that the relative prices of the sector with a higher capital share (modern sector) decreases as the aggregate capital stock grows. The implications of the much higher rate of returns on capital in the modern sector, following a reallocation of factors can also be interpreted using the Rybczynski theorem. According to that theorem, at constant relative good prices, a rise in the endowment of one factor will lead to a more than proportional expansion of output in the sector which uses that factor intensively. In addition to sustained human and physical capital accumulation, the structural transformation of Asian economies has been attributed to the effectiveness of entrepreneurship in terms of their ability and speed of response to profit-making opportunities. In the proposed African development framework, this constraint is expected to be alleviated by the activist role of the state which supports the growth of entrepreneurship in several ways, including through the provision of infrastructure, improvement of the business and investment climate, import-substitution industries and greater exploitation of increasing returns to scale in a context of deepening regional integration and strengthening governance and regulatory framework for efficient allocation of capital.

15.6  Concluding Remarks A review of the history of development in Africa over most of the post-independence era shows that development models applied to the region have been divorced from the local context and reductionist in their approach. In general, these models have assimilated development to growth in national income measured by changes in per capita income—the sufficient statistics expected to capture both the dynamics and complexity of the development process in any given country in the region. The overemphasis on economic dimensions of development reflected the ascendency of the neoliberal ideology and rise of market forces, but also the vulnerability of African countries in an era where the conformity with the global trend of thought became a precondition for accessing resources for countries facing recurrent balance of payments crises. Besides establishing that these models failed to meet expectations strictly on the income growth objective, this paper also shows that the pre-eminence of exogenous development models over the African development and policy space was not necessarily a reflection of a relatively larger deficit of endogenous models rooted in Africa’s institutional and traditional realities. Over the years, several attempts have been made to conceive reflexive development models that were organic and more comprehensive in the region. The cornerstone of development in these Africa-led frameworks has been self-reliance. And economic

The Idea of Economic Development    295 growth—not just measured by income growth, but by broad-based growth and structural transformation—has been advocated as one out of several means to achieve that higher development goal. Other means equally central to the pursuit of self-reliance under these frameworks include a holistic human resource development and effective utilization, acceleration of capital accumulation and regional integration to expand opportunities, both in terms of increasing returns to scale and employment. Human resource development is a means to promote economic growth through the productivity channel, but also an end in itself. The complementarity between the market and the state is another important feature of these models. While the state is the ultimate agent of development (expected to be active in all the different phases of development, from aspirational settings to monitoring of development outcomes), the market is the instrument for optimizing the allocation of public resources under the structural budget constraint. Hypothetically, the analysis suggests that development outcomes in the region would probably be superior under African-led development frameworks which emphasize structural transformation and technology-led productivity growth. The concurrent pursuit of generalized balanced growth and structural transformation as simultaneous development outcomes advocated under these frameworks would provide the insurance against the recurrent adverse TOT shocks which for decades have made sustainable growth elusive in the region. At the same time, it would foster the integration of African countries into the new global economy where the globalization of trade is increasingly synonymous to the globalization of manufacturing exports. However, to successfully embrace the structural transformation path and enhance the competitiveness of their economies to become global players, African countries should first and foremost reclaim their policy space and establish the right conditions (both in terms of valuation of indigenous knowledge and domestic resource mobilization) to avoid the development pitfalls which are associated with the globalization of the development thought. Two pitfalls that loomed large in the past and potentially stifled the development process in the region are the acquiescence to the globally dominant ideology and development mimicry. Under the present international financial architecture, these development pitfalls are not likely to disappear anytime soon. Still, although the risk exposure to these pitfalls remains much higher for low-income countries subject to recurrent balance of payments crises, the rise of Asian emerging market economies over the last few decades suggests that they can be overcome in the quest to self-reliance. However, in the African context where implementation has been the main binding constraint, this is not likely to happen unless the development aspirations are collectively shared and all the national treasures (human and financial resources) are strategically deployed to serve the overarching development objective—self-reliance.

References Acemoglu, D., Johnson, S., and Robinson, J.A. (2005). Institutions as a fundamental cause of long-run growth, in Handbook of Economic Growth, Vol. 1, Part A. North Holland: Elsevier. Acemoglu, D., and Guerrieri, V. (2008). Capital deepening and non-balanced economic growth. Journal of Political Economy, 116:467–498.

296   Concepts Adedeji, A. (1999). Structural adjustment policies in Africa. International Social Science Journal, 51(162):521–528. Adedeji, A. (2002). From the Lagos Plan of Action to the New Partnership for African Development and from the Final Act of Lagos to the Constitutive Act: Whiter Africa? Keynote address prepared for presentation at the African Forum for Envisioning Africa, Nairobi, Kenya. Akyeampong, E., and Fofack, H. (2014). The contribution of African women to economic growth and development in the precolonial and colonial period: historical perspectives and policy implications. Economic History of Developing Regions Journal. Amin, S. (1972). Underdevelopment and development in black Africa—origins and contemporary forms. Journal of Modern African Studies, 10(4):503–524. Amsden, A. (1989). Asia’s Next Giants:  South Korea and Late Industrialization. New York: Oxford University Press. Artadi, E., and Sala-i-Martin, X. (2003). The Economic Tragedy of the 20th Century: Growth in Africa. NBER Working Paper 9865. Cambridge, MA: National Bureau of Economic Research. Bahadur, C., Schmidt-Traub, G., Sachs, J.D. et al. (2004). Ending Africa’s Poverty Trap. Brookings Papers on Economic Activity, 2004, No. 1. The Brookings Institution, Washington DC. Basu K. (2001). On the goals of development, in G. Meier and J. Stiglitz (eds), Frontiers of Development Economics: The Future in Perspective. Washington, DC: World Bank. Blanchard, O., Dell’Ariccia, G., and Mauro, P. (2010). Rethinking Macroeconomic Policy. Staff Position Note SPN/10/03. Washington, DC: International Monetary Fund. Bonfatti, B., and Poelhekke, S. (2013). From Mine to Coast:  Transport Infrastructure and the Direction of Trade in Developing Countries. Working Paper Series 4170. Munich: CESifo. Boyce, J., and Ndikumana, L. (2012). Capital Flight from Sub-Saharan African Countries: Updated Estimates 1970-2010. PERI Report. Amherst, MA: University of Massachusetts. Brett, E. (2009). Reconstructing Development Theory:  International Inequality, Institutional Reform and Social Emancipation. Basingstoke: Palgrave Macmillan. Bujra, A. (2004). Pan-African Political and Economic Vision of Development:  From OAU to the AU, from the Lagos Plan of Action (LPA) to the New Partnership for African Development (NEPAD). DPMF Occasional Paper, No. 13, Development Policy Management Forum, Addis Ababa. Chang, H.-J. (2002). Kicking Away the Ladder: Development Strategy in Historical Perspective. London: Anthem. Currie-Adler, B., Kanbur, R., Malone, D., and Medhora, R. (2013). The State of Development Thought. Working Paper 2013-13. Ithaca, NY:  Charles H.  Dyson School of Applied Economics and Management, Cornell University. Devarajan, S., and Shetty, S. (2010). Africa: Leveraging the Crisis into Development Takeoff. Economic Premise. World Bank PREM. Washington, DC: World Bank. Easterly, W., and Levine, R. (1998). Troubles with neighbors: Africa’s problem, Africa’s opportunity. Journal of African Economies, 7(1):120–142. Easterly, W., and Levine, R. (2003). Tropics, germs, and crops: the role of endowments in economic development. Journal of Monetary Economics, 50(1):3–39. Easterly, E. (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.

The Idea of Economic Development    297 Easterly, E. (2013). The Tyranny of Experts: Economists, Dictators, and the Forgotten Rights of the Poor. New York: Basic Books. Fetter, B. (1979). Colonial Rule in Africa:  Readings from Primary Sources. Madison, WI: University of Wisconsin Press. Fine, B., Lapavitsas, C., and Pincus, J. (eds) (2003). Development Policy in the Twenty-First century: Beyond the Post-Washington Consensus. Milton Park: Routledge. Fofack, H. (2008). Technology Trap and Poverty Trap in Sub-Saharan Africa. Policy Research Working Paper 4582. Washington, DC: World Bank. Fofack, H. (2010). Africa and Arab Gulf States: divergent development paths and prospects for convergence. Journal of African Development, 12(1):97–130. Fofack, H. (2012). Stolen asset recovery: the need for a global effort. ACAS Bulletin, 87:29–35. Fofack, H. (2013). A Model of Gendered Production in Colonial Africa and Implications for Development in the Post-Colonial Period. Policy Research Working Paper 6438. Washington, DC: World Bank. Fofack, H. (2014). Retrospective Analysis of Africa Post-HIPC Growth Resurgence: Overcoming the Risk of Immiserizing Growth. Africa Affairs. Forthcoming. Fofack, H., and Ndikumana, L. (2010). Capital flight repatriation: investigation of its potential gains for sub-Saharan African countries. African Development Review, 22(1): 4–22. Fofack, H., and Ndikumana, L. (2014). Capital Flight and Monetary Policy in African Countries, in I. Ajayi and L. Ndikumana (eds), Capital Flight from Africa: Causes, Effects, and Policy Issues. Oxford: Oxford University Press, forthcoming. Fokam, P.K. (2013). Quelle Afrique a l’Horizon 2050? Un Espace Transforme en Ilot de Prosperite. Yaoundé: Afredit. Gerschenkron, A. (1962) Economic Backwardness in Historical Perspective. A Book of Essays. Cambridge, MA: Belknap Press of the Harvard University Press. Harriss, J. (2014). Development theories, in B. Currie-Adler, R. Kanbur, D.M. Malone, and R. Medhora (eds), International Development: Ideas, Experience and Prospects. Oxford: Oxford University Press. Harvey, D. (2005). A Brief History of Neo-Liberalism. Oxford: Oxford University Press. Hermes, N., and Lensink, R. (2014). Financial liberalization and capital flight, in I. Ajayi and L. Ndikumana (eds), Capital Flight from Africa: Causes, Effects, and Policy Issues. Oxford: Oxford University Press. Herrendorf, B., Rogerson, R., and Valentinyi, A. (2013). Growth and Structural Transformation. Working Paper Series 18996. Cambridge, MA:  National Bureau of Economic Research. Isham, J., Woolcock, M., Pritchett, L., and Busby, G. (2005). The varieties of resource experience: natural resource export structures and the political economy of growth. World Bank Economic Review, 19(2):141–174. Justesen, M.K., and Bjornskov, C. (2012). Exploiting the Poor: Bureaucratic Corruption and Poverty in Africa. Working Paper 139. Afrobarometer. Kohli, A. (2004). State-Directed Development:  Political Power and Industrialization in the Global Periphery. Cambridge: Cambridge University Press. Krueger, K. (2004). Meant Well, Tried Little, Failed Much:  Policy Reforms in Emerging Market Economies. Roundtable Lecture (March 23), Economic Honor Society, New York University. Lewis, L. (1954). Economic development with an unlimited supply of labour, Manchester School, 22: 139–191.

298   Concepts Lewis, L. (1955). Theory of Economic Growth. Milton Park: Routledge. Lin, J.Y., and Monga, C. (2011). Growth identification and facilitation: the role of the state in the dynamics of structural change. Development Policy Review, 29(3):264–290. Marx, K. (1867). Das Kapital:  Kritik der Politischen okonomie. Verlag von Otto Meisner, Germany. Mkandawire, T. (2004). Can Africa have development states? in S. Bromley et al. (eds), Making the International: Economic Interdependence and Political Order. London: Pluto Press (in association with Open University). Moyo, d. (2009). Dead Aid: Why Aid Is Not Working and How There Is a Better Way for Africa. New York: Allen Lane. Myrdal, G. (1972). Asian Drama:  An Inquiry into the Poverty of Nations. New York: Random House. Nellis, J. (2003). Privatization in Africa: What Happened? What is to be Done? Working Paper 25. Washington, DC: Center for Global Development. Nelson, R., and Pack, H. (1999). The Asian miracle and modern growth theory. Economic Journal, 109(457):416–436. Norman, A., and Stiglitz, J. (2012). African development prospects and possibilities, in E. Aryeetey et al. (eds), The Oxford Companion to the Economics of Africa. Oxford: Oxford University Press. North, D.C. (1990). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. Organisation of African Unity (OAU) (1980). Lagos Plan of Action for the Economic Development of Africa. Addis Ababa: OAU. OAU (2001). The New Partnership for African Development (NEPAD). Addis Ababa: OAU. Ocampo, A., and Ros, J. (eds) (2013). The Oxford Handbook of Latin American Economics. Oxford: Oxford University Press. Prebisch, R. (1950). The Economic Development of Latin America and Its Principal Problems. New York: United Nations. Richmond, S., and Alpin, C. (2013). Governments Falter in Fight to Curb Corruption: The People Give Most a Failing Grade. Afrobarometer. Rodney, W. (1982). How Europe Underdeveloped Africa. Washington, DC:  Howard University Press. Rodrik, D. (2006). Goodbye Washington Consensus, hello Washington confusion? Journal of Economic Literature, 44(4):973–987. Rodrik, D. (2008). One Economics, Many Recipes:  Globalization, Institutions and Economic Growth. Princeton: Princeton University Press. Rostow, W.W. (1959). The Stages of Economic Growth. Economic History Review 12(1):1–16. Sachs, J.D., and Warner, A. (2001). Natural resources and economic development: the curse of natural resources. European Economic Review, 45:827–838. Seers, D. (1969). The Meaning of Development. IDS Communication 44. Brighton: IDS. Sen, A. (1988). The concept of development, in H. Chenery and T.N. Srinivasan (eds), Handbook of Development Economics, vol 1. North-Holland: Elsevier. Singer, H. (1950). The distribution of gains between investing and borrowing Countries. American Economic Review, 40(2):473–485. Stiglitz, J. (1998). An agenda for development in the twenty-first century, in B. Pleskovic and J.E. Stiglitz (eds), Annual World Bank Conference on Development Economics 1997. Washington DC: The World Bank.

The Idea of Economic Development    299 Stiglitz, J.Y. (2002). Globalization and its Discontents. New York: WW Norton. Stiglitz, J., Sen, A., and Fitoussi, J.P. (2011). Report by the Commission on the Measurement of Economic Performance and Social Progress. www.Stiglitz-Sen-Fitoussi.fr. Stiglitz, J., Lin, J., and Monga, C. (2013a). The Rejuvenation of Industrial Policy, Policy Research Working Paper 6628. Washington, DC: World Bank. Stiglitz, J., and Lin, J.Y. (eds) (2013). The Industrial Policy Revolution I: The Role of Government Beyond Ideology. New York, Palgrave Macmillan. Stiglitz, J., Lin, J.Y., Monga, C., and Patel, E. (2013b). Industrial Policy in the African Context. Policy Research Working Paper 6633. Washington, DC: World Bank. UN (2009). The Economic Commission for Africa:  Fighting to be Heard. UN Intellectual History Project, Briefing Note N0. 21, July 2009. United Nations, New York. UN Economic Commission for Africa (UNECA) (2013). Making the Most of Africa’s Commodities:  Industrializing for Growth, Jobs and Economic Transformation. Addis Ababa: UNECA. Wade, R. (1990). Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization. Princeton, NJ: Princeton University Press. Weber, M. (1905). The Protestant Ethic and the Spirit of Capitalism. London: Unwin Hyman, 1930; Translated: Talcott Parsons and Anthony Giddens. World Bank (1997). World Development Report 1997:  The State in a Changing World. Washington, DC: World Bank. World Bank (2002). Globalization, Growth, and Poverty: Building an Inclusive World Economy. A World Bank Policy Research Report. Washington, DC: World Bank. World Bank (2005a). Economic Growth in the 1990s:  Learning from a Decade of Reform. Washington, DC: World Bank. World Bank (2005b). Country Assistance Strategy for the Gabonese Republic for the Period FY05-FY09. Group Report 31882-GA. Washington, DC: World Bank. World Bank (2011). Improving the Odds of Achieving the MDGs:  Heterogeneity, Gaps and Challenges. Global Monitoring Report 2011. Washington, DC: World Bank. World Bank (2013). World Development Indicators 2013. Washington, DC: World Bank. Williamson, J. (1990). What Washington means by policy reform, in J. Williamson (ed.), Latin American Adjustment: How Much Has Happened? Washington, DC: Institute of International Economics.

Pa rt  I I

M E T HOD OL O G IC A L I S SU E S

Chapter 16

Principles of E c onomi c s African Counter-Narratives Célestin Monga

16.1 Introduction Some of the most pervasive epistemological issues in the humanities and the social sciences concern the validity, legitimacy, and transferability of concepts, theories, and frameworks across places and time. Such issues have always sparked fierce debates among researchers and thinkers. Writers and poets have also joined in, offering their own views. Victor Hugo, who believed in a single human destiny beyond a community of subjectivities among all people, famously wrote in the Preface to his poetry volume The Contemplations: “It begins with a smile, continues with a sob, and ends with a trumpet blast from the abyss. A destiny is written there day by day. Is that the life of a man? Yes, and the lives of other men too. None of us has the honor of having a life all his own. My life is yours, your life is mine, you live what I live; destiny is one. Take this mirror and in it look at yourself. People sometimes complain about writers who say I. Speak to us about us, they cry. Alas! When I speak to you about myself, I am speaking to you about yourself. How is it that you don’t see that? Ah, you madman, who think that I am not you!” (1856: 59–60). Mainstream economists have long positioned themselves—perhaps unwittingly—as followers of Hugo, opting to promote the idea of globally representative agents, which allows for simple and elegant modeling techniques to describe and predict human thinking and decision-making. The French writer’s reflections can indeed be interpreted as laying the ground for the main assumptions of economic theory: if human beings basically share the same life journey, it can be inferred that whatever differences they may have in their thought processes and actions are negligible, and that their sometimes different assumptions and patterns of behavior are more or less irrelevant for comparative social studies. But wait, not so fast. Gates, a prominent African American literary theorist, who also happens to be a strong believer in cosmopolitanism and the intrinsic commonality of all broad human values widely

304   Methodological Issues shared by people across civilizations and time, takes a different view. While he agrees that people may all be driven by the same dreams and fears, he also suggests that their particular historical journeys and life experiences often lead them to conceptualize their thoughts and behavior differently, and to make decisions that may reflect time-specific and place-specific “cultural” backgrounds and imaginaries. In his magnum opus with the evocative title The Signifying Monkey, he makes the case for differentiating the variety of human experiences, and understanding the implications for research. He eloquently argues that it is possible and indeed necessary to locate within a particular social group a system of rhetoric and interpretation that could be drawn upon both as figures for a genuinely “black” criticism and as frames through which one can “read” theories of contemporary literary criticism. He writes: “the challenge of my project, if not exactly to invent a black theory, was to locate and identify how the ‘black tradition’ had theorized about itself.” (1988: ix). That statement raises the possibility of an “ethnic” approach to the quest for knowledge, or at least the need for adopting ethnic-specific lenses in the study of rhetoric and practices. Economists have long been either dismissive of or nervous about such thinking, which they perceive as unjustified legitimation of socially constructed differences in humans (agents)—and for good reason: accepting the prevalence of intrinsically different modes of thinking and patterns of behavior among people or social groups raises serious methodological questions and challenges the very boundaries of economics (Malinvaud 1991). At a more abstract level, it can also open the door to ethical questions and crude prejudices of the type that led many to believe that the “economic man” did not exist in Africa (Jones 1960). Fortunately, the terms of the debate have been clarified in recent years, polemics has subsided, and there has been much intellectual progress on such questions. Mainstream economists have done a much better job explaining the methodological reasons why “economic imperialism” should not be a source of shame but of pride in the social sciences (Lazear 2000). In doing so, they convincingly expanded the scope of their discipline. Even leaving aside the advent of the behavioral revolution in economics, it is now clear that economics has opened up (Colander et al. 2004) and enriched its traditional methodological tools by studying non-monetary interactions (Glaeser and Scheinkman 2003), social interactions, and social norms (Manski 2000; Loury 1977), and making the case for a more rigorous exploration of the variety of human motivations and behavior (Akerlof and Kranton 2010; Basu 2010; Akerlof 1984; Sen 1977). This chapter tackles some of these epistemological dilemmas and policy issues through a reconsideration of the basic principles of economics. It follows Mankiw’s didactic approach but comes up with a list of ten principles different from the one he proposes.1 Building on theoretical and empirical works carried out in the African context this chapter offers a series of counter-narratives to conventional economic thinking. It also highlights how some of the recent developments in economics are consistent with analyses made in the study of Africa’s economic experience.

1 

Since its release in 1997, Mankiw’s introductory textbook Principles of Economics, which features a list of ten principles (Appendix 1), has been widely seen as a consensual manifesto for mainstream economics. It has been a best-seller (deservedly so), selling over a million copies and has been translated into twenty languages.

Principles of Economics    305

16.2  A Reconsideration of  the Ten Principles Principle 1: The goal of the economic agent is not always to maximize profit—“Dogs may well have four legs, but they can’t go down two paths at the same time.” African proverb.

The question of the fundamental motivation of households and firms as agents is at the heart of economic theory. The dominant models of economic thought emerge from Adam Smith’s famous comment: “It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard for their own interest” (1977 [1776]: 26). Let us leave aside the veracity of this assertion in the framework of the Western societies of which Smith was speaking.2 What about the African context? Practically since the Portuguese navigator Vasco da Gama “discovered” the African coasts at the end of the fifteenth century, successive generations of Western explorers, colonists, traders, missionaries, anthropologists, sociologists, and economists have regularly described African economic agents as people with unusual ways of reasoning, irrational goals in life as well as irrational spiritualities, cosmogonies, and motivations, and therefore usually adopting methods of decision-making that economic science and theory would consider incoherent. Already in 1960, Jones pointed out with mordant irony the rapid and erroneous conclusions of the many researchers in various disciplines who found in the “strange” behaviors of the economic agents in Africa the “proof ” that those people had peculiar motivations. Commenting on the analyses of African behaviors reported by Western experts, he wrote: They have come home with stories of all sorts of peculiar, seemingly irrational responses to economic stimuli and innovation. Stories of farmers who refuse to harvest the cotton crop the government has required them to plant, or who refuse even to plant the food crops needed for their very subsistence, of cattle that are valued for the shape of their horns rather than for their flesh, of laborers who cannot be induced to work overtime even by attractive bonuses, of consumers who refuse to buy unfamiliar but nutritious foods available at low prices although their supply of the traditional staple is exhausted, of producers who react to higher prices or lower production costs by reducing their output, of prices determined by custom that are stubbornly resistant to changes in supply or demand, and of consumers who spend their money on spectacle frames without glasses, shoes to be worn slung over the shoulders, and nostrums that have no effect other than to turn the urine purple. (1960: 107)

Jones’ assessment led him to stress the parallel between the perception that these “Africanist” specialists had of African economic agents and the irreducible otherness that once allowed certain Westerners to deny the humanity of Negroes and slaves: Economists who have studied the societies of tropical Africa may find it particularly easy to attribute economic backwardness to the absence of economic motivation—of a desire for material things—in a people apparently as different from Europeans and Americans as the 2  Much has been written concerning this old question. For a detailed critical review, see particularly Sen (1977) and Basu (2010).

306   Methodological Issues African. Just as some nineteenth century Americans denied the Negro a soul in an attempt to justify slavery, so have some twentieth century Europeans denied the African Negro an economic spirit in an attempt to justify colonial rule. (1960: 108)

Entrepreneurship and trade exchanges have existed in Africa since time immemorial and a market culture has been very important there for millennia. No matter: skeptics continue to proclaim that the African markets have no historic depth, or that the transactions taking place there do not have the rationality observed elsewhere.3 Thus myths and fantasies still persist today concerning the supposed inability of Africans to conceptualize the future seriously and thus to save, concerning their propensity to live in the present and therefore to prioritize consumer goods, and concerning the weight of customs and traditions, which supposedly make workers incapable of understanding the notion of productivity or even of theorizing about laziness. It is assured that in a context the chronic low level of savings would reduce investment levels to those of savings coming from the exterior—particularly development aid; and the price mechanisms regulating offer and demand would almost never work. Nothing is farther from the truth. Like most economic agents elsewhere in the world, households, workers, and companies moving in the African context make decisions they see as optimal, based on the quality and cost of the information available to them and according to the circumstances and changes in supply and demand. But contrary to the suppositions of microeconomic theory, they are capable of simultaneously having multiple objectives that do not fit into the simplifying modeling techniques imposed by the maximization postulate. Their utility functions are much richer and more complex than what appears in textbook diagrams. Individual interest is not always their sole motivation. Altruistic, social, spiritual, or even philosophical objectives are equally at work in business relations. For instance, people give much of their time and money to social activities with the goal of helping improve the quality of life in their communities—this even when it is not necessarily in their own interest to do it. Throughout the continent, entire villages are equipped with rural paths, wells, schools, libraries, and dispensaries financed anonymously by citizens whose only profit is the feeling of having done what seemed logical to them in a context of chronic government failure. Even in the West, where the frenzied cult of profit is said to dominate social interactions and justify the theoretical postulate of the maximization of a utility reduced to its financial aspects, notions like the public good or the general interest are greatly valued in relations between economic agents, justifying the dictum often attributed to Winston Churchill: “We make a living by what we get. We make a life by what we give.” Resistant to the idea of having to seriously integrate these sorts of non-egoistical behaviors into microeconomic theory at the risk of invalidating it, some neo-classical economists try to trivialize their importance. So they explain that apparent acts of altruism are in reality self-interested and should be conceptualized as such for they simply express another 3  The very influential anthropologist M.J. Herskovits popularized the thesis that the populations of Eastern Africa had no notion of markets before the arrival of Europeans. And yet this thesis was invalidated by a great many travelers who had passed though the region before colonial times. See for example the accounts of sixteenth-century travelers reported by Quiggin (1949), or the travel journals of Speke (1863) and Livingstone (1875).

Principles of Economics    307 dimension of the satisfaction of those who commit them. In other words, economic agents who do not maximize their profits do all the same, through their generosity, maximize their own satisfaction and therefore their utility function. What is the actual fact of the matter? Might hedonism be the hidden motivation of acts of generosity on the part of many African economic agents? What drives African economic agents to the altruistic behaviors of not wanting to systematically maximize profit as anticipated by neo-classical theory? Is their altruism sincere or disguised? Are they guided by a pure and totally disinterested ethical ambition or rather by the enjoyment of a self-satisfaction that they derive from their own acts of generosity? Recourse to a bit of algebra will help to clarify the problem to be elucidated. Let us imagine two individuals, Amadou (A) and Binta (B), and the good x. Amadou maximizes his preferences as follows:

U t ( x ) ≡ (1 − r ) . δ A ( x ) + r . δ B ( x ) (1)

where δA(x) represents the level of material well-being that the consumption of the good x gives to Amadou, and δB(x) that which it gives to Binta. Recourse to the coefficient r allows us to introduce into the utility function the degree of importance that a person accords to the well-being of others, and thus to measure the satisfaction that Amadou derives from Binta’s happiness. Supposing that r is positive (>0), the moments when Binta’s preoccupations affect Amadou’s well-being and behavior could be identified. The greater the coefficient r, obviously the smaller (1 – r), which reflects the high degree of Amadou’s altruism. This altruism should be deciphered more precisely in order to understand whether it is simply the reflection of a will to please oneself, or whether it truly represents a pure wish to conform to an ethic of sharing. Indeed, Amadou is perhaps interested in Binta’s satisfaction because this gives himself pleasure—in which case his act of altruism actually corresponds to egoism, to the maximization of his own utility. To disentangle the motivations of the economic agent, the preceding equation must be formulated even more explicitly. Amadou’s utility function would then take the form:

U A ≡ (1 − r ) . δ A + r . f A (δ A , δ B ) (2)

where ƒA represents Amadou’s view of optimal sharing out, δA as above, the personal satisfaction he derives from the whole operation, and r (between 0 and 1) measures the relative importance he gives to his own interest, this with respect to a more equitable sharing out. In other words, in order to decipher Amadou’s behavior with precision and determine whether it originates in a genuine, principled adherence to equity or whether, on the contrary, it is a matter of the egoistical ambition to please himself by being generous (make others admire him or make him feel good about himself, for example), one must be able to measure the coefficient r precisely and also understand the nature of the function ƒA. Empirical proofs enabling the systematic elimination of one thesis or the other do not exist in economic literature on Africa. But work in other areas of the social sciences tend to invalidate the old theory of archaic exchange formulated by Mauss (1925), suggesting that the gift is never free. Ela (1982), for example, shows on the contrary that this strictly materialistic view of exchange relationships described by Mauss does not correspond to what can be observed in African villages. Principle 2: Preferences are not stable—“When you’ve eaten salted food, you can no longer eat without salt.” Bantu proverb.

308   Methodological Issues Microeconomic tradition is based on the postulate of preference stability, that is on the idea that the choices made daily by economic agents are the result of a stable evaluation of alternatives, according to the degree of satisfaction, happiness, and utility they offer. The logical implication of this postulate is that the decisions of consumers, for instance, are necessarily optimal from their point of view (whether real or imaginary) for they reflect an underlying organization of the values and priorities accorded to each good or service. Economists do not judge the pertinence of the preferences thus expressed or the agents’ choices, but consider them to be sufficiently stable to constitute the base of the theory of the decision to act. When they formulate models of economic behavior, they therefore assume that preferences are given and stable. And when an analysis of the empiric data indicates that the consumers’ or producers’ preferences have changed, this fact is attributed to a change in circumstances—traditionally, the economist thus concludes that people make different decisions simply because they are faced with different choice options. In actuality, this postulate of preference stability adopted by microeconomic theory is above all imposed by the methodological coherence it offers: from an econometric point of view, it is indeed easier to model the behavior of an economic agent if one assumes that his preferences are not subject to random change. But careful attention to the modes of decision-making of economic agents in the African context and elsewhere shows that this methodological postulate is far too simplifying. “When you’ve eaten salted food, you can no longer eat without salt,” asserts a Bantu proverb, as if to make explicit the economics of a feeling of habituation suggested by the idea of a baseline of well-being. One implication is that people often tend to give a greater negative value to material and monetary losses than to value gains of the same size positively. Thus, a farmer will be unhappier about losing a few acres of his cotton production than he will appreciate the gain of an equivalent area of land. Equally, a man with ten heads of cattle would see far more harm in the idea of losing five of them than he would appreciate the possibility of acquiring five more. This kind of attitude can be explained by the fact that very broad segments of populations have experienced the violence of complete destitution and are always apprehensive about operations that imply material or financial loss. This kind of attitude is not limited to Africa. Thaler (1980) called it the endowment effect to explain the fact that people tend to give greater value to what they possess. An empirical experiment conducted by Tversky and Kahneman (1991) confirmed the finding that when it comes to money and in many other fields, people accord about twice as much importance to losses as to gains of the same size. An essential aspect of human nature emphasized by psychologists, but which neo-classical economics is slow in taking into account is the fact that people tend to judge their material situation not in absolute terms, but in accordance with a precise reference level. Rabin explains this with a nice metaphor: “The same temperature that feels cold when we are adapted to hot temperatures may appear hot when we are adapted to cold temperatures” (1998: 13). In other words, people are more sensitive to changes in their well-being than to absolute levels measuring it. Now it so happens that the traditional concave utility function found in all works of microeconomics is unable to illustrate such an attitude toward risk. This bias in favor of the status quo (the fact that some African economic agents have propensity to avoid taking even limited risks) ought to be expressed graphically in a sudden change in slope of the function at the reference point. The existence of standard points of reference or levels of well-being to which many economic agents in Dakar or Djibouti refer has a second implication: marginal changes in the quality of satisfaction they experience (and thus in their utility level) do not all have the same

Principles of Economics    309 importance. These changes weigh more heavily on the scale of measure when the economic agent is close to the initial baseline. They are less important as one moves away from it. In other words, the slope of the utility function flattens as one’s level of wealth moves away (positively or negatively) from the point of reference. Rabin attempts to explain the same idea with a bit of exaggeration: “We are more likely to discriminate between 3° and 6° changes in room temperature than between 23° and 26° changes” (1998: 15). But it is much more than just that. The fact that one is less and less sensitive to marginal changes in well-being has interesting implications for microeconomic theory in Africa: it ought to take into account the way in which today’s behaviors affect points of reference in the future, and also the economic agent’s feelings when faced with changes to his baseline utility level. These observations could significantly enrich microeconomic theory—and also make it more complex. It ought to be less rigid and static, and integrate into the classic analysis of utility other factors like the habitual level of consumption to which the economic agent is accustomed, or even the consumption level of certain members of the community to whom the individual compares himself when he seeks to increase his standing. For the degree of satisfaction or utility at the given moment t depends not only on the level of well-being at that particular moment w t, but also on a baseline that enables one to measure the weight of the change, the progress not only in comparison to one’s own initial situation but also to that of another economic agent to whom one compares oneself as a benchmark. The utility function of the “typical” African economic agent should then not simply be stated as u(w), as formulated in textbooks, but rather u(st;w t), with st representing a point of reference determined by the degree of satisfaction in the past as well as by anticipations of the future level of well-being. As additional proof that the behaviors of African economic agents are actually not at all incongruous, this approach to the study of preferences as dynamic processes including points of reference is comparable to those adopted by Duesenberry (1949) and Ryder and Heal (1973). Principle 3: Poor people face difficult Cornelian trade-offs that lead to intergenerational traps. “Love is a despot who spares no one.” Namibian proverb.

Microeconomic theory is based on the idea that every economic agent is permanently confronted with choices and must thus constantly make trade-offs, even when he decides to abstain—for doing nothing is also a choice. This obligation to express preferences and make decisions that are at times difficult but that best satisfy each person’s utility function has led Milton Friedman to assert that “there is no free lunch out there.” In other words, to obtain one thing, it is always necessary to renounce another. Like many founding principles of economics, this postulate is simply a matter of common sense. The constraint constituted by the limitation of the resources at the disposal of every household or firm (financial and time constraints, or others) automatically drives each to engage in conscious or subconscious trade-offs when expressing choices and making decisions. From this point of view, African economic agents are not particularly different from those in other regions of the world. Nevertheless, certain particular and often age-old social and political practices have bequeathed them a collective imaginary that sometimes defies the forecasts of microeconomic theory. It is important to emphasize that these particular social and political practices are not unchangeable, and that this collective imaginary is not immutable. No generalization would therefore be warranted here.

310   Methodological Issues But the particularities of the economic environment, institutional organization, and structure and regulation of markets in Africa often drive agents there to adopt “deviant” positions and behaviors that surprise the traditional microeconomist. This is so because in an environment where risk is the dominant rule and uncertainty imposes a particularly high premium on any economic operation, estimating the opportunity cost of each decision is a difficult challenge for economic agents. In practice, it is always inevitably necessary to choose between several alternatives—without even realizing it. In the African context, the choice is often an implicit non-choice imposed by circumstances but clearly accepted, and not a deliberate act of renunciation of one thing in favor of another. An example of deviation from the behavior ideal predicted by theory is the illusion often held by certain agents of being able to spare themselves the difficulty of choosing, the illusion of believing that by giving up the idea of choosing between unattractive and even less attractive options one can escape the chore. Great African novelists like Kane (1963), Achebe (1959), and Labou Tansi (1983) have popularized these ways of thinking and acting in best-selling novels. In popular imagery, this (utopian) “refusal” to clearly formulate a preference in certain situations is presented as the elegant decision to passively and serenely accept a fate that is not considered fortuitous. For “things are the way they should be.” The justification of such an attitude is at times even theorized: it is the permanence of destiny. This is not necessarily fatalism, an attitude that includes an intrinsic share of sadness and regrets. It is rather a voluntarism of inaction, deliberately accepted and considered an attitude of wisdom and intelligence that minimizes futile gesticulating and useless efforts. For, it is said, the worst should never be feared because, although it’s probable, it’s never sure … This principle of non-choice—or more exactly of choice not to choose—expresses itself in modes of decision-making and manners of interaction that surprise uninformed people. It is a sort of nihilism that at times gives rise to behaviors that would be judged antieconomic in the West. Another illustration of African deviations from the microeconomic theory of choice concerns the question of child labor and the apparently surprising decisions adopted by some heads of family in the poorest segments of society. This situation can be defined as being the dilemma of the poor. To understand it, it is useful to know that in Africa as elsewhere, poor and non-poor families organize themselves differently to manage the risks that confront them. When a high-income household is faced with a negative shock like a sudden drop in income (salary or capital gains), its typical reaction consists in recourse to savings or borrowing in order to more or less maintain its usual level of consumption, even if this falls slightly. Its initial level of wealth and well-being and its social status thus allow it to resort temporarily to access to the credit market in order to soften the shock. Once the crisis has passed, this well-to-do household can even reduce its consumption level a bit to reconstitute its savings or pay off the credit obtained. In the end, its consumption is therefore only rather marginally affected, and the mid- or long-term consequences of this crisis are negligible. Things are totally different for poor households. In general they have no (or very little) savings and do not have access to credit because the banking and financial market does not function correctly for them. Faced with a negative shock, these poor households are often forced to make two sorts of decisions that worsen their already precarious situation: they must drastically reduce their consumption and often even sell their means of production to continue to survive. Now even if such decisions allow them to maintain a certain level of consumption, they deprive them of future sources of income.

Principles of Economics    311 The crisis therefore creates a permanent loss of productive capital for poor farmers who must, for instance, sell their cattle in order to continue to eat. For this segment of the population, the negative effects of the crisis persist even when the initial shock has passed. As Bourguignon notes, “a household in that situation can very well fall below the poverty line, not only monetarily at the time of the shock but for as long as it has not reconstituted its lost productive assets—if it is ever able to do so. That is an illustration of the poverty trap, or of the persistence of the effects of a negative shock on poor households” (2006). In some cases, the poverty trap can have intergenerational ramifications. This is notably seen in poor farm families when a drop in income forces them not only to sell productive assets like cattle but also to take their minor children out of school. In other words, the main response of the poor to economic hardship is to work longer hours and, in some rural areas where children constitute an important fraction of the labor force, to pull them—especially young girls—out of school. It is a Cornelian tradeoff for these poor families, especially when their income decreases below a level where they can barely survive. Survey results indicate that the probability of pulling children out of school increases rapidly for families near the poverty line. This is not because of some strange behavior of the parents (Basu and Van 1998). In many poor provinces of the country, children constitute indispensable additional manpower and sources of family income. In the absence of social safety nets for the elderly, they are also the only real assets in the community. On Burkina Faso’s Mossi Plateau, where deforestation has necessitated greater time input by females to obtain enough traditional energy for cooking, in the northern provinces where environmental degradation has the same effect for water, similar effects have been observed. The Cornelian tradeoff facing the poor can be expressed as in Figure 16.1. The situation of the poor Burkinabè household facing economic hardship and having a probability of P* (or higher) to pull their children from school presents public policy with a challenge. On the one hand, the goal should be to convert the CC curve into a straight line closer to the x-axis (CC) so that more children can stay in school irrespective of the Probability of pulling children out of school

Poverty Line

Median Income

C C’

P∗

Pm U

Pr

C

Prevailing situations: the sad reality

C’

Realistic expectations from policy U Utopian ideal

Family Income

Figure  16.1  Maintaining subsistence income or schooling children? Cornelian tradeoffs in Burkina  Faso.

312   Methodological Issues family income. This would imply strong and well-targeted social safety nets that are effective enough to induce important positive changes in the behavior of poor households, even when confronted with negative shocks. On the other hand, or perhaps simultaneously, the average income should be improved to raise the poverty line to levels where even households considered poor are less likely to pull children from school (from P* to Pm). It is reasonable to assume that such a goal can only be achieved over time, with P* decreasing to Pr in the medium-term, when family income increases enough for children to stay in school and when returns to education are widely acknowledged by society. Principle 4: To be rational is at times to choose stupidity—“The cook does not have to be a beautiful woman.” Shona proverb.

At least since Adam Smith, one of the fundamental topics of theoretical discussion in economics has been whether and to what extent egoistic behavior by individuals can achieve general good. In trying to address that central question, researchers have debated the related issues of utility maximization, stability of preferences, rationality, and market discipline, which constitute the core methodological assumptions of modern economic theory. Thinking about the conception of man in economic theory and models, some have adopted the so-called “first principle of economics” articulated clearly by Edgeworth, according to which “every agent is actuated only by self-interest”4 (1881: 16). Others have rejected that blind faith in a rational representative agent and the debate has being going on for centuries (Hartley 1996). Perhaps a clear indication that these discussions have not yet been settled convincingly is the fact that many good economists who have studied these matters at length still weigh in on different sides of the argument. Shiller raises the following intriguing questions: Are people really rational in their economic decision making? (…) Does it make sense to suppose that economic decisions or market prices can be modeled in the precise way that mathematical economists have traditionally favored? Or is there some emotionality in all of us that defies such modeling? This debate isn’t merely academic. It’s fundamental, and the answers affect nearly everyone. Are speculative market booms and busts—like those that led to the recent financial crisis—examples of rational human reactions to new information, or of crazy fads and bubbles? Is it reasonable to base theories of economic behavior, which surely has a rational, calculating component, on the assumption that only that component matters?5 (Shiller 2014) 4  Despite that bold statement, Edgeworth appeared torn by the issue, and contradicted himself. He also wrote: “the concrete nineteenth century man is for the most part an impure egoist, a mixed utilitarian.” (1881: 104). See Sen (1977) for a critical analysis of Edgeworth’s views. 5  Shiller also offers a quick survey of the opinions of his fellow Nobel laureates, observing that even among the three researchers who shared the 2014 Nobel in economics, there were fairly different opinions on that issue: he argued that aggregate stock price movements are mostly irrational, and that ideas about non-rational or irrational behavior from other social sciences (psychology, sociology, political science, and anthropology), should be incorporated into economics. Similar arguments were made in Monga (2011). By contrast, Eugene Fama presented the findings of his many years of empirical work in strong support of the notion of economic rationality. For instance, he offered evidence suggesting that share prices respond almost perfectly to information about stock splits and that interest rates contain rational forecasts of inflation. Lars Peter Hansen took a centrist position in the debate, referred to the “distorted beliefs,” “animal spirits,” and “overconfidence” that, in his view, explain some otherwise incongruous empirical evidence about the behavior of actors in financial markets. “I have been studying Nobel lectures

Principles of Economics    313 Convincing, clear-cut responses to such questions are not straightforward. Part of the problem is the definition and scope of rationality. Economic agents in Africa often display too well the kind of “rational inattention” (Sims 2003) that leads them to deliberately ignore or neglect information that could help them make “better” decisions. They do so because they have better things to do with their time and their capacity to process all relevant information at their disposal is limited. They believe that the highest levels of efficiency in decision-making always require accepting a healthy dose of inefficiency. “Economists have long recognized the need to relax the assumption of optimizing agents by introducing an assumption that translation of observed external random signals by households and firms into actions must represent a finite rate of information flow; that is, economic agents are finite-capacity channels” (Sims 2010). That line of thinking is certainly valid but insufficient to analyze the wide variety of meta-rational reasons that underline the patterns of behavior of agents in Africa. True, if drivers and motorists in Douala, Cameroon, do not stop at red lights because they have little time and the likelihood of being caught by a policeman or a speed camera is nearly zero—though they know they will risk causing accidents and getting hurt themselves, that is a form of stupid and perverse rationality (or foolish rationality à la Amartya Sen). Others would even argue that the red lights, stop signs, and other traffic rules symbolize a failed state that had been incapable of delivering the most basic public services to the people. In that case, violating almost any rule and regulation enunciated by an authoritarian and illegitimate government is a heroic act of defiance and rebellion, a proclamation of freedom (Monga 1996). However, many of those bad drivers are simply lazy and bad citizens who believe that complying with traffic law is an unpleasant waste of time. This makes the classification of their behavior more complex, and raises difficult questions about the true determinants of their decision-making processes. Is all this still rational inattention? Another type of inattention often observed with bad drivers in African cities is even more perplexing: it is the propensity to deliberately run through red lights, stop signs, and yellow lines, knowing that the probability of causing injuries to themselves and mortal accidents is quite high. Clearly, microeconomic theory has not yet found answers to such patterns of behavior and may not be able to do so without drawing from research in other disciplines.6 Principle 5: Social interactions determine economic decisions and outcomes—“The true poor person is he who has no one.” Serer proverb (Senegal). of our predecessors, and the debate doesn’t seem new, write Shiller. Judging from their words, many laureates—including Herbert Simon in 1978, Maurice Allais in 1988, Daniel Kahneman in 2002, Vernon Smith in 2002, Elinor Ostrom in 2009 and Olivier Williamson in 2009—have questioned whether economic actors are rationally pursuing self-interest, as traditional economic theory assumes” (2014). 6  For instance, developments in the neurosciences over recent decades reveal the consistent links between emotions and rational thinking. Ernst Fehr’s work, which focuses on the proximate patterns and the evolutionary origins of human altruism and the interplay between social preferences, social norms and strategic interactions, provide empirical evidence for such linkages. His research combines game theoretic tools with experimental methods and the use of insights from psychology, sociology, biology and neuroscience for a better understanding of human social behavior. He has convincingly shown the impact of social preferences on competition, cooperation and on the psychological foundations of incentives, and highlighted the role of bounded rationality in strategic interactions and on the neuroscientific foundations of social and economic behavior. See Fehr et al. (2013, 2005) for an empirical examination of the behavioral consequences of authority and power and their motivational origins,

314   Methodological Issues One of the uncomfortable contradictions of modern microeconomic theory lies in its conflicting treatment of social interactions. On the one hand, the theory acknowledges that individuals and firms are indirectly influenced by the decisions and behavior of other agents—such influence determines relative prices in all markets (goods and services, labor, and capital). On the other hand, because of its foundational principle of methodological individualism, the theory assumes that is individual decisions and behavior are not directly influenced by what other people do … Again, some of the mythical characters of African literature such as Achebe’s (1959) Obi Okonkwo or Kane’s (1963) Samba Diallo epitomize the direct influence of social thinking and rules on individual preferences, actions, and performance. Anthropologists and sociologists have long presented evidence that social norms, prevailing cultures, and the collective consciousness reflecting particular times and places, all affect and determine individual behavior directly—not just indirectly. People everywhere in the world very often make consumption, production, and even investment decisions based on what others are doing, not as a strategic reaction but as a “rational” act of mimicry. That is why movie star George Clooney’s pictures are used in advertisement campaigns to sell perfume. That is why soccer star Lionel Messi and basketball legend Kobe Bryant are paid large amounts of money to appear in television commercials for an airline company. Potential consumers know very well that neither athlete is an aerospace engineer or a travel expert. Still, the public is happy to associate with them and many educated people would fly an airline company simply because it features celebrities in commercials. The economic force of the prevailing social norms is also obvious in the African context. For instance, the South African Smoking Survey 2013 confirms what has been observed elsewhere: most smokers start young, in their adolescence or early 20s, when peer pressure is felt most acutely. And while nearly all respondents saying they associated the habit with lung cancer and other respiratory diseases, 46 percent of the respondents said they continue to smoke simply because it is an important social ritual for them (Rose-Innes 2013). An additional indication that individual behavior on this matter is directly determined by social norms is the finding that a good support network has been shown to be enormously helpful in quitting. Likewise, the main reason given by respondents for relapse was being in the company of others smoking. Such survey results are similar to those carried out elsewhere in Africa and, indeed, around the world. The entire advertisement industry is built upon the ample empirical evidence that people’s tastes of what is perceived to be a “good” product or service often depends on beliefs held by others. These truths, which are inconvenient to many traditionalist economic modelers, go well beyond consumption and other marketable goods and services. Even in the most private sphere of their lives (to use the categorization offered by Habermas 1991), people think and act consciously or unconsciously in a way that reflect constant and deep social influences: they often pick their religious affiliations in conformity with family “traditions” and culture. Their political opinions and practices are often influenced by the preferences of others whom they trust or even peer pressure (friends, family, colleagues, etc.). Various

and the neuroeconomic foundations of trust and social preferences. Rubinstein, who is “not a big fan of neuroeconomics” but finds it “appealing,” warns about the rush by “some brain researchers to use economic terms without fully understanding their subtleties.” (2008, p. 486).

Principles of Economics    315 institutions—from families to religions and governments regularly shape the preferences of individuals and lead them to think in a particular way or adopt particular norms. In sum, social influences and pressures affect the African homo economicus and agents everywhere in the world in ways that that challenge and invalidate the basic microeconomic principles. While economists have long recognized such inconsistencies of the fundamental microeconomic framework (Veblen 1934; Schelling 1978), they are only starting to devote the appropriate attention to their analytical implications and how they should affect theoretical modeling and policymaking. But the richness of the African economic experience suggests that there is still a lot of research to be carried out before one can be satisfied that the microeconomic theoretical corpus appropriately reflects the scope of human economic behavior and a good understanding of decision-making processes. To be sure, economists have long studied externalities and situations in which the actions of some agents affect the set of feasible actions available to other agents. They have also explored interactions between agents in small groups or networks. Interesting work has been made on the study of such issues since Becker and Murphy (2000) under the title of social economics.7 But economists restrict that label to “the study, with the methods of economics, of social phenomena in which aggregates affect individual choices.” (Benhabib et al. 2010: xvii). Yet the problem is precisely with the traditional methods of economics, which are still sacrosanct in social economics. It is therefore not surprising that much of the work done under the label of social economics rely mostly, although not exclusively, on rational choice theory. As a result, intellectual progress in the study of social norms, practices, and conventions, has aimed at maintaining the analytical difference between the sub-field of social economics and economic sociology, which focuses on the study of economic phenomena using the methods of sociology. Excessive reliance on the rational choice paradigm (with its foundational assumptions of utility maximization and equilibrium in the behavior of groups) makes social economics as currently conceived a narrow field, with analyses often dominated by market fundamentalism. Even important new steps aiming at studying social interactions among agents—when their actions at equilibrium are social in the sense that they are not mediated exclusively by markets—tend to take their preferences as given (Jackson 2010). Microeconomic theory would be enriched and strengthened by the study of peer, family, and neighborhood effects in various African contexts. Principle 6: Decent employment is a prerequisite to economic security, stability, and social peace. “You learn how to cut down trees by cutting them down.” Bateke proverb.

For centuries there was wide consensus among economists that employment is a critical pillar for any growth and poverty reduction strategy. With technological progress and increasing productivity gains from capital and new processes there is now strong belief that the world economy has entered a new era in which labor (especially low-skilled labor) contributes less and less to the growth process. MIT’s Brynjolfsson and McAfee have argued convincingly that impressive advances in computer technology—from improved industrial 7  In their thematic breakdown of their New Palgrave Dictionary of Economics, Eatwell et al. (1989) devote a volume to social economics. Their general approach is similar to Becker and Murphy’s and builds on Wicksteed’s (1910) contention that there is no “economic motive as such,” distinct from other motives, and that a person’s allocation of time or physical and mental effort to all her activities (including, for example, her prayers and her devotions) is governed by precisely the same principles as her allocation of monetary expenditure among market commodities.

316   Methodological Issues robotics to automated translation services—are opening up an era of even higher growth rates for the world. They contend that the global economy is on the verge of a substantial growth spurt driven by smart machines that take full advantage of advances in computer processing, artificial intelligence, and networked communication. They also warn that these new developments are largely behind the sluggish employment growth of the last 10–15 years (Brynjolfsson and McAfee 2014). Is the era of steady, full-time employment indeed over? Should the world economy be prepared to accommodate the notion that increasingly large fractions of the labor force will never be able to find decent, formal sector employment? The success of a handful of countries that are able to use resources from extractive industries as a source of financing for generous and well-functioning welfare systems (i.e. the Scandinavian countries) or social safety nets (i.e. several oil-rich Gulf countries) also seems to give credence to those predictions. Africa’s experience and economic trajectory are unique and likely to remain so for the foreseeable future: Most workers are trapped in very low productivity activities in subsistence agriculture and the informal sector; about two-thirds of the continent’s population is under the age of 24 and is underemployed—including those with college and university degrees. With population growth projected to be 2.2 percent in the next 25 years, the African private sector faces the challenge of creating employment opportunities to absorb the youth bulge—no amount of mineral wealth would suffice to provide high-quality, sustainable Gulf countries-type social programs and public sector pensions to such a large number of people. Given Africa’s level of economic development, economic, social and demographic structure (with a large youth bulge), full employment of low-skilled workers will remain critical concerns across the continent—for policymakers, the business community, and the more than 1 billion people striving to provide for their families. Beyond the traditional economic rationale, the stability of African polities and social structures also require that the largest fractions of the continent low-skilled labor force be put into decent employment, mainly in the formal sector. Yet sub-Saharan Africa has the lowest rate of formal sector employment in the world.8 Decent formal sector jobs generate not only income and GDP growth but also provide platforms for improving skills and building human capital, and constitute the main vectors for people’s ardent quest for dignity, which help them become better citizens (Figure 16.2). While the old-fashioned, industrial times depicted in classic movies such as Charlie Chaplin’s (1936) Modern Times may be features of the past, it can be said that Africa is still the place where labor-intensive industries that create employment for low-skilled workers have good perspectives for expansion. To understand why, one must look beyond official definitions and statistics of unemployment. Africa’s labor markets typically display very low unemployment rates, yet most of the labor force feels unemployed. Economists have long struggled to make sense of involuntary unemployment and to come up with convincing theories that explain a phenomenon still hard to grasp conceptually.9 The main intellectual obstacle they have not been able to overcome satisfactorily is to reconcile 8  Gallup Surveys conducted in 129 countries in 2009 and 2010 indicated that less than 20 percent of sub-Saharan Africa’s labor force was employed full time for an employer, compared to 35 percent in Asia, nearly 40 percent in the Middle East and North Africa, and well over 50 percent in the Americas, Europe, and the countries of the Former Soviet Union. 9  This section draws on Monga (2013a).

Principles of Economics    317

Income

Skills

Jobs

Dignity

Citizenship

Figure  16.2   Good jobs generate a virtuous circle of economic and social benefits. the popular, intuitive conception of unemployment with the official definitions commonly used by government statistical agencies and the International Labor Organization (ILO). The ILO defines an unemployed as a member of the labor force who meets the following criteria: not employed during the past seven days, even for one hour; looking for work; and available for work. The underemployed are the unemployed plus those who are employed part time (less than 30 hours per week) and want to work full time.10 One realizes that something is wrong with the official numbers of unemployment when countries such as Nigeria, Cameroon, Kenya, Zambia, or the Central African Republic score much better on that criterion than the United States, France, Italy, Spain or South Africa. The story changes when the coin is flipped and one measures the portion of the labor force employed full time for an employer (those working for an employer at least 30 hours per week): Africa’s general performance is suboptimal. Survey results in Africa show consistently that most people define involuntary unemployment as situations where workers are unsuccessfully looking for jobs at the prevailing wages when they are as qualified as those holding these jobs, or where workers are willing to work 10 

According to ILO’s standard definition from the 1982 Thirteenth International Conference of Labour Statisticians, the “unemployed” comprise all persons above a specified age who during a reference period were: (i) “without work”, i.e. were not in paid employment or self-employment; (ii) “currently available for work”, i.e. were available for paid employment or self-employment during the reference period; and (iii), “seeking work”, i.e. had taken specific steps in a specified reference period to seek paid employment or self-employment. The specific steps may include registration at a public or private employment exchange; application to employers; checking at worksites, farms, factory gates, market or other assembly places; placing or answering newspaper advertisements; seeking assistance of friends or relatives; looking for land, building, machinery or equipment to establish own enterprise; arranging for financial resources; applying for permits and licenses, etc. The national definitions used vary from one country to another as regards inter alia age limits, reference periods, criteria for seeking work, treatment of persons temporarily laid off and of persons seeking work for the first time.

318   Methodological Issues at less than the prevailing wages for jobs which they could usefully fulfill, but are unable to find such jobs. Such situations (let’s call them Type A) raise issues of economic inefficiency and social injustice: large segments of the population, often with useful skills, are kept out of the productive economy, which obviously makes the process of national wealth creation suboptimal. This is true even if the unemployed are less productive than those who hold employment. Furthermore, the people left outside the economic system are often a burden to those who are employed and feel disenfranchised, which makes them a potential source of social instability and a constant threat to social peace. Government agencies and international organizations usually approach and count unemployment differently. Broadly speaking, they only focus on people of “working age” (whatever they choose it to be), who are out of work (whatever is considered “work”) and capable of submitting evidence (whatever is deemed acceptable) of having looked for work in the recent past. Clearly, it is possible for a person to qualify for these Type B situations and not qualify for Type A; for example it is possible for a skilled worker who has been laid off and is looking for a job not to qualify for inclusion in unemployment statistics because he/she is temporarily helping in a family business (and therefore technically not out of work) or has not been able to produce “acceptable evidence” of having looked for a job. And vice versa: it is possible for a young college graduate who is officially included in unemployment statistics to actually be seeking work at real wages in excess of the threshold of his/her market value, regardless of his/her potential contribution to society. Such plausible scenarios raise several questions that economic theories of unemployment have attempted to answer over the decades, generally with little success: Why are market economies often unable to provide all the jobs that people are looking for? Why are certain economic systems unable to attract workers who would be willing to demand real wages that fall short of their potential contribution to society? Why are firms unwilling or unable to capture the economic opportunities that are associated with the existence of a large pool of unemployed workers? Why are involuntary unemployed workers unwilling or unable to employ themselves (and others) by starting new firms or to underbid their employed counterparts? etc. Defining unemployment requires that one implicitly assigns value to various types of work, occupation patterns, and sources of income. This supposes that one also makes subtle value judgments about topics that are actually outside the typical boundaries of economics, as it implies dealing with issues of interpersonal comparisons of well-being: What is to be considered “work”? Who defines it, and for whom? When does any activity rise to that qualification? Do subjective perceptions have a role in the definition of work? Should a worker who thinks of him/herself as unemployed but occupies part of his/her days in temporary, unfulfilling pastime activities (say, to “stay away from trouble”) be truly considered an active member of the labor force? Does any occupation necessarily allow the person who holds it to develop useful “soft skills” that may be of use over the course of their lifetime? Such questions fall far beyond the realm of the economic discipline and answering them adequately would require that researchers be willing to venture into new territories and engage deep philosophical questions. While that may be the right approach to economic research and policy, mainstream economics has so far stayed away from methodological adjustments that imply embracing cross-disciplinary complexity (Monga 2011).

Principles of Economics    319 Unemployment statistics shed little light on the anatomy of African labor markets (types of jobs available, fastest/slowest job-creating industries, relative shares of full-time and part-time workers, formal/informal sectors, wage earners and self-employed, social groups, and gender balance in the labor force, etc.) and their dynamics over time. It is therefore necessary to refocus the analysis of labor market issues from unemployment to employment. Employment appears to be less complicated to measure, especially in the high under-unemployment environment of the developing world. Another reason for focusing on employment is its strong conceptual and economic appeal: analyzing job creation instead of experience with unemployment offers better insights to policies that may be necessary to sustain employment growth. That change of focus is important in general and particularly in the African context. Moreover, the global picture of unemployment generally shows a weak relationship with GDP per capita, and the year-over-year change in the unemployment rate has a weak relationship with GDP growth across countries. By contrast, employment appears to be strongly correlated to with GDP growth. If the relationship between unemployment and economic performance is not easily tractable, at least in the short term, unemployment statistics and theories may be less meaningful in the African context than often thought, and the analytical focus should be on tracking information on employment and wages. Because the search for solutions to unemployment and underemployment has too long focused on the wrong questions, traditional, mainstream remedies have failed to provide useful answers to developing country policy makers. Researchers should now draw lessons from economic history and the experience of other countries where structural change has involved a variety of industrial processes. The transformation of the world economy and the emergence of large developing countries open up new possibilities for latecomers. African countries can accelerate the shift of labor from low-productivity jobs in agriculture and the informal sector to higher-productivity jobs in agro-industry, manufacturing or tradable services and achieve sustained growth and poverty reduction. But in order to do so, new and more strategic forms of industrial policies that avoid the pitfalls of the past must be designed and implemented. Principle 7: Sustainable and inclusive growth requires structural transformation and vice versa. “A good deed reaps another.” Kikuyu proverb.

Over the past decades, the deep intuition of theoretical research and the true ingenuity of empirical work in growth economics have taught us much about the mechanics of growth and its broad determinants. But while their theoretical macroeconomic analyses have provided important clues to the mystery of growth, major gaps remain in our knowledge. In fact, despite all the production of very sophisticated theoretical models and the many insights from carefully designed cross-country regressions, policymakers in developing still do not always have convincing answers to their most burning questions about actionable policy levers. Rarely has a country evolved from a low- to a high-income status without sustained structural transformation from agrarian or resource-based towards an industry or services-based economy. Yet few African countries have been successful in this transformation. Even those that have slowly reached middle-income status are facing the so-called “middle income trap” and are struggling to break out of it. Africa’s economic experience suggests that the fruitful era of growth research based on traditional macroeconomic models and regressions may be over, and that there is a need to shift the focus towards more systematic analyses

320   Methodological Issues of the micro-foundations of growth. Understanding the mechanics of dynamic, sustained, and inclusive growth requires a careful study of the process of how an economy’s structure evolves over time. Economists have learned a great deal from various waves of growth research (Monga 2012). On the theoretical front, the analysis of endogenous technical innovation and increasing returns to scale has provided economists with a rich general framework for capturing the broad picture and the mechanics of economic growth. Robert Solow’s work has shown the importance of the role of capital accumulation (both physical and human) and technical change in the growth process. From contributions by Becker, Heckman, Lucas,11 and many others, we also learned about the importance of human capital through diffusion of new knowledge or on-the-job learning, often stimulated by trade, and the so-called college wage premium. From work by North (1981), with supporting theoretical and empirical analyses exemplified by the works of Acemoglu et al. (2001) and Greif (1993), we have learned that growth is in large part driven by innovation and institutions that have evolved in countries where innovative activity is promoted and conditions are in place for change to take place. From endogenous growth theorists, we have understood the need to change the focus of growth theory from accumulation to knowledge creation and innovation. In sum, we know quite a lot about some of the basic ingredients of growth. On the empirical side, the availability of standardized data sets—especially the Penn World tables—has stimulated interest in cross-country work that highlights systematic differences between high-growth and low-growth countries with regard to initial conditions (such as productivity levels, human capital, demographic features, infrastructure, financial development, and inequality), institutional variables (such as rule of law, protection of property rights, and governance indicators), and policy variables (such as macroeconomic stability, financial regulation, or trade openness). However, growth research still faces significant challenges in identifying actionable policy levers to sustain and accelerate growth in specific countries. A major challenge is the very large number of possible determinants of growth. The availability of data has led to an emphasis on cross-country regressions, which only identifies “average” relationships. Yet, micro-studies and anecdotal evidence reveal that countries vary with regard to the conditions under which they are able to generate and sustain growth. Because many empirical studies only include a small subset of appropriate variables, cross-country analyses are too often plagued by important left-out variable errors. The obvious solution, which consists of including a more comprehensive set of exogenous variables in regressions, often leads to an open-ended list of determinants of growth. The classic framework, which consists of running cross-sectional regressions of the form Y = α + β1 x1 + β2 x2 + ... + βn xn + ε (3)



where Y is the vector of rates of economic growth, and x1 , x2 , … xn , are vectors of explanatory variables, has yielded results that are too generic to inform policymaking. In a famous paper, -Sala-i-Martin (1997) proclaimed, with humor and trepidation: “I have just ran two million regressions”! He then identified about 60 variables which have been found to be

11 

See, in particular, Becker (1992); Heckman (2006); Lucas (2004).

Principles of Economics    321 significant in at least one regression. That was nothing, compared to the 108 subsequently identified by Durlauf (2005), which led him to conclude that “approximately as many growth determinants have been proposed as there are countries for which data are available”! This reflects the bigger issue of model uncertainty that confronts growth researchers and the incentive that each of them has to emphasize the validity of a single tool from which they can generate a convincing story. But traditional inference procedures rooted on a single model carry the risk of overstating the truth, as the standard errors often understate the degree of uncertainty about the estimated parameters.12 This explains why the economics of growth has generated such a wide variety of methodological approaches, many of them contributing to some understanding of growth but also carrying their own set of unresolved new issues. Even the wide consensus among growth researchers on the importance of building knowledge and sustaining technological change has not translated into a common understanding of issues of non-rivalry and non-excludability, and how to address them. In high-income countries, it involves difficult trade-offs. Striking the right balance between policy measures to encourage technological change by setting the appropriate incentive system in place for innovation and profits, and at the same time making new ideas and discoveries available (freely or cheaply) to other agents in the economy is not an easy exercise. Granting patents for instance to stimulate innovation also poses a difficult trade-off problem in an economy where private agents should be encouraged to exploit all the knowledge available. The problem is easier to resolve for traditional private goods that are excludable and rival: the price mechanism represents a powerful tool for producing and allocating rival goods—strong property rights and anonymous markets generally constitute effective institutions. The case of non-rival goods is much more challenging because of the high level of uncertainty about the “right” institutions (those that favor wide and low-cost diffusion of technological advances while preserving incentives for creativity and innovation). One option has been to work out trade-offs where patents rights are allowed but in a narrow and finite way—that is, partial excludability. While this is an effective strategy for certain goods or services, it cannot constitute a general solution: certain types of non-rival goods have been given no property rights whatsoever; a typical example is mathematical formulas, which any knowledgeable private agent can use freely; other forms of non-rival goods such as books or movies are granted strong intellectual property protection. Another option is to create other mechanisms such as subsidies for R&D. Non-profit and government-supported institutions such as universities are then encouraged to produce new ideas. But this also raises new questions about the proper rule for distinguishing ideas from human capital. Becker (1993) has argued that human capital is just another ordinary factor of production and a private good comparable to capital or land, and that some of its claims for externalities are exaggerated—the logical conclusion being that there should be no government subsidies for the production of human capital. Yet, it is obvious that human capital constitute a precondition for the generation of ideas. Therefore, in order to stimulate the production of ideas, one needs to subsidize either the ideas themselves, or the inputs for their production.

12  Even the use of Bayesian or pseudo-Bayesian techniques for data analysis does not always address properly issues of model uncertainty. See Brock et al. (2003).

322   Methodological Issues Countries that find themselves in a second-best situation where social returns from new ideas are higher than private returns (a distortion often observed in African economies) may have legitimate reasons to use additional distortions such as subsidies to train the type of human capital their economies need. By helping the economy create a larger pool of engineers for instance, government intervention could actually lower their cost for businesses willing to use their talent for new ideas, and that would offset the initial distortion. The next frontier for growth research is therefore to conceptualize the various mechanisms and processes that encourage new ideas and the expansion of endowments (mainly labor and capital). As Romer (1999) puts it, “the economies that will really do well in the next 100 years will be the ones that come up with the best institutions for simultaneously achieving the production of new ideas and their widespread use.” It is therefore crucial for growth theory to differentiate between various types of non-rival goods, and adjust institutional structures and degrees of property protection accordingly. The disappointments of growth research—most notably from the perspective of policymakers seeking specific action plans to generate prosperity—have led to a reassessment of the validity and usefulness of existing knowledge, and to the development of radically new approaches. An important study by the World Bank (2005) focused on lessons of the 1990s highlighted the complexity of economic growth and noted that the reforms carried out in many developing countries in the 1990s focused too narrowly on the efficient use of resources, not on the expansion of capacity and growth. While they enabled better use of existing capacity, thereby establishing the basis for sustained long-run growth, they did not provide sufficient incentives for expanding that capacity.13 The report concluded that there is no unique, universal set of rules to guide policymakers. It recommended less reliance on simple formulas and the elusive search for “best practices,” and greater reliance on deeper economic analysis to identify each country’s one or two most binding constraints on growth. That line of research is exemplified by the growth diagnostics framework, which aims to identify the one or two most binding constraints on any developing economy, and then focus on lifting those (Hausmann, Rodrik, and Velasco 2008). It offers a decision tree methodology to help identify the relevant binding constraints for each country but it is concerned mainly with short-run constraints. In this sense, its focus is on igniting growth and identifying constraints that inevitably emerge as an economy expands, not on anticipating tomorrow’s constraints on growth.14 That approach sheds light on the need to prioritize reforms. Still, when applied in the African context, it does not provide sufficient guidance on specific policies to foster the process of industrial upgrading and structural change, which are at the core of all successful development strategies. Historical evidence suggests that the growth process followed a similar pattern in developing economies such as the four East Asian dragons (Korea, Singapore, Taiwan, Hong Kong), which converged to the income levels of advanced western countries in the second half of the twentieth century. A similar strategy subsequently allowed countries as diverse as China, Vietnam, Botswana, or Mauritius to achieve rapid and sustained growth in the 1980s

13  Pritchett (2006) suggests that economists abandon the quest for a single growth theory, and focus instead on developing a collection of growth and transition theories tailored to countries’ particular circumstances. 14  See Lin and Monga (2011) for a critical discussion.

Principles of Economics    323 and 1990s. These experiences highlight the need to understand how developing countries can create the conditions for facilitating the flow of technologies and unleash growth, even in the context of distorted microeconomic policies, weak institutions, and the absence of full-fledged private property rights. In view of Africa’s growth experience, there should be a rethinking of economic development that refocuses attention to the deeper issues of structural transformation. A transformational growth process has four key components that are equally important and mutually reinforcing: • First, growth is to be understood as the cause of the sustained rise in the supply of goods; thinking about it as such (and not just the result of the dynamics higher levels of demand and supply) helps focus on the sources of that process, and the respective and complementary roles that markets mechanisms and state action that may be needed to facilitate it. • Second, growth should be the reflection of a carefully sequenced diversification process; clearly, many economies have done quite well by relying for a long period of time on the exploitation of a single commodity, or on a small set of products. But they tend to be small, or they eventually manage to use the income derived from the initial products to develop additional sources of growth. • Third, technological progress (through appropriation, imitation, or innovation) is the most sustainable source of long-term economic growth. Countries willing and capable of continuously exploiting technological advances must prepare their factor endowment (labor, capital) to respond to macroeconomic and sectoral policies that induce the accumulation of human and physical capital. • Fourth, for technological progress to serve its purpose in the growth process, it must take place within an environment that is conducive to business development and innovation. This implies constant rethinking, validation and update of the prevailing intellectual frameworks, rules and regulations that govern the business climate in the country. Such a “structural” understanding of growth is at the core of New Structural Economics (Lin 2012a). It sheds light to the dynamics of economic development, whose main engines are productivity growth, factor accumulation, and constant learning. The process can then be analyzed from several different perspectives: changes in sectoral contributions to growth and industrial structure, technological upgrading, and diversification (Monga 2012). Principle 8: Sustainable industrialization is an essential ingredient to inclusive growth and stable societies. “Always being in a hurry does not prevent death, neither does going slowly prevent living.” Ibo proverb.

Mankiw’s proclamation that “a country’s standard of living depends on its ability to produce goods and services” (principle no. 8 on his list) is certainly one of the most valid basic ideas in economics. There is little disagreement on the fact that “countries whose workers produce a large quantity of goods and services per unit of time enjoy a high standard of living. Similarly, as a nation’s productivity grows, so does its average income.” But the formulation of that very important economic truth in traditional textbooks is too vague when analyzed from the perspective of the mostly low-income countries. A key lesson learned by African policymakers from the elusive pursuit of the Millennium Development Goals (MDGs) that framed development thinking and policy in the 1990s is the need to set more specific

324   Methodological Issues goals and to identify key policy priorities that would facilitate their reach. Such goals should carry clear benefits and create win–win opportunities and incentives to all social groups. Sustainable industrialization is the most important of such goals. It provides an integrative framework for thinking more systematically about the other important national and global goals (productivity growth, economic growth, employment creation, human capital and infrastructure development, the inclusion of women and minorities, effective governance, etc.) and for engineering an orderly process of structural transformation. Countries that have achieved sustainable industrialization have also improved the quality of life for poor people, achieved shared prosperity, and become stable societies. Sustainable industrial development is therefore the most reliable engine of growth and a key ingredient for global peace and prosperity. The reason is at least threefold: first, modern economic growth is by definition a process of continuous technological innovation, industrial upgrading and economic diversification. No country in the world has been able to move from low- to middle- and high-income status without undergoing the process of industrialization. The kind of structural transformation dynamics that brings about human welfare only takes place because of changes in technology, in comparative advantage, and in the global economy. Inclusive and sustainable industrial development is the dynamics that offers guiding principles on how “best” any society should move its human, capital and financial resources from low- to high-productivity sectors. For that process to be efficient, coordination issues among economic agents (households, firms, and government entities) and externalities issues (the difference between private and social benefits) must be addressed. Markets typically do not manage such structural transformations on their own well. Inclusive and sustainable industrial policy is necessary for the process to unfold optimally. It is therefore not a speculative intellectual exercise for academic debates, but an effective strategy to address the pervasive discrepancies between private gains and social returns and to correct major sectoral or other misallocations. Second, economic theory has established that most increases in standard of living are related to the acquisition of knowledge, to “learning” (Joseph Schumpeter). Most increases in per capita income arise from advances in technology—about 70 percent of growth comes from sources other than factor accumulation. In developing countries, a substantial part of growth arises from closing the technology (or knowledge) gap between themselves and those at the frontier. And within any country, there is enormous scope for productivity improvement simply by closing the gap between best practices and average practices. If improvements in standards of living mainly come from diffusion of knowledge, learning strategies must be at the heart of the development strategies (Robert Solow). Successful economies—and prosperous and peaceful societies—are those where individuals and firms learn constantly, even as they compete against each other. It follows that understanding how economies best learn—how economies can best be organized to increase the production and dissemination of productivity-enhancing knowledge—should be a central part of the study of development and growth. But markets on their own fail to “maximize” learning. They ignore important knowledge spillovers. Sectors where knowledge is important tend to be imperfectly competitive, with the result that output is restrained. In fact, the production of knowledge is often a joint product with the production of goods, which means that the production of goods themselves will not in general be (intertemporally) efficient. In sum, markets, by themselves, are not likely to produce sufficient growth enhancing investments, such as those associated with learning, knowledge accumulation, and research.

Principles of Economics    325 These are the elements of a new intellectual consensus that provide further justification for promoting inclusive and sustainable industrial development—well beyond the traditional theoretical discussion of market failures based on coordination and conventional externalities (Stiglitz et al. 2013a,b). That emerging consensus among economists and policymakers has grown wider on the need for governments to focus on promoting learning, infant industries and economies, exports, and the private sector, not only in manufacturing but also agriculture and services such as, health, information technology, or finance. Industrial development is therefore not just about promoting manufacturing. It should be acknowledged that industrial development had many failures and only few successes around the world in the second half of the twentieth century. New analytical work has shed light on these past experiences (Lin 2009, 2012a, b; Ocampo and Ros 2011). There is now a much better understanding of the reason why, in many places, industrial development was mired with policy mistakes: the main determinant of success or failure is whether industrial development aims at encouraging industries that are consistent with comparative advantage so that capital-scarce, labor-intensive economies do not devote their meager resources trying to build capital-intensive industries (Lin and Monga 2013, 2011). Principle 9: Trade and integration into global value chains are the main routes to economic prosperity “Money, if you use it, comes to an end; learning, if you use it, increases.” Swahili proverb.

It is almost tautological to note that African low-income countries have little room for sustainable economic growth and employment generation within their borders. Their low-income status is a reflection of limited potential sources of growth in domestic demand—few of them have markets that are large enough to propel growth and generate the kind of decent, formal sector employment that would allow the eradication of poverty. In such contexts, trade is the main engine for achieving prosperity. Therefore, integrating African economies into global value chains and fostering their international trade should be among the most important items on Africa’s development agenda. The reason is that a tectonic shift has occurred in global commerce in recent years but has not yet been fully accounted for and integrated into development thinking, policies and operations. For decades, multilateral lending institutions and advisory entities devoted much of their resources, intellectual and political capital trying to convince the world of the importance of reforms to open up trade. It was certainly a worthwhile effort, especially in the second half of the twentieth century. But things have changed dramatically in the past decade: research by the OECD and the WTO (World Economic Forum 2013) shows that tariff reductions and market access have become much less relevant for economic growth than it was the case even a generation ago. Trade is no longer about manufacturing a product in one country and selling it elsewhere but cooperating across boundaries and time zones to minimize production costs and maximize market coverage. Value chains are therefore the dominant framework for trade. Estimates suggest that reducing supply chain barriers around the world could increase global GDP up to six times more than removing all import tariffs. Simulations indicate that improvements on just two key bottlenecks to supply chains (border administration and transport/communications infrastructure) to all countries’ performance only halfway to that of Singapore would yield an increase of US $2.7 (4.7 percent) in global GDP and US $1.6 trillion (14.5 percent) in global exports. These staggering numbers compare with much smaller gains from complete worldwide tariff elimination, which would only lead to US $400 billion

326   Methodological Issues (0.7 percent) in global GDP and US $1.1 trillion (10.1 percent) in global exports. That new information suggests that global prosperity and social peace are now within reach, provided that the low-income regions of the world take appropriate policy measures to integrate their economies into the dynamics of global value chains. Successful experiences of sustained growth with employment creation (most notably in Singapore and China) confirms the centrality of economic clusters (defined as geographical concentrations of interconnected companies with close supply links, specialist suppliers, service providers, and related industries and institutions) as tools for boosting competitive industries even in poor business environments. By developing a few, well-functioning islands of excellence in countries with bad infrastructure and weak governance, even the poorest countries can circumvent many of the constraints and bottlenecks on private sector development, while attracting and building in a pragmatic and targeted fashion the type of human capital that is immediately needed. The theory underlying the benefits of clusters dates back to Marshall, whose Principles of Economics (1920 [1890]) helped think systematically about agglomeration externalities (the notion that the concentration of production in a particular geographic area brings major external benefits for firms in that location through knowledge spillovers, labor pooling, and close proximity of specialized suppliers).15 However, economics generally considers clusters as being almost banal phenomena that randomly occur whenever private firms gather by accident in some place, start trading together, and eventually realize that it is more profitable even for competitors to stick together in a specific location. Clusters have thus been viewed merely as byproducts of economic development and their story is often presented as evidence of randomness in the emergence of clusters—the so-called “economics of QWERTY” (Krugman 1994). From that perspective, it not surprising that most countries that have attempted to proactively build clusters to reap the economic benefits of agglomeration have relied on chance. Even those that have chosen to achieve that goal through the creation of free zones and other special economic zones (SEZs) have largely elected to let the “invisible hand” of the market make things happen. But success has been scarce: there have been too few high-performing clusters—which generate economic growth and good employment opportunities—especially in the developing world. Luck has either been too random, or the “economics of QWERTY” has not yielded its magic. A close analysis of recent developments challenges conventional economic knowledge. The eruption of new clusters in countries like China, often in the most unlikely places,16 and as the result of strong and deliberate government action, provides a roadmap on how low-income countries in Africa could build well-functioning SEZs in the form of cluster-based industrial parks, and use them to promote competitive industries and integrate their economies into global value chains (Monga 2013b). Thanks to these startling 15 

Just like Adam Smith before him, Marshall (1920) offered several historical examples of clusters, which all seem to emerge accidentally in various places in Britain or Italy in the eighteenth and nineteenth centuries. Modern examples of agglomeration include Silicon Valley software industry, Detroit car manufacturing, Dalton (Georgia, United States) carpets, or Massachusetts Route 128 high-tech corridor. 16  Interesting cases include those of Qiaotou, Wenzhou and Yanbu are all relatively small regions in China that account for 60 percent of world button production, 95 percent of world cigarette lighter production, and dominate global underwear production, respectively (Lyn and Rodriguez-Clare 2011). Industrial clusters have also emerged in places such as Dongguan, Guangdong (electronic products) or Shandong (transport equipment).

Principles of Economics    327 developments, economic theorists and policymakers around the world are being forced to reassess the validity of theories of agglomeration. Principle 10: Coordinated government and market actions determine optimal changes in industrial structure, and economic prosperity—“A single hand cannot tie up a package.” African proverb.

Economic policy in the African context—must start with the observation that the main feature of modern economic development is continuous technological innovation and structural change. It also requires that the optimal industrial structure, that is, the industrial structure that will make the economy most competitive domestically and internationally at any specific time, be the basis for thinking and strategy (Lin 2009, 2012a; Lin and Monga 2011, 2010). The optimal industrial structure in an economy is endogenous to its comparative advantage, which in turn is determined by the given endowment structure of the economy at that time. Economies that try to grow simply by adding more and more physical capital or labor to the existing industries eventually run into diminishing returns; and economies that try to deviate from their comparative advantage are likely to perform poorly. Because the optimal industrial structure at any given time is endogenous to the existing factor endowments, a country trying to move up the ladder of technological development must first change its endowment structure. With capital accumulation, the economy’s factor endowment structure evolves, pushing its industrial structure to deviate from the optimal determined by its previous level. Firms then need to upgrade their industries and technologies accordingly in order to maintain market competitiveness. If the economy follows its comparative advantage in the development of its industries, its industries will be most competitive in domestic and world markets. As a result, they will gain the largest possible market share and generate potentially the largest surplus. Capital investment will also have the largest possible return. Consequently, households will have the highest savings propensity, resulting in an even faster upgrade of the country’s endowment structure. A developing country that follows its comparative advantage to develop its industries can also benefit from the advantage of backwardness in the upgrading process and grow faster than advanced countries. Firms in developing countries can benefit from the industrial and technological gap with developed countries by acquiring industrial and technological innovations that are consistent with their new comparative advantage through learning and borrowing from developed countries. The main question then is how to ensure that the economy grows in a manner that is consistent with its comparative advantage. The goal of most firms everywhere is profit maximization, which is, ceteris paribus, a function of relative prices of factor inputs. The criterion they use to select their industries and technology is typically the relative prices of capital, labor and natural resources. Therefore, the precondition for firms to follow the comparative advantage of the economy in their choice of technologies and industries is to have a relative price system which can reflect the relative scarcity of these production factors in the endowment structure. Such a relative price system exists only in a competitive market system. In developing countries where this is not usually not the case, it is necessary that government action be taken to improve various market institutions so as to create and protect effective competition in the product and factor markets. In the process of industrial upgrading, firms need to have information about production technologies and product markets. If information is not freely available, each firm will need to invest resources to search for it, collect and analyze it. For individual firms in

328   Methodological Issues African countries, industrial upgrading is therefore a high-reward, high-risk process. First movers who attempt to enter new industries can either fail—because they target the wrong industries—or succeed—because the industry is consistent with the country’s new comparative advantage. In case of success, their experience offers valuable and free information to other prospective entrants. They will not have monopoly rent because of competition from new entry. Moreover, these first movers often need to devote resources to train workers on the new business processes and techniques, who may be then hired by competitors. First movers generate demand for new activities and human capital which may not have existed otherwise. Even in situations where they fail, their bad experience also provides useful knowledge to other firms. Yet, they must bear the costs of failure. In other words, the social value of the first movers’ investments is usually much larger than their private value and there is an asymmetry between the first movers’ gain from success and the cost of failure. Successful industrial upgrading in an economy also requires new types of financial, legal, and other “soft” (or intangible) and “hard” (or tangible) infrastructure to facilitate production and market transactions and allow the economy to reach its production possibility frontier. The improvement of the hard and soft infrastructure requires coordination beyond individual firms’ decisions. Economic development is therefore a dynamic process marked with externalities and requiring coordination. While the market is a necessary basic mechanism for effective resource allocation at each given stage of development, governments must play a proactive, facilitating role for an economy to move from one stage to another. They must intervene to allow markets to function properly. They can do so by (i) providing information about new industries that are consistent with the new comparative advantage determined by change in the economy’s endowment structure; (ii) coordinating investments in related industries and the required improvements in infrastructure; (iii) subsidizing activities with externalities in the process of industrial upgrading and structural change; and (iv) catalyzing the development of new industries by incubation or by attracting foreign direct investment to overcome the deficits in social capital and other intangible constraints (Lin and Monga 2010). It appears that a successful strategy obviously requires strong collaborative work between the state and the private sector in the identification of new sectors or lines of business and prioritization of infrastructure investment. Beyond utilities, transportation, and well-functioning port and airport facilities, different industries require different types of infrastructure, some of which must be provided by the state: fruit exporters need refrigeration facilities; garment and textiles producers need specific storage facilities, etc.; and they all need various types of workforce development and skills training programs. Pragmatic government intervention is needed to overcome issues of coordination and externalities, which no individual firm can address alone effectively. Unfortunately, too many economic development strategies in general require the state to be stretched too thin across too many industries—and eventually not being able to accommodate any single industry with the type of first-rate infrastructure needed in a competitive global economy. As a result, they do not take advantage of their low wages in the agribusiness, light manufacturing, or services sector, and they do not join the international supply chains. Summing up: Africa’s economic experience has much to offer to development thinking and policy, just as the lessons from success and failure elsewhere can enrich the continent’s march towards prosperity and social peace.

Principles of Economics    329

References Acemoglu, D., Johnson, S., and Robinson, J.A. (2001). The colonial origins of comparative development: an empirical investigation. American Economic Review, 91(5):1369–1401. Achebe, C. (1959). Things Fall Apart. New York: Fawcett Crest Books. Akerlof, G.A. (1984). An Economic Theorist’s Book of Tales. Cambridge:  Cambridge University Press. Akerlof, G.A., and Kranton, R.E. (2010). Identity Economics: How Our Identities Shape Our Work, Wages, and Well-Being. Princeton: Princeton University Press. Basu, K. (2010). The moral basis of prosperity and oppression: altruism, other-regarding behaviour and identity. Economics and Philosophy, 26(2):189–216. Basu, K., and Van, P.H. (1998). The economics of child labor. American Economic Review, 88(3):412–427. Becker, G.S. (1993). Human Capital:  A  Theoretical and Empirical Analysis, with Special Reference to Education, 3rd edn. Chicago: University of Chicago Press. Becker, G.S. (1992). Education, labor force quality, and the economy: The Adam Smith Address. Business Economics, 27(1):7–12. Becker, G.S., and Murphy, K.M. (2000). Social Economics:  Market Behavior in a Social Environment. Cambridge, MA: Harvard University Press. Benhabib, J., Bisin, A., and Jackson, M.O. (2010). Social economics: a brief introduction to the handbook, in: Handbook of Social Economics, Vol. 1A. North Holland: Amsterdam, pp. xvii–xxi. Bourguignon, F. (2006). Stability, Security, and Development:  An Introduction. Keynote Address at Annual Word Bank Conference on Development Economics Europe 2006. Washington, DC: World Bank. Brock, W.A, Durlauf, S.N., and West, K.D. (2003). Policy evaluation in uncertain policy environments. Brookings Papers on Economic Activity, 1:235–322. Brynjolfsson, E., and McAfee, A. (2014). The Second Machine Age:  Work, Progress, and Prosperity in a Time of Brilliant Technologies. New York: W. W. Norton & Co. Colander, D., Holt, R., and Rosser, B. Jr. (2004). The changing face of mainstream economics. Review of Political Economy, 16(4):485–499. Duesenberry, J. (1949). Income, Saving, and the Theory of Consumer Behavior. Cambridge: Cambridge University Press. Durlauf, S.N. (2005). Growth econometrics, in P. Aghion and S. Durlauf (eds), Handbook of Economic Growth, Vol. 1A. North-Holland: Amsterdam, pp. 555–677. Eatwell, J., Milgate, M., and Newman, P. (eds) (1989). The New Palgrave:  Social Economics. New York: W.W. Norton and Co. Ela, J.-M. (1982). L’Afrique des villages. Paris, Karthala. Fehr, E., Herz, H., and Wilkening, T. (2013). The lure of authority: motivation and incentive effects of power. American Economic Review, 103(4):1325–1359. Fehr, E., Fischbacher, U., and Kosfeld, M. (2005). Neuroeconomic foundations of trust and social preferences. American Economic Review, 95(2):346–351. Gates, H.L. (1988). The Signifying Monkey: A Theory of African American Literary Criticism. New York: Oxford University Press. Glaeser, E., and Scheinkman, J.A. (2003). Non-market interactions, in M. Dewatripont, L.P. Hansen, and S.J. Turnovsky (eds), Advances in Economics and Econometrics: Theory and Applications, Eighth World Congress, Vol. I. Cambridge: Cambridge University Press, pp. 339–370.

330   Methodological Issues Greif, A. (1993). Contract enforceability and economic institutions in early trade: the Maghribi traders’ coalition. American Economic Review, 83:99–118. Habermas, J. (1991). The Structural Transformation of the Public Sphere:  An Inquiry into a Category of Bourgeois Society. Cambridge, MA: MIT Press. Hartley, J.E. (1996). Retrospectives: the origins of the representative agent. Journal of Economic Perspectives, 10(2):169–177. Hausmann, R., Rodrik, D., and Velasco, A. (2008). Growth diagnostics, in N. Serra and J.E. Stiglitz (eds), The Washington Consensus Reconsidered: Towards a New Global Governance. New York: Oxford University Press, pp. 324–354. Heckman, J.J. (2006). Skill formation and the economics of investing in disadvantaged children. Science, 312(5782):1900–1902. Hugo, V. ([1856] 1967). Préface, Contemplations, Paris, éd. Massin du Club Français du Livre, tome IX. Jackson, M.O. (2010). An overview of social networks and economic applications, in J. Benhabib, A. Bisin, and M.O. Jackson (eds), Handbook of Social Economics, Vol. 1A. Amsterdam: North Holland, pp. 511–585. Jones, W.O. (1960). Economic man in Africa. Food Research Institute Studies, 1(2):107–134. Kane, C.A. (1963). Ambiguous Adventure. New York: Walker. Krugman, P. (1994). Peddling Prosperity: Economic Sense and Nonsense in the Age of Diminished Expectations. New York: W.W. Norton and Co. Lazear, E.P. (2000). Economic imperialism. Quarterly Journal of Economics, 115:99–146. Lin, J.Y. (2009). Economic Development and Transition:  Thought, Strategy and Viability. New York: Cambridge University Press. Lin, J.Y. (2012 a). New Structural Economics: A Framework for Rethinking Development and Policy. Washington, DC: World Bank. Lin, J.Y. (2012b). The Quest for Prosperity: How Developing Economies Can Take Off. Princeton, NJ: Princeton University Press. Lin, J.Y., and Monga, C. (2010). The Growth Report and New Structural Economics, World Bank Policy Research Working Paper no. 5336. Washington DC: World Bank. Lin, J.Y., and Monga, C. (2011). Growth identification and facilitation: the role of the state in the dynamics of structural change. Development Policy Review, 29(3):259–310. Lin, J.Y., and Monga, C. (2013). Comparative advantage: the silver bullet of industrial policy, in J.E. Stiglitz and J.Y. Lin (eds), The Industrial Policy Revolution: the Role of Government Beyond Ideology. New York: Palgrave Macmillan, pp. 19–38. Livingstone, D. (1875). The Last Journals of David Livingstone in Central Africa. New York. Loury, G. (1977). A dynamic theory of racial income differences, in P. Wallace and A. LaMond (eds), Women, Minorities, and Employment Discrimination. Lexington, MA: Lexington Books. Lucas, R.E. (2004). Lectures on Economic Growth. Cambridge, MA: Harvard University Press. Lyn, G., and Rodriguez-Clare, A. (2011). Marshallian Externalities, Comparative Advantage, and International Trade, mimeo, University of California Berkeley, http://emlab.berkeley. edu/~arodeml/Papers/LR_Marshallian_Externalities_Trade.pdf. Malinvaud, E. (1991). Voies de la recherche macroéconomique. Paris: Editions Odile Jacob. Mankiw, N.G. (2014 [1997]). Principles of Economics, 7th edn. Boston: Cengage Learning. Manski, C.F. (2000). Economic analysis of social interactions. Journal of Economic Perspectives, 14(3):115–136. Marshall, A. (1920 [1890]). Principles of Economics, London: Macmillan and Co. Ltd. Mauss, M. (1925). Essai sur le don: Forme et raison de l’échange dans les sociétés archaïques. Paris: Alcan.

Principles of Economics    331 Monga, C. (2013a). Winning the jackpot: job dividends in a multipolar world, in J.E. Stiglitz, J.Y. Lin, and E. Patel (eds), The Industrial Policy Revolution II: Africa in the 21st Century. New York: Palgrave Macmillan, pp. 135–172. Monga, C. (2013b). Theories of agglomeration: critical analysis from a policy perspective, in J.E. Stiglitz, J.Y. Lin, and E. Patel (eds), The Industrial Policy Revolution II: Africa in the 21st Century. New York: Palgrave Macmillan, pp. 209–224. Monga, C. (2012). Shifting gears: igniting structural transformation in Africa. Journal of African Economies, 21(Suppl 2):ii19–ii54. Monga, C. (2011). Post-macroeconomics: lessons from the crisis and strategic directions ahead. Journal of International Commerce, Economics and Policy, 2(2):1–28. Monga, C. (1996). The Anthropology of Anger:  Civil Society and Democracy in Africa. Boulder: Co. Lynne Rienner Publishers. North, D.C. (1981). Structure and Change in Economic History. New York: W.W. Norton. Ocampo, J.-A., and Ros, J. (eds) (2011). The Oxford Handbook of Latin American Economics. New York: Oxford University Press. Pritchett, L. (2006). The quest continues. Finance and Development, 43(1). Quiggin, A.H. (1949). Trade Routes and Currency in East Africa. Occasional Papers of the Rhodes-Livingstone Museum, No. 5. Livingstone: Northern Rhodesia. Rabin, M. (1998). Psychology and economics. Journal of Economic Literature, 36(1):11–46. Romer, P.M. (1999). Interview with Paul Romer, in B. Snowdon and H. Vane (eds), Conversations with Economists: Interpreting Macroeconomics. Cheltenham: Edward Elgar. Rose-Innes, O. (2013). The Great South Africa Smoking Survey 2013: Results, Health24. com, November, http://www.health24.com/Lifestyle/Stop-smoking/News/The-Great-SASmoking-Survey-2013-results-20131105. Rubinstein, A. (2008). Comments on neuroeconomics. Economics and Philosophy, 24:485–494. Ryder, H.E. Jr., and Heal, G.M. (1973). Optimal growth with intertemporally dependent preferences. Review of Economic Studies, 40(1):1–33. Sala-i-Martin, X.X. (1997). I just ran two million regressions. American Economic Review, 87(2):178–183. Schelling, T.C. (1978). Micromotives and Macrobehavior. New York: Norton. Sen, A. K. (1977). Rational fools: a critique of the behavioral foundations of economic theory. Philosophy & Public Affairs, 6(4):317–344. Shiller, R.J. (2014). The Rationality Debate, Simmering in Stockholm. The New York Times, January 18. Sims, C.A. (2010). Rational inattention and monetary policy, in B.M. Friedman and M. Woodford (ed.), Handbook of Monetary Policy, 1st edn, Vol. 3. New York: Elsevier, pp. 155–181. Sims, C.A. (2003). Implications of rational inattention. Journal of Monetary Economics, 50(3):665–690. Smith, A. (1977 [1776]). An Inquiry into the Nature and Causes of the Wealth of Nations. Chicago: The University of Chicago Press. Speke, J.H. (1863). Journal of the Discovery of the Source of the Nile. Edinburgh and London, William Blackwood. Stiglitz, J., Lin, J.Y., and Monga, C. (2013a). Rejuvenation of industrial policy, in J. Stiglitz and J.Y. Lin (eds), The Industrial Policy Revolution I: The Role of Government Beyond Ideology. New York: Palgrave Macmillan, pp. 1–18. Stiglitz, J., Lin, J.Y., Monga, C., and Patel, E. (2013b). Industrial policy in the African context, in J. Stiglitz, J.Y. Lin, and E. Patel (eds), The Industrial Policy Revolution II: Africa in the 21st Century. New York: Palgrave Macmillan, pp. 1–24.

332   Methodological Issues Tansi, S.L. (1983). L’anté-peuple. Paris: Seuil. Thaler, R.H. (1980). Toward a positive theory of consumer choice. Journal of Economic Behavior and Organization, 1(1):39–60. Tversky, A., and Kahneman, D. (1991). Loss aversion in riskless choice: a reference-dependent model. Quarterly Journal of Economics, 106(4):1039–1061. Veblen, T. (1934). The Theory of the Leisure Class:  An Economic Study of Institutions. New York: Modern Library. Wicksteed, P. (1910). Common Sense of Political Economy. London: Macmillan and Co. World Bank (2005). Economic Growth in the 1990s:  Learning from a Decade of Reform. Washington, DC: World Bank. World Economic Forum (2013). Enabling Trade: Valuing Growth Opportunities (in collaboration with Bain & Company and the World Bank), Geneva, http://www3.weforum.org/docs/ WEF_SCT_EnablingTrade_Report_2013.pdf.

Appendix 1: N. Gregory Mankiw’s Ten Principles of Economics Although the study of economics has many facets, the field is unified by several central ideas. The Ten Principles of Economics offer an overview of what economics is all about. How People Make Decisions 1. People Face Trade-offs. To get one thing, you have to give up something else. Making decisions requires trading off one goal against another. 2. The Cost of Something is What You Give Up to Get It. Decision-makers have to consider both the obvious and implicit costs of their actions. 3. Rational People Think at the Margin. A rational decision-maker takes action if and only if the marginal benefit of the action exceeds the marginal cost. 4. People Respond to Incentives. Behavior changes when costs or benefits change. How the Economy Works as a Whole 5. Trade Can Make Everyone Better Off. Trade allows each person to specialize in the activities he or she does best. By trading with others, people can buy a greater variety of goods or services. 6. Markets Are Usually a Good Way to Organize Economic Activity. Households and firms that interact in market economies act as if they are guided by an invisible hand that leads the market to allocate resources efficiently. The opposite of this is economic activity that is organized by a central planner within the government. 7. Governments Can Sometimes Improve Market Outcomes. When a market fails to allocate resources efficiently, the government can change the outcome through public policy. Examples are regulations against monopolies and pollution.

Principles of Economics    333 How People Interact 8. A Country’s Standard of Living Depends on Its Ability to Produce Goods and Services. Countries whose workers produce a large quantity of goods and services per unit of time enjoy a high standard of living. Similarly, as a nation’s productivity grows, so does its average income. 9. Prices Rise When the Government Prints Too Much Money. When a government creates large quantities of the nation’s money, the value of the money falls. As a result, prices increase, requiring more of the same money to buy goods and services. 10. Society Faces a Short-Run Tradeoff Between Inflation and Unemployment. Reducing inflation often causes a temporary rise in unemployment. This tradeoff is crucial for understanding the short-run effects of changes in taxes, government spending and monetary policy.

Chapter 17

Ec onomics and C u lt u re in Afri c a Felwine Sarr

17.1 Introduction Economics and culture are two powerful mainsprings of human action. Economics is an order of efficiency turned toward the optimal allocation of resources. It has also become, as a discipline, a space for reflection on the theory of human action and what founds it. Culture is defined by anthropologists as a body of practices and values, of distinctive material and spiritual traits identifying a given social group. It is a polysemic notion, used in a variety of senses. Already in 1952, Kroeber and Kluckhohn had listed more than 164 definitions of culture.1 In this chapter, culture is understood as the totality of sociological and anthropological

1 

Kroeber and Kluckhohn (1952) listed more than 164 definitions of culture. The following are a few of these: According to Tylor, “Culture or civilization, taken in its wide ethnographic sense, is that complex whole which includes knowledge, belief, art, morals, law, custom, and any other capabilities and habits acquired by man as a member of society.” Tylor distinguishes three degrees in the evolution of societies: the savage state, the barbarian state, and the civilized state. From this point of view, civilization proper is no longer simply identified with culture, but with a certain advanced type of culture. It proceeds from an evolutionist view of culture. “To avoid the dangers of ethnocentric value judgments and arbitrary hierarchies between savage and civilized peoples, some sociologists and anthropologists prefer to base themselves on an analysis of the manifestations of social life and classify certain elements under the heading ‘civilization,’ and other elements under the heading ‘culture,’ which avoids making the former a superior form of the latter. These two orders should thus be able to coexist in any kind of society, however advanced it may be in this or that domain. According to R. M. Mac Iver, culture consists of the ‘expressions of life.’ This should be understood as ideologies, religions, arts, and literatures. As for civilization, it represents the creations of society to ensure its control of its own conditions of life, which involves social organization as well as techniques. This conception, fairly widespread among German sociologists, is founded partly on distinctions made by Alfred Weber and also agrees with that established by Kroeber between two kinds of culture: that of value and that of reality, the latter corresponding for the most part to civilization. It makes it possible to avoid the pitfalls linked to value judgments.” In Jean Cazeneuve, “Civilisation,” Encyclopaedia Universalis: http://www.universalis.fr/encyclopedie/civilisation.

Economics and Culture in the African Context    335 traits that distinguish a given social group, but also as the totality of the works of the mind necessitating creativity in their production, that are linked to the education of the mind and the production and communication of symbolic significations: artistic, scientific, and literary culture, various arts, ways of life, and ways of inhabiting time. Culture is therefore a space of perpetual creation, a desire and a way of grasping and exploring reality in all its dimensions; it is thus a moving object, in constant redefinition. As a result it has the double dimension of fact and process. Although it can be a locus of creation of value, the finalities of culture are first and foremost symbolic. They are chiefly a matter of the production of meaning and significations. This chapter explores the relationship between these orders having visibly opposing finalities by situating its reflection in the context of Africa. Since the accession of its countries to independence, this continent has encountered great difficulties in satisfying the fundamental needs of its populations in terms of food, education, health, and security and in providing them with economic and social well-being. Although there has been recent progress in economic growth, these difficulties remain, in spite of the implementation of a great many economic programs designed to resolve them.2 The principal characteristic of the economic models implemented on the African continent in the past half-century is the extraversion of their origin. They are not the result of internal economic practices or production. Hence the dualism of these systems characterized by the coexistence of an economy known as formal with a popular economy based on a culture and labeled informal, which, however, ensures the subsistence of a majority of the African populations and contributes the main part of the GDP (according to Charmes [2000], 54.2 percent in sub-Saharan Africa). In traditional African societies, economic matters were included in a much larger social system.3 They were, of course, governed by their classical functions (subsistence, allocation of resources) but were above all subordinated to social, cultural and civilizational finalities. This seems to be less the case in contemporary societies (African and others) where the economic order has become hegemonic and, overflowing its natural space, attempts to impose its significations and principles on all dimensions of human existence.4 In its functional dimensions, culture produces value, and the products of artistic and esthetic creations can become economic goods. The object of this study is, however, not to evaluate the contribution of African cultural industries to the creation of wealth in their respective countries.5 Rather, this chapter considers the question of the articulation of two powerful determiners of individual and collective actions—culture and economics in the 2  Explanations differ in nature: some are linked to external dynamics, relationships of domination that are unfavorable to the African continent in international economic competition; others to internal causes, linked to political instability and above all to the inefficiency of internal models and modes of economic management. 3  This analysis is confined to traditional sub-Saharan societies. 4  Economics has become a discipline that has lost sight of its place in the whole (anomie), overflows its natural space and invades the totality of social relations by imposing on them its significations and profit principles (privative appropriation of nature, space, public goods, transformation of social relationships by wage payment). 5  The contribution of cultural industries to the GDP of African countries is becoming increasingly important. The example of Nollywood in Nigeria is significant in more ways than one. Due to the emendation of the Nigerian GDP by a better evaluation of the Nigerian film industry’s contribution, Nigeria has replaced South Africa as the leading African economy.

336   Methodological Issues African context—with an intention that could be called civilizational, that is, which would make it possible to determine the finalities judged best by the individual and the group.6 This makes it necessary to consider the social vision from every angle, analyzing the multiple interactions of its environmental dimensions, that is, those intended to ensure the conditions of existence (the economy) with those whose goal it is to work on the significations of existence itself (culture, meaning, aims and finalities valued by individuals and groups). For cultural factors do indeed influence economic performances. Culture has an impact on the perceptions, attitudes and habits of consumption, investment and saving, and on individual and collective choices. And economics as a system of discourse and practice is the result of a cultural process that elaborates a worldview and produces a scale of particular values. More precisely, the double interrogation pursued in this chapter is to determine in what measure the efficiency of an economic system is linked to its degree of adequation with its cultural context on one hand, and on the other, whether the efficiency of the resulting social system is dependent on respect of the functions assigned to each order (economy and culture) by the group. Thus this chapter’s working hypothesis is that an articulation of the economic and cultural orders that would avoid confusing their respective finalities would make society’s visions (economic and social) more coherent and more efficient. The first section of this chapter analyzes the different levels of interaction between economy and culture in Africa in order to bring to light their reciprocal determinations. The cultural foundations of agents’ economic choices are studied and the economy is analyzed as a cultural process. In human groups, nuclei of the imaginary are fundamental to social relationships, including the most material ones. The economic relationship is first a social relationship. The play between the imaginary and the symbolic is part of its production. The second section outlines exploratory paths regarding the way to root the African economies in their sociocultures so that they become productions of their respective cultural contexts. It also analyzes how culture works as a force of adaptation and change.

17.2  Economy and Culture: Interactions and Reciprocal Determinations 17.2.1  The cultural foundations of economic choices Individual decision-making processes are strongly influenced by the individual’s cultural environment, which conditions preferences and regulates behaviors. The satisfaction of the fundamental needs of food, shelter, clothing, etc., inevitably entails choices that are not solely dictated by the existing alternatives, that is, by the quantity of goods and services available to satisfy these needs. An individual’s cultural matrix, made of social conventions, religious beliefs (food prohibitions, dress codes), culinary culture, esthetic conceptions and ethical prescriptions functions to shape the desires (needs) of the individual as well as the circumstances (temporality, place) of their satisfaction. The temporality of a need’s satisfaction or 6 

Those to which individuals and groups attach value.

Economics and Culture in the African Context    337 even the transformation of a need into a desire, that is, of a need whose satisfaction implies the mediation of work and time, is conditioned by culture. In addition, the structure of so-called fundamental needs evolves and some needs, although induced culturally (television, cigarettes, cell phones, internet, need of culture, entertainment) nonetheless become fundamental for individuals. Economic anthropology finds that the behaviors of saving, investment, and accumulation, as well as the logic or rationale governing certain modes of consumption, are culturally determined. Herskovits (1952) emphasized the fact that some West African peoples (the Yoruba) expend their reserves of food ostentatiously and liberally between the end of the dry season and the beginning of the rainy season. One would rather have expected then to conserve this food to tide them over. These behaviors are not, however, due to a lack of foresight on the part of these communities, but are dictated by cultural considerations that value expenditures linked to rituals, ostentatious and prestigious expenditures, and inscribe the significations of the act of consumption beyond the biological need to nourish oneself and the caloric contribution necessary for physical work. Herskovits (1952) noted the same fact among the Tallensi of the African Gold Coast (present-day Ghana), where the supply of food is the lowest at the time when, because of physical labor in the fields (May–June) the most calories are needed; and yet this people has mastered the technique of storing foodstuffs. Among the Zulus, Ashantis, Balubas, and Bagandas the economy of prestige plays an equally important role in maintaining the social status of individuals. Custom demands that guests be served quality food, even if what is eaten in private is of lesser quality.7 An analysis of family and individual budgets and the allocation of resources shows, for example, that in the Ibo families of northern Nigeria the amounts allocated to symbolic goods (expenses for religious purposes, funerals, and so forth) are greater than those allocated to tangible things. There is certainly a universal propensity to satisfy nutritional needs as well as personal tastes, but among the Ibos, religious and social obligations are as important for the life of the group as is nourishment for the organism. As concerns the need for clothing, this too does not obey the sole function of protecting the body. The choice of fabric, its design, the differentiation of clothing according to gender and social status are also a matter of cultural choices. In the land of the Ashantis, even the right to wear certain kinds of clothing was granted by the king. Thus the different conceptions of the economic management of households is not a matter of chance, but the result of different social evaluations of time, work, leisure and social and religious obligations. In conjunction with the quantity of resources available, these define the communities’ productivity and standard of living. The preceding thoughts pertain to traditional societies, whose social and behavioral structures show a stability and durability authorizing such analysis. Two dominant, distinctive traits can be singled out: a consumption of material goods that is certainly utilitarian, but above all symbolic, as well as a mentality and practices of sharing rooted in the cultural substratum of these traditional communities. Individuals belong to a group and a historically constructed social formation. Social interactions founded on the group’s culture explain a large portion of their choices. 7 

Meat stew and yam loaf are provisions stocked in expectation of visitors.

338   Methodological Issues

17.2.2  Economics as a cultural process David Throsby (2001) emphasizes the fact that economics as a scientific discipline and constituted field of knowledge is the result of a process that can be called cultural. He notes that the various schools of thought making up economics can be distinguished by a body of practices and beliefs that constitute them. Just as shared values produce the basis of the cultural identity of a group, the coalescence of Marxist, Keynesian, neo-classical, neo-institutional, etc., schools of thought can be interpreted as cultural processes. Moreover, the impact of culture on the thought of economists is important because the cultural values which they inherit or acquire have a profound and often overlooked influence on their perceptions and attitudes. Thus, the cultural presuppositions inherited by economists influence their ability to explain the economic reality that they observe, particularly when it comes from an environment different from their own. The cultural context of economics is a matter of social organization as well as a system of thought. In “The Limitation of the Special Case,” (1963), Dudley Seers challenges, based on historic and empiric facts, the claim to universal validity made by dominant economics.8 He questions the capability of concepts stemming from Western economies to analyze the economies of developing countries. Seers (1963) shows that the dominant economics, taught in universities, was constructed based on phenomena observed in the countries that are today “developed” (those that are precisely the special case) and that it is consequently inapplicable to the general case, which is that of the “underdeveloped” countries. He enumerates the fundamental differences separating the economies of industrialized countries from the others—in almost all domains. According to Seers, therefore, economics must be restructured, based on the idea that economics is the study of economies rather than that of economic models. Thus it seems clear that dominant economic thought conveys a culture, a particular view of the world and of man (homo economicus). Cultural considerations affect the manner in which economists practice their discipline, develop their postulates and model interactions between agents. Economic thought elaborates an episteme: a body of beliefs and values that it defends and promotes. Analyzing the processes by which economic ideas are generated, discussed, approved and transmitted, Throsby (2001) found that they belong to the order of discourse as analyzed by 8  “A book is not called ‘Principles of Astronomy,’” he says, “if it refers only to the earth or the solar system … We justifiably expect a lecture course on geology to deal with other continents besides the one on which the author happens to live …” So there is deception when economists set forth principles or laws that supposedly apply to everyone everywhere, for it is not legitimate to deduce a “general theory” from “special cases.” Seers enumerates the fundamental differences that separate the industrialized countries from the others, in almost all domains: the factors of production, the structure of the economy, public funds, the role of foreign trade, household spending, the structure of savings, investment capabilities, etc. He proposes founding a new discipline that would take into account the specific problems posed by the entrance of so-called traditional societies into an international system governed by other rules. This would result in considering economics as a “local” discipline and not a science with universal pretentions. And it would mean putting into place a generalized economics within which the industrialized countries would be an atypical province where special rules would apply. According to Seers, it is a matter of acknowledging the diversity of historical situations.

Economics and Culture in the African Context    339 Foucault.9 Their acceptance by the community of researchers is founded more on processes of legitimation and intellectual persuasion than on the faithful rendering of an observed reality.10 In this context, economic discourse functions as a language that ensures the establishment of the common symbolic code through which the group’s manner of saying, thinking and experiencing the real is worked out. It defines the links between facts and concepts, distinguishes the essential from the accessory, and selects the materials of the real that it makes available for assembly. This language is not neutral, for it carries out a particular structuring of space and time, chooses to reveal certain aspects and passes over others, establishes a particular hierarchy of values, and shapes modes of thinking and thus of acting. Mudimbé (1982) emphasized in this connection that the discourses of the social sciences in Sub-Saharan Africa ought to present another order and other rules of structuring. The Africanization of the sciences is often conceived of only as application. The practice of economic science seems to be merely a reproduction of the inegalitarian relations of dependence between the international “mother countries” and the African countries. From the viewpoint of establishing truly African social and human sciences, that is, practices and knowledge that would be in harmony with the gradients of African cultures, he raises the question of the conditions in which a different and original order of discourse could appear. He considers, in the light of Senghor’s work on the civilization of the Universal, that a new, ethnically diverse, rich universe would emerge. An avenue which he suggests is that changing the linguistic instrument of knowledge and scientific production would bring about an epistemological break and open the way to a new destiny for Africa, just as the promoters of Greek thought, in transplanting into their language the techniques, methods and uses of the knowledge received from Egypt, set in motion a reorganization of knowledge and life whose essential order is still current and in use—an order that through the West is now leaving its mark on Africa. Economic thought also produces culture in the sense that it disseminates and promotes practices and ways of doing things in the social body. Additionally, it projects its own criteria of evaluation onto all human activities, including those which do not have as their primary function participating in the market exchange networks. Economics and culture as spaces of human thought and action have always been concerned by the question of the “value” of things. The question of the adequate evaluation of 9  Episteme is defined as the body of scientific learning, the knowledge of a historical period and its presuppositions. More generally, it is the manner of thinking, of visualizing the world: a body of the dominant values and beliefs of a period that very broadly cover the entire culture. In The Order of Things [Les mots et les choses, 1966] and The Archaeology of Knowledge [L’archéologie du savoir, 1969], Foucault describes three successive epistemes: that of the Renaissance, the classical period, and the modern period. From a scientific point of view, each historical period can be defined, according to Michel Foucault, by an episteme, that is, a body of problematics, hypotheses, and research methods, which constitute an invariant for this period. For example, the search for “similitude” governs thought from the Renaissance until the classical age, whereas it is the search for “order” that organizes the modern period. For Foucault there is no “progress” in the cultural process in the course of history; the changes that appear from one century to the next are produced by the passage from one episteme to another. 10  The effectiveness of competitive markets, the fact that they are in pure and perfect competition, is more a matter of belief in a paradigm than of observation of a reality whose perception is influenced by the position of the observer.

340   Methodological Issues the value of a good is central in economics. Since the eighteenth century, different conceptions of the value of a good have been in use and confront each other in controversies of theory.11 The debates around value that take place within the economic paradigm link value to the notion of utility, or the price (thus the value) that individuals and markets assign to goods. It is a matter of quantifying the value of the goods to be integrated into an economic system. In the sphere of culture, value concerns an intrinsic property of certain cultural objects or phenomena, which can be expressed as the quality of a work (literary masterpiece, sculpture, painting, piece of music), or of an experience. It is a social construct whose evaluation cannot be isolated from the context of its production. The notion of value is often attached to the singular dimension of an object or a work, bearing significations beyond its practical utility or its materiality. The quality of a musical composition by Richard Bona, of a novel by Cheikh Hamidou Kane or Boubacar Boris Diop comes from the singular dimension and sensitivity of the existential experience it translates and transmits. Adorno and Horkheimer (1947)12 as well as Arendt (1972)13 note that by applying the logic of economic practice to objects linked to culture, the culture industry effects the degradation of any cultural object into a perishable consumable object, resulting in the production of an entertainment culture for the needs of its economy.14 The commodities offered by the entertainment industry are not cultural objects in the primary sense of the term, as defined by Hannah Arendt, that is, objects whose function is to sustain and enrich the vital process, to establish permanence and duration (functions that transcend needs), to be traces intended to remain in the world after we have left it in order to testify to our human adventure. Rather, they are consumer goods to be used and replaced by others through a production cycle that ensures the perpetuity of an industry needing continuous production

11 

In the eighteenth century, Adam Smith distinguished the use value of a good from its exchange value. Economists of the nineteenth century (Ricardo, Marx) calculated the value of a good according to the cost of its manufacture: the quantity of inputs necessary to its production or the amount of work necessary to its making. Beginning with the marginalist revolution, prices were seen as the means by which the market coordinates the different evaluations arrived at by the actors of the economic system, and neo-classical analysis likened the price theory to a value theory. There followed a critique of the marginalist theory of utility which led to a value theory according to which individuals can formulate a revealed preference within which they classify goods according to their tastes, needs and utility gained, this without being influenced by the environment, institutions, social interactions, and the processes that regulate and govern the exchanges. This critique claims that in a context dominated by market imperfections, asymmetry of information, imperfect competition and incomplete information, prices are only an indication of value, but not necessarily a measure of it. 12  La dialectique de la raison [Dialectic of Enlightenment]. 13  La Crise de la Culture [Between Past and Future]. 14  The notion of culture industry originates with Theodor Adorno and Max Horkheimer of the Frankfurt School. In Dialectic of Enlightenment (1947), they assert that the diffusion of mass culture endangers true artistic creation. The culture industry (film, radio, press, television) tends, not toward the emancipation of the individual, but on the contrary toward a uniformization of his modes of life and the domination of economic logic. The culture industries degrade culture into entertainment. Wage-earners think they can escape the alienation of their work through entertainment, but in reality it is in entertainment that the individual is disciplined and prepared to come to grips with work. The careers of professional people are determined more by belonging to culture than by technical knowledge. It is in this culture that allegiance to power and to social hierarchies is manifested.

Economics and Culture in the African Context    341 to ensure its maintenance. Hence a culture of material production, optimization, production at minimal cost, transaction based for the most part on market exchange and exclusion by prices, as well as the assimilation of the notion of value to that of price. Such a culture, applied to all areas of social life can result in a destruction of value and a perversion of the finality of objects.

17.3  Articulating the African Economies with their Cultural Context If there is a space where Africa’s power of dissemination and irradiation has remained intact, full and entire, despite the convulsive movements of a turbulent recent history, it is that of culture. Can this order constitute the foundation of an economics that would be more efficient because it would be better linked to its cultural context?

17.3.1  Economic project and mythological universe At the origin of every community, there is the establishment of a common symbolic code that enables its members to think, say and experience the real in a relatively univocal way. Anthropology has shown that societies are built on a founding story, a myth, which shapes a certain conception of the world and its organization, and puts into place a particular hierarchy of values often conceptualized by a social and linguistic code internalized by its members.15 Modern societies, more particularly Western industrial society, are not an exception to this structure.16 The latter must legitimize its evolution and its appropriation of the future through a mythology that reflects its social cosmologies and ideologies. From this viewpoint, economic discourse functions as an ecomyth in industrial society, guaranteeing maintenance 15  This internalization of the social and linguistic code by the social group takes place though the establishment of institutions that refer to the myth established by the story. 16  In “Le vrai du faux, fondements mythiques de la pensée ordinaire du développement,” (1986), Fabrizio Sabelli considers the myth either as a story of origins or as a product of history. Founding mythical stories generally take the form of a narrative account that functions as a Truth or a True word, even if they are in reality false (that is, the fruit of a sort of collective storytelling), for in constituting the categories within which cultures take root, in establishing the fundamental rules of social existence, in laying the foundations of signification and communication, and in ruling on the order of society, they furnish the tangible proof of their real effectiveness. The paradox is that the myth is false in its reality and true in the social imaginary. Therefore, the truth of a myth is measured by its degree of social effectiveness. Contemporary myths, even if their roots descend into the depths of origins, are essentially historic, thus constructed. Historic accounts are constructs whose components are not consciously chosen by the citizens. Some sociologists speak of stratified modern mythology: the oldest mythical elements functioning as archetypes would be the foundation of a whole mythical architecture formed in strata where the oldest stories would make the myths of the modern world appear. Myths are necessary to human groups, for they are the expression of the struggle of individuals and societies against non-fulfillment; in modern societies, however, they can be selected, imposed, and manipulated by the state, information companies, private capital, and so forth.

342   Methodological Issues of the industrial social order. It fosters representations of the universe and society, legitimizes institutions, ensures beliefs, shapes modes of life and thinking that make possible the organization and arrangement of reality in conformity with the message initially given by the story. Wellbeing, Progress, Growth and Equality are key concepts of the Western cosmology that condition its reading of the real and that it has imposed on other peoples through one of its most powerful myths: development. Thus, through the developmentalist ideology, it was proposed to Africans that they reproduce a prefabricated model of society where there was no place planned for their local culture and where it is generally evaluated negatively.17 This while omitting the fact that Western development is an economic project but above all cultural, produced by a particular universe. Ze Belingua (2009) deconstructs the ideological and paradigmatic frameworks that, consciously or not, relegate the strategic African cultural advantage to mere decoration, principally by treating the material and immaterial African cultures as folklore. Thus the modernization proposed to Africans has imposed significations often absent from their everyday lives and at times produced misconstructions in societies pressured to reproduce a history that they did not experience. A great many social categories were not invited to the discussion on their future, and meanwhile had no choice but to invent survival strategies that took the form of a popular economy which, though it ensures the subsistence of a large proportion of African populations, is struggling to be articulated efficiently into the formal economy. This transposition of the Western myth of Progress has resulted in a deconstruction of the basic personality of African social groups and existing solidarity networks with their systems of signification, but above all in locking populations into a value system that is not their own. This ecomyth has become hegemonic on one hand, by projecting the Western view of the finalities of the social enterprise onto African societies, and on the other, by being tempted to shape all social practices. Hence the necessity for most African countries to work out an economic and social plan, basing it on their socioculture and coming from their own mythological universe and worldview. This poses the question of the dialectics between the particular and the universal in the economic and social field. All societies share a universal requirement to attempt to answer the fundamental needs of a social group by an approach, a plan. The practical modalities to achieve this, defining and establishing a hierarchy of needs, and creating a scale of values are, however, multiple and varied and are dependent on the particular specificity of each human group. It must nonetheless be stressed that this desire of an articulation between economics and culture raises a few difficulties. There are different systems of culture in the African societies: a popular mass culture that seems to oppose an elite culture chiefly founded on models imitating extraverted consumption; an urban globalized youth culture that is a consumer of the global mass sub-culture (tubes, clips, videos), and a so-called traditional culture. There are also power relations within groups that often result in the imposition of a dominant culture by the elite. Nevertheless, in different individuals, culture is a palimpsest composed of different superposed strata and borrowing from different universes of reference, which Quiminal (1991) called the contemporaneousness of several worlds. 17  Axelle Kabou (1991) is one of the figures of Afro-pessimism who explains the poor development of the African continent by an inadequacy of African cultures to progress.

Economics and Culture in the African Context    343 If culture is considered to be a constantly redefined transactional concept, a constantly renegotiated dynamic notion, the difficulty is to identify stable distinctive traits that could be considered the framework of the basic personality of groups and communities. However, though there is no immutable essence or unalterable identity, it is obvious that this basic personality of communities does exist and that it changes slowly. Go to the Betis in Cameroon or the Serers in Senegal and you will find shared habits and customs, references and values, a way of grasping relations with others, and all these cultural traits influence their relation to economic matters. It can therefore be stressed that in contemporary African societies, despite the weakening of the ability of “tradition” to regulate behaviors, certain traits of traditional culture have survived—particularly those linked to prestige expenditures, to investment in symbolic goods, to the injunction to generosity, and the internalized notion of a duty to render aid. The latter could partially explain the large sums of funds transferred by migrants that are noted in sub-Saharan Africa (these could also be analyzed along the lines of gifts and counter-gifts). These survivals are, however, challenged by utilitarian and individualistic principles imposed by the transformation of social relationships brought about by a societal modernity that now makes of individual success the driving force of the social enterprise. Nonetheless, an observation of the behaviors of the economic agents shows that self-interested rationality does not always prevail in their choices. They do not necessarily seek to optimize their utilities or their profits and a variety of rationalities are at work in their decision-making processes. Homo africanus is not a homo economicus in the strict sense of the expression. His choices are, among other things, motivated by principles of honor, redistribution,18 subsistence, gifts and counter-gifts, ritual gifts, and so forth. A great many activities take place off the market or in a setting of primary sociability (families, personal relationships) or secondary sociability (social system that includes an injunction to generosity). Despite the various mutations taking place, however, the cultural context of contemporary African societies, although composite and borrowing from many universes of reference and constantly renegotiating and reevaluating its values, remains a powerful determiner of the economic choices of its members. To better base the economic domain on the sociocultural domain the first thing needed is a theoretical point of view aimed at better understanding and analyzing the determiners of the individual and collective choices of the societies in question by applying to them adequate conceptual tools. A thorough understanding of these choices is a necessary preliminary of any policy aimed at improving the living conditions of these groups. In order to do this, the theoretical gap in dominant economics must first be filled. Analytical frameworks founded only on the postulate of methodological individualism fail to account for the determiners of Homo africanus’s economic behaviors. The examples in Section 1.1 clearly show this and lead one to doubt the explicative ability of this postulate as the foundation of the choices of individuals in African societies. In

18 

It has been shown that in Africa in general and in Senegal in particular, wages profit not only the wage-earner and his immediate family (spouse and children). A non-negligible portion is redistributed to close relatives (parents, siblings, aunts, uncles, and grandparents). These can be called solidarity expenditures.

344   Methodological Issues its formal analyses, dominant economics has minimized the impact of the culture of individuals by considering their behaviors as manifestations of universal characteristics that could be fully captured by the rational model of the individual maximizer independently of cultural and historic contexts.19 Behavioral economics shows that although agents react to incentives or changes in the price of goods, their decisions are not always founded on a cost–benefit calculation. Thus a better understanding of the causes of their behavior—their psychology, culture and social realities—would make economic policies more efficient.20 Areas never thought of in the neo-classical paradigm must be taken on by including holistic and systemic approaches in the reasoning, by taking into account collective strength, socially constructed institutions, the historic structures of social formations, and by identifying other modes of regulating production that allow for collective choices influenced by culture and not only by individual preferences. Recourse to works of economic anthropology, socioeconomics, institutional psychology and economics can be of help in including economics in a greater social and natural system. This first challenge is theoretical: it is a matter of producing an economic discourse that grasps the foundations of observed economic practices.

17.3.2  Assigning economics and culture to their respective finalities 17.3.2.1  Avoiding the confusion of orders The second challenge consists in taking better account of the finalities aimed at and valorized by individuals and groups through all the mechanisms of social exchange and, in the interest of efficiency, assigning to them adequate instruments. From this viewpoint, culture would be asked to define the goals valorized by individuals and groups, and economics as a discipline would have as function to explore the means of allocating resources to achieve these goals. Thus economics would renounce its hegemonic temptation to shape all social practices and would again become an order of means subject to the finalities valorized by individuals and groups. Since the eighteenth century, philosophers, jurists, and economists have been interested in measuring social well-being, exploring its arguments, and maximizing it (Jeremy Bentham 1825; John Stuart Mill 1871; Rawls 1971; Sen 2009; Nussbaum 2000). Welfare economics took as its objective maximizing the wellbeing of the greatest number. Jeremy Bentham and John Stuart Mill principally founded this on the concept of utility. It was a question of maximizing individual and social utility, income being considered one of its principal arguments. Sen (1980) critiqued the limits of an exclusively utilitarian view of wellbeing, based chiefly on income, by proposing a capabilities approach.21 19  Even when it attempts to capture these effects, it does it on its own terms. Guido Cozzi (1998) interprets culture as a social capital that is a factor in the classical production function. 20  See the World Bank’s Report on World Development 2015. 21  The basic concepts of the capabilities approach are functionings and capabilities. Functionings concern multiple aspects of life that a person values and has a reason to value. They are accomplishments or achievements in terms of activities and states. Functionings can be very elementary in nature—being

Economics and Culture in the African Context    345 Capabilities define freedom or the aptitude to be and do what one values and has reason to value; it is a theoretical framework based on real freedoms. In other words, the evaluation of wellbeing is focused not only on the means of existence, like income, but on the real possibilities to live that individuals have (Sen 2012a).22 It is a matter of pursuing the possibility of leading a good life according to one’s own evaluation criteria. Martha Nussbaum23 shifts the reflection further to public policies and their evaluation, advocating an economics of care that would succeed in seeing to the idiosyncratic needs and preferences of each individual by offering them a space of satisfaction.24 Sen and Nussbaum’s approaches seem interesting, for they make it possible to avoid the pitfall of economicism and reintegrate economics into a larger social system. Thus the view of wellbeing that consists in the possibility of leading a good life according to one’s own criteria of evaluation makes it possible to avoid projecting myths and themes coming from other universes of reference onto groups or individuals. Economic public policies would thus be put in charge of favoring the satisfaction of the psychosocial functions of groups. The matter of defining the dimensions to valorize in order to improve the wellbeing of individuals can nonetheless pose a problem. Sen and Nussbaum diverge on the most appropriate approach. Nussbaum considers that one must start with a predefined list of functionings while Sen thinks that this list should be particular to the contexts studied and be determined by public reasoning. One of the solutions could consist in defining a minimum ethical standard shared by all as finalities to be valued:  education, health, nutrition and fundamental rights, and then drawing up lists of functionings by formulating normative hypotheses based on societal values reflected by the prevailing social and religious theories. For example, if it seems important for the wellbeing, equilibrium and cohesion of a group (the members of a Sufi brotherhood for instance) to devote an entire day of the week to religious ceremonies or devotion, an evaluation consisting in valuing this non-working day exclusively in terms of loss of productivity, or of less economic value, would be partial and would not take into account the fact that by taking charge of a psychosocial function of this group, the subjective wellbeing of its members is increased. In order to avoid the trap of the dominant ideologies of a group, a second path could consist in engaging a discussion in the course of which individuals would deliberate on the in good health (health is a capability); having enough to eat; being spared an early death, etc.—or very complex—being happy; engaging in the life of one’s community; retaining self-respect, etc. Rather than states or activities, capabilities define the freedom or ability to do what one values and has a reason to value. In the space of functionings, capabilities can be conceived as being an n-tuple of functionings that symbolize the different kinds of life the individual can lead and among which he chooses one. Sen (2012b) contends in this sense: “Capabilities are defined by derivation from functionings and they include, among other things, all the information on the combinations of functionings that a person can choose.” Furthermore, a central element in the definition of capabilities is freedom in its positive sense, that is, as opportunity that a person has to lead a good life. Therefore capabilities indicate, in a body of vehicles of functionings, whether the individual is free to lead this or that kind of life. 22  Poverty and inequalities remain objectives of economic policy, but it ought to go beyond them by recognizing the multidimensional character of the notion of wellbeing. 23  The choice of capabilities can be based on an existing, predefined list like that of Nussbaum (2000). 24  Nussbaum’s list consists of: a long life, bodily health, bodily integrity, senses, imagination and thought, emotions, practical reason, sociability (affiliation), being able to live with other species, play, and control over one’s environment. As for Sen, he maintains that any list should be particular to the contexts studied and be determined by public reasoning (2004: 78).

346   Methodological Issues dimensions of their lives that can be valued. In short, they could base themselves on empirical evidence by analyzing behavioral data, or the beliefs of the individuals, in order to construct a body of dimensions that seem to represent their values. With the goal of deciphering the underlying motivations of economic behavior, anthropologists, sociologists and researchers in the social and human sciences25 interested in the African continent have emphasized that in this geographic zone objects, goods and services circulate in a relational economy that seems to give primacy to interpersonal and intercommunity relationships.26 This economy, which Maurice Obadia (2012) called a relational economy, seems to be the most powerful determiner of exchanges and the framework of the material economy. Although this explanatory sketch needs better empiric support, it is worth looking at more closely. Maurice Obadia (2012) defined relational economy as being founded on the production and exchange of authentic relationships. This economy is primary and precedes material economy. Relationship is a tie that individuals voluntarily establish between themselves, that is to say with particular material structures, independent of their market value (example: the attachment or aversion that an individual can have for an object, a being, a place). The entire range of positive or negative relationships that individuals can create between themselves—producing, exchanging, and perpetuating over the course of time outside of any true material consideration and interest—constitutes the sub-stratum of relational economy. The internal and external relational network thus constituted acquires such quality and strength that it constitutes a value in itself, not needing the imperative presence of the material in order to exist, and capable of functioning outside the determiners of classical economics. Money can be a consequence; it is not the objective. This relational economy can be at the root of a collective understanding within a community (group, venture, farmers’ cooperative) and be a source of value addition. All creation of wealth, therefore, assumes an interaction with this economy. Obadia notes that the material and relational economies can mutually feed each other, provided that they clearly understand that neither of the two is the objective of the other and that they mutually recognize their respective territories as well as their rules of functioning. Thus the objective of the relational economy is to produce quality relationships between individuals, relationships that in themselves constitute values.27 The example of the Mourides in Senegal can illustrate the ties between a relational and a material economy, the first appearing as the framework of the second. The Mourides are a Sufi brotherhood of Senegal whose founder, Sheikh Amadou Bamba (1853–1927), led a peaceful cultural resistance to colonization, relying for support on Islamic values reinterpreted by sub-Saharan culture. In this community, there is a culture of work that some have compared to the Protestant work ethic analyzed by Max Weber (1964). It is founded on a hadith (statement) of the Prophet Mohammed taken up by the brotherhood’s spiritual 25 

Herskovits (1952). Widespread practices of tontines, gifts and counter-gifts, and so forth seem to attest to this. 27  To produce a relationship worthy of the name, rare, limited, and exhaustible factors are called upon: energy and information over a defined period of time. These are the same factors involved in the production of material things, to which must be added raw materials and capital. It is nevertheless more costly to produce a quality relationship, for it requires energy, time, and diversified information over a significant time period. 26 

Economics and Culture in the African Context    347 guide and set forth as follows: “Work for this life as if you were eternal and for the hereafter as if you were to die tomorrow,” as well as on various recommendations of the brotherhood’s founder taken from his sermons/poems: “I recommend to you two things and do not add to them a third: they are work and the worship of God. Thus you will obtain tranquility …” This community certainly has a culture of work and effort, but also of engagement, of a gift of self, and obedience to the ndiguels, which are the prescriptions of the community’s spiritual guide. The latter is able to mobilize a large, unpaid work force for various tasks of community interest, for example to clear a 175-square-mile forest for agricultural production. So most of the economic interactions are based on ties that unite the members of this brotherhood. This is an example of a flourishing material economy whose determining principles are the relational economy. The latter is characterized by the existence of an intra-brotherhood solidarity that makes it possible to carry out basic economic operations while minimizing the transaction costs, as the relationships are based on confidence and respect of word once given. There is thus, among the members of this community, a system of transfer of funds by compensation when they are on business trips, thus avoiding costs of the classical banking system, an establishment of economic networks and solidarities, and a tradition of making capital available without cost when an economic activity is launched, and well as ease of reimbursement. Mouride tradesmen form the main part of Senegal’s informal economic sector, notably in commerce, construction, transportation, textiles, processing, etc. Their social and economic success is due to a very great solidarity characterized by a common ideal and the conviction of belonging to a community. This popular economy, founded on sociocultural and religious values shared by a group, is dynamic and results in this community’s controlling large sectors of Senegal’s so-called informal economy, employing 60 percent of the working population and representing 54.2 percent of the GDP.28 The city of Touba, headquarters of the brotherhood, is Senegal’s second city, both demographically and economically.29 The preceding example illustrates the goal, which is to preserve the foundations of this relational economy and to avoid transpose to it the mechanisms of classical economics. According to Obadia, the spirit of classical economics can negatively influence a relational economy and result: those who calculate their relational production at minimum cost will end up in a negative relational economy.30 28 

El HadjI Ibrahima Sakho Thiam (2010) notes that five generations ago, commerce in Senegal was in the hands of the Lebanese and Syrians. The situation has changed since then and the Mourides now have a monopoly on trade. Until the 1970s the Mouride brotherhood was considered a theocratic, agrarian movement because of its large production of peanuts. In the early 1980s they invested in the small retail trade and have now gone beyond the stage of large retail and wholesale to reach an industrial and entrepreneurial level. 29  Brunau Lautier (2004) emphasizes that the principle of this informal, relational economy is first and foremost social, the objective here being to satisfy the needs of the family in its broad sense, but especially to give work to relatives, cousins, and nephews. The profit realized is redistributed within the network and does not result in an accumulation allowing the enterprise to grow. The principle of the reproduction of the social group seems to have priority and determines all the aspects of this economy. Taking the example of the Mouride community, it is, however, apparent that this kind of economy can go beyond this first stage and become part of a process of growth and industrialization. 30  A positive relational economy needs energy in quantity and quality, as well as diversified knowledge.

348   Methodological Issues

3.2.2  Culture as a force of social adaptation and change Culture refines the perception and comprehension of the world in a given social milieu; it can also organize and rationalize the social experience of the individuals concerned. This being the case, it is possible to direct the work of identity production toward purposes of collective mobilization around societal projects or issues. The social and/or cultural identities to be mobilized can move into new fields of competence and expertise. This is a matter of grasping the way in which culture works as a force of adaptation, but also of social change. African societies are subjected to many dynamics and profound mutations are taking place. Important social, economic, demographic, political and cultural recompositions are occurring and are radically transforming social relations and practices. Sainsaulieu (1987: 229) notes that culture “articulates, on one hand, the body of representations of a society through systems of values and symbolic codes, indeed even myths, tending to reproduce and maintain the social order. On the other hand, collective representations of change propose another view of the world, another conception of rationality, and of the role of social groups and the distribution of power.” It then appears as that which makes a society capable of “reacting to the event and transforming its social structures in the face of external pressures.” Faced with mutations and changes, the challenge here would be practical. It is a matter of improving the efficiency of existing economic practices in order to better augment the value of available resources and meet the needs of societies. There are in the African societies reservoirs of knowledge, endogenous expertise,31 modes of social organization of work, distribution of goods and allocation of resources that have demonstrated their effectiveness and survived the deconstruction of social systems. The function of culture is also to ensure the preservation and transmission of the technical, intellectual and moral assets of a given social group. Consequently, the efficiency of the modes of economic production could be improved by taking inventory of existing modes of organization and production, preserved by culture, adapting them to new contexts, optimizing their effectiveness by incorporating new technologies, and disseminating innovations that could build on the expertise of the populations.

17.4 Conclusion This study has explored the possibility of a fruitful interaction between economics and culture in the African context (but not only there). It has brought to light the cultural foundations of the agents’ economic choices and emphasized the fact that individual decision-making processes are strongly influenced by the individual’s cultural environment, which conditions his preferences and regulates his behaviors. The function of the cultural matrix of individuals is to model their desires (needs) as well as the circumstances (temporality, place) of their satisfaction. It thus appears that the different conceptions that 31  A great range of knowledge and traditional crafts could be concerned: weavers, basket makers, shoemakers, expertise in food processing and preservation, various arts, and phytotherapeutic knowledge.

Economics and Culture in the African Context    349 individuals and households have of economic management are the result of different social evaluations of time, work, leisure, and social and religious obligations coming from their respective cultures. It has furthermore emerged that cultural considerations also affect the way in which economists practice their discipline, elaborate their postulates and model interactions between agents. Economic discourse functions as a language that ensures the establishment of the common symbolic code through which the group’s way of speaking, thinking and experiencing the real is developed. This language is not neutral; it implements a particular structuring of time and space. This chapter shows that a fruitful articulation between economics and culture could be achieved through a better rooting of African economies in their sociocultures. This study defends the idea that this could be done only by limiting the power of economics; by assigning to it a strict respect of its functions and its technical role of exploring the knowledge and expertise linked to the allocation of resources and the production of economic goods. By thus entrusting the production of values and significations to culture, culture becomes the producer of regulating myths and the organizer of the social enterprise. The ability to reappropriate its future and invent its own teleologies, to order its values, and to find a harmonious equilibrium between the different dimensions of existence will depend upon the ability of the African cultures to conceive of themselves, as Mudimbé (1982) suggested, as projects taking on the present and the future and having as their goal the promotion of freedom in all its expressions. From this point of view, a fruitful articulation between economics and culture would be achieved by assigning each order to the finality for which it is the most efficient.

References Arendt, H. (1972). La crise de la culture. Translation by Patrick Levy of Between Past and Future. Paris: Gallimard. (Original edition 1961.) Bentham, J. (1825). Théorie des peines et des récompenses. Edited by Etienne Dumont. Paris: Bossange Frères, Libraires. Cazeneuve, J. Civilisation. In Encyclopaedia Universalis. Accessed October 31, 2014. http:// www.universalis.fr/encyclopedie/civilisation. Charmes, J. (2000). The Contribution of Informal Sector to GDP in Developing Countries:  Assessment, Estimates, Methods, Orientations for the Future. OECD EUROSTAT State Statistical Committee of the Russian Federation, Non-Observed Economy Workshop, Sochi (Russia) 16, 20 October 2000. Cozzi, G. (1998). Culture as a bubble. Journal of Political Economy, 106(2):376–394. Foucault, M. (1969). L’archéologie du savoir. Paris: Gallimard. [English title: The Archaeology of Knowledge.] London and New York: Routledge (1972). Foucault, M. (1966). Les mots et les choses. Une archeology des sciences humaines. Paris: Gallimard. [The Order of Things: an archaeology of the human sciences]. Herskovits, M. (1952). Economic Anthropology. New York: Knopf. Horkheimer, M., and Adorno, T. (1974). La dialectique de la raison. Translated from the German by Éliane Kaufholz. Paris: Gallimard. (Original edition 1947.) Kabou, A. (1991). Et si l’Afrique refusait le développement? Paris: L’Harmattan.

350   Methodological Issues Kroeber, A.L., and Kluckhohn, C. (1952). Culture: A Critical Review of Concepts and Definitions. New York: Vintage Books. Mill, J.S. (1871). Utilitarianism. London: Longmans, Green, Reader, and Dyer. Mudimbé, V.Y. (1982). L’odeur du Père: essai sur des limites de la science et de la vie en Afrique noire. Paris: Présence Africaine. Nussbaum, M.C. (2000). Women and Human Development:  The Capabilities Approach. Cambridge: Cambridge University Press. Obadia, M. (2012). Economie relationnelle et économie matérielle. Cahiers de Sol 9:  L’intelligence Collective. Quiminal, C. (1991). Gens d’ici, gens d’ailleurs:  migrations Soninké et transformations villageoises. Paris: C. Bourgois. Rawls, J. (1971). Theory of Justice. Cambridge, MA:  Belknap Press of the Harvard University Press. Sabelli, F. (1986). Le vrai du faux, fondements mythiques de la pensée ordinaire du développement, in F. Sabelli and G. Rist (eds), Il était une fois le développement. Lausanne: Editions d’en bas. Sainsaulieu, R. (1987). Sociologie de l’organisation et de l’entreprise. Paris, Dalloz/FNSP. Seers, D. (1963). The limitation of the special case. Bulletin of the Oxford University Institute of Economics & Statistics, 25:77–98. Sen Amartya K. (2004). Dialogue capabilities: lists and public reason: continuing the conversation. Feminist Economics, 10(3):77–80. Sen Amartya K. (1980). Equality of What? French translation in Ethique et économie. 5th ed. Paris: Quadrige, 2012, pp. 189–213. Sen Amartya K. (2009). The Idea of Justice. London: Penguin Books, Ltd. Sen Amartya K. (2012a). L’idée de justice. 2nd ed. Translation by Paul Chemla and Eloi Laurent of The Idea of Justice. Paris: Flammarion. Sen Amartya K. (2012b). Ethique et Economie. 5th ed. Translation by Sophie Marnat of On Ethics and Economics. Paris: Quadrige. Thiam, El hadji Ibrahima Sakho. (2010). Les Aspects du mouridisme au Sénégal. PhD diss., Université de Siegen. Throsby, D. (2001). Economics and Culture. Cambridge: Cambridge University Press. Weber, M. (1964). L’éthique protestante et l’esprit du capitalism. Translated from the German by Jacques Chavy. Paris: Plon. [The Protestant Ethic and the Spirit of Capitalism]. Ze Belingua, M. (2009). Aggiornamento! Pour une approche alternative de la culture en Afrique contemporaine. Présence Africaine, 1–2:179–180.

Chapter 18

The Ec onomi c s of Non-c o gniti v e  Sk i l l s Laura Camfield

18.1 Introduction Non-cognitive skills and related concepts such as “life skills” or “social skills” are becoming increasingly important to policymakers; however, they are rarely clearly defined or comprehensively measured (Duckworth et al. 2009). When these concepts and measures are used in applied fields such as development studies, it is essential to understand what they mean and how they work: specifically, how individuals learn and acquire non-cognitive skills in different contexts, and how these develop over time, including their transmission across generations. In this chapter I first define non-cognitive skills, acknowledging the multiplicity of definitions (Duckworth et al.: 51–53). I explain why they are measured and the potential they offer in understanding socioeconomic outcomes. I note the problems with taking an individualized approach to a global economic phenomenon, namely youth unemployment, or assuming that the skills captured by these measures are universally valued or valuable. Finally, I use data from a study of young African entrepreneurs’ non-cognitive skills to highlight concerns with how these skills are conventionally measured. My interest in non-cognitive skills is framed within comparative anthropology, which asks what is the same and what is different in different contexts, and sociology, which asks how particular skills become capitals, that is, acquire market-value in different contexts.1 Using these measures outside the global North highlights the fact that high self-esteem and tolerance for risk are not universal goods, and may reflect the demands of contemporary Western societies. In international policy contexts non-cognitive skills may be becoming a form of embodied cultural capital,2 that is, legitimated cultural attitudes, preferences, and 1  Bourdieu (1976) uses the analogy of card games where players have different cards, but the outcomes are also dependent on the rules of the game and the skill with which it’s played. 2  I use Lamont and Lareau’s (1988, p. 156) understanding of cultural capital which draws on Bourdieu and Passeron’s work on social reproduction: “Institutionalised i.e. widely shared, high status cultural signals (attitudes, preferences, formal knowledge, behaviours, goods and credentials) used for social and cultural exclusion.”

352   Methodological Issues behaviors or practices that are internalized during socialization processes, which exacerbate and legitimize existing inequalities. For example, non-cognitive skills are used as an explanation for the unemployment of educated young people in urban Africa (e.g. Langevang 2008, Mains 2013) and entrepreneurship is presented as a potential solution. The reason for this is the role of small and medium-sized enterprises (SMEs) in Asian growth (Beck et al. 2005). However, there are few entrepreneurs in Africa compared to Asian and Latin American countries at similar stages of development (Chigunta et al. 2005; Naude 2008) and SMEs currently represent a “missing middle” between large firms and microenterprises (Kingombe et al. 2010). Can a lack of non-cognitive skills offer an explanation for this complex and entrenched problem? Or does this explanation elide structural factors, including global inequalities, in favor of narratives that focus on individual deficits, which create “a problem without a cause” (Izzi 2013).

18.2  Economics Approaches to Non-cognitive Skills Within economics the concept of non-cognitive skills developed from a realization that human capital did not provide a full explanation of how labor markets function, who got jobs and how they were paid (Bowles and Gintis 1976). Even with an expansive definition of human capital (e.g. Becker and Tomes 1986), there was still unexplained variation in how labor markets value people’s skills. Non-cognitive skills originated to fill this gap, which accounts for their broad specification. James Heckman (2008: 298), a key figure in this field, defines non-cognitive skills as “motivation, socio-emotional regulation, time preference, personality factors, and the ability to work with others.” Arguably these are examples of non-cognitive skills rather than a definition of the concept, that is, what links these diverse attributes together (I return to this point later in the chapter). The examples of non-cognitive skills provided are wide-ranging, even within papers by the same author (Table 18.1). This lack of precision enables variation in the type and quality of measures used as authors are justified in using almost any indicator to represent this concept. The value of knowing what you are measuring is illustrated by Mischel et al.’s (1972) infamous “marshmallow test” with young children which purportedly established the predictive value of a willingness to delay gratification. A subsequent study by Kidd et al. (2013) identified that the main determinant of children’s waiting was not self-control, but whether they believed the promise would be fulfilled, favoring children brought up in more stable environments. The main conclusions of the studies presented in Tables 18.1 and 18.2 are that non-cognitive skills often correlate weakly with IQ or school achievement and are sometimes a stronger predictor of achievement. When young people from the global North leave school, non-cognitive skills are more important than test scores or years of schooling in securing employment (Bowles and Gintis 1976, 2002) and ultimately higher incomes—in fact, Bowles et al. (2001) argue that cognitive skill only accounts for 20 percent of effect on earnings. This point is illustrated by the performance of graduates of the General Educational Development (GED) testing program3 in the United States (Heckman and Rubenstein 2001). Although GED 3  GED is a “second-chance program” that administers cognitive tests to self-selected high-school dropouts to determine whether they are the academic equivalents of high-school graduates.

The Economics of Non-Cognitive Skills    353 Table 18.1  Examples of non-cognitive skills Examples

Reference

Motivation, perseverance, trustworthiness, adaptability, task persistence, thinking ahead and self-discipline Motivation, perseverance and tenacity Socio-emotional skills, physical and mental health, perseverance, attention, motivation, and self confidence Perseverance, motivation, time preference, risk aversion, self-esteem, self-control, preference for leisure, patience, risk aversion and time preference “Social skills” Social adaptability, motivation “Characteristics,” “soft skills” and “personality traits” “Behavioral traits” “Soft” skills

Heckman and Rubinstein 2001 Heckman 2006 Heckman, Stixrud, and Urzua, 2006 Cunha and Heckman, 2008

Carneiro et al. 2007 Heckman 1999 Blanden et al. 2006 Bowles et al. 2001 Holzer et al. 2001

certificate holders have the same cognitive skills as graduates, they typically command a lower income than either graduates or other drop-outs. Heckman and Rubenstein (2001) suggested their certificate “signals” to employers that they have fewer non-cognitive skills, essentially they are “smart but unreliable” (149), and so employing them might entail higher transaction costs. Non-cognitive skills have a genetic component, but are more responsive to “nurture” and more malleable than cognitive skills (Knudsen et al. 2006; Cunha and Heckman 2008). For example, the Perry Preschool program, discussed later, demonstrates that educational interventions can have positive effects even if they do not increase IQ (Heckman, Stixrud, and Urzua 2006). In the following sections I look at whether non-cognitive skills are genetic, what we know about their development and effect on outcomes, and how they relate to personality. Are non-cognitive skills genetic? The greater malleability of non-cognitive skills is partly due to a longer “sensitive” period for their development.4 This extends to adolescence and secondary schooling, giving them a larger “window” for intervention (Carneiro and Heckman 2003). The neurological explanation for this is that the pre-frontal cortex which governs emotion and self-regulation is malleable until people’s early 20s. Nonetheless, early investment, for example, by age 8, is still considered desirable as the multiplier effects described later in the chapter mean that investments will generate higher economic returns (Peterson and Zill 1986). The Perry Preschool Project (Heckman et al. 2009) shows the value of early intervention as it produced good economic and social outcomes in adulthood, especially among women (Heckman 2005). Table 18.2 summarizes other economic studies showing the predictive role of non-cognitive skills. The main conclusions are that the timing of parental “investment” or external intervention is crucial; parents’ own non-cognitive skills influence their children’s (de Coulon et al. 2011), suggesting some degree of hereditability (Duncan et al. 2001); and there are gender differences in both the transmission of non-cognitive skills and their effects (e.g. mothers’ skills have a greater influence on daughters than on sons, 4  The estimated cut-off for non-cognitive skills ranges from age 13 to early 20s vs. middle-childhood or even earlier for cognitive skills (Cunha and Heckman 2008).

354   Methodological Issues Duncan et al. 2001). These conclusions need to be tested outside the global North where the greater availability of panel data has made these analyses possible. Further points to note are that while some non-cognitive skills such as extraversion and conscientiousness—one of the “Big five”5 personality traits—are stable in childhood, others reach stability in adulthood. Most non-cognitive skills are self-productive, which means that skills acquired at one stage in the lifecycle enhance skill formation at later stages (e.g. the more self-control you have now, the more self-control you will have in the future). There are also synergies or “dynamic complementarity” between skills so self-control and emotional security may reinforce intellectual curiosity and promote learning (Shiner, Masten, and Roberts 2003). For example, Cunha et al. (2006) observed that students with greater early cognitive and non-cognitive abilities are more efficient in learning both cognitive and non-cognitive skills later in life and consequently less likely to engage in problem behaviors. They explain in Cunha and Heckman (2008) that non-cognitive skills make children more open to learning, increasing their cognitive skills. These “multiplier effects” explain why even investments made relatively late in childhood are beneficial. Heckman and others argue, on the basis of data from North America, that the high social costs of failing to invest in young people outweigh the expense of, for example, life skills programs. The growth of these programs in the developing world suggests that their arguments are influential. They also note that early investments need to be followed up to create a virtuous cycle of increasing skills; the lack of follow-up accounts for the weak long-term effect of early childhood care and development programs in Europe and the USA.6 How do non-cognitive skills relate to other psychological measures like personality? The heterogeneity among non-cognitive skills, which was a concern I flagged at the start of the chapter, has been recognized by economists. For example, Blomeyer et al. (2009) found two distinct factors (intrapersonal and interpersonal), Heckman, Stixrud, and Urzua (2006) found three, and Borghans et al. (2008) noted low correlations between different non-cognitive skills relative to correlations between tests of cognitive ability. Not only are non-cognitive skills diverse, but they may also have different effects, for example, Roberts et al. (2005) showed that lower-level facets of the personality factor “Conscientiousness” (e.g. industriousness) are better predictors of labor market outcomes. Concerns about the scope of non-cognitive skills have prompted a growing interest among economists in personality, defined by Borghans et al. (2008: 3) as “patterns of thought, feelings, and behaviour,” as a distinct and measurable subset of non-cognitive skills (e.g. Nyhus and Pons 2005; Almlund et al. 2011). The importance of personality is intuitively plausible given that sociability, empathy, and the capacity to get along with others, which are related to the Big Five factors of Agreeableness and Extraversion, are predictors of success in many activities worldwide. The five-factor model is widely used by economists due to its inclusion in panel and cohort studies, which enable economists to track the influence of personality across the life course. It has been tested in Africa,7 and particularly in South Africa where there is a lively debate as to its validity relative to indigenous personality models (e.g. Laher 2008). 5

  The Big Five model of personality comprises Extroversion, Neuroticism, Openness to Experience, Conscientiousness, and Agreeableness. 6   There is little comparable evidence on the long-term effects of these programs in Sub-Saharan Africa as they are a new form of intervention, although there is comparable data from Latin America. 7   The Neo-PRI bibliography lists four validated translations into African languages (Costa and McCrae 2011).

The Economics of Non-Cognitive Skills    355 Economists also have some concerns about using personality as a measure of non-cognitive skills, for example—are personality “traits” less malleable than non-cognitive “skills”?8 Are personality traits available to all activities, or does greater use in one activity means there is less available to use in another? Do they directly influence decision making or only through their effect on moods and emotions? (Loewenstein et al. 2001) Can they even be characterized as non-cognitive given that they also influence cognitive processes? (Borghans et al. 2008). In the following section I briefly discuss the potential benefits of focusing analyses of non-cognitive skills on personality, before outlining some of the problems. Borghans et al. (2008) reviewed an impressive body of evidence on the predictive power of different personality traits vs. IQ (Borghans et al., Fig. 3) and finds that Conscientiousness, which controls risk aversion, leisure preference (i.e., orientation towards work), and time preference, is most important, followed by Openness to experience. Personality and IQ are said to be weakly correlated, with the exception of Openness to experience, which has been shown to influence learning, and Conscientiousness, which might affect performance on IQ tests. Specific personality factors may moderate the effect of cognitive skills and resources on outcomes, for example, Judge and Hurst (2007) identified a new factor called “core self-evaluations,” which encompasses high self-esteem, high generalized self-efficacy, internal locus of control, and high emotional stability. This factor enables young people to capitalize on resources such as socioeconomic status, academic performance, and educational attainment. While these resources are normally predictive of high income later in life, the authors found that this relationship did not apply when respondents had below-average core self-evaluations. Nested within the five factors are lower-level personality traits such as impulsivity which may be valuable predictors in their own right. However, it is difficult to reliably assign them to factors, for example, is impulsivity a facet of Neuroticism or Conscientiousness? They may also be situational rather than traits, in the same way that risk preferences are domain specific (e.g. attitudes towards health risks may be different to attitudes to financial risk). The benefits from personality traits do not increase monotonically like IQ so, e.g. more impulsivity is not always better. Their effect may also depend on the amount that people have of them, for example, according to Borghans et al. (2008) only very large or very small amounts of conscientiousness show effects. Finally, solely focusing on personality would exclude other non-cognitive skills such as motivation and time preference.

18.3  Measurement of Non-cognitive Skills in Economic Studies Non-cognitive skills are becoming increasingly important in development as interventions (e.g. life skills programs in the Dominican republic and the slums of Bombay evaluated by 8

  Borghans et al (2008) argue that measures change in response to life events even when people are in their fifties or sixties, however, Bouchard and Loehlin (2001) estimate that the heritability of personality is between forty and sixty percent.

356   Methodological Issues Ibarran et al. 2012 and Krishnan and Krutikova 2012 respectively), measures used in impact evaluations (e.g. Alfonso et al. 2010, teacher placement in Chile), and basic research on topics such as the best incentive structures to support desired behaviors (e.g. reduction of teacher absenteeism in India, Duflo et al. 2010; increasing project participation in Malawi, Lilleor, and Lonberg 2010). They are also implicitly used as an explanation for the persistence of poverty over generations and specifically the poverty of young people (Krishnan and Krutikova 2012). The theoretical background to this is the work of Bowles and Gintis (1976 [2002]) and more recently Cunha and Heckman (2010) on the relationship between non-cognitive skills and adult earnings (see also Heineck and Anger 2010). It also reflects a growing interest among development economists in psychological and sociological concepts such as aspirations and learned behaviors (e.g. Bernard et al. 2014; Laroche 2010) and how they can be used in interventions (e.g. Duflo et al’s (2011) use of “nudge” techniques to increase fertilizer uptake in Kenya). Some economic studies in sub-Saharan Africa use measures of non-cognitive skills (e.g. Bauer and Chytilova 2007 on patience and time discounting in Uganda; Bernard et al. 2014 on enhancing aspirations among Ethiopian farmers) and they are also used within health psychology in HIV-affected areas (e.g. Betancourt et al. 2011, who identify perseverance and self-esteem as protective factors for Rwandan adolescents). There are few panel studies in sub-Saharan Africa (Baulch 2011) and fewer still include measures of non-cognitive skills, although Young Lives (www.younglives.org.uk/ what-we-do/research-methods/methods-guide,downloaded25/08/13) has developed its own measures of agency/self-efficacy, trust, pride/self-esteem, and inclusion. These have been analyzed to assess the extent to which nutrition, material poverty, and gender affect non-cognitive skills in Ethiopia and other developing countries (respectively Sanchez and Dercon 2011; Dercon and Krishnan 2009; Dercon and Singh 2011). The novelty of these measures means that their data cannot be compared with other studies or contextualized—there are no population norms to use as a reference point. The authors admit there is some confounding because the measure of pride/self-esteem asks about experiences such as going to school without shoes, which are more common for poorer children. The findings that poorer children experience lower levels of self-esteem are therefore likely to be driven by what Dercon9 called the “shoe factor” rather than inherent deficits. Comparative cross-sectional studies cannot capture change over time; however, they can show the importance of context, as in the World Bank’s STEPS Skills Measurement study, which surveys young people and adults (aged 15–64) in urban areas in thirteen countries, two of which were in Sub-Saharan Africa. STEPS aims to establish the level of skills in different contexts and the fit between people’s skills, including non-cognitive skill measures such as the Big Five, grit, and decision-making, and those perceived as important by potential employers.10 While there are international skills tests in OECD countries such as PIAAC, this is the first to include non-cognitive skills measures, focus on developing economies, and look specifically at the relationship between what employers want and job seekers have to offer. 9 

Pers. comm., June 2009. http://siteresources.worldbank.org/EXTHDOFFICE/Resources/5485726-1281723119684/STEP_ Skills_Measurement_Brochure_Jan_2012.pdf, downloaded 12/01/14. 10 

The Economics of Non-Cognitive Skills    357

18.4  Sociological Critiques of Non-cognitive Skills Sociologists are also interested in what economists call non-cognitive skills because attitudes, communication, motivation, and personality account for much of the variance in educational progress and employability. But seen from a sociological perspective the economic approach presents both methodological and ideological problems—methodological in that measures of what are likely to be culturally and context-specific constructs are used with minimal adaptation or pre-testing, and ideological as data from these measures support narratives that focus on individual deficits rather than, for example, the structure of the labor market. Two key omissions are class and culture: economists claim that one set of personality traits and behaviors are better and make people more productive, while sociologists argue that these are not inherently better but are valued by the labor market because they replicate the behaviors of the dominant classes in ways that serve their interests. For example, schools contribute to young people’s success in the workplace not only through encouraging intellectual development, but also by acting as a forum for socialization (Lareau and Weininger 2003). They enable social reproduction by creating a hierarchical environment that matches that found within a western workplace—what sociologists call a “correspondence” relationship between the structural relations of production and those in school. The school develops the types of discipline and demeanor that are required in the workplace, and in the workplaces of different social classes (for example, manual workers vs. professionals11). Being socialized into certain patterns of behavior is vital to labor market success and certain traits are more highly valued than others in the labor market (Bowles, Gintis, and Osborne 2001), although some traits, such as working hard are universally valued. This explains the role of non-cognitive skills in socioeconomic mobility and social stratification (Farkas 2003). Similar processes of social reproduction operate in the developing world (e.g. Froerer 2011), although these are less well studied. Nonetheless, although some non-cognitive skills appear to have universal salience, many do not. In fact, the skills and ways of behaving that are appropriate for formal western workplaces may be less useful in the street or the market place (cf. Hoechner 2013, peripatetic Quranic scholars in Nigeria). De Weerdt (2010: 340), for example, identifies successful young people in Tanzania who are “young ambitious apprentices who act as couriers, or go-betweens between the trader and the farmers … over the years they did not only build up assets and financial capital, but also a network of people outside the village, experience in the trade and are exposed to new ideas from outside.” The importance of relationships, being “trustworthy,” and “exposure” come up again in the data from young entrepreneurs, as I discuss later. Formal schooling does not emphasize, and may devalue, these skills. It may also deskill pupils for agricultural employment, which represents the majority of work available to young people in Africa (Katz 2004; Mains 2013). 11  For example, Edwards (1976) suggests that dependability and consistency are valued more by blue collar supervisors than cognitive ability or independent thought and Bowles et al. (2001) identify three types of markets that require and reward different skill-sets.

358   Methodological Issues Few people would deny that a white middle-class North American man is likely to have different experiences, and outcomes of those experiences, to a Ugandan girl who lives just below the poverty line, or even a black North American male. Why then do we assume that the skills that play a part in explaining those outcomes are the same? This point has been made by Schoon (2009), who argued for the importance of understandings that are context specific and focused on the role of specific competencies (skills). These skills are a moving target, not only because individual needs change across the life course, but also because the societies around them change as well, as Wiegratz (2010) has observed in relation to social structures in Uganda after a period of economic transformation. I propose that non-cognitive skills are also adaptive—otherwise they wouldn’t be “skills”—and this adaptability can be seen in ethnographic analyses of the changing value of “human capital” in China, where the “focused, motivated student” is expected to “transform into the entrepreneurial, self-actualizing, pragmatic and technologically sophisticated consumer and modern citizen of a globalized China” (Crabb 2010: 388) and Brazil, where “sociability and manners” are seen as the most important outcome of adult literacy classes because this enables students to build economically productive relationships (Bartlett 2007). Are there universally applicable ways of measuring non-cognitive skills? The ideological implications of the growing interest in non-cognitive skills are not a reason to reject them. However, there is a more substantial methodological critique which I summarize below. Firstly, definitions of non-cognitive skills and choice of measures to capture them are not consistent, making it difficult to compare the results of different studies. We might even ask how it is possible to study non-cognitive skills if we don’t have a clear idea of what they are, although arguably clear definitions and taxonomies of how different skills relate are the outcome of research rather than the starting point. Secondly, even Heckman and Rubenstein (2001:145) admit that “many different personality and motivational traits are lumped into the category of non-cognitive skills,” which makes it difficult to identify the pathways by which they operate and develop successful interventions. Thirdly, cognitive and non-cognitive are not always clearly separated (e.g. the examples of non-cognitive skills in Table 18.1) and some authors argue that they are not actually separable. Farkas (2003: 60) notes that while “agreeableness” and “helpfulness” are examples of non-cognitive skills, “both are dispositions which require knowledge of how to behave in appropriate ways, and of the circumstances in which such behaviour is called for.” Fourthly, the importance of non-cognitive skills may have been overestimated in some economic analyses as a common method of estimation is to subtract the effect of test scores (measured cognitive skill) from the total effect of education and assume that the effects of all unmeasured educational factors are attributable to non-cognitive skills, rather than, say, unmeasured cognitive skills. Finally, measures used in surveys may not be subject to the same psychometric standards as psychological measures and it is common practice to reword and remove items without checking that the scales continue to function in the same way. The focus on non-cognitive skills may misdirect policy attention as both cognitive and non-cognitive inputs are required in, for example, improving exam results, due to the complementarities discussed earlier. There is also a question of whether non-cognitive skills are skills at all, something I am acutely aware of when discussing them in interviews. Is self-esteem a skill? Is it purely non-cognitive, or does it also involve thought and judgment? Is it malleable (and teachable), or innate and reinforced by the processes of social reproduction described earlier? Finally, there is an automatic assumption that non-cognitive skills are

The Economics of Non-Cognitive Skills    359 beneficial, whereas sociological analyses of cultural capital have used micro-interactional studies to establish which cultural practices and skills should be deemed “capital.” The qualitative data presented in the next section suggests this might also be valuable in measuring non-cognitive skills in Uganda and South Africa.

18.5  Studying Non-cognitive Skills with Entrepreneurs in  South Africa and Uganda In this section I  illustrate the potential for contextualized studies of non-cognitive skills with a small selection of quantitative data on the non-cognitive skills of young entrepreneurs in South Africa and qualitative data on how these measures were interpreted and the skills perceived necessary to succeed as an entrepreneur (South Africa and Uganda). These countries were chosen as they have high youth unemployment and underemployment and government-sponsored entrepreneurship programs. Additionally, South Africa has a large body of indigenous research on the applicability (of otherwise) of non-cognitive measures, including scales of entrepreneurial attitudes (Robinson et al. 1991). I used the World Bank’s STEPS measures as the basis for a longer measure of non-cognitive skills which initial qualitative work and consultation had suggested might be important for young South African entrepreneurs.12 This was administered to 683 respondents participating in the Awethu entrepreneurship program.13 Thirty-one percent were female with a mean age of 32.4. 99 percent were black African and the main language spoken was IsiZulu (42 percent). Sixty-five percent were already entrepreneurs and 37 percent had parents who were business owners, illustrating the often dynastic nature of African entrepreneurship. The majority were educated to Grade 12 (52 percent) and >5 percent had lower than Grade 10, which may be an artifact of the selection criterion for the program of fluency in English. There were few significant different between groups by gender or occupational status, and these were largely in the expected direction, for example, women were higher in neuroticism and extroversion. Perhaps surprisingly respondents who were unemployed or casual workers showed higher self-esteem and extroversion than either existing business owners or people with salaried employment. There were also interesting differences when I compared this sample with others; for example, they had lower than average self-esteem relative to both international studies (e.g. Schmitt and Allik 2005) and ones of the general population in South Africa (Maluka and Grieve 2009). This may be due to the predominance of business owners in the sample. They also showed higher than average hope, “agency” and “pathway” thinking relative to both North American College students, who were the sample Snyder et al.’s (1991) hope scale was 12  The additions to the STEPS scales were Rosenberg’s Self-esteem scale (1979), Snyder’s Hope scale (Snyder et al 1991), and Pearlin and Schooler’s mastery scale (1978), all of which have been used in South Africa. 13  http://awethuproject.co.za/,downloaded 13/01/14.

360   Methodological Issues developed with, and South African adults (Boyce and Harris 2013). This suggests that as a group Awethu entrepreneurs are resourceful and well motivated. The most interesting comparison was with the Five-factor personality profile found to be predictive of entrepreneurship in USA, Germany, and UK (highest possible value on extraversion, conscientiousness and openness, lowest possible value on agreeableness and neuroticism, Obschonka et al. 2013). While my sample scored high on openness and low on neuroticism, unlike Obschonka et al’s entrepreneurs they also scored high on agreeableness and relatively low on extraversion and conscientiousness. This finding obviously needs to be tested in other African samples; however, it suggests entrepreneurs’ skills vary across contexts, for example, a greater emphasis on building relationships through being agreeable and even humble, which was mentioned frequently in the focus groups. I used cognitive interviewing to understand what the measures were capturing (Willis 1999). This involved observing people while responding to the questionnaire, noting signs of difficulty or annoyance, and asking them to either describe how they are responding to the measure as they complete it (“think aloud”) or recall what they thought each question was asking and how they chose their answer (retrospective “verbal probing”). I found that respondents’ biographies influenced the way they responded to questions about their non-cognitive skills in ways that were not predictable. For example, respondents who were committed to their business and consequently very self-critical responded in a way that (erroneously) suggested low self-esteem. Similarly, devout Christians among the sample were more likely to say that they were satisfied with themselves because—as Glen, a middle-aged business owner explained “I do not think I should not be satisfied with what God has created.” They were also less likely to say that they were proud, because they understood pride as a sin (e.g. Professor, a young man starting up an internet-based business). This is not a technical problem that can be resolved by tweaking the wording of items (Fahmy et al. 2011), but something that requires an understanding of “higher order normative contexts” (White 2010: 162) where the measures are being administered, suggesting a role for complementary qualitative research on this topic. Finally, I  asked open-ended questions in individual and group interviews with young male and female entrepreneurs about the skills entrepreneurs use in their work or are developing to support their future success.14 Once a list of skills had been generated I asked participants to rank these using pairwise ranking. The most important skills in Uganda were customer care and reliability (mentioned by at least a quarter of the sample). For example, one entrepreneur described how he “entices [customers] by chatting and making fun with them, peel bananas for them, and by the time he/she comes back tomorrow, you tell them a different story.” Market knowledge was ranked highest in South Africa—one salon owner said that “If you are not there on your business then your ears will not hear anything, but if you are there … I can be on track with what is happening outside the world.” Reliability, passion and perseverance were also mentioned as important, although no other skills were mentioned by more than 20 percent of the sample. Although the analysis is at an early stage, it is already clear that gender, age/ level of experience, and type of occupation influence the types of skills that are useful—as found by Borghans, ter Weel and Weinberg (2008) in their 14  The interviews were conducted by me in South Africa in English, N = 41, and with a research assistant in Uganda in Luganda, N = 42, typically at the respondents’ workplaces.

The Economics of Non-Cognitive Skills    361 analyses of British and German skills surveys. Findings such as these question the value of searching for the “holy grail” of transversal skills, as argued by Billing (2003) in relation to higher education.

18.6 Conclusion In this chapter I highlight the increasing interest in non-cognitive skills and traction among policymakers for arguments made using data from their measures. While the concept of non-cognitive skills is appealing—you don’t have to be a psychologist to recognize that people behave in ways that in the right contexts will support their success—we do not yet know enough about what they are, how they develop, how people use them, and how they are valued. This lacuna is particularly visible when measuring non-cognitive skills with young people in Africa. Nonetheless, non-cognitive skills offer some important insights, for example, that cognitive measures also have limitations and national exams or international assessments may tell us even less about young people’s futures in Africa than they do in the global North. Entrepreneurship is promoted in Africa as a possible solution to a combination of high levels of education, relative to previous decades, and equally high unemployment (nearly 50 percent in Southern Africa, PRB 2013). However, the small study reported here supports qualitative evidence that entrepreneurship cannot substitute for salaried employment and may not increase young people’s wellbeing (e.g. business owners reported significantly lower self-esteem). While the findings from non-cognitive skills measures in other settings endorse the value of expenditure on social protection, early childhood and universal primary education, they also emphasize the need to build on early investment, perhaps by supporting vocational education and apprenticeship schemes where skills specific to occupations can be passed to the next generation. Ethnographies such as Mains (2013) suggest the most valuable skill for African youth is patience. However, for young people to successfully “navigate” complex societies and generate their own opportunities they also need skills such as openness and agreeableness to build relationships across generational hierarchies. There are measures of these constructs that need further testing, including in translation, and economic techniques such as social network analysis that could be used to understand how young people progress. Focusing solely on individuals without looking at their contexts and the networks of relationships in which they are embedded may direct our attention away from both structural constraints and the relational resources that enable people to overcome these.

References Alfonso, M., Santiago, A, and Bassi, M. (2010). Estimating the Impact of Placing Top University Graduates in Vulnerable Schools in Chile. IADB technical note IDB-TN-230. Almlund, M., Duckworth, A.L., Heckman, J., and Kautz, T. (2011). Personality psychology and economics. Handbook of the Economics of Education, 4:1–181. Ange, S. (2012). The Intergenerational Transmission of Cognitive and Non-Cognitive Skills during Adolescence and Young Adulthood. SOEP papers on Multidisciplinary Panel Data Research, pp. 473–2012.

362   Methodological Issues Bartlett, L. (2007). Human capital or human connections? The cultural meanings of education in Brazil. Teachers College Record, 109(7):1613–1636. Bauer, M., and Chytilová, J. (2007). Does Education Matter in Patience Formation? Evidence from Ugandan Villages. IES Working Paper 10/2007. Charles University. Baulch, B. (2011) Household Panel Data Sets in Developing and Transition Countries. Downloaded from http://www.chronicpoverty.org/uploads/publication_files/Annotated_ Listing_of_Panel_Datasets_in_Developing_and_Transitional_Countries.pdf 11/9/14. Beck, T., Demirguc-Kunt, A., and Levine, R. (2005). SMEs, growth, and poverty: Cross-country evidence. Journal of Economic Growth, 10(3):199–229. Becker, G.S., and Tomes, N. (1986) Human capital and the rise and fall of families. Journal of Labor Economics, 4:S1–S39. Bernard, T., Dercon, S., Orkin, K., and Taffesse, A.S. (2014). Learning with Others: A Field Experiment on the Formation of Aspirations in Rural Ethiopia. In press. Betancourt, T.S., Meyers-Ohki, S., Stulac, S.N. Barrera, A.E., Mushashi, C., and Beardslee, W.R. (2011). Nothing can defeat combined hands (Abashize hamwe ntakibananira): Protective processes and resilience in Rwandan children and families affected by HIV/AIDS. Social Science & Medicine, 73:693–701. Billing, D. (2003). Generic cognitive abilities in higher education: an international analysis of skills sought by stakeholders. Compare, 33(3):335–350. Blanden, J., Gregg, P., and Macmillan, L. (2006). Explaining Intergenerational Income Persistence: Non-cognitive Skills, Ability and Education. Bristol: Centre for Market and Public Organisation. Blomeyer, D., Coneus, K., Laucht, M., and Pfeiffer, F. (2009). Initial risk matrix, home resources, ability development and children’s achievement. Journal of the European Economic Association, 7(2–3):638–648. Borghans, L., Duckworth, A.L., Heckman, J.J., and ter Weel, B. (2008). The economics and psychology of personality traits. Journal of Human Resources, 43(4):972–1059. Borghans, L., ter Weel, B., and Weinberg, B.A. (2008). Interpersonal styles and labor market outcomes. Journal of Human Resources, XLIII:4. Bouchard, T.J., and Loehlin, J.C. (2001). Genes, evolution and personality. Behaviour Genetics, 31(3):243–273. Bourdieu, P. (1976). Marriage strategies as strategies of social reproduction. Boyce, G, and Harris, G. (2013). Hope the Beloved Country: Hope Levels in the New South Africa. Social Indicators Research, 113(1), pp. 583–597. Bowles, S., and Gintis, H. (1976). Schooling in Capitalist America. New York: Basic Books. Bowles, S., and Gintis, H. (2002), “Schooling in Capitalist America” revisited. Sociology of Education, 75:1–18. Bowles, S., Gintis, H., and Osborne, M. (2001). The determinants of earnings: a behavioral approach. Journal of Economic Literature, 39(4):1137–1176. Carneiro, P., Crawford, C., and Goodman, A. (2007) Impact of Early Cognitive and Non-cognitive Skills on Later Outcomes. London: Department for Children, Schools and Families. Carneiro, P., and Heckman, J.J. (2003). Human Capital Policy. IZA Discussion Paper No. 821. Chigunta, F., Schnurr, J., James-Wilson, D., and Torres, V. (2005). Being “Real” about Youth Entrepreneurship in Eastern and Southern Africa: Implications for Adults, Institutions and Sector Structures, ILO Series on Youth Entrepreneurship, SEED Working Paper No. 72.

The Economics of Non-Cognitive Skills    363 Coneus, K., and Laucht, M. (2011). The effect of early non-cognitive skills on social outcomes in adolescence. Education Economics, 22(2):112–140. Coneus, K., Laucht, M., and Reuß, K. (2012). The Role of Parental Investments for Cognitive and Noncognitive Skill Formation—Evidence for the First 11 Years of Life. ZEW Discussion Paper No. 10–028. Crabb, M.W. (2010) Governing the middle-class family in urban China: educational reform and questions of choice. Economy and Society, 39(3):385–402. Cunha, F., and Heckman, J. J. (2008). Formulating, identifying and estimating the technology of cognitive and noncognitive skill formation. Journal of Human Resources, 43(4):738–782. Cunha, F., Heckman, J. J., Lochner, L.J., and Masterov, D.V. (2006). Interpreting the evidence on life cycle skill formation, in E.A. Hanushek and F. Welch (eds), Handbook of the Economics of Education. Amsterdam: North-Holland, pp. 697–812. Cunha, F., Heckman, J., and Schennach, S. (2010). Estimating the technology of cognitive and non-cognitive skill formation. NBER working paper 15664. Cunha, F., and Heckman, J.J. (2010). Investing in our young people. NBER Working Paper 16201. de Coulon, A., Meschi, E., and Vignoles, A. (2011). Parents’ skills and children’s cognitive and non-cognitive outcomes. Education Economics, 19(5):451–474. De Weerdt, J. (2010) Moving out of Poverty in Tanzania: Evidence from Kagera. Journal of Development Studies, 46(2):331–349. Dercon, S., and Krishnan, P. (2009). Poverty and the Psychosocial Competencies of Children: Evidence from the Young Lives Sample in Four Developing Countries. Children, Youth and Environments, 19(2):138–163. Dercon, S., and Singh, A. (2011). From Nutrition to Aspirations and Self-Efficacy: Gender Bias over Time among Children in Four Countries. Young Lives working paper #71. Duckworth, A.L., and Seligman, M.E.P. (2005). Self-discipline outdoes IQ in predicting academic performance in adolescents. American Psychological Society, 16(12):939–944. Duckworth, K., Akerman, R., MacGregor, A., et al. (2009). Self-regulated learning: a literature review. Centre for Research on the Wider Benefits of Learning Research Report No.33. Duflo, E., Hanna, R., and Ryan, S.P. (2010) Incentives Work: Getting Teachers to Come to School. Mimeo, http://economics.mit.edu/files/5995 (accessed 15/8/13). Duflo, E., Kremer, M., and Robinson, J.A. (2011). Nudging Farmers to Use Fertilizer: Theory and Experimental Evidence from Kenya. American Economic Review, 101(6):2350–2390. Duncan, G., Boisjoly, J., and Harris, K.M. (2001). Sibling, peer, neighbor, and schoolmate correlations as indicators of the importance of context for adolescent development. Demography 38:437–447. Duncan, G., Dowsett, C., Classens, A. et al. (2007). School readiness and later achievement. Developmental Psychology, 43:1428–1446. Edwards, R. (1976). Individual traits and organisation incentives: what makes a good worker? Journal of Human Resources, XI:51–68. Fahmy, E., Pemberton, S., and Sutton, E. (2011). Cognitive Testing of the UK Poverty and Social Exclusion Survey. Working Paper 17. Falch, T., Nyhus, O.H., and Strom, B. (2012). Performance of young adults: The importance of cognitive and non-cognitive skills. Unpublished paper. Farkas, G. (2003). Cognitive Skills and Noncognitive Traits and Behaviors in Stratification Processes. Annual Review of Sociology, 29:541–562.

364   Methodological Issues Froerer, P. (2011). Education, inequality and social mobility in central India. European Journal of Development Research, 23:695–711. Glewwe, P., Huang, Q., and Park, A. (2010). Cognitive Skills, Non-Cognitive Skills, and the Employment and Wages of Young Adults in Rural China. Unpublished paper. Heckman, J. (2008). Schools, skills and synapses. Economic Inquiry, 46(3):289–324. Heckman, J.J. (2005). Invited comments, in L.J. Schweinhart, J. Montie, Z. Xiang, et al. (eds), Lifetime Effects: The High/Scope Perry Preschool Study Through Age 40. Ypsilanti, MI: High/ Scope Press, pp. 229–233. Heckman, J.J. (1999) Education and job training: doing it right. Public Interest, 135:86–107. Heckman, J.J. (2006) Skill formation and the economics of investing in disadvantaged children. Science, 312:1900–1902. Heckman, J.J., Moon, S.H., Pinto, R. et al. (2009). The Rate of Return to the High/Scope Perry Preschool Program. NBER Working Papers 15471. Heckman, J.J., and Rubenstein, Y. (2001). The importance of noncognitive skills: lessons from the GED testing program. American Economic Review, 91(2):145–149. Heckman, J.J., Stixrud, J., and Urzua, S. (2006). The effects of cognitive and noncognitive abilities on labor market. Journal of Labor Economics, 24(3):411–482. Heineck, G., and Anger, S. (2010). The returns to cognitive abilities and personality traits in Germany. Labour Economics, 17:535–546. Helmers, C., and Patnam, M. (2011). The formation and evolution of childhood skill acquisition: Evidence from India. Journal of Development Economics, 95:252–266. Hoechner, H. (2013). Mobility as a contradictory resource: Peripatetic Qur’anic students in Kano, Nigeria. Children’s Geographies. doi: 10.1080/14733285.2013.827876. Holzer, H.J., Stoll, M.A., and Wissoker, D. (2001). Job Performance and Retention among Welfare Recipients. Washington, DC: The Urban Institute. Assessing the New Federalism Discussion Paper 01–06. Ibarran, P., Ripani, L., Taboada, B. et al. (2012). Life Skills, Employability and Training for Disadvantaged Youth: Evidence from a Randomized Evaluation Design. Unpublished report, IADB. Izzi, V. (2013). Why youth anyway? Examining the rationale for a youth-specific focus in donor-funded employment programmes. Presentation at Entrepreneurship and Employment in the Global South Conference, June 24–25, 2013. Judge, T.A., and Hurst, C. (2007). Capitalizing on One’s Advantages:  Role of Core Self-Evaluations. Journal of Applied Psychology 92(5):1212–1227. Katz, C. (2004). Growing up Global: Economic Restructuring and Children’s Everyday Lives. University of Minnesota Press, Minnesota. Kidd, C., Palmeri, H., and Aslin, R.N. (2013). Rational snacking: Young children’s decision-making on the marshmallow task is moderated by beliefs about environmental reliability. Cognition, 126:109–114. Kingombe, C, Bateman, M, and Willem te Velde, D. (2010) Review of the most recent literature on Entrepreneurship and SMEs. Input to DFID’s Wealth Creation Agenda: Making British International Development Policy more focused on Boosting Economic Growth and Wealth Creation. Knudsen, E.I., Heckman, J.J., Cameron, J.L., and Shonkoff, J. P. (2006). Economic, Neurobiological and Behavioral Perspectives on Building America’s Future Workforce. NBER working paper # 12298.

The Economics of Non-Cognitive Skills    365 Krishnan, P., and Krutikova, S. (2012). Non-cognitive skill formation in poor neighbourhoods of urban India. CWPE Working Paper series, 1010. Laher, S. (2008). Structural equivalence and the Neo-Pi-R: Implications for the applicability of the five-factor model of personality in an African context. Theoretical research, 34(1):76-80. Lamont, M., and Lareau, A. (1988). Cultural Capital: Allusions, Gaps and Glissandos in Recent Theoretical Developments. Sociological Theory, 6:153–168. Lareau, A., and Weininger, E.B. (2003). Cultural capital in educational research: A critical assessment. Theory and Society, 32:567–606. Laroche, C. (2010). Reforming Development Economics With a Nudge: How Can the Policies Proposed by Behavioral Economics be Used to Improve International Development Policy? http://dx.doi.org/10.2139/ssrn.1888758, downloaded 12/01/14. Langevang, T. (2008): ‘We are managing!’: uncertain paths to respectable adulthoods in Accra, Ghana, Geoforum, 39(6):2039-2047. Lilleor, H.B., and Lonberg, J.H. (2010). Does Personality Influence Project Participation? Evidence from Rural Malawi. The Rockwool Foundation Research Unit. Study paper 31. Lleras, C. (2008). Do skills and behaviors in high school matter? The contribution of non-cognitive factors in explaining differences in educational attainment and earnings. Social Science Research, 37:888–902. Loewenstein, G.F., Weber, E.U., Hsee, C.K., and Welch, N. (2001). Risk as Feelings. Psychological Bulletin, 127(2):267–286. Mains, D. Hope is Cut:  Youth, Unemployment, and the Future in Urban Ethiopia (Global Youth). US, Temple. Maluka, C.S., and Grieve, K. (2009). Determining the suitability of the Satisfaction With Life and the Rosenberg Self-esteem scales in a cross-cultural setting. New Voices in Psychology, 1(4):41-50. Martin, R., Olejnik, S., and Gaddis, L. (1994). Is temperament an important contributor to schooling outcomes in elementary school? Modeling effects of temperament and scholastic ability on academic achievement, in W.B. Carey and S.C. McDevitt (eds), Prevention and Early intervention: Individual Differences as Risk Factors for the Mental Health of Children—A Festschrift for S. Chess and A. Thomas. New York: Brunner/Mazel, pp. 59–68. Mischel, W., Ebbesen, E.B., and Raskoff Zeiss, A. (1972). Cognitive and attentional mechanisms in delay of gratification. Journal of Personality and Social Psychology, 21(2):204–218. Naudé, W. (2008). “Entrepreneurship in Economic Development”, UNU WIDER Research Paper No. 2008/20 Nyhus, E.K., and Pons, E. (2005). The effects of personality on earnings. Journal of Economic Psychology, 26(3):363–384. Pearlin LI, and Schooler C. (1978). The structure of coping. Journal of Health and Social Behaviour, 18:2–21. Peterson, J.L., and Zill, N. (1986). Marital disruption, parent-child relationships, and behavior problems in children. Journal of Marriage and the Family, 48(2):295–307. Population Reference Bureau (PRB) (2013). The World’s Youth 2013 Data sheet. Downloaded 11/09/14, from http://www.prb.org/pdf13/youth-data-sheet-2013.pdf. Obschonka, M., Scmitt-Rodermund, E., Silbereisen, R.K., Gosling, S.D., and Potter, J. (2013). The regional distribution and correlates of an entrepreneurship-prone personality profile in the United States, Germany, and the United Kingdom: A socioecological perspective. GSOEP working paper 550–2013.

366   Methodological Issues Roberts, B.W., Chernyshenko, O.S., Stark, S., and Goldberg, L.R. (2005). The structure of conscientiousness: An empirical investigation based on seven major personality questionnaires. Personnel Psychology, 58:10–139. Robinson, P.B., Stimpson, D.V., Huefner, J.C., and Hunt, H.K. (1991). An attitude approach to the prediction of entrepreneurship. Entrepreneurship Theory & Practice, Summer, 15(4):13–30. Rosenberg, M. (1979): Conceiving the Self. New York: Basic books. Sanchez, A., and Dercon, S. (2011). Long-term implications of undernutrition on non-cognitive skills: evidence from four developing countries. Young Lives working paper #72. Schoon, I. (2009). Measuring social competencies. RatSWD Working Paper Series #58. Schmitt, D.P., and Allik, T. (2005). Simultaneous Administration of the Rosenberg Self-Esteem Scale in 53 Nations: Exploring the Universal and Culture-Specific Features of Global Self-Esteem. Journal of Personality and Social Psychology, 89(4):623–642. Shiner, R.L., Masten, A.S., and Roberts, J.M. (2003). Childhood personality foreshadows adult personality and life outcomes two decades later. Journal of Personality, 71(6):1145–1170. Snyder, C.R., Irving, L., and Anderson, J.R. (1991). Hope and health: measuring the will and the ways, in C.R. Snyder and D.R. Forsyth (eds), Handbook of Social and Clinical Psychology: The Health Perspective. Elmsford, New York: Pergamon Press, pp. 285–305. White, S. C. (2010). Analysing wellbeing: a framework for development practice. Development in Practice, 20(2):158–172. Wiegratz, J. (2010). Fake capitalism? The dynamics of neoliberal moral restructuring and pseudo-development: the case of Uganda. Review of African Political Economy, 37(124):123–137. Willis, G. (1999). Cognitive Interviewing A “How To” Guide. Short course presented at the 1999 Meeting of the American Statistical Association.

Table 18.2  Examples of economic analyses of measures of non-cognitive skills in longitudinal datasets Study

Dataset

Age-group

Indicators

Purpose/ findings

Mannheim Study of 3 months and Children at Risk then at 2, 4.5, 8, (epidemiological cohort 11, 15 and 19 study of 384 children, Rhine–Neckar, Germany) Coneus et al., Mannheim Study of 3 months and 2012 Children at Risk then at 2, 4.5, 8, 11, 15 and 19 Ange, 2012 German Socio-economic Adolescence (age Panel Study 17) and young adulthood (18–29) Falch et al., 2012 Norwegian register data Age 16 and 22

Parent and expert assessment of child’s attention span, reaction to new things, prevailing mood and distractibility

Duncan et al., 2007

Six large-scale longitudinal Ages 5–6 and studies from US, UK 10–15, and Canada depending on the study Gansu, China Children and young people

Mother and teacher reports of attention and socio-emotional behaviors using Rutter A and B and Child Behavior Checklist Internalizing and externalizing behavior, self-esteem, depression, resilience

Young Lives, India, Ethiopia, Vietnam, and Peru

Educational aspirations, self-efficacy, trust, self-esteem, inclusion

Non-cognitive skills, especially attention span, predicted educational performance, tobacco and alcohol use, delinquency and autonomy in adolescence. Boys with low non-cognitive skills had worse social outcomes than girls. Non-cognitive skills weakly correlated with IQ. Parental investment in cognitive and non-cognitive skills most effective after birth, less effective by age 8, and ineffective by 11 (c.f. Cunha and Heckman, 2008) Non-cognitive skills were transmitted less strongly than cognitive skills, although transmission was stronger than in equivalent studies in the US. Transmission of both sets of skills was stronger at older ages. High levels of cognitive and non‐cognitive skills are equally important for high school graduation, while low levels of non‐cognitive skills are more predictive of receiving welfare benefits or being inactive in the labor force at age 22. Attention skills are the best predictors for educational attainment; socio-emotional behaviors failed to predict attainment, even among children with behavioral problems Measures in middle childhood and adolescence predict labor force participation and wages via decision-making around school leaving/ work (see also Heckman, Stixrud and Urzua 2006). Cognitive measures moderately correlated with non-cognitive. Gender differences in children’s educational aspirations and parents aspirations for their children, self-efficacy and trust; no gender differences in self-esteem and inclusion

Coneus and Laucht, 2011

Glewwe et al., 2010

Dercon and Singh, 2011

Ages 7–8, 12–13 and 14–15

Persistence, emotion, adaptability and temperament Big Five, locus of control

Proxied by grades in behavioral and practical subjects

(continued)

Table 18.2 Continued Study

Dataset

Helmers and Young Lives, Patnam, 2011 India

de Coulon et al., National Child 2011 Development study (NCDS), UK

Carneiro et al., 2007

NCDS, UK

Martin, Olejnik, USA and Gaddis, 1994 Duckworth and USA, North East Seligman, 2005

Age-group

Indicators

Purpose/ findings

Ages 4–5, 7–8, and 12–13

Non cognitive skills proxied by Evidence of cross-productivity from cognitive to pre-school attendance and fluency non-cognitive skills for ages 8 to 12. Parental in native language (age 5), child investment has a positive effect on skill levels work and Strength and Difficulties at all ages. Questionnaire (age 8), friendliness, group membership, self-pride, determination, social trust (age 12) Aged 3–6 and Strength and Difficulties Explored contribution of parental literacy and numeracy adults Questionnaire to cognitive and non-cognitive skills of children aged 3–6. Found strong correlation between adults and children’s skills, controlling for adults’ own skills when they were children. Parents’ early non-cognitive skills do not influence their children’s cognitive outcomes, but their current skills do. Some evidence of gender-specific transfer of skills as mothers have greater influence on daughters and fathers on sons. Strong correlation between parents’ early emotional and behavioral problems and children’s. Aged 7 and 11 Teacher-report Bristol Social Social adjustment, a dimension of non-cognitive skills, Adjustment Guide is important for schooling, teenage motherhood, involvement in crime, and labor market outcomes. Early home environment determines social skills which appear to be more malleable than cognitive skills between the ages of 7 and 11. Studied from 1st to Task orientation, flexibility, Activity, distractibility, and persistence are five times 5th grade reactivity better at predicting school success than IQ. 8th grade students Self-discipline, measured by self, (measures parent and teacher-report and administered monetary choice questionnaires in Autumn and Spring)

Self-discipline better predictor of final grades, high school selection, school attendance and time spent studying than IQ. Effect on final grades held even when controlling for first year grades and IQ.

Study

Dataset

Age-group

Indicators

Purpose/ findings

Lleras, 2008

National Educational Longitudinal Study, USA

Age 15–16 and 25–26

Teacher report on conscientiousness, motivation, sociability; participation in extra-curricular activities.

Students whom teachers perceived to get on well with others, have better work habits, and participate in extracurricular activities had higher educational attainment and earnings, even after controlling for cognitive skills. Non-cognitive skills in high school explain a substantial portion of the socio-economic, gender, and ethnic gaps in educational attainment and earnings in adulthood. Non-cognitive skills influence wages via the pathway of schooling decisions and affect behavioral outcomes such as risky behavior. Non-cognitive skills more useful for females than males in four-year college market. Steeper gradient for non-cognitive skills than cognitive in relation to schooling/ reproduction decisions and illegal activities, indicating their greater influence. Positive self-evaluations of self-esteem, locus of control, and related traits predict income in mid-life and their predictive power equals or exceeds the predictive power of cognitive traits for schooling, occupational choice, wages, health behaviors, teenage pregnancy, and crime. 12 percent of the variation in educational attainment is explained by non-cognitive skills vs. 16 percent explained by cognitive. Non-cognitive skills are cross-productive for cognitive skills in the first time period of childhood, but effect less obvious in the second period, possibly because cognitive skills are fixed. Similarly, parental non-cognitive skills are important in the first period but not in the second. Motivation, conscientiousness, surgency (reactivity) and agreeableness predict adult autonomy two decades later.

Heckman, Stixrud National Longitudinal and Urzua Study of Youth, 1979 2006 cohort (NLSY79), USA

Children and Self-esteem, locus of control young people (aged 12 and above), followed up at 18 and 30

Cunha, NLSY79, USA Heckman, and Schennach, 2010

Children and young people (aged 12 and above)

Shiner, Masten, and Roberts, 2003

8–12 year olds Multidimensional Personality and 28–32 year Questionnaire, Young adult olds competence

Longitudinal study of competence, adversity, and resilience, USA

Cognitive ability, temperament, social development, behavior problems, and self-competence

Chapter 19

Modeling A fri c a n Ec onomi e s  A DSGE Approach

Andrew Berg, Shu-Chun S. Yang, and Luis-Felipe Zanna

19.1 Introduction Dynamic stochastic general equilibrium (DSGE) models provide a coherent framework for macroeconomic analysis and policy discussions. The seminal work of Smets and Wouters (2003, 2007) and Christiano et al. (2005) showed that a canonical New Keynesian DSGE model can fit well the data of advanced economies, providing credibility to these models as a policy tool. Since then, many central banks in advanced and emerging economies have developed and adopted DSGE models for monetary policy analysis (e.g., Erceg et al. 2006; Christoffel et al. 2008; García-Cicco 2011), and multilateral institutions, like the IMF, and academics have been using them to study fiscal issues (e.g., Kumhof et al. 2010; Coenen et al. 2012). Despite this, there has been an underinvestment in DSGE models to undertake macroeconomic analysis in low-income countries (LICs), particularly in sub-Saharan Africa (SSA). A typical concern is that these models are not capable of capturing the specificities of African economies. In this chapter, we argue that a stylized DSGE model can be tailored to address several macroeconomic policy issues frequently encountered in these economies. Admittedly, DSGE models contain many rough approximations and shortcuts, and they will not on their own reflect all the complex macroeconomic mechanisms. They can at best complement the large body of existing empirical and narrative work. However, adapting DSGE models to African economies is an important step in the research agenda of development economics. There are broader objections than lack of specificity to African economies. Many of the common assumptions—rational expectations, perfect competition, linearization, representative agents—are uncomfortable in many contexts, and to some observers particularly so in LICs. This is not the place to discuss the larger debate, but it is clearly a research priority

Modeling African Economies    371 to better understand which assumptions are most in need of modification. On the other hand, there are also substantial LIC-specific difficulties with the application of alternative methodologies, such as a paucity of data to conduct more theory-free empirical analyses. These objections remind us to be modest and careful in our application of these tools. We do not believe that we are uncovering “deep parameters.” Rather, we are trying to efficiently focus on the key mechanisms and issues for a particular question. After years of mostly empirical work and close engagement with policy-making in LICs, we have felt the need for coherent frameworks in which supply curves slope up, demand curves slope down, behavior is forward-looking, agents cannot be systematically fooled forever, and budget constraints add up across time and sectors. The pervasiveness of such models in standard macroeconomics may make it easy to forget what it is like when they are scarce. We would emphasize three specific advantages of the methodology in our context. First, they help organize thinking, educating intuition about how various economic features and policies come together and about what features and parameters matter for what outcomes. Second, they help systematically incorporate various sorts of empirical evidence. We may have a view on say the likely volatility of the oil price and of the rate of return to public investment, but it is still hard to get a feel for how these interact to determine the riskiness of a particular strategy for public investment/wealth accumulation during a natural resource boom. And third, they provide a vehicle for transparently producing alternative macroeconomic and policy scenarios. It is hard to think quantitatively about the benefits of improving the efficiency of public investments on growth when the role of public capital in growth was not explicit in the baseline forecast. The “art” lies in picking the right model for the right purpose. For example, perfect competition may be useful when analyzing 30-year debt trajectories, but not so when looking at fiscal multipliers at short horizons. All this implies that these models are not generally “true.” Empirical validation is critical but also requires careful tailoring the question at hand. We present a DSGE model that incorporates several relevant features of SSA. These features are supported by empirical evidence and are important for macroeconomic transmission mechanisms. • A large share of model households is financially constrained—they consume all disposable income each period. This feature, which amplifies the effect of demand shocks and breaks Ricardian equivalence, captures the fact that most households in SSA do not have an account in a financial institution and that many do not even interact significantly with informal financial markets. • We assume a closed private capital account, which makes monetary and sterilization policies effective. While private capital flows to SSA increased significantly from US$16.1 billion in 2001 to US$55.2 billion in 2011 (Hou et al. 2013), measures of de jure restrictions on cross–border financial transactions by Schindler (2009) suggest that the capital account for the median SSA country is fairly closed.1 • Following Pritchett (2000), public investment is inefficient: a dollar of public investment spending does not always yield a full dollar of public capital. This helps to achieve both a high productivity of infrastructure and a low return on public investment in the 1 

In some of the work we cite, we make the closure of the capital account a matter of degree.

372   Methodological Issues model. Furthermore, because of low governance quality, we introduce a wedge between the tax burden and revenues accrued to a government. These inefficiencies and governance problems seem to be pervasive in SSA, as reflected by international comparisons using the public investment management index of Dabla-Norris et al. (2011) and the Worldwide Governance Indicators of Kaufmann et al. (2013).2 This also helps account for low levels of private investment despite capital scarcity. • We introduce a central bank balance sheet to capture the interaction of monetary, fiscal, and foreign exchange reserves policies. Combined with the closed private capital account, this captures the notion that sterilized foreign exchange intervention can have real effects on the risk premium and hence the exchange rate. This emerges as a natural way to capture some of the macroeconomic implications of different policies with respect to the management of aid flows and natural resource windfalls. The evidence for the effectiveness of sterilized intervention is controversial; however, central banks in developing countries believe that sterilized foreign exchange interventions are effective (Bank for International Settlements 2005; Neely 2011). This is more likely to be the case in LICs such as in Africa, where domestic and foreign assets are particularly likely to be imperfect substitutes and markets are relatively thin. To illustrate the DSGE approach for policy analysis, we focus on four macroeconomic experiments. A common theme in the experiments is the effects that monetary and fiscal policies have on the private sector (e.g., crowding out), which ends shaping the final impact of aid, public investment, and natural resource shocks on the whole economy. First, we study the effects of reserve accumulation policies in response to aid surges following Berg et al. (2010) and Berg et al. (forthcoming). The fiscal authority and the central bank can follow policies that are not necessarily coordinated:  as the government spends the domestic currency counterpart of aid provided by donors, the central bank can decide whether to accumulate some of the foreign currency proceeds of aid in international reserves. We show that when aid is not accumulated in reserves, the model predicts a sizeable nominal and real appreciation, sectoral reallocations between the traded and the non-traded sector, and a crowding in of private investment due to the positive effects of higher infrastructure on the return to private capital. Instead, accumulating part of aid in reserves can have negative medium-term effects on growth, because of the crowding out of private consumption and investment—not selling the aid inflow in the FX market constrains private demand—reducing the expansionary effects of aid. This combination of policies (spending the local currency counterpart but accumulating the aid dollars in reserves) along with nominal rigidities can explain a perhaps puzzling combination of higher real interest rates and depreciated real exchange rates observed in response to several major aid inflow episodes (Berg et al. 2007). Most previous work on the effect of aid is based on the Keynes–Ohlin transfer problem—fully spending aid must also result in higher domestic absorption and an increase in the current account deficit net of aid—which implies that aid cannot be both spent and

2  For example, the government effectiveness and corruption control indices in the Governance Indicators for SSA countries excluding South Africa are −0.82 and −0.67 for 2012, compared to the averages of −0.24 and −0.29 for developing countries on a scale of −2.5 to 2.5.

Modeling African Economies    373 saved as reserves at the same time. Adam and Bevan (2006), Agénor et al. (2008), Agénor and Yilmaz (2013), Arellano et al. (2009), Cerra et al. (2008), and Chatterjee and Turnovsky (2007), among others, explore the macroeconomic effects of aid in the context of real growth models, abstracting from fiscal and reserve policy interactions. Adam et al. (2009) and Buffie et al. (2004, 2008, 2010) are important exceptions as they extensively analyze the role of monetary policy and exchange rate regimes. They do so in models that ignore capital accumulation, but feature currency substitution and fiscal dominance, which conflate the direct impact of aid with passive changes in monetary policy and currency-demand induced capital flows. Unlike our results, they find benefits to reserve accumulation following an aid surge.3 Second, we use the stylized DSGE model to study how the type of spending financing affects the magnitude of the fiscal multiplier. We compute the fiscal multipliers distinguishing between government consumption and public investment and show that, with restricted capital mobility, external aid financing increases the resources available for the economy and generally produces a larger multiplier than domestic financing. Although, this result is certainly rooted in the private investment crowding out effects of domestic financing, external financing is nevertheless associated with a more pronounced real appreciation and therefore more drastic decline in traded output. In contrast to the abundant work on fiscal multipliers in advanced economies that uses structural macro models (e.g., Coenen et al. 2012), work in LICs using DSGEs is scant. Shen et al. (2013) is an exception. Their work complements empirical work on the subject (e.g., Kraay 2012), by disentangling the macroeconomic transmission mechanisms that may help explain the econometric estimates for fiscal multipliers in LICs. Third, another issue that has recently resurfaced in SSA is how to manage natural resource flows (e.g., International Monetary Fund 2012). Capital scarcity calls for investing natural resource revenues domestically to build productive capital and improve the living conditions of the current generation, as suggested by van der Ploeg and Venables (2011). However, in light of restricted absorptive capacity and volatile revenue flows, our model simulations endorse the benefits of partially saving resource revenues externally to support a stable fiscal regime, as proposed in Berg et al. (2013), Collier et al. (2010), and van der Ploeg (2010). Finally, we assess debt sustainability of public investment scaling-ups. Collier (2007) and Easterly et  al. (2008), among others, have argued for the need of recognizing that scaling up public investment in developing countries may have different short-term versus long-term effects on the public sector balance, debt, and growth. Borrowing to finance public investment can certainly increase fiscal deficits and debt-to-GDP ratios in the short run. However, over the medium to long term, this investment may be able to pay for itself through higher user fees and taxes, as a result of higher growth. We show that even if the long run looks good, transition problems can be formidable when concessional financing does not cover the full cost of investment: tax increases require sharp macroeconomic adjustments, crowding out private demand, and delaying the growth benefits of public investment. External non-concessional borrowing can smooth these difficult tax adjustments but is also risky, as public debt may become unsustainable, especially when the 3  This analysis is closely connected to recent work that characterizes managed floating regimes as a two-instrument/two-target problem (Benes et al. 2013; Ostry et al. 2012).

374   Methodological Issues economy faces difficulties over time in raising sufficient revenue to repay the debt and the return to investment projects turns out to be low. The fiscal reaction function emerges as a critical feature: sustainability is extremely sensitive to the degree and speed with which the primary balance can adjust to adverse debt dynamics.4 These results contribute to the vast literature on the macroeconomic effects of public investment (see Agénor 2012 and references therein), which often assumes government balanced budget rules, thereby abstracting from public debt accumulation. We proceed to lay out the model in stages, calibrate it, and discuss the experiments.

19.2  A Stylized DSGE Model The framework, adapted from Berg et al. (2010), is a small open New Keynesian model with non-traded (N) and traded (T) good sectors and a closed private capital account. It includes various policy and exogenous shocks depending on the application.

19.2.1 Households The economy is populated by two types of households:  a fraction f, representing savers (a) who have access to assets, and a fraction 1−f, representing hand-to-mouth consumers (h) who are liquidity constrained.

19.2.1.1 Savers The representative saver chooses consumption (cta ), real money balances (mta ), labor (lta ), investment (itN ,a and itT ,a ), capital (ktN ,a and ktT ,a ), and domestic government debt (bta ) to maximize the expected utility

( )

( )

( )

1+ ψ 1− ξ   c a 1− σ lta mta t  , (1) − + Et ∑ β   1− σ 1+ ψ 1− ξ  t =0   ∞



t

where β is the discount factor, σ, ψ, and ξ are the inverse of the elasticity of intertemporal substitution for consumption, labor, and money. The saver’s budget constraint is



Rt −1bta−1 πt (2) mta−1 N N ,a T T ,a a * = (1 − τt )(wt lt + rt kt −1 + rt kt −1 ) + + st rm + zt + Ωt . πt

cta + mta + itN ,a + itT ,a + bta + acti ,a −

4 

Mauro et al. (2013) emphasize this perspective in another context.

Modeling African Economies    375 Domestic government debt bta pays a nominal rate of Rt, and πt is the domestic inflation. A tax rate τt is levied on labor income (wt lta ) and capital income rtN ktN−,1a + rtT ktT−,1a . Foreign remittances rm* are assumed to be constant. zt is government transfers to house-holds, Ωt is firms dividends, and st is the real exchange rate in units of domestic consumption per foreign good unit. Capital is sector specific, subject to adjustment costs acti ,a ≡



2 2    iT , a  k  itN ,a  N ,a − δ ktN−,1a +  tT ,a − δ ktT−,1a  , and accumulated through investment as 2  kt −1   kt −1   

ktj ,a = (1 − δ) ktj−,a1 + itj ,a , j ∈{N ,T }, (3)

where δ is the depreciation rate. Consumption and investment are CES aggregates of non-traded and traded goods, for example,



χ

 1 ct = ϕ χ ctN 

( )

χ −1 χ

1 χ

( )

+ (1 − ϕ ) ctT

χ −1 χ

 χ −1  . (4) 

with χ and φ denoting the intratemporal elasticity of substitution and the degree of home bias. Given this, the consumer price index (CPI) corresponds to

( )

Pt = ϕ PtN 

1− χ

( )

+ (1 − ϕ ) P

1− χ

T t

1

 1− χ , (5) 

where PtN and PtT are the prices for non-traded and traded goods. The relative prices of PN S P* non-traded and traded goods to the CPI are ptN ≡ t and st ≡ t t , where St is the nomiPt Pt * nal exchange rate, and Pt is the price of foreign goods. Households supply labor lta to both sectors where



 lta =  ϕl  

( )

−

1

χl

1 + χl

(lta , N )

χl

(

+ 1 − ϕl

1

1 + χl

) (l ) −

a ,T t

χl

χl

   

χl 1+ χl

(6)

φl is the steady-state share of labor in the non-traded good sector, and χl > 0 is the elasticity of substitution between the two types of labor. The aggregate real wage index corresponds then to

( )

wt = ϕl wtN 

1 + χl

(

+ 1− ϕ

l

)(w ) T t

where wtN and wtT are the real wage rate in each sector.

1 + χl

1

 1+ χl , (7) 

376   Methodological Issues

19.2.1.2.  Hand-to-mouth households h h They have an inelastic labor supply ( lt = l ∀t ) and consume all the disposable income each period as determined by the budget constraint

cth = (1 − τt ) wt l h + st rm* + zt . (8)



19.2.2 Firms Non-traded good firms are assumed to be monopolistically competitive, while traded good firms are perfectly competitive.

19.2.2.1 Non-traded sector The monopolistic producer i ∈ [0,1] uses the following technology: ytN (i ) = z N ktN−1 (i )



1− α N

ltN (i )

αN

(k ) G t −1

αG

, (9)

where zN is the sectoral total factor productivity (TFP) and ktG−1 is public capital with an output elasticity αG. The monopolistic producer i faces a demand function for the variety i  p N (i)  ytN (i) =  t N   pt 



−θ

ytN , (10)

where ptN (i) is the price it charges, ytN is aggregate non-traded demand, and θ is the elasticity of substitution. It chooses price PtN (i) , labor ltN (i) , and capital ktN (i) to maximize its expected net present-value profits Et ∑ t = 0 βt λ ta Ωt (i) subject to the production function ∞

(9) and the demand function (10), where λ ta is the savers’ marginal utility and

Ω t (i) = (1 + u)(1 − ι)  ptN (i) ytN (i) − actp (i) − wtN ltN (i) − rtN ktN−1 (i) + (ι − u − ιu)  ptN ytN − actp  .



2

ζ  π N (i)  N N Price rigidity is introduced by adjustment costs actp (i) ≡  Nt − 1 p y à la Rotemberg 2  πt −1 (i)  t t ptN πt is non-traded good inflation and total price adjustment costs, ptN−1 1 1 non-traded output, and dividends correspond to actp = ∫ actp (i )di, ytN = ∫ ytN (i ) di, (1982), where πtN ≡ 1

0

0

Ωt = ∫ Ωt (i) di, and , respectively. To capture additional distortions (other than taxes on 0

Modeling African Economies    377 factor income), we introduce an implicit cost (tax) ι, which discourages firms from producing at a higher level. Unlike income taxes, the revenue collected by ι does not enter the government budget but remains in the private sector. For simplicity, we assume that the price markup is zero in the steady state by having a subsidy u, and the implicit cost is rebated back to firms in a lump-sum fashion.5

19.2.2.2  Traded good sector A representative traded good firm i produces with technology

ytT (i) = z T ktT−1 (i)

1 − αT

ltT (i)

αT

(k ) G t −1

αG

(11)

and chooses labor l(i)Tt and capital k(i)Tt to maximize profits (1 − ι)st ytT (i) − ωTt ltT (i) − rtT ktT−1 (i) + ι st ytT , (12)

1

where  ytT = ∫ ytT (i)di. 0

Total output produced in the economy at period t is  yt = ptN ytN + st ytT .

19.2.3 Public sector The model allows for flexible policy specifications. We describe here the common setup across applications. The public sector consists of a government and a central bank. In each period, the government receives taxes and foreign aid (at* ) and contracts domestic and foreign  debt (btd and btg * ). Total expenditures include government consumption ( g tc ) , public investment ( g tI ), transfers (zt), and debt services. Government purchases g t = g tC + g tI are CES baskets of traded and non-traded goods with a degree of home bias φG, an elasticity of substitution χ, and a relative price to the CPI ptG . Public investment generates capital according to:

ktG = (1 − δ G )ktG−1 + g tI , (13)

where inefficiencies 0 < ϵ ≤ 1 may be present.6 The government is subject to the budget constraint

5  The implicit cost is a modeling shortcut to rationalize why, given the high marginal return to capital implied by capital scarcity, investment to output ratios are low in SSA. 6  To capture the costs associated with absorptive capacity constraints when scaling-up public investment, efficiency can be modeled such that it depends on investment levels, as assumed in the application of resource revenue management.

378   Methodological Issues



taxt + btd + btcb + st btg * + st at* = ptG ( g tC + g tI ) + zt +

Rt -1btd-1 btcb-1 R *btg-1* + + st , π* πt πt

(14)

where taxt = τt (wt lt + rtN ktN−1 + rtT ktT−1 ). We assume a constant nominal interest rate R* for borrowing externally.7 The central bank conducts monetary and reserve policies. Many central banks in LICs still target money in practice, so we assume a money growth rule, where nominal reserve money grows at a constant rate μ. On the other hand, reserve rest* accumulation policy is described as

( )



rest* − res * =

1 (rest*−1 − res * ) − ω s (πtS − π S ), (15) π*

St is the change in the nominal exchange rate, and a variable without a time St −1 subscript indicates its steady-state value. Different exchange rate regimes can be captured by this policy rule depending on the parameter ωs, including a flexible regime (ωs = 0) and a fixed regime (ωs >> 0). When the central bank decides to accumulate reserves, we assume full sterilization of this accumulation, which implies that open market operations must adjust by S where πt ≡



btcb −

 btcb−1 m res *  = (µ − 1) t −1 + st  rest* − tS−1  , (16) πt πt πt  

so that money keeps growing at the rate μ.

19.3  Solution and Calibration To solve the model, we log-linearize the equilibrium system and use Sims’s (2001) solution method. Given that most countries in SSA do not have long series of quarterly data, we cannot estimate the model structural parameters and back out historical values of various shocks. Therefore, we calibrate the model to an average African economy, based on 2005-2012 data of SSA countries, including CPI inflation, interest rates, and the ratios to GDP of private consumption, private investment, government consumption, public debt, aid, remittances, and tax revenues. Table 19.1 summarizes the baseline calibration at the quarterly frequency. We discuss the calibration of the key parameter values below. The quarterly discount factor β = 0.98 is consistent with an annual real interest rate of 8 percent. This relatively high discount rate—compared to the typical value assumed for 7  Later in the application of debt sustainability, we assume that there is an interest rate risk premium, which rises when the debt-to-output ratio increases.

Modeling African Economies    379 advanced economies (β  =  0.99)—could reflect that uncertain institutional outlooks can lower life expectancy in most African economies. Based on the estimates for developing countries by Ogaki et al. (1996), the intertemporal elasticity of substitution is set to 0.34, implying σ = 2.94. Without empirical evidence for the Frisch labor elasticity for SSA economies, we calibrate ψ = 1 for savers. Together with hand-to-mouth households’ inelastic labor supply, the average Frisch labor elasticity is set to 0.25.8 The inverse of the intertemporal elasticity of real money balances (ξ = 11.22) is endogenously determined given the calibrated nominal interest rate and the money-output ratio.

Table 19.1  Baseline calibration Parameters

Values

β σ ψ ξ

0.98 2.94 1 11.29

φ φG χ θ u f κ δ φl χl αN αT αG δG ζ ι ∈ R R* π π* μ τ ρτ γ ρG ωs

0.6 0.7 0.44 6 0.2 0.25 1.4 0.025 0.59 0.6 0.45 0.60 0.11 0.012 11.22 0.38 0.4 1.04 1.02 1.019 1.006 1.019 0.17 0.9 0.04 0.9 0

8 

The discount factor Inverse of intertemporal elasticity of substitution for consumption, savers Inverse of Frisch labor elasticity, savers Inverse of intertemporal elasticity of substitution for real money balance, savers Degree of home bias in private consumption and investment Degree of home bias in government consumption and investment Elasticity of substitution between traded and non-traded goods Elasticity of substitution among non-traded goods Subsidy rate to remove mark-up of non-traded goods in the steady state Fraction of savers Investment adjustment cost parameter Depreciation rate of private capital Steady-state labor share in non-traded good sector Elasticity of substitution between labor of two sectors Labor income share in non-traded output Labor income share in traded output Output elasticity with respect to public capital Depreciation rate of public capital Price adjustment cost parameter Implicit production cost parameter Steady-state public investment efficiency Quarterly domestic nominal interest rate Quarterly foreign nominal interest rate Quarterly CPI inflation Quarterly foreign inflation Quarterly nominal money growth rate Income tax Tate AR(1) coefficient in tax rule Tax response parameter to changes in debt-to-output ratio AR(1) coefficient in gC, gI rule Flexible nominal exchange rate regime

Goldberg (2013) estimates that the intertemporal elasticity of working probability in a daily labor market in rural Malawi is 0.15–0.17.

380   Methodological Issues We set f = 0.25 since, based on 2011 data, Demirguc-Kunt and Klapper (2012) report that on average about 24 percent of adults in SSA has an account in a financial institution. The intra-temporal elasticity of substitution between labor of the two sectors is set to χl = 0.6. Horvath (2000) estimates this elasticity to be 1 using the US sectoral data but work by Artuc et al. (2013) suggests that labor is less mobile in developing countries than in developed countries. To calibrate public investment efficiency, we follow Pritchett (2000). With a zero TFP growth rate, he estimates that the public investment efficiency is 0.49 for SSA. Our baseline calibration assumes a slightly smaller value (ϵ  =  0.4). The output elasticity with respect to public capital is selected to be αG  =  0.11, so in the steady state the annual net rate of return to public capital is 24 percent, matching the median rate of World Bank projects in 2008 (International Bank for Reconstruction and Development and the World Bank (2010)). Given the investment-GDP ratio at 4.5 percent, the implicit tax rate is ι= 0.38. We assume that government purchases have a relatively high degree of home bias than private consumption (φG  =  0.7 vs. φ  =  0.6), because generally a large share of government spending goes to pay service of civil servants, classified as non-traded goods.

19.4 Applications To show how the stylized model can be used for policy analysis in Africa, we adapt the model to address four issues related to (i) reserve accumulation policy responses to aid surges, (ii) government spending, financing schemes, and fiscal multipliers, (iii) management of natural resource revenues, and (iv) public investment surges and debt sustainability.

19.4.1  Aid surges and reserve accumulation policies Out of fear of “Dutch disease,” central banks have frequently responded to aid surges by accumulating much of the additional aid in international reserves, even as governments spend the local currency counterpart on domestic goods. This is supposed to contain real appreciation pressures and therefore ameliorate the harm on the traded good sector. But what are the macroeconomic consequences of such responses? To provide an answer, we follow Berg et al. (forthcoming) and analyze the effects of different reserve accumulation policies under full sterilization and a flexible exchange rate regime.9 The policy rule (17) is then modified to

9  See Berg et al. (2007) for case studies and Berg et al. (2010) and Berg et al. (forthcoming) for a full analysis on the interactions of fiscal, monetary, and reserve policies in response to aid flows, including partial spending, no sterilization, and managed exchanged rate regimes.

Modeling African Economies    381

rest* − res * =

1 (rest*−1 − res * ) + (1 − ω)(at* − a * ) − ω s (πtS − π S ), (17) π*

with ω representing the extent to which aid is accumulated in reserves, while setting ωs = 0 to reflect a flexible exchange rate regime. Aid respects the exogenous process:

log

sta sta−1 = p log + εta , (18) a sa sa

sa * g* where sta ≡ yt (the aid share of steady-state GDP). We also set fiscal variables— τt , bt , and zt—to their steady-state values and keep the stock of domestic government debt, btd + btcb , constant. The shares of debt holding between the central bank and households are then determined by open market operations. Given aid flows, total government spending each period g tC + g tI is endogenously determined by (14), and the increase in government consumption and public investment has the same relative shares as in the steady state. Figure 19.1 compares the impulse responses to an aid surge under two different reserve accumulation policies: no accumulation (ω = 1) depicted by the solid lines and partial accumulation (ω = 0.5) by the dotted-dashed lines. Consider the case of no reserve accumulation first. Government spending rises by more than the aid increase. The short-run expansionary effects of spending aid generates more tax revenue, which supports a higher level of spending. Because government spending has a higher content of non-traded goods, the demand pressure drives up inflation of these goods as well as overall CPI inflation. With a flexible exchange rate regime, aid inflows lead to a nominal appreciation and, therefore, a real appreciation. Consequently, Dutch disease concerns manifest through falling traded output. Despite these concerns, output spikes due to higher public demand for non-traded goods and the fact that the hand-to-mouth consumers receive higher wage income, increasing substantially their consumption and hence aggregate consumption. Accumulating some aid in reserves successfully reduces the magnitudes of both nominal and real appreciation, and thus the decline of traded output. As the central bank does not sell aid-related foreign currency to the private sector, trade deficits do not rise as much as under the no accumulation policy. Because government spending still rises by a similar magnitude, accumulating aid in reserves implies that private demand must be crowded out. Under full sterilization, savers’ holdings of government debt increases, while private consumption and investment are negatively affected (even over time), relative to the case of no reserve accumulation. Thus, limiting foreign exchange sales of aid comes at a real cost in terms of reduced private consumption and investment. As in Berg et al. (forthcoming) it is possible to show that reserve accumulation policies may be welfare reducing. This underscores that foreign exchange policies have real effects (which would emerge even in a model without nominal rigidities) and suggests a role for monetary-fiscal cooperation. Berg et al. (2010) show, for instance, that if public investment is very inefficient and there are large learning-by-doing externalities in the traded sector capturing Dutch disease effects, reserve accumulation may be actually welfare enhancing. In this case, however, it is even better to limit the amount of spending so that government savings coincide with the accumulation of reserves.

382   Methodological Issues

7 6 5 4

aid (% of GDP)

0

5

25

10

international reserve (% of GDP)

25

0

5

10

−1

−5 0

5

10

inflation

−4 10

trade deficit (% of GDP)

0

5

10

3 2 1 0 −1

2 1 0 −1

5

10

real interest rate

1

0

5

10

−1

0

consumption

−2 5

0

0

real exchange rate

0

nominal exchange rate

0

0

0

8 7 6 5

20

1

20 15

govt spending (% of GDP)

5

10

investment 0 −10 −20

0

5

10

output

0

5

no reserve accumulation

10

0

1 0 −1 −2 −3

5

10

traded output

0

5

10

partial accumulation

Figure  19.1   Responses to an aid surge under different reserve accumulation policies. y-axis is in percentage deviation from a trend-growth path unless specified otherwise in parentheses. x-axis is in number of  years.

19.4.2  Fiscal multipliers: the role of external and domestic financing Government spending is an important countercyclical tool for all countries. In most SSA economies, pressing capital needs also give government spending another role to promote economic growth. As few efforts have been devoted to study fiscal policy effects in SSA, we use our model to calculate fiscal multipliers, paying particular attention to the role of financing: domestic versus external (aid). Fiscal policy is then assumed to follow the rules:

log

g tj gj (19) = ρG log t −j 1 + εGj t , j ∈{C , I } , j g g



log

τt τ sb = ρτ log t −1 + γ log t −b1 , γ > 0, (20) τ τ s

Modeling African Economies    383 b where st −1 =

btd−1 + btcb−1 + st −1btg−*1 y t −1

When domestic debt (aid) is used for financing,

at* = a * ∀t (btd = bd ∀t ) and btd (at* ) is endogenously determined by (14).10 The present-value, cumulative multipliers at various horizons are reported in Tables 19.2 and 19.3. With domestic financing, government spending increases generate substantial crowding out in private consumption and investment, as shown by the negative multipliers. As government domestic borrowing increases, savers demand a higher interest rate to hold government debt, discouraging investment. In the longer horizon, higher government debt triggers persistently higher income taxes, prolonging negative investment responses. With aid financing, instead, the output multiplier is generally larger: the 10-year government consumption (public investment) multiplier under the baseline calibration is 0.15 (0.66), compared to −1.33 (−0.99) with domestic debt. In addition, the multiplier of private consumption turns positive. With aid and a larger share of hand-to-mouth consumers, increasing government consumption may not require higher taxes and therefore private consumption may also rise. The main reason for the multiplier differences is that aid financing expands the resource envelope, alleviating the crowding-out effects of domestic financing. Moreover, since aid financing does not trigger higher taxes, the longer-horizon multiplier is larger than that under domestic financing. While aid-related output multipliers are bigger, Dutch disease concerns are present. The aid-related multipliers for traded output and trade balance are more negative than with domestic financing, in both types of spending. Despite assuming a high return to public capital, the public investment multipliers for output with aid financing are small. This is due to the low investment efficiency—one dollar of investment expenditure only delivers 40 cents of public capital (ϵ = 0.4). Thus, improvements in efficiency over time may pay off: doubling efficiency implies that the cumulative

Table 19.2  Government consumption multipliers: baseline calibration. yt Domestic debt financing Impact 0.26 2 years –0.07 10 years –1.33

ytN

ytT

ct

it

tbt

0.25 0.12 –0.58

0.01 –0.19 –0.75

–0.04 –0.34 –1.32

–0.70 –0.74 –1.05

–0.00 0.02 0.02

0.75 1.14 1.06

–0.31 –0.98 –0.90

0.52 0.06 0.05

–0.57 –0.01 0.01

–0.51 –0.90 –0.92

Aid financing Impact 2 years 10 years

10 

0.44 0.15 0.15

The government can also borrow externally to finance spending increases. Below we study this g* g* option in the context of public investment scaling-ups and debt sustainability. Here, bt = b in either financing scenario.

384   Methodological Issues Table 19.3  Public investment multipliers: different investment efficiency. Baseline efficiency assumes ϵ = 0.4. The higher marginal efficiency assumes investment above the steady-state level has efficiency 0.8. Domestic debt, baseline efficiency 2 years 10 years 20 years

yt –0.03 –0.85 –0.99

ytN 0.15 –0.30 –0.36

ytT

–0.17 –0.55 –0.63

ct –0.23 –0.90 –1.09

it –0.82 –1.00 –0.96

tbt 0.02 0.02 0.04

1.21 1.35 1.51

–0.99 –0.68 –0.48

0.17 0.45 0.71

–0.07 0.07 0.14

–0.93 –0.93 –0.86

ytT

ct 0.27 0.85 1.35

it –0.13 0.13 0.27

tbt –0.93 –0.91 –0.77

Aid, baseline efficiency 2 years 10 years 20 years

0.21 0.66 1.02

Aid, higher marginal efficiency 2 years 10 years 20 years

yt 0.26 1.17 1.87

ytN

1.24 1.59 1.92

–0.97 –0.42 –0.04

long-term output multiplier increases from 1.02 to 1.87 in the baseline, as shown in the bottom panel of Table 19.3.11

19.4.3  Fiscal management of natural resource revenues Natural resource (NR) revenues provide a valuable source to finance public investment in SSA. Recent NR discoveries have called for a fiscal framework where sustainable growth and stability can be achieved, while investing resource windfalls. In this application, we use the model to provide important insights for such a framework. We analyze, in particular, two public investing approaches:  spend-as-you-go versus sustainable investing. Spend-as-you-go assumes that all the windfall is allocated between government consumption and public investment, leading to a procyclical fiscal policy. Sustainable investing combines a relatively stable path of public investment with an external resource fund. We follow Berg et al. (2013) to add a NR sector to the model, whose value is assumed to be 20 percent of GDP in the steady state. We also introduce absorptive capacity constraints that can affect investment efficiency, a learning-by-doing externality in traded production, and rising public capital depreciation rates when investment is insufficient to cover recurrent costs. Learning-by-doing implies that TFP of the traded good sector evolves as 11  In addition to financing sources and public investment efficiency analyzed here, the framework can potentially be used to study the role of other factors in government spending effects, e.g., sectoral labor mobility (χl), home bias in government purchases (φG), and the share of hand-to-mouth households (f).

Modeling African Economies    385



log

ztT zT yT = ρzT log t T−1 + d log t T−1 , T z z y

(21)

with ρzT = d = 0.1 capturing a small degree of externality. We borrow the assumptions about oil price and quantity shocks from Richmond et al. (2013) for Angola. To accommodate the resource fund ( ft* ) and NR revenues, the government constraint (14) is modified as taxt + btd + btcb + st btg * + st a * + st

R f * ft*−1 Rt −1btd−1 btcb−1 R *btg−*1 G s = p g + z + + + + st f t* , (22) t t t πt πt π* π*

where taxt includes NR revenues, and the nominal return of the fund (Rf*) is consistent with the annual average real return of 2.7 percent for the Norwegian Government Pension Fund from 1997 to 2011 (Gros and Mayer 2012). To capture the low absorptive capacity, we assume that public investment efficiency drops from the baseline ϵ = 0.4 to 0.3 when the additional investment spending rises 60  percent above the initial steady-state level. Figure 19.2 compares the macroeconomic outcomes for the two investing approaches under two resource revenue projections. The left column assumes that resource prices are relatively non-volatile, and the right one assumes large negative resource price shocks as those observed in 2008–2009. Sustainable investing (solid lines) yields greater macroeconomic stability than spend-asyou-go (dotted-dashed lines), mainly because public investment paths are delinked from NR revenue flows. Government consumption is also reduced to make additional funding available to support public investment. The investment path analyzed here only scales up investment gradually. As the calibrated economy has little resource fund in the initial steady state, ramping up investment slowly can shore up a resource fund to provide a fiscal buffer. When large negative price shocks hit (right column), the resource fund can be drawn down to support the stable investment path. Comparing the two investment approaches also highlights the importance of building fiscal buffers to protect the economy from the boom-bust cycles driven by NR revenue shocks. In our simulations, spend-as-you-go outperforms sustainable investing in non-resource GDP for most periods under a non-volatile resource price scenario. However, the volatility of NR revenue can be damaging when investment is pro-cyclical, as shown with spend-asyou-go. Over-spending beyond absorptive capacity during a boom may increase the costs, as investment efficiency drops. Underspending during a bust may result in insufficient investment to maintain existing capital, driving up the depreciation rate and, hence, creating a smaller public capital stock. The model-based analysis can also inform policy allocation decisions between capital spending and external saving under uncertainty. A large number of simulations that account for historical volatility of NR revenues can be conducted to construct confidence intervals for variables of interest (see Richmond et  al. 2013). A  high probability that a resource fund cannot support a public investment plan suggests that it may be too ambitious and thus unlikely to be sustained. On the other hand, a high probability that the economy ends up accumulating a big resource fund indicates a high opportunity cost of external savings in terms of foregone growth from more productive capital.

386   Methodological Issues resource revenue

resource revenue

20 0 −20

20 0 −20 2014

10 5 0

2016 2018 2020 stabilization fund (% of GDP)

2022

2014 2016 2018 2020 2022 public investment (level, % of GDP)

6

2022

2014 2016 2018 2020 2022 public investment (level, % of GDP)

4 2014

2016 2018 2020 public capital

2022

2014 2016 2018 2020 2022 government consumption (level, % of GDP)

16

2014 4 2 0 −2

2016 2018 2020 public capital

2022

2014 2016 2018 2020 2022 government consumption (level, % of GDP)

16

15

15 2014

4 2 0 −2 −4

10 5 0

2016 2018 2020 stabilization fund (% of GDP)

6

4

4 2 0 −2

2014

2016 2018 2020 non-resource GDP

2022

2014 2016 2018 2020 2022 non-volatile secnario of resource prices sustainable investing

2014 4 2 0 −2 −4

2016 2018 2020 non-resource GDP

2022

2014 2016 2018 2020 2022 adverse scenario of resource prices spend-as-you-go

Figure  19.2  Effects of two investing approaches for natural resource revenue. y-axis is in percentage deviation from a trend-growth path unless specified otherwise in parentheses.

19.4.4  Public investment surges and debt sustainability Many African economies face dire infrastructure gaps. For the first time in decades, many also have substantial growth momentum, low debt levels, and access to non-concessional foreign credit. Meanwhile, aid resources are not increasing as promised, making the opportunity to borrow non-concessionally to meet infrastructure needs very tempting. Following Buffie et al. (2012), we use the model to assess the debt sustainability implications of a permanent public investment increase (from 4.5  percent to 6  percent of GDP), which is partly financed by external non-concessional borrowing. Government consumption, transfers, and domestic debt are kept at the initial levels, and external debt adjusts to satisfy the government budget constraint (14). The in-come tax rate is

Modeling African Economies    387 determined by the rule (20) that depends on public debt and intends to maintain debt sustainability. To highlight the fiscal risk of external non-concessional borrowing, we assume that its real R* interest rate rt* = t increases with the public debt-to GDP ratio stb , as in García-Cicco et al. π* (2010): sb − sb

(rt* − 1) = (r * − 1) + η(e t



− 1), (23)

where sb is the steady-state level of this ratio and η > 0 is the debt elasticity. In the baseline calibration, r* = 1.015 or an annual real rate of 6 percent. By assuming a constant risk premium of 200 basis points at the steady state and an annual real risk-free rate of 4 percent, the elasticity η is set to 0.2.12 Also, with higher risk premia, the fiscal adjustment parameter γ in (20) increases from 0.04 to 0.08 to maintain debt sustainability. Figure 19.3 compares three scenarios. The first scenario (solid lines) assumes that taxes follow the fiscal rule (20) and can adjust freely to stabilize the debt to GDP ratio. This is clearly

7

public investment (% of GDP)

external debt (% of GDP)

100

tax rate

0.22

80 6

0.2

60 40

5

0.18

20 4

0

5

10

15

public capital

20

0

0

10

−0.5

5

−1 0

5

10

5

15

no tax ceiling

−1.5

10

15

output

0.5

15

0

0

0.16

0

5

10

15

private demand

1 0 −1 −2 −3

0

5 tax ceiling

10

15

−4

0

5

10

15

tax ceiling and lower efficiency

Figure  19.3  Debt sustainability of public investment scaling-up under external commercial debt. y-axis is in percentage deviation from a trend-growth path unless specified otherwise in parentheses. x-axis is in number of  years. 12 

When the specification has r* set at the risk-free rate, Akitoby and Stratmann (2008) estimate η = 1 with the data of 32 countries.

388   Methodological Issues unrealistic but serves as a benchmark. The second scenario (dotted-dashed lines) assumes gradual tax adjustment: it combines the same fiscal rule with a ceiling of 0.21 on the tax rate.13 The third one (dashed lines) assumes a slightly lower ceiling but the government is much less efficient in building public capital. In the first scenario, with unconstrained taxes, the investment plan implies a drastic fiscal adjustment which crowds out private demand. The second scenario smooths the tax adjustment, so the negative impact of higher taxes on output and private demand is reduced. And although this implies more external commercial borrowing, debt sustainability is still achieved. If, however, the cap on the tax rate is further reduced (because of difficulty in mobilizing revenues) and the return to investment expenditures is lowered (because of low efficiency), the outlook for debt sustainability becomes grim, as the debt to GDP ratio now explodes. Therefore, the model simulations reveal that external commercial borrowing can smooth difficult fiscal adjustments under public investment scaling ups, but involve some tangible risks. With poor execution (e.g., low efficiency) and sluggish fiscal policy reactions (e.g., tight caps on taxes), this borrowing can lead to unsustainable public debt.

19.5  Concluding Remarks In this chapter, we construct a DSGE model that incorporates several important features for most African economies—a large poor population without access to financial markets, restricted international capital mobility, low governance quality, and central bank balance sheet effects. We believe the DSGE approach can serve as a useful and flexible framework to address various macroeconomic policy and academic issues that are important for Africa or LICs in general. To illustrate how such a DSGE model can be used, we present four applications that are mostly related to fiscal and monetary (reserve) policy responses in the context of aid surges, public investment, scaling ups, and natural resource booms. The issues we analyzed with a DSGE approach are only a subset of important macroeconomic questions in African economies. The literature has also employed this approach to address other topics, including the following. First, the model can be enriched with other features to analyze the trade-offs associated with monetary policy frameworks (inflation targeting, exchange rate pegs, and money targeting, among others) as in Baldini et al. (2012), Buffie et al. (2013), and Peiris and Saxegaard (2007), among others.14 This analysis could include specific questions about inflation dynamics and the appropriate inflation targeting measure in the context of food price shocks, following Anand and Prasad (2012), Andrle et al. (2013), Catao and Chang (2013), and Portillo and Zanna (2013), among others. Second, the model can be simulated to study the macroeconomic effects of remittances along the lines of Mandelman (2013). Third, by adding an informal sector, as in Anand et al. (2013); Senbeta (2011), the model could shed light on the implications of labor or product market deregulation on unemployment and growth. Finally, one could study issues related to structural transformation, income

13  In this linearized model we implement the cap with a sequence of shocks that keep taxes below the cap. In Buffie et al. (2012), we solve the model without linearizing. 14  See also the Chapter “Monetary Policy Issues in Sub-Saharan Africa” (Handbook, volume 2).

Modeling African Economies    389 inequality and fiscal policy, as in Peralta-Alva (forthcoming), or structural transformation and monetary policy, as in O’Connell et al. (2013). A further promising avenue might be to introduce heterogeneous agents and exploit household and labor force data, bringing some elements from traditional computable general equilibrium models into the DSGE context.

Acknowledgments We thank Alan Gelb, Jaime de Melo, an anonymous referee, and participants in the Authors’ Meeting for helpful comments, and Manzoor Gill and Pranav Gupta for data assistance. This paper is part of a research project on macroeconomic policy in low-income countries supported by UK’s Department for International Development. This paper should not be reported as representing the views of the IMF or of DFID. The views expressed in this paper are those of the authors and do not necessarily represent those of the IMF or IMF policy, or of DFID.

References Adam, C., and Bevan, D. (2006). Aid and the supply side:  Public investment, export performance, and Dutch disease in low-income countries. Work Bank Economic Review, 20(2):261–290. Adam, C., O’Connell, S., and Buffie, E. (2009). Monetary policy rules for managing aid surges in Africa. Review of Development Economics, 13(3):464–490. Agénor, P.-R. (2012). Public Capital, Growth and Welfare:  Analytical Foundations for Public Policy. Princeton: Princeton University Press. Agénor, P.-R., Bayraktar, N., and El Aynaoui, K. (2008). Roads to poverty? Assessing the links between aid, public investment, growth and poverty reduction. Journal of Development Economics, 36(2):277–295. Agénor, P.-R., and Yilmaz, D. (2013). Aid allocation, growth, and welfare with productive public goods. International Journal of Finance and Economics, 18(2):103–127. Akitoby, B., and Stratmann, T. (2008). Fiscal policy and financial markets. Economic Journal, 118(533):1971–1985. Anand, R., and Prasad, E. S. (2012). Core vs headline inflation targeting in models with incomplete markets. NBER Working Paper No. 16290, National Bureau of Economic Research. Cambridge, MA. Andrle, M., Berg, A., Morales, R.A., et al. (2013). Forecasting and monetary policy analysis in low-income countries: Food and non-food inflation in Kenya. IMF Working Paper 13/61, International Monetary Fund. Arellano, C., Bulir, A., Lane, T., and Lipschitz, L. (2009). The dynamic implications of foreign aid and its variability. Journal of Development Economics, 88(1):87–102. Artuc, E., Lederman, D., and Porto, G. (2013). A mapping of labor mobility costs in developing countries, Policy Research Working Paper 6556, the World Bank. Baldini, A., Benes, J., Berg, A., et al. (2012). Monetary policy in low income countries in the face of the global crisis: The case of Zambia. IMF Working Paper 12/94, International Monetary Fund, Washington, DC.

390   Methodological Issues Bank for International Settlements (2005). Foreign Exchange Market Intervention in Emerging Markets: Motives, Techniques, and Implications. Basel, Switzerland, BIS Papers No. 24. Benes, J., Kumhof, M., Laxton, D., et al. (2013). The benefits of international policy coordination revisited. IMF Working Paper 13/262, International Monetary Fund, Washington, DC. Berg, A., Gottschalk, J., Portillo, R., Zanna, L.-F., (2010). The macroeconomics of medium-term aid scaling-up scenarios. IMF Working Paper 10/160, International Monetary Fund, Washington, DC. Berg, A., Hussain, M., Roache, S.K., et  al. (2007). The Macroeconomics of Scaling Up Aid: Lessons from Recent Experience. International Monetary Fund, Washington, DC, IMF Occasional Papers 253. Berg, A., Portillo, R., Yang, S.-C.S., Zanna, L.-F., 2013. Public investment in resource-abundant developing countries. IMF Economic Review 61(1):92–129. Berg, A., Portillo, R., Zanna, L.-F. (forthcoming). Policy responses to aid surges in countries with limited international capital mobility:  the role of the exchange rate regime. World Development. Buffie, E., Adam, C., O’Connell, S. (2010). Fiscal inertia, donor credibility, and the monetary management of aid surges. Journal of Development Economics, 52(8):287–298. Buffie, E., Adam, C., O’Connell, S., Pattillo, C. (2004). Exchange rate policy and management of official and private capital flows in Africa. IMF Staff Papers, 51(1):126–160. Buffie, E., Adam, C., O’Connell, S., Pattillo, C. (2008). Riding the wave: Monetary responses to aid surges in low-income countries. European Economic Review, 52(8):1378–1395. Buffie, E., Airaudo, M., Zanna, L.-F., 2013. Inflation targeting and exchange rate management in less developed countries. Manuscript, International Monetary Fund. Buffie, E.F., Berg, A., Pattillo, C., et al. (2012). Public investment, growth, and debt sustainability: Putting together the pieces. IMF Working Paper 12/144, the International Monetary Fund, Washington, DC. Catao, L., Chang, R. (2013). World food prices, the terms of trade-real exchange rate nexus, and monetary policy. IMF Working Paper 13/114, International Monetary Fund, Washington, DC. Cerra, V., Tekin, S., Turnovsky, S. (2008). Foreign aid and real exchange rate adjustments in a financially constrained dependent economy. IMF Working Paper 13/114, International Monetary Fund, Washington, DC. Chatterjee, S., and Turnovsky, S. (2007). Foreign aid and economic growth: the role of flexible labor supply. Journal of Development Economics, 84(1):507–533. Christiano, L.J., Eichenbaum, M., and Evans, C. L., 2005. Nominal rigidities and the dynamic effects of a shock to monetary policy. Journal of Political Economy, 113(1):1–45. Christoffel, K., Coenen, G., and Warne, A. (2008). The new area-wide model of the Euro area: A micro-founded open-economy model for forecasting and policy analysis. Working Paper Series No. 944, European Central Bank. Coenen, G., Erceg, C.J., Freedman, C., et al. (2012). Effects of fiscal stimulus in structural models. American Economic Journal: Macroeconomics, 4(1):22–68. Collier, P. (2007). The Bottom Billion: Why the Poorest Countries are Failing and What Can Be Done About It. New York: Oxford University Press. Collier, P., van der Ploeg, R., Spence, M., and Venables, A.J. (2010). Managing resource revenues in developing economies. IMF Staff Papers, 51(7):84118.

Modeling African Economies    391 Dabla-Norris, E., Jim, B., Kyobe, A., et  al. (2011). Investing in public investment:  an index of public investment efficiency. IMF Working Paper 11/37, International Monetary Fund. Demirguc-Kunt, A., Klapper, L. (2012). Measuring financial inclusion:  The global Findex. Policy Research Working Paper No. 6025, the World Bank, Washington, DC. Easterly, W., Irwin, T., Servén, L. (2008). Walking up the down escalator: Public investment and fiscal stability. World Bank Research Observer, 23(1):37–56. Erceg, C.J., Guerrieri, L., Gust, C. (2006). Sigma: A new open economy model for policy analysis. International Journal of Central Banking, 2(1):1–50. García-Cicco, J. (2011). On the quantitative effects of unconventional monetary policies in small open economies. Journal of International Central Banking, 7(1):53–115. García-Cicco, J., Pancrazi, R., and Uribe, M. (2010). Real business cycles in emerging countries? American Economic Review, 100(5):2510–2531. Goldberg, J. (2013). Kwacha gonna do? Experimental evidence about labor supply in rural Malawi. Unpublished Manuscript, Economics Department, University of Maryland. Gros, D., and Mayer, T. (2012). A Sovereign Wealth Fund to Lift Germany’s Curse of Excess Savings. CEPS Policy Brief No. 280, August 28, Centre for European Policy Studies. Horvath, M. (2000). Sectoral shocks and aggregate fluctuations. Journal of Monetary Economics 45(1):69–106. Hou, Z., Keane, J., Kennan, J., et  al. (2013). The changing nature of private capital flows to sub-Saharan Africa. Working Paper 376, Overseas Development Institute. International Bank for Reconstruction and Development and the World Bank (2010). Cost Benefit Analysis in World Bank Projects. Washington, DC:  International Bank for Reconstruction and Development/The World Bank. International Monetary Fund (2012). Macroeconomic Policy Frameworks for Resource-Rich Developing Countries. Washington, DC: International Monetary Fund. Kaufmann, D., Kraay, A., and Mastruzzi, M. (2013. Worldwide governance indicators, the World Bank Group. Kraay, A. (2012). How large is the government spending multiplier? Evidence from world bank lending. Quarterly Journal of Economics 127(2). Kumhof, M., Laxton, D., Muir, D., and Mursula, S. (2010). The global integrated monetary and fiscal model (GIMF) theoretical structure. IMF Working Paper 10/34, International Monetary Fund. Mandelman, F.S. (2013). Monetary and exchange rate policy under remittance fluctuations. Journal of Development Economics 120(May):128–147. Mauro, P., Romeu, R., Binder, A., and Zaman, A. (2013). A modern history of fiscal prudence and profligacy. IMF Working Paper 13/5, International Monetary Fund. Neely, C.J. (2011). A foreign exchange intervention in an era of restraint. Federal Reserve Bank of St. Louis Review, 93(5):303–324. O’Connell, S., Portillo, R., and Zanna, L.-F. (2013). On the implications of structural transformation for inflation and monetary policy. Manuscript, International Monetary Fund. Ogaki, M., Ostry, J., and Reinhart, C. (1996). Saving behavior in low- and middle-income developing countries: A comparison. IMF Staff Papers 43(1):38–71. Ostry, J.D., Ghosh, A.R., Chamon, M., and Qureshi, M.S. (2012). Tools for managing financial stability risks from capital inflows. Journal of International Economics, 88(2):407–421. Peiris, S.J., and Saxegaard, M. (2007). An estimated DSGE model for monetary policy analysis in low-income countries. IMF Working Paper 07/282, International Monetary Fund.

392   Methodological Issues Peralta-Alva, A. (forthcoming). Infrastructure and income inequality in a structural transformation setting. IMF Working Paper, International Monetary Fund. Portillo, R., and Zanna, L.-F. (2013). On the first round effects of food price shocks. Manuscript, International Monetary Fund. Pritchett, L. (2000). The tyranny of concepts:  Cudie (cumulated, depreciated, investment effort) is not capital. Journal of Economic Growth, 5(4):361–384. Richmond, C., Yackovlev, I., and Yang, S.-C.S. (2013). Investing volatile oil revenues in capital-scarce economies: An application to Angola. Pacific Economic Review, 20(1): 193–221. Rotemberg, J.J. (1982). Sticky prices in the United States. Journal of Political Economy, 90(December):1187–1211. Schindler, M. (2009). Measuring financial integration:  a new dataset. IMF Staff Papers 56(1):222–238. Senbeta, S.R. (2011). How applicable are the new Keynesian DSGE models to a typical low-income country. Research Paper 2011-016, Department of Economics, University of Antwerp. Shen, W., Yang, S.-C.S., and Zanna, L.-F. (2013). Government spending effects in low-income countries. Manuscript, International Monetary Fund. Sims, C.A. (2001). Solving linear rational expectations models. Journal of Computational Economics, 20(1-2):1–20. Smets, F., and Wouters, R. (2003). An estimated dynamic stochastic general equilibrium model of the Euro area. Journal of the European Economic Association, 1(5):1123–1175. Smets, F., and Wouters, R. (2007). Shocks and frictions in US business cycles: A Bayesian DSGE approach. American Economic Review, 97(3):586–606. van der Ploeg, F. and Venables, A.J. (2011). Harnessing windfall revenues: Optimal policies for resource-rich developing economies. Economic Journal, 121(551):1–30. van der Ploeg, R. (2010). Aggressive oil extraction and precautionary saving: Coping with volatility. Journal of Public Economics, 94:421–433.

Chapter 20

M easuring Ec onomi c Pro gress i n t h e African C ont e xt Morten Jerven

20.1 Introduction The debates on measurement of economic progress in Africa and beyond have been resurfacing with different intensity through time. The need for comparative statistics and global standards of measuring progress is fundamental to the operation of international organizations, national governments, and to academic and popular discourse on the success and failure of development in Africa and elsewhere.1 The absence of comparable metrics in the analysis of economic progress would preclude meaningful statements of how the region and countries in the region is progressing. The question that appears is of course what is the most relevant metric for economic progress? This question should not be confused with the question of what is the most appropriate measure of development. For some purposes it may certainly make sense to rank countries according to observance of human rights, the gains made in social development and an evaluation of political and economic governance. Here, a narrow view of economic progress is adopted. Measuring economic progress then is a matter of aggregating information on the totality and movement in the value of goods and services produced or consumed. The metric in question is gross domestic product (GDP). The interpretation of GDP must of course be tempered with an evaluation of trends in poverty and inequality, but overall an expansion in the capacity to produce and consume more goods and services is how economic progress is measured. The key question in this chapter is how economic progress has been measured in practice, but to discuss this in a meaningful way, we first need to understand how GDP is aggregated theoretically.

1 

Jerven, M. (2013a). Poor Numbers: How We Are Misled by African Development Statistics and What to Do about It. Ithaca, NY: Cornell University Press.

394   Methodological Issues In theory, there are three distinct ways of aggregating GDP: the income method, the expenditure method, and the production method. Again in theory, these are supposed to be reached independently and their respective results should be balanced. The first approach adds up profits, rents, interest, dividends, salaries, and wages. In practice, this approach has not been suitable for estimating the GDP of African economies. The main component of the method would be profits earned by farmers, and this information is not directly available. The expenditure approach is more feasible, at least at first glance. Its components are private consumption, investment, government consumption, and the balance of exports and imports. The problem here is personal consumption and the part of capital formation related to rural and small-scale economic activities. The production method totals estimates of value added (output minus intermediate consumption) per sector (agriculture, mining, manufacturing, construction, and different services) to equal total value added, or GDP. It is the production method that has been preferred in official national income accounting in post-colonial Africa. While the System of National Accounts suggests that all three methods should be estimated independently, thus providing a check on the accuracy of each estimate, this practice is not often followed. Post-colonial national accounts have typically been estimated using the production method, while expenditure on private consumption has typically not been estimated independently but has been derived as what is called a “residual.” In practice, this means that instead of reaching an independent estimate of this important component, an estimate is reached by subtracting all other components of expenditures from the GDP estimate that was reached using the production approach. And when the production approach is used, there is still a meager data basis to reckon with. The quality of the GDP estimates is a result of the combined quality of the activities at the statistical office. The national accounts division depends on data that are produced in different parts of the statistical office—particularly for data on population, agricultural, and industrial production and information on prices. The supply of data from these sub-divisions is subject to manpower and funds available for data collection and processing. Frequently, the statistical offices rely on data made available from other public and private bodies. For example, the agricultural data will typically come from the equivalent of the ministry of agriculture. In some sectors that are dominated by a few large operators, such as construction, mining, electricity, water, finance, communications, transport, the office will depend on the supply of data from these private or public entities. There is a distinction between “survey data” and “administrative data.” A survey is a specific tool in which the statistical office is collecting responses from individual agents. The ability to conduct surveys is conditional on specific funding as the normal budget allowance normally only covers the basic operation costs of the office. The administrative data are data that are collected by the public bodies in order to facilitate day-to-day governance, and will reflect the ambitions and extent of the activities of the state. In this manner, availability of data varies from country to country and according to the circumstances at a given time, which in turn determines the quality of the final estimates. Thus the basic questions that determine the quality of the GDP numbers are whether the statistical office has any data, how good they are and what the national accountants do when data are missing. The first step in the aggregation process is to make a baseline estimate or a benchmark year. The most exhaustive instrument here is a census. This can be a census of the

Measuring Economic Progress in the African Context    395 population, the agricultural production or the transport sector. If a census is absent, a survey may be available. A survey contains some information about a sample of the total. If there ever was a census, then one can aggregate these results, assuming that the sample is representative. If there is no total to relate the survey to, the statistician will have to make a guesstimate, literally making up the missing information without any official guidelines. Often there is no data. When data on levels of economic activity are missing GDP compilers have to rely on estimation by proxy, or assumed relationships. A classic example is when one has no data on food production and then assumes a per capita caloric intake which is multiplied by an estimate of the farming population. Data are usually missing for parts of the service sector and a common method is to assume a proportional relationship with the production of other physical goods. When a level estimate for a given year has been reached, the wealth of the nation is measured. The next step is to measure economic growth, so that the progress of the nation can be monitored. It is easy to get the impression that this would simply entail aggregating all available data once more and comparing the current year with the previous. However, the way this is done in practice is quite different. The level estimates for the individual sector are already made and form the basic starting point. In some sectors, such as government expenditures and turnover for larger businesses, one is able to compare the total for one year with another, but for large parts of the economy one usually relies on so called “performance indicators” or “proxies.” The annual data collected from public bodies and private businesses are utilized. These are supplemented by data on exports and imports. Typical examples are the use of cement production and/or imports as a proxy for growth in the construction sector, the number of new official licenses for transport sectors, and a reliance on population growth for sectors where little adequate data are available. Even if all the data are acquired, GDP would not be a perfect measure to capture improvements in living standards across time and space. This is not a problem specific to either African or low-income countries. As pointed out in the report by Stiglitz, Sen, and Fitoussi (2010) many aspects of living standards are missed by conventional income measures. Not only are there problems of direct comparisons of production—deriving from changes in quality of goods and changes in consumption baskets across time and space, but there are other issues that are left out completely. These are issues that greatly impact human welfare such as leisure, health, education and political freedom to mention a few, and also other issues such as environmental sustainability and security. It was suggested by Stiglitz, Sen, and Fitoussi that one could include survey results from subjective evaluations on these questions, as well as to a greater degree make imputations for leisure, social activities, or activities that are not marketed. The principle of “invariance” means that GDP compilers need to make imputations to correct for these systemic differences from country to country that may lead us astray when using the measure to compare living standards.2 In the recent report from the Commission on Measurement of Economic Performance and Social Progress there is an enlightening discussion on the scope for adding household work and the value of leisure to the conventional GDP measure: 2  Stiglitz, J., Sen, A., and Fitoussi, J.P. (2010). Mismeasuring our Lives: Why GDP doesn’t add up. New York: New Press.

396   Methodological Issues But imputations come at a price. One is data quality:  imputed values tend to be less reliable than observed values. Another is the effect of imputations on the comprehensibility of national accounts. Not all imputations are perceived as income-equivalent by people, and the result may be a discrepancy between changes in perceived income and changes in measured income. This problem is exacerbated when we widen the scope of economic activity to include other services that are not mediated by the market. Our estimates below for household work amount to around 30 percent of conventionally-measured GDP. Another 80 percent or so are added when leisure is valued as well. It is undesirable to have assumption-driven data so massively influencing overall aggregates.

It is worthwhile to remind ourselves that this is very much the state of play for national accounts estimation in many sub-Saharan African economies. In the year-to-year estimates one is relying heavily on imputations. But only very occasionally can the process of going from scanty observations to a time series of numbers actually be retrospectively observed. One opportunity is provided thanks to Pius Okigbo, the Nigerian economist commissioned by the Nigerian federal government to prepare the national accounts for Nigeria for 1950–1957. Commenting on his own imputations he wrote, “It is impossible to overstate the arbitrariness of the process of “quantification.”3 When preparing the estimates of agricultural production he had only a few unreliable observations from agricultural officers’ subjective reports, which varied in detail: “An occasional officer ventures a guess at the total acreage and yield since the previous year. Others guess at the percentage changes in acreage and yield since the previous year. Most restrict themselves to such remarks as ‘average’, ‘no change’, ‘1952 plus’, ‘1954 minus’, or even ‘very poor.’ ”4 Based on these data, Okigbo prepared a time series for 1950–1957. Okigbo warned that the series should not be used for comparisons of living standards, because it contained so little information of the most important sector of the economy.5 Admirably, Okigbo further noted that it “would be unfair to the officials who have responsibility for statistics if we ended on a critical note” and therefore he emphasized that he had no doubt “that many of the gaps indicated in this report will be filled in the very near future.” Just a few years later though, Helleiner wrote in 1966 that “the Nigerian national accounts remain in a sorry condition,” and that the changes in the estimation procedures made comparisons for the early years “unsuitable.” Nevertheless, he concluded, with a touch of a positive spin, that “the estimates inevitably involve so wide a margin of error that the lack of consistency in the aggregates needs not to be viewed so seriously.” This evaluation was further expanded upon in a report prepared by a team led by Professor O. Aboyade, who revised the Nigerian national accounts in 1981, and the final paragraph of the official report is worth quoting in full, because it is uniquely candid about the shortcomings of statistical methods: Our experience has shown that in a setting where weights and measures are amorphous and in a highly variegated landscape with contrasting political geography, the more mundane nuts and bolts approach of the economic anthropologist may advance the course of development of economic statistics more than the sophisticated discourses of the systems designer and sampling theorist.6 3 

Okigbo, P.N.C. (1962). Nigerian National Accounts, 1950-57. Enugu: Government Printer, p. 65. Okigbo, P.N.C. (1962). Nigerian National Accounts, p. 63. 5  Okigbo, P.N.C. (1962: 174). 6  Federal Republic of Nigeria, Federal Ministry of Planning. (1981). National Accounts of Nigeria, 1973–1975. Lagos: Nigeria, p. 53. 4 

Measuring Economic Progress in the African Context    397 Thus, one may have arrived at a sensible recommendation. A final judgment on development must of course be tempered by an appreciation of the inherent limitations of quantification. As argued by Harriss in making a case for interdisciplinary methods in development studies, the quest for quantitative resolution in development studies must be enriched with qualitative rigor.7 However, states need and demand these and other metrics to function, but as scholars of economic development in Africa we must remind ourselves not to ask too much of the measure. The debates on GDP and the adequacy of the measure have been greatly enriched by the history of measuring African progress.

20. 2  Colonial Statistics In the 1950s and 1960s the most vocal skeptic in development economics regarding the preparation of national income estimates and particular the comparison of resulted aggregated GDP levels was Dudley Seers who argued that:8 In the hands of authorities, such international comparisons may yield correlations which throw light on the circumstances of economic progress, and they tell us something about relative inefficiencies and standards of living, but they are very widely abused. Do they not on the whole mislead more than they instruct, causing a net reduction in human knowledge? (1952–53: 160)

The debates surrounding the measurement of economic progress has then typically focused on the applicability of the GDP measure to local conditions. Here—in the debates on GDP and its suitability or the tension between universal standards and local particulars—one might say that economist’s research on Africa has been at the frontier of these questions. First, early pioneering attempts at compiling GDP took place as early as in the interwar period, before the universal standards, as set out on the first UN Standard of National Accounts were published in 1952.9 This section will discuss some of the key arguments in those debates and revisit some of the key lessons from early estimates of GDP that were attempted in Rhodesia, Nigeria, and Tanganyika10 before independence.11 National income estimates were prepared in African countries following the Second World War. In theory this was done according to the United Nations’ universal Standard of National Accounts; however, in practice the local application varied considerably. In 1945, the only African country to publish national accounts was South Africa. Southern and 7  Harriss, J. (2002). The case for cross-disciplinary approaches in international development. World Development, 30:487–496. 8  Seers, D. (1952–1953). The role of national income estimates in the statistical policy of an under-developed area. Review of Economic Studies, 20(3):1952–1953. 9  Speich, D. (2011). The use of global abstractions: national income accounting in the period of imperial decline. Journal of Global History, 6(1): 7–28. 10  Peacock, A.T., and Dosser, D.G.M. (1958). The National Income of Tanganyika 1952-54. London: Her Majesty’s Stationery Office. 11  As discussed in more detail in Jerven, M. (2013). Poor Numbers: How We Are Misled by African Development Statistics and What to Do about It. Ithaca, NY: Cornell University Press.

398   Methodological Issues Northern Rhodesia followed suit beginning in 1949 and by 1958, Ghana, Kenya, Uganda, and the Congo had all published annual estimates. National income was estimated for Nigeria in 1951, but the next estimates were not prepared until independence in 1960.12 The first estimates made for the colony of Southern Rhodesia and the British protectorates of Northern Rhodesia and Nyasaland were characteristic of colonial accounting in that they did not initially include an estimate of the value added by “African” producers. From 1949 onwards “a nominal figure of £5 million for African subsistence income was included in the value of national income of Northern Rhodesia.”13 The same amount was reported as unchanged in the accounts between 1949 and 1953, thus de facto assuming that the value of total food production from African producers was decreasing quite rapidly (when population growth and inflation are taken into consideration). The first estimates ignored the “subsistence” product altogether while the later estimates acknowledged it, with a marginalized role in the accounts. Meanwhile, there was a vigorous scholarly debate concerning the issue of the “subsistence economy.” In a report on an experiment of preparing income estimates for Nyasaland, Northern Rhodesia, and Jamaica in 1941, Phyllis Deane noted that “when working out national income tables for Central Africa (as compared to Jamaica) it soon became clear that a more comprehensive and direct knowledge of the social and economic structure of Central African peoples was essential if a satisfactory framework was to evolve.” Therefore, it was felt necessary to discard the formal tables and envisage a new system and thereby abandon “the income classification according to profits, interest, rents, wages and salaries, and substitute a classification according to nationality.”14 Alan Prest and Ian Stewart, who prepared income estimates for Nigeria in 1951, also noted problems with the application of “Western” concepts:  “for a start, the distinction between production and living, the distinction between working and not working, is something reasonably tangible in the ‘West’; it is often nebulous in Nigeria.”15 Prest and Stewart ended up accounting for transactions that took place within Nigerian households as market transactions, arguing that the extended household in Africa had to be interpreted differently than the Western household. A striking diversion from conventional methods was the inclusion of intra-household services in the estimates, which even involved evaluating the value of the service of procreation as provided by wives to husbands. Data on bride wealth was used as a proxy for the market values for this intra-household service. Pius Okigbo who prepared estimates for 1950–1957, discarded this approach and favored a less inclusive approach than that taken by Prest and Stewart.16 Eke, who reviewed the two estimation methods, noted that “this excursion by Prest could easily be dismissed as ludicrous, but it is much more serious than that.”17 He 12 

Ady, P. (1962). Use of national accounts in Africa. Review of Income and Wealth, 1962(1):52–65. For a particular focus on French West Africa, see Bonnecasse, V. 2014, Des revenus nationaux pour l’Afrique? La mesure du développement en Afrique occidentale française dans les années 1950, Canadian Journal of Development Studies, forthcoming. 13  Federation of Rhodesia and Nyasaland. (1955). Monthly Digest of Statistics, Salisbury. 14  Deane, P. (1948). The Measurement of Colonial National Income: An Experiment. Cambridge: Cambridge University Press, p. 127. 15  Prest, A.R. and Stewart, I.G. (1953). The National Income of Nigeria, Colonial Office. Colonial Research Studies 11. London: Her Majesty’s Stationery Office. 16  Okigbo, P.N.C. (1962). Nigerian National Accounts, 1950-57. Enugu: Government Printer. 17  Eke, I.I.U. (1966). The Nigerian National Accounts—A Critical Appraisal. Nigerian Journal of Economic and Social Studies, 8(3):334.

Measuring Economic Progress in the African Context    399 argued that it was a fundamental misconception that national accounts could fully capture all the processes that contribute to the welfare of human beings.18 Prest and Stewart, who estimated the income of Nigeria, and Peacock and Dosser, who provided estimates of the income of Tanganyika, all argued that the provision of total aggregates was necessary, and that they would help in informing the government and the international community regarding prospects for economic progress.19 In the same vein, Billington argued that the United Nations System of National Accounts was the best approach to measure the progress of African economies, and that the system of standardization of measurement was the right path forward.20

20. 3  Independence and Development Planning At independence the GDP, very much in vein of the general emphasis on investment, planning, and growth, became a centerpiece of national economic development plans in the 1960s and 1970s. Some related key challenges appeared for the statistical offices that were compiling GDP in this period. One was the need for exhaustive estimates. In essence this meant that instead of having estimates that were representative of the aggregate national income—the total monetized transactions that were recorded by the colonial state, the post-colonial states aspired to have a GDP measure that had a more inclusive measure, and thus with the help of some new household budget surveys, and also some brave assumptions on the value of non-monetary food consumption in the rural sector, a more exhaustive measure of GDP was compiled in the 1960s. In the 1960s and 1970s the national statistical offices in many countries benefitted from an increased availability of administrative data. As part of the day to day operation, governments, ministries and parastatal agencies such as marketing boards, utility and transport companies did collect data that were made available for statistical office to compile guesstimates of the growth of the economies in these decades.21 18 

At this time there was a further debate whether formal economic concepts could capture and described all societies, as in Frankel, S.H. (1953). The Economic Impact on Under-Developed Societies: Essays on International Investment and Social Change. Cambridge, MA: Harvard University Press. For further discussion, see Jerven (2012a). An unlevel playing field: national income estimates and reciprocal comparison in global economic history. Journal of Global History, 7(1):107–128; and Jerven (2011b). Users and producers of African income: measuring African progress. African Affairs, 110(439):169–190. 19  Prest, A.R., and Stewart, I.G. (1953). The National Income of Nigeria, Colonial Office. Colonial Research Studies 11. London: Her Majesty’s Stationery Office; and Peacock, A.T. and Dosser, D.G.M. (1958). The National Income of Tanganyika 1952-54. London: Her Majesty’s Stationery Office. 20  Billington, G.C. (1962). A minimum system of national accounts for use by African countries and some related problems. Review of Income and Wealth, 1962(1). For a review of the earliest estimates, see Deane (1961). Review of three publications: “Domestic Income and Product in Kenya: A Description of Sources and Methods with Revised Calculations from 1954-1958”; “The National Income of the Sudan, 1955-1956,” by C.H. Harvie and J.G. Kleve; “Comptes Économiques Togo, 1956-1957-1958” by G. Le Hégarat. Economic Journal, 71(283):630–631. 21  Jerven, M. (2014). Economic growth and measurement reconsidered in Botswana, Kenya, Tanzania, and Zambia, 1965-1995. Oxford: Oxford University Press.

400   Methodological Issues Independence meant new priorities and new statistical needs. Before independence in former Northern Rhodesia, now Zambia, national accounts were prepared by the Central Statistical Office (CSO) in Salisbury. At the beginning of 1964 this responsibility was transferred to the CSO in Lusaka, where the national accounts for 1964 onwards were prepared; “Economic Planning was an important task for the Government and the need for statistical information had therefore increased considerably.”22 Under the new economic and political conditions there was a need to revise the data for the level of private consumption and other categories of expenditure. Essentially this meant estimating the magnitude of total production as compared to monetary demand. In other words, the national accounts had to be based on the “production approach” rather than the “income approach” adopted during the colonial period. This implied an upward revision compared to earlier years as non-monetary activities such as production for own consumption and smaller-scale transactions were included in the new national income estimates. A part of the population neglected earlier was now seen as economically and politically, and therefore statistically, important.23 Despite the aim of establishing a new basis for the accounts, the available basic statistics were not sufficient. The estimates of agriculture in the first national account reports for Zambia covered commercial farming (non-African) and officially registered sales from African farms, while “African subsistence farming and hunting was estimated mainly in accordance with information given by the Food and Agricultural Organization (FAO) for per capita consumption of different kinds of commodity.”24 This assumption of proportional agricultural growth to rural population growth was made in many African countries.25 In one of the very few empirical studies of African national income statistics, Derek Blades noted that for the growth estimates of subsistence agriculture, “the basic assumption is that output grows at the same rate as the rural population,” thus assuming a 1 to 1 labor productivity ratio in the rural sector.26 A notable change in many African economies, particularly prominent in Zambia and Tanzania in the 1970s, was a centralization of the economy and the growing power of the parastatal companies. In both Zambia and Tanzania this was paralleled by an emphasis on socialism. Accordingly, the change was more pronounced in these two, and other socialist countries. This does not mean that these countries should be considered extreme outliers. In other so-called “capitalist” countries the state was also deeply involved in the economy conducting trade, marketing and transport of agricultural crops (both for food and for exports) and was engaged directly or indirectly in manufacturing and construction through newly

22 

Republic of Zambia, 1967. National Accounts 1964-1967. Lusaka: Central Statistical Office, 37. The example of population census in Nigeria in Chapter 3 (Handbook, this volume) illustrates the numerical effect of the upside of being counted in Nigeria following independence, versus the downside of being counted before independence. 24  Republic of Zambia (1967). National Accounts 1964-1967. Lusaka: Central Statistical Office, 37. 25  Similarly, Killick reports that for Ghana the assumption for the 1960s was a constant real per capita consumption of locally produced goods; see Killick (2010). Development Economics in Action: A Study of Economic Policies in Ghana. 2nd ed. New York: Routledge, p. 93. 26  Blades, D. (1980). What do we know about levels and growth of output in developing countries? A critical analysis with special reference to Africa. Economic growth and resources: Proceedings of the Fifth World Congress, International Economic Association, Tokyo, Vol. 2, Trends and factors, R.C.O. Mathews, ed., New York: St. Martin’s Press: pp. 60–70. 23 

Measuring Economic Progress in the African Context    401 formed development corporations.27 This also eased economic recording. In Tanzania in the 1970s the data used for the national accounts on the trade, finance and industry sectors were largely drawn from parastatal enterprises, while data on crops were largely drawn from state marketing boards. This might be interpreted as a choice of convenience, but in the case of Tanzania, there was a correspondence between legitimate and recorded economic activity. Marketing of commodities outside state or parastatal channels was illegal and could not, as such, be thought of as making a contribution to the national income.

20. 4  The Lost Decades The late 1970s and 1980s were a watershed in the economic, social, and political development in sub-Saharan Africa, and the 1980s and 1990s have been referred to as the “lost decades” in terms of economic development, and permanent crisis with regards to politics.28 For measurement of economic progress was this a momentous period. First of all, there was a change in the relative importance of formal and therefore recorded economy, towards informal and therefore unrecorded markets. Second, the state and its agencies were fundamentally challenged by the power of external economic shocks that saw real wages fall and funds for budgets dwindle. Third, there was a gradual change away from focus on economic growth to a focus on poverty reduction. There was a fundamental change in the outlook of the state following the period of structural adjustment policies introduced in the 1980s and 1990s. The mandate of the state was drastically reduced and former interventionist states were liberalized and curtailed. Access to data for the statistical office changed, and so did political priorities. In turn, the statistical office made some adjustment in their assumptions following revisions in accounting methods. It is important to read direct causality into this argument. Statistical capacity deteriorated in line with the general economic and political crisis in the 1980s. The fault of the IMF and the World Bank is rather that of omission than commission. That is: statistical reform has been slow and incomplete—and some unintended consequences of some structural adjustment reforms include the deterioration of statistical capacity. It is clear that liberalization and cutting back the state does limit the ability of states to collect information—meanwhile the data needs—in terms of international monitoring of development—have not lessened. Thus, both the ability and incentive for the states to monitor development themselves has lessened. When Killick wrote a new afterword to his new edition book on his Ghana, Development in Practice,29 he reissued the general health warnings relating to the macroeconomic data. Meanwhile, he also noted that “this condition has not improved over time. In fact, it may 27 

Bates, R.H. (1981). Markets and States in Tropical Africa: the Political Basis of Agricultural Policies. Berkeley and Los Angeles, CA: University of California Press. 28  Van de Walle, N. (2001). African Economies and the Politics of Permanent Crisis, 1979-1999. New York: Cambridge University Press. 29  First published in 1978: Killick, T. (1978). Development Economics in Action: A Study of Economic Policies in Ghana. London: Heinemann; and then in 2010: Killick, T. (2010). Development Economics in Action: A Study of Economic Policies in Ghana, 2nd ed. New York: Routledge.

402   Methodological Issues well be that in some areas the statistical services today are less dependable than in the early-1960s.”30 This mirrors Ward’s observation that the World Bank was not ready to wait for the slow process of producing the national income estimates and rather moved towards “the adaptation of these more fragile but ‘up to date’ figures prepared by the Bank.”31 These were “agreed upon numbers” and were results of negotiations between the Bank and country representatives,32 according to Ward: In effect, to many statisticians working “at the coal face,” this development of data artifacts, which replaced more robust but time-consuming procedures in some countries, seemed dangerous because it rolled assumptions and hypotheses into official numbers. Genuine measurement took a back seat. In reality, there were no recent official GNP estimates based on actual detailed source data in many countries, especially the poorest.33

Thus, economic crisis is correlated with poor data. Disruption shook administrative cap­ acity, and statistical systems. The response of both governments and international financial systems was to withdraw from serious data collection and rather rely on negotiated numbers that could play the role needed in drawing up the policy papers, but that did not yield good information on development.34

20. 5  Liberalization and The Informal Sector The focus on poverty reduction and the millennium development goals (MDGs) meant a re-shift in focus in development, and also in the data needs for measuring progress. While the estimation of GDP was largely a result of combining all available administrative data and making use of some survey data, the data for MDG and poverty lines are largely dependent on survey data. There has been an expansion in social statistics, and some crowding out in the provision of economic statistics. It was not until the late 1990s that the statistical offices in some countries were able to begin to adjust to new economic and political realities.35 A Zambian report on a national income estimate revision for a new series based in 1994 starts by stating the obvious: “inflation rates of more than 200 percent in the early 1990s had adverse effects on the provision of macroeconomic statistics.”36 Creating meaningful data on 30 

Killick, T. (2010). Development economics in Action, 397. Ward, M. (2004). Quantifying the World: UN Ideas and Statistics. Bloomington: Indiana University Place, p. 100. 32  Which is the manner in which data are produced today, as described in Chapter 4 (Handbook, this volume). 33  Ward, M. (2004). Quantifying the World: UN Ideas and Statistics. Bloomington: Indiana University Place, p. 100. 34  Jerven, M. (2011a). Growth, stagnation or retrogression? On the accuracy of economic observations, Tanzania, 1961-2001. Journal of African Economies, 20(3):377–394; Morten Jerven (2010b). Random Growth in Africa? Lessons from an Evaluation of the Growth Evidence on Botswana, Kenya, Tanzania and Zambia, 1965–1995. Journal of Development Studies, 46(2):274–294. 35  For more recent GDP revisions in Nigeria and Ghana, see Jerven, M. (2013b). For Richer, for Poorer: GDP Revisions and Africa’s Statistical Tragedy. African Affairs, 112(446):138–147. 36  Republic of Zambia. (1994). National Accounts Statistics GDP Revision of Benchmark 1994 Estimates. Central Statistical Office: Lusaka. 31 

Measuring Economic Progress in the African Context    403 year-to-year real economic growth under such circumstances is complicated. Furthermore, structural adjustment entailed massive change in the structure of production and “the break-up of the former large parastatals meant that previous sources of data were not available.”37 A revision and a rebasing were overdue as the accounts were still based in 1977 prices and the benchmarks were “becoming inadequate, and over time provided less accurate estimates.”38 The previous estimates had largely “excluded [the] informal sector and therefore impaired the value of GDP estimates over time, in all sectors except agriculture.”39 After incorporating the informal sector activity in the total GDP, the formal sector share was estimated at 58 percent in terms of value added with a corresponding 42 percent share for the informal economy. To this estimate, the statistical office gave the following warning: “we wish to caution that including the informal sector activity in the Zambia National Accounts may tend to exaggerate the GDP of the nation, relative to other countries or even to the previous estimates which mostly excluded it. It must also be recognized that it will be difficult to up-date the sector relation based on indicators in the absence of surveys to monitor the activity in the future.”40 In Tanzania the report accompanying the new constant price series at 1992 prices held that “strong efforts were made to determine what is the story behind the figures, whether the data applies to what is experienced as happening in the industry. This has not been emphasized earlier”; thus indicating that rather than letting the data speak for themselves, the resulting figures were compared to what was otherwise known or assumed regarding economic trends.41 Structural changes, especially in the later part of the 1980s, were not reflected in the available statistics, resulting in an underestimation of value added. “Estimates of the size of this deficiency ranged from 30 percent to as much [as] 200 percent of GDP.”42 The new level estimates were reached by incorporating all available data into the accounts, including the results of new surveys of the transport, trade and construction undertaken as part of the project. In the previous estimation methods of 1976 the “private sector was under covered—sometimes not covered at all—and the growing informal sector was not generally accounted for.”43 A  time series was developed by extrapolating these data on trends backwards. The assumptions were changed: the informal economy was expected to increase when the formal sector was in decline, rather than to move with it. This question is analogous to the issues related to “subsistence” or “traditional” output raised in the 1950s. The “discovery” of the “informal sector” is usually credited to the ILO in 1972 and Keith Hart in 1973,44 but there is still an unresolved scholarly question regarding 37  Republic of Zambia. (1994). National Accounts Statistics GDP Revision of Benchmark 1994 Estimates. 38  Republic of Zambia. (1994). 39  Republic of Zambia. (1994). 40  Republic of Zambia. (1994). 41  United Republic of Tanzania, Bureau of Statistics. (1997). Report on the Revised National Accounts of Tanzania 1987-96. Dar es Salaam: Bureau of Statistics, p. 1. 42  United Republic of Tanzania, Bureau of Statistics. (1997). 43  United Republic of Tanzania, Bureau of Statistics. (1997). 44  International Labour Organization. (1972). Employment, Incomes and Equality: A Strategy for Increasing Production Employment in Kenya. Geneva: International Labour Organization, pp. 97–108; and Hart, K. (1973). Informal Income Opportunities and Urban Employment in Ghana. The Journal of Modern African Studies, 11(1):61–89.

404   Methodological Issues the productive potential of this sector.45 Is the “informal sector” an independent source of economic growth; is it dependent on the “formal” economy for demand and supply; or, is it a parasitic sector, profiting from the demise of the formal sector? The facts that are available in the national account statistics are expressions of assumed relationships between the measured and unmeasured economy, and the assumptions are often not transparent for the data user. The resulting figures therefore need to be questioned and treated critically as historical evidence rather than as raw data that can be used as empirical observations to, for instance, test the relationship between the “formal” and “informal” economy.

20. 6 Conclusion It is difficult to give justice to all the issues surrounding the measurement of economic progress in African countries in the twentieth century. In this chapter I have focused on national accounting and the measurement of GDP per capita and economic growth. Already in colonial times it was remarked that these GDP estimates very soft numbers, and that they were not really fixed data points, but rather an expression of a guess with a large error of margin on each side. This issue remains with us today.46 As described in this chapter, Africa economies have been at the frontier of measurement debates—from the pioneering national account attempts in the 1950s, through the “discovery” of the informal sector in the 1970s and towards evidence based policy on social indicators in the twenty-first century.47 In 1997, in a book on the social sciences and international development, Packard and Cooper note that “there is no notion of ‘field economics’ comparable to the status of fieldwork in anthropology, so that economists work with data mediated through state collection apparatuses and categories that are not fully examined.”48 Not only has economists who have been working in development in African countries had these notions fully challenged, there has also been decisive movements towards economists collecting their own data, in field experiments and household surveys.

45 

For recent reviews of the informal sector literature, see King, K. (2001). Africa’s Informal Economies: Thirty Years On. SAIS Review, 11(1): 97–108; Hart, K. (2009). On the Informal Economy: The Political History of an Ethnographic Concept. Working Paper CEB, 09/04; Centre Emile Bernheim; and Meagher, K. (2010). Identity Economics: Social networks and the informal economy in Nigeria. London: James Currey. 46  In particular, see Jerven. M. (2013b). For richer, for poorer: GDP Revisions and Africa’s statistical tragedy. African Affairs, 112(446):138–147; Jerven. M. (2012b). Comparability of GDP Estimates in sub-Saharan Africa: The Effect of Revisions in Sources and Methods since Structural Adjustment. Review of Income and Wealth, 59(1):16–36; Jerven, M. (2010a). The Relativity of Poverty and Income: How Reliable Are African Economic Statistics? African Affairs, 109(434): 77–96, and Jerven, M. and Duncan, M.E. (2012). Revising GDP Estimates in Sub-Saharan Africa: Lessons from Ghana. African Statistical Journal, 15:12–24. 47  Sanga, D. (2011). The challenges of monitoring and reporting on the millennium development goals in Africa by 2015 and beyond. African Statistical Journal, 12(118):104–118. 48  Cooper, F. and Packard, R., “Chapter 1: Introduction,” in International development and the social sciences: essays on the history and politics of knowledge, Frederick Cooper and Randall Packard (eds), pp. 1–63. Berkeley, CA; London: University of California Press, 1997, 27.

Measuring Economic Progress in the African Context    405 The measurement of GDP in the African context is an insightful meeting between global standards and local applicability. GDP as an economic concept and method of aggregating information has at times given us a misleading picture of African economies, but the exchange between the complexities of economic transactions and social life and the rule based standards of accounting continues to be a fruitful intellectual challenge. Yet one of the fundamental insights from the study of economics in Africa, and this goes beyond Africa, is that no single metric can fully capture the human pursuit of progress.

References Ady, P. (1962). Use of national accounts in Africa. Review of Income and Wealth, 1962(1): 52–65. Bates, R.H. (1981). Markets and States in Tropical Africa:  the Political Basis of Agricultural Policies. Berkeley and Los Angeles, CA: University of California Press. Billington, G.C. (1962). A minimum system of national accounts for use by African countries and some related problems. Review of Income and Wealth, 1962(1):1–51. Cooper, F., and Packard, R. (1997). International Development and the Social Sciences: Essays on the History and Politics of Knowledge. Berkeley, CA; London: University of California Press, 1997. Deane, P. (1961). Review of three publications:  “Domestic Income and Product in Kenya: A Description of Sources and Methods with Revised Calculations from 1954-1958”; “The National Income of the Sudan, 1955-1956,” by C.H. Harvie and J.G. Kleve; “Comptes Économiques Togo, 1956-1957-1958” by G. Le Hégarat. Economic Journal, 71(283):630–631. Deane, P. (1948). The Measurement of Colonial National Income:  An Experiment. Cambridge: Cambridge University Press. Eke, I.I.U. (1966). The Nigerian National Accounts—A Critical Appraisal. The Nigerian Journal of Economic and Social Studies, 8(3):333–360. Federal Republic of Nigeria, Federal Ministry of Planning. (1981). National Accounts of Nigeria, 1973-1975. Lagos: Nigeria. Federation of Rhodesia and Nyasaland. (1955). Monthly Digest of Statistics, Salisbury. Frankel, S.H. (1953). The Economic Impact on Under-Developed Societies: Essays on International Investment and Social Change. Cambridge, MA: Harvard University Press. Harriss, J. (2002). The case for cross-disciplinary approaches in international development. World Development, 30:487–496. Hart, K. (2009). On the Informal Economy: The Political History of an Ethnographic Concept. Working Paper CEB, 09/04; Centre Emile Bernheim. Hart, K. (1973). Informal income opportunities and urban employment in Ghana. Journal of Modern African Studies, 11(1):61–89. International Labour Organization. (1972). Employment, Incomes and Equality: A Strategy for Increasing Production Employment in Kenya. Geneva: International Labour Organization. Jerven, M. (2014). Economic Growth and Measurement Reconsidered in Botswana, Kenya, Tanzania, and Zambia, 1965-1995. Oxford: Oxford University Press. Jerven, M. (2013a). Poor Numbers: How We Are Misled by African Development Statistics and What to Do about It. Ithaca, NY: Cornell University Press. Jerven, M. (2013b). For richer, for poorer: GDP revisions and Africa’s statistical tragedy. African Affairs, 112(446):138–147.

406   Methodological Issues Jerven, M. (2012a). An unlevel playing field: national income estimates and reciprocal comparison in global economic history. Journal of Global History, 7(1):107–128. Jerven, M. (2012b). Comparability of GDP estimates in sub-Saharan Africa: the effect of revisions in sources and methods since structural adjustment. Review of Income and Wealth, 59(1):16–36. Jerven, M. (2011a). Growth, stagnation or retrogression? On the accuracy of economic observations, Tanzania, 1961-2001. Journal of African Economies, 20(3):377–394. Jerven, M. (2011b). Users and producers of African income:  measuring African progress. African Affairs, 110(439):169–190. Jerven, M. (2010a). The relativity of poverty and income: how reliable are African economic statistics? African Affairs, 109(434):77–96. Jerven, M. (2010b). Random growth in Africa? Lessons from an evaluation of the growth evidence on Botswana, Kenya, Tanzania and Zambia, 1965–1995. Journal of Development Studies, 46(2):274–294. Jerven, M., and Duncan, M.E. (2012). Revising GDP estimates in sub-Saharan Africa: lessons from Ghana. African Statistical Journal, 15:12–24. Killick, T. (2010). Development Economics in Action: A Study of Economic Policies in Ghana, 2nd ed. New York: Routledge. Killick, T. (1978). Development Economics in Action: A Study of Economic Policies in Ghana. London: Heinemann. King, K. (2001). Africa’s informal economies: thirty years On. SAIS Review, 11(1):97–108. Meagher, K. (2010). Identity Economics: Social networks and the informal economy in Nigeria. London: James Currey. Okigbo, P.N.C. (1962). Nigerian National Accounts, 1950-57. Enugu: Government Printer. Peacock, A.T., and Dosser, D.G.M. (1958). The National Income of Tanganyika 1952-54. London: Her Majesty’s Stationery Office. Prest, A.R., and Stewart, I.G. (1953). The National Income of Nigeria, Colonial Office. Colonial Research Studies 11. London: Her Majesty’s Stationery Office. Republic of Zambia. (1994). National Accounts Statistics GDP Revision of Benchmark 1994 Estimates. Lusaka: Central Statistical Office. Sanga, D. (2011). The challenges of monitoring and reporting on the millennium development goals in Africa by 2015 and beyond. African Statistical Journal, 12(118):104–118. Seers, D. (1952–1953). The role of national income estimates in the statistical policy of an under-developed area. The Review of Economic Studies, 20(3):1952–1953. Speich, D. (2011). The use of global abstractions: national income accounting in the period of imperial decline. Journal of Global History, 6(1):7–28. Stiglitz, J., Sen, A., and Fitoussi, J.P. (2010). Mismeasuring our Lives: Why GDP doesn’t add up. New York: New Press. United Republic of Tanzania, Bureau of Statistics. 1997. Report on the Revised National Accounts of Tanzania 1987-96. Dar es Salaam: Bureau of Statistics. Van de Walle, N. (2001). African Economies and the Politics of Permanent Crisis, 1979-1999. New York: Cambridge University Press. Ward, M. (2004). Quantifying the World:  UN Ideas and Statistics. Bloomington:  Indiana University Place.

Chapter 21

M easu ring St ru c t u ra l Ec onomic Vu lne ra bi l i t y in Afri c a Patrick Guillaumont

21.1  Introduction: The Challenge of  Structural Economic Vulnerability for African Development In 2006, at a time when growth had clearly resumed in Africa, the opening speech at the first African Economic Conference organized by the African Development Bank and AERC was entitled “Economic vulnerability, still a challenge for African growth” (Guillaumont 2007, 2008). Eight years on, including a global recession, food and fuel price spikes, and recent state crises in Africa—although in many countries growth has continued—vulnerability remains an issue to be addressed. Both cross-country econometrics and case studies have documented the impact of external, climatic, and political shocks on Africa’s growth, development, and poverty reduction (Guillaumont 2007, 2008; see also the chapter by Xubei Luo, Handbook, this volume). Although some progress has been recorded in addressing economic vulnerability in Africa, it remains limited; moreover, the scope of vulnerability itself has been changing with the emergence of new—social and environmental—dimensions. Addressing the vulnerability of African economies requires an identification of its sources and determinants, including a conceptual clarification in view of the broadening scope. Section 2 proposes a conceptual framework where structural vulnerability is distinguished from general vulnerability, from physical vulnerability to climate change, and from state fragility as well. Section 3 analyzes the main features and evolution of structural economic vulnerability in Africa on the basis of an economic vulnerability index, highlighting not only higher structural economic vulnerability, but also a slower decline than in other developing economies. It then appears (section 4) that African economic vulnerability is reinforced by higher physical vulnerability to climate change, as shown by a specific index, and that Africa is the continent with the highest proportion of fragile states, suggesting a link between the

408   Methodological Issues various forms of vulnerability in Africa. Finally (section 5), measuring the structural vulnerability of African countries provides a useful tool for the international allocation of resources and not just to guide policies aimed at structural transformation and sustainable development. Adequately measured, structural vulnerability, as it is exogenous to current policy, may be a relevant criterion for the international allocation of concessional resources.

21.2  A Conceptual Framework for Measuring Structural Vulnerability in Africa Vulnerability, at the macro-level (as at the micro-level), is the risk for a country to be hampered by exogenous shocks, either natural (e.g. droughts) or external (e.g. fall in terms of trade). Structural vulnerability includes only factors that do not depend on a country’s current policies, being entirely determined by exogenous and persistent factors; while general vulnerability also includes the effect of current and future policies, and therefore changes more rapidly (Guillaumont 2001, 2006).

21.2.1  Size of the shocks, exposure, and resilience There are three main sources of country vulnerability: the size of exogenous shocks; the country’s exposure to those shocks (e.g. a small population size); and its capacity to cope with them, also named capacity to adapt or resilience. Structural vulnerability mainly results from the size of the shocks and the country’s exposure to them. General vulnerability also depends on resilience, more linked to current policy, and less to structural factors. There are indeed structural factors in the resilience of a country (such as its level of human capital and more generally its level of development or income per capita). However, most often these factors are not taken into account in the measurement of structural economic vulnerability, because, as we will see later on, they are considered separately. For instance, the Economic Vulnerability Index (EVI), an index of structural economic vulnerability devised by the United Nations for the identification of the Least Developed Countries (LDCs), includes neither per capita income nor the level of human capital, as their levels are also and separately used as LDC identification criteria.

21.2.2  The dynamic design of structural economic vulnerability A country’s structural economic vulnerability can be understood in a dynamic manner as the risk for this country to see its economic growth, and more generally its development rate, durably slowed down by exogenous shocks, independently of its will. It is not only a risk of static loss of welfare. Thus, factors to be taken into account in the design and measurement of structural economic vulnerability should be likely to lower the rate of economic growth. An even broader meaning of structural economic vulnerability would include the risk that

Measuring Structural Economic Vulnerability in Africa    409 the country’s development become unsustainable, again because of shocks and factors independent of its will. Since the meaning of sustainability, as reflected in the preparatory works of the post-2015 agenda, now covers several dimensions—not only economic, but also environmental and sociopolitical—it is useful to examine together, but distinctly from structural economic vulnerability (using its original meaning mentioned above), vulnerability to climate change and what is commonly called state fragility.

21.2.3  Structural economic vulnerability and physical vulnerability to climate change Indeed, some of the climatic sources of economic vulnerability that can be taken into account in the design of an index of economic vulnerability (as has been the case with the EVI—see section 3.1, are related to structural and permanent economic and geographical features, but not to climate change per se. Vulnerability to climate change stems from a risk of long-term change in geo-physical conditions rather than to a growth handicap in the medium term. In other words, it is more physical than economic, and refers to a longer time horizon. As with structural economic vulnerability, and even more so, physical vulnerability to climate change is independent of present (and future) country policy. For this reason, its measurement should be based only on physical characteristics and trends, as is done in the “Physical Vulnerability to Climate Change Index” (PVCCI) set up at Ferdi. Leaving aside the resilience components, physical vulnerability to climate change, like structural economic vulnerability, should reflect two main components:  shock intensity (due to climate change) and exposure (for instance the sea level rise and the share of areas likely to be flooded). The lack of socioeconomic components in the design of a physical index of vulnerability to climate change is all the more legitimate given that any assessment of future adaptation capacity is highly uncertain. Blending the measurement of structural economic vulnerability and physical vulnerability to climate change is conceivable, but it would risk blurring information about the type of vulnerability a given country is facing.

21.2.4  Structural economic vulnerability and state fragility A third dimension of vulnerability, but one that is highly dependent on the policy and current will of countries, is sociopolitical. State fragility is often presented as a form of vulnerability, although it is conceptually quite different. State fragility is designed and identified from present policy and institutional factors (lack of state capacity, political will, and political legitimacy), with many changing definitions, most often from an assessment of policies and institutions through the World Bank’s CPIA (Country Policy and Institutions Assessment). On the contrary, structural economic vulnerability is designed from factors supposed to be independent of policy (Guillaumont and Guillaumont Jeanneney 2009). Structural economic vulnerability significantly influences state fragility, as shown, for instance, by the impact of the UN’s EVI index on the CPIA (Guillaumont, McGillivray, and Wagner 2013). Accordingly, most of the countries identified as “fragile states” are

410   Methodological Issues also “least developed countries,” a category based on structural economic vulnerability (in addition to levels of income per capita and human capital). This holds for Africa, as explained later. On the basis of this conceptual framework, we now consider the measurement of structural economic vulnerability in Africa.

21.3  Comparative Levels and Trends of Structural Economic Vulnerability in Africa One approach to the analysis of economic vulnerability in Africa could be to consider growth volatility, an indicator that is widely used on account of its apparent simplicity and alleged impact on average growth (as evidenced by Ramey and Ramey 1995). Growth volatility is generally proxied by the standard deviation of the annual growth rate of gross domestic product (GDP) per capita over a given number of years (9–10 years in the World Development Report 2014). However, this approach is not appropriate for the measurement of structural economic vulnerability for two reasons. The main one is that growth-rate instability may result not just from structural factors but also from transitory ones and domestic policy, dependent on the will of the country. Second, its measurement is highly sensitive to the length of the period covered—it should cover a minimum number of years to reflect a structural feature, but the longer the period, the higher the risk that the standard deviation simply reflects a trend change.1

21.3.1  The UN Economic Vulnerability Index In order to examine the levels and trends of structural economic vulnerability in Africa, it is convenient to refer to the EVI set up by the UN’s Committee for Development Policy. The present structure of this index (initially introduced in 2000) was designed in 2005 (history and details in Guillaumont 2009a,b). It has been used for the triennial reviews of the list of LDCs in 2006, 2009, and, after a revision, in 2012 (on the scope and relevance of the revision, see Guillaumont 2013). Its principle is to combine with equal weights a group of three sub-indices reflecting the intensity of recurrent shocks, natural and external, and a group of four of five sub-indices reflecting exposure to those shocks. The structure of the index is shown in Figure 21.1 in its 2006–2009 and 2012 (revised) versions. The main change in the revised version was to add an “environmental” new component, the share of population living in low coastal areas, compensated by a reduced weight on population smallness (as argued by Bruckner 2012).

1  According to the statistics of the World Development Report 2014, the average volatility of the GDP growth in the 1990s and 2000s has been, respectively, 5.8 and 3.6 in Africa and 3.4 and 3.3 in other developing countries.

Measuring Structural Economic Vulnerability in Africa    411 Economic vulnerability index (EVI) Exposure index (1/2)

Size index 1/4

Population 1/8

2005 2011

Location index 1/8

Remoteness 1/8

Shock index (1/2)

Structural index 1/8

Environment index 1/8

Natural shock index 1/4

Share of agriculture, forestry and fisheries 1/16

Instability of agricultural production 1/8

Merchandise export concentration 1/16

Homeless due to natural disasters 1/8

Share of population in low elevated costal zones 1/8

Trade shock index 1/4 Instability of exports of goods and services 1/4

Instability of agricultural production 1/8 Victims of natural disasters 1/8

Figure  21.1  The Economic Vulnerability Index, 2005–2009 and 2011–2012 versions compared.

21.3.2  A high average level in Africa, a slower decline Given our choice to examine separately structural economic vulnerability and physical vulnerability to climate change, we use the 2006–2009 definition, all of whose components can be considered as potential contributors to slower growth. For that, we used new calculations of the EVI made at Ferdi on the basis of the 2006–2009 definition (see Table 21.1). In the same way, we will consider the EVI’s evolution using a constant definition (Ferdi “Retrospective EVI”), either the 2006–2009 one or the 2012 one (Cariolle 2011, Cariolle, and Guillaumont 2011; Cariolle and Goujon 2013).2 From Table 21.1, it clearly appears that African countries have a (significantly) higher EVI than other developing economies (whatever the definition, 2005–2009 or 2011–2012), due both to the shock and the exposure components of the index with the 2006–2009 definition, though only to the shock components with the 2012 definition (mainly because the share of population in low elevated coastal zones (LECZs) is noticeably lower in Africa, due to the number of landlocked countries, where it is zero). When only sub-Sahara Africa (SSA) is considered, the gap is even more important and observable for the exposure as well as for the shock components. Among African countries, the African LDCs, as a group, show an even higher structural economic vulnerability than other African countries. The same holds for the group African Landlocked, but to a lesser extent. 2  Several other improvements could be brought to the measurement of the EVI, in particular in the way by which the components are averaged (presently an arithmetic average) (see Guillaumont 2009a, b).

412   Methodological Issues Table 21.1  Economic Vulnerability Index (EVI): level in 2011 and change from 2000 to 2011. Africa and African subgroups compared to other developing countries. Components of EVI 2011 Country Category All Developing Countries(130) All African Countries(53) Sub Saharan Africa Countries(48) African LDCs(33) African Landlocked LDCs(11)

EVI 2011EVI 2000 (2006-2009 definition*)

Exposure

Shock

EVI 2011 (2011 definition)

EVI 2011 (2006-2009 definition)

37.2

36.3

36.7

37.2

−3.8

36.8

41.5

38.9

39.7

−2.6

38.8

43.0

40.6

41.2

−2.6

39.9 38.7

47.8 48.0

43.9 43.3

43.1 42.6

−3.1 0.2

*  The calculation of change (EVI 2011–2000) has been done without three countries: Eritrea, Ethiopia, and Timor-Leste because data for these countries are not available in 2000.

Has the structural vulnerability gap between African countries and other developing countries been shrinking or widening? The level of EVI measured according to a constant definition has been declining in African countries, evidencing some structural change. But the decrease in EVI observed between 2000 and 2011 has been deeper in non-African countries than in African ones, widening the structural gap between Africa and the rest of the developing world. In landlocked African countries, there has been no decrease, in spite of rapid population growth, a factor dampening the impact of small size.

21.3.3  Heterogeneity in vulnerability sources and in country levels Unsurprisingly, there are large differences among African countries, particularly obvious when looking at the EVI’s components, indicating the heterogeneity of sources of structural vulnerability among the African countries (see Table 21.2 and Appendix 1). As for the shock index’s components, when all African countries are considered, the three component sub-indices are higher in Africa than in other developing countries, the instability of exports (of goods and services) being dramatically higher. When only SSA countries are considered, the number of victims of disasters is significantly higher, while the instability of agricultural production is no longer higher. For only African LDCs and for only landlocked African countries, the three sub-indices are even higher, the former evidencing the highest average for export instability, the latter for the victims of disasters. At the country level, the highest levels are shown by oil exporters (for which export concentration is high, as well as the instability of exports) and by some agricultural countries (Eritrea and the Gambia).

Table 21.2  Components of the Economic Vulnerability Index (EVI) in 2011. Africa and African subgroups compared to other developing countries. Components of EVI 2011 Exposure Index

Shock Index

Country Category

Export Share of Share of Population Remoteness concentration Agriculture LECZ

Instability of Agricultural Instability production of Export

Victims of disasters

All developing countries (130) All African countries (53) Sub-Saharan Africa countries (48) African LDCs (33) African landlocked LDCs (11)

42.8 39.2 40.9 39.1 33.8

24.1 25.1 24.4 26.3 27.8

60.8 61.6 66.0 69.1 75.0

55.9 55.7 61.1 59.3 72.3

33.7 40.1 41.2 42.8 46.0

27.3 40.3 42.9 52.7 51.5

19.5 12.5 11.4 13.6 0.0

30.1 38.4 39.4 48.0 44.5

414   Methodological Issues As for the exposure sub-indices, when all African countries are considered, only the export concentration and the share of agriculture, forestry, and fisheries in GDP are significantly higher than in other developing countries, while the share of LECZ is of course significantly lower. When only SSA countries are considered, remoteness is also higher. For only landlocked African countries (and, to a lesser extent, African LDCs), the same three sub-indices are even higher, while the smallness of population is lower, and the LECZ component is nil. Finally the highest levels of the exposure index are observed in small coastal or island countries (Guinea Bissau, Comoros, Sao Tome and Principe, and the Gambia), and the lowest levels in the three Maghreb countries. These measures should be taken as proxy indicators of structural economic vulnerability, likely to be discussed and improved in their composition and calculation.

21.3.4  Structural economic vulnerability is even higher when considered more broadly It should be remembered that the assessment of structural economic vulnerability through the EVI, as it is presently designed, can only give a partial view, since it does not take into account the structural components of resilience, which are numerous and depend on the overall level of development. They are considered separately for the identification of the LDCs through the Human Assets Index (HAI), a composite index of health and education indicators, and the level of income per capita. Since the levels of human capital and income per capita are on average lower in Africa than in other developing countries, this reinforces the diagnosis of higher structural economic vulnerability in Africa. In this respect, it seems clear that the African countries with the lowest income and human capital face very high structural vulnerability:  these poor and vulnerable countries, the category that the term “least developed countries” tries to capture, are more likely than other to be “caught in a trap” (unless they receive special support (and/or achieve bold reforms); Guillaumont 2009a). The risk of getting trapped, as supposed in the LDCs identification criteria, results from the conjunction of structural economic vulnerability (stricto sensu) and low human capital in countries with low income per capita. For this reason, these three criteria are taken into account separately, in a complementary manner, for the identification of the LDCs. It is conceivable to aggregate the EVI and the HAI in a composite index of structural handicaps (a structural economic vulnerability lato sensu), allowing for limited substitutability between them to remain consistent with the initial hypothesis of complementarity or even to combine these two indicators with the smallness of income per capita, in order to obtain an index of “least development” (see details in Guillaumont 2009a). It then remains to distinguish between the African LDCs, most of them are small or medium population countries (except Ethiopia and Tanzania), and the other African countries. Some of those countries may appear to be vulnerable with regard to their EVI, but not with respect to their level of income per capita (mainly the case of oil exporters) or their level of human capital. Finally, it should be noted that the EVI may not seem to capture the new forms of vulnerability evidenced by the political events linked to the “Arab Spring” in North Africa. This vulnerability, easier to measure ex post, for instance through its impact on the instability of

Measuring Structural Economic Vulnerability in Africa    415 exports, does not fit the definition of structural vulnerability adopted above, since it mainly results from policy and institutional factors. It rather reflects a state fragility.

21.4  Structural Economic Vulnerability and other Measures of Vulnerability in Africa: Climate Change and State Fragility In the introduction of this chapter, we juxtaposed structural economic vulnerability with two other forms of vulnerability, physical vulnerability to climate change and state fragility. For both there are attempts at measurement, which leads, now, to consider two issues. First, how structural economic vulnerability in Africa, measured with a medium-term horizon, is enhanced in the long term by physical vulnerability to climate change, according to the “Physical Vulnerability to Climate Change Index (PVCCI)” and what are the main sources and patterns of vulnerability to climate change in Africa? Second, do African countries suffer from state fragility, according to current assessments, and is this fragility the result of structural vulnerability?

21.4.1  Africa vulnerability to climate change Because it is highly controversial to assess the likely socioeconomic consequences of climate change, there is a rationale for setting up an index of vulnerability relying solely on physical components. As shown in Figure 21.2, this index combines the physical impact of two kinds of shocks, the progressive shocks, namely the sea level rise and the aridification, and the intensification of recurrent shocks, in temperature and rainfall, respectively (see details in Guillaumont and Simonet 2011a, b). For these four kinds of shocks, we combine an indicator of the size of the shock with an indicator of the country exposure to the shock. Moreover, to better capture the vulnerability to any kind of shock linked to climate change we use a quadratic average of the main components instead of an arithmetic one. In Table 21.3, we give the average value of this index and its main components for the same groups of countries as the EVI. On average, all African countries, as well as sub-Saharan countries, have a higher average PVCCI than the other developing countries (Table 21.3), a difference more significantly demonstrated when a quadratic average is used, instead of the arithmetic one. The difference increases when only African landlocked countries are considered. This higher level of African countries is due both to the impact of the increasing intensity of recurrent shocks (weighted by corresponding exposure indices) and to the impact of progressive shocks. As for the risk associated with progressive shocks (see Tables 21.3 and 21.4), there is first a rather low impact of the sea level rise in Africa, since Africa includes very few small islands (more threatened by this trend) and many landlocked countries. But this lower vulnerability to sea level rise is compensated by a greater vulnerability to “increasing aridity” (7 points

416   Methodological Issues Physical Vulnerability to Climate Change Index PVCCI

Risks related to the intensification of recurrent shocks

Risks related to progressive shocks

Flooding due to sea level rise (1/4)

Increasing aridity (1/4)

Share of flood areas (1/8)

Share of dry lands (1/8)

Size of likely rise in sea level (1/8)

Rainfall (1/4)

Trend in – Temperature (1/16) – Rainfall (1/16)

Temperature (1/4)

Rainfall instability (1/8)

Temperature instability (1/8)

Trend in rainfall instability (1/8)

Trend in temperature instability (1/8)

Figure  21.2   The Physical Vulnerability to Climate Change Index (PVCCI). Source: Guillaumont and Simonet 2011b.

Table 21.3  The Physical Vulnerability to Climate Change Index: Africa and African sub-groups compared to other developing countries. Arithmetic average Country Category

Progressive shocks

Increasing recurrent shocks

Developing countries (118) African countries (53) Sub-Saharan Africa countries (48) African LDCs(33) African landlocked (15)

24.4

Quadratic average

PVCCI

Progressive shocks

Increasing recurrent shocks

PVCCI

47.9

36.2

32.4

48.9

42.5

26.2

49.4

37.8

35.9

50.1

44.7

24.7

50.5

37.6

33.8

51.2

44.3

24.2 26.1

51.4 51.2

37.8 38.7

33.4 36.9

52.3 52.1

44.8 46.1

difference compared to all developing countries), itself due to a stronger (increasing) trend in temperature (rather than to the (decreasing) trend in rainfall), combined with a larger share of drylands. Finally, when an arithmetic average is used, the index of the risk of progressive shocks is not significantly different in Africa and in other developing countries, because of these two opposing effects, but becomes so when a quadratic average is used.

Measuring Structural Economic Vulnerability in Africa    417 Table 21.4  Components of the Vulnerability to Climate Change Index: Africa and African subgroups compared to other developing countries. PVCCI, risk related to Country Category

Progressive shocks

Intensification of recurrent shocks in

Sea Level Rise

Increasing aridity

Rainfall

5.3 2.0 2.1 1.3 0

43.5 50.3 47.3 47.1 52.2

Developing countries (118) African countries (53) Sub-Saharan Africa countries (48) African LDCs (33) African landlocked LDCs (11)

42.6 45.8 47.2 47.4 46.4

Temperature 53.3 53.0 53.8 55.5 56.0

Number of countries between brackets.

As for the index of increasing recurrent shocks, the higher average level for all African countries is due to the trend of rainfall instability, from a high initial level. For African LDCs or African Landlocked countries, it is also due to the intensification of temperature instability. The index exhibits a large heterogeneity in the levels and types of vulnerability among countries (see Appendix 2). The ten most vulnerable African countries with regard to the quadratic PCCVI are Sudan, Namibia, Mauritania, Niger, Botswana, Somalia, the Gambia, Zimbabwe, Senegal, and Burkina Faso. This is generally due to a high level of several components of the index. The ranking is sensitive to the kind of average used, due to the fact that the sources and profile of vulnerability differ from one country to another. The risk of “progressive shocks” is, for some African countries (Botswana, Chad, and Mali), due to an increase of aridity, at the highest level in the world, while some African countries also face the other type of progressive shock, the sea level rise (Senegal or the Gambia). The “risk of intensification of recurrent shocks,” high on average for African countries, is at the highest level for some African countries (Burundi, Sao Tome and Principe, Guinea Bissau, Sudan, Zimbabwe, and Angola are among the ten developing countries at the highest level in the world). Thus, although many African countries seem to be highly vulnerable to climate change for physical reasons, the precise reason or channel of this (physical) vulnerability may differ significantly from one country to another. It would be conceivable to combine the PCCVI with the EVI measure (taking its 2005–2009 definition, to avoid partial overlapping through the share of population living in low coastal zones, the new EVI component introduced in 2001). Owing to the different horizons of EVI and PCCVI, the weight given to each of them would reflect some time preference. With equal weights, the highest averages would be obtained by Sudan, the Gambia, and Eritrea. Without calculating such a heterogeneous average, a picture of the two vulnerabilities can be given by representing the two indices on the same graph, as done in Appendix 3, where the oblique line corresponds to the 40 percent of countries with the highest average combined index. If African countries are compared to other developing countries with regard to this combined index, they evidence a significantly higher index (42.2 versus 37.9), even higher in SSA (42.7) and African LDCs (43.9) or African landlocked (43.7). Fourteen African countries are among the 20 developing countries with the highest combined index.

418   Methodological Issues

21.4.2  State fragility State fragility seems to be on the opposite side of structural (or physical) vulnerability, since it is designed from an assessment of present policy, as explained above. Among the various (and often changing) definitions of “Fragile States” and measurements of “state fragility” (see Guillaumont and Guillaumont Jeanneney 2009), let us take the list of Fragile States used at the OECD in 2013, and presented as a list harmonized between OECD and the multilateral development banks. For 2012, it includes 47 countries, 27 of which are African countries, all south of the Sahara, all but four of which are LDCs. It is worth noting that it did not include Mali, Burkina Faso, Libya, Egypt—showing how volatile and uninformative it can be in terms of the real political risks faced by the countries. It is more a tool for designing curative measures, than to prevent the occurrence of failing states. The source of this paradox, well exemplified by the case of African countries, is to be found in the way in which the list is set up. Roughly speaking, the Fragile States are countries with a CPIA (World Bank Country Policy and Institutional Assessment) lower than 3.2 (on a scale from 1 to 6) or with UN peace-keeping operations. Until recently, Mali was not considered as a Fragile State, because, due to policy improvement, its CPIA was above 3.2, although it was highly vulnerable; Niger, also highly vulnerable, is no longer considered “fragile,” also thanks to an improved CPIA. In fact, the category of Fragile State has been introduced to solve the problem met by the multilateral development banks in the allocation of their aid, based on so-called “performance.” The strict application of “performance based allocation” (PBA) appeared to require an exception for the states considered by various names as “fragile”; below a given CPIA threshold an exception to the strict PBA rule was applied. Without discussing here the consistency of the rule and its exceptions (see Guillaumont, Guillaumont Jeanneney, and Wagner 2010; Guillaumont 2013), let us note from a methodological point of view, it suggests the built-in weakness of the category of Fragile States. State fragility is indeed a huge issue, particularly in Africa, but it necessitates a qualitative assessment, allowing observers and donors to adapt their diagnostic and support, rather than a quantitative measurement (Collier 2012). On the contrary, structural economic vulnerability, as well as physical vulnerability to climate change, can be roughly measured, and legitimately used for international allocation of resources, as we see now.

21.5  The Measurement of Structural Vulnerability as a Tool for Designing International Policies towards Africa The measurement of structural vulnerabilities through indices such as EVI or PVCCI is useful, first, in the design of policies aimed at tackling their main sources, which, as we have seen, differ according by country. From that perspective, the components of the indices are more important than the composite index. For another use of such a measurement, the international allocation of concessional resources, the composite index is needed and matters more.

Measuring Structural Economic Vulnerability in Africa    419 Let us first consider the allocation of ODA. As already noted, PBA has been challenged by the Fragile State cases. Actually, the challenge is broader. We had previously argued that the allocation of international development assistance should take into account the structural economic vulnerability of recipient countries, for several reasons (Guillaumont 2009b, 2010b, 2013; Guillaumont, Guillaumont Jeanneney, and Wagner 2010). One such reason is equity. Taking into account the structural handicaps to development is a way by which development opportunities are made more equal between countries. At the same time, it increases the consistency of the reference to “performance,” which in its true meaning cannot be assessed without a consideration of these handicaps. Moreover, taking them into account is favorable to aid effectiveness, which has been shown to be higher in more vulnerable countries (thanks to the stabilizing impact of aid) (see, for instance, Chauvet and Guillaumont 2009; Collier and Goderis 2009). Finally, it allows for treating state fragility in an integrated framework, preventively as well as curatively. All these reasons particularly apply in the African context. This view has been supported by a recent resolution of the UN General Assembly (A/C.2/67/L.53 of 4 December 2012 on the smooth transition of graduating LDCs), inviting the “development partners to consider least developed countries indicators, gross national income per capita, the human assets index, and the economic vulnerability index as part of their criteria for allocating official development assistance,” (United Nations 2012). This has been recently implemented by the European Commission in programming the allocation of European assistance to ACP countries (Africa, Caribbean, Pacific) for the coming years. In this context, the measurement of structural economic vulnerability becomes particularly important. It may point to the need for a new refinement of the EVI. A similar argument can be advanced for the allocation of assistance for adaptation to climate change. It is indeed legitimate to consider vulnerability to climate change as a relevant allocation criterion (Füssel 2010; Guillaumont and Simonet 2011a; Wheeler 2011; Birdsall and De Nevers 2012). The main reason is that the recipient developing countries are not responsible for the physical vulnerability they face. From that perspective, the PVCCI, since it does not depend on recipient policy, may be a relevant allocation criterion. Of course, as previously noted for the EVI, a refinement of the PCCVI index (ongoing at Ferdi) may be useful for this aim. Moreover, other allocation criteria are needed alongside, such as the level of income per capita, as a complementary indicator of need, and an indicator of the likely effectiveness of assistance. Thus, in both cases (ODA and concessional resources for adaptation), it is clear that a structural or physical measure of vulnerability is a relevant allocation criterion. It also appears that structural resilience, not captured through the components of the EVI or the PVCCI, should also be taken into account through indicators such as per capita income or the level of human capital. While weak structural resilience, as a part of structural vulnerability, should lead to a higher allocation, a low resilience due to policy, if measurable, would act in the opposite direction, for reasons of effectiveness.

21.6 Conclusion This chapter tried to present the main methodological issues involved in the measurement of structural economic vulnerability in Africa. This meant designing a conceptual framework

420   Methodological Issues capable of disentangling (i) “structural” vulnerability from vulnerability linked to present policies, and (ii) medium-term structural economic vulnerability from long-term physical vulnerability to climate change. In this framework, there are many options for measurement, depending on the expected use. Although the indices presented here may be considered as tentative, they highlight both Africa’s high average level of structural or physical vulnerability and the strong heterogeneity of its sources across African countries. While the diversity of sources should be taken into account in the design of the domestic policies addressing structural vulnerability, synthetic measures given by composite indices of structural economic vulnerability and of physical vulnerability to climate change should be used as criteria in the international allocation of development assistance and of concessional adaptation resources, respectively.

Acknowledgment Olivier Cadot is strongly acknowledged for useful remarks and editing, without being committed by any opinion expressed in this chapter.

References Birdsall, N., and De Nevers, M. (2012). Adaptation Finance. How to Get Out from between a Rock and Hard Place. CGD Policy Paper 001, February. Bruckner, M. (2012). Climate change vulnerability and the identification of the least developed countries, UN DESA. CDP Background Paper No.15, ST/ESA/2012/CDP/15, June Cariolle, J. (2011). “The Economic Vulnerability Index: 2010 Update”, Ferdi Working Paper I09 Innovative Indicators Series. Cariolle, J., and M. Goujon (2013). “A Retrospective Economic Vulnerability Index, 1990-2011: Using the 2012 UN-CDP definitions”, Ferdi Working Paper I17 Innovative Indicators Series. Cariolle, J., and Guillaumont, P. (2011). A Retrospective Economic Vulnerability Index: 2010 update Policy Brief/17, March, Ferdi. Chauvet, L., and Guillaumont, P. (2009). Aid, volatility and growth again. when aid volatility matters and when it does not. Review of Development Economics, 13(3):452–463. Collier, P., and Goderis, B. (2009). Does Aid Mitigate External Shocks? Review of Development Economics, 13(3):429–451. Collier, P. (2012). How to Spend it. The organization of public spending and aid effectiveness, UNU-WIDER, Working Paper No. 2012/05, January. Füssel, H.M. (2010). How inequitable is the global distribution of responsibility, capability, and vulnerability to climate change: A comprehensive indicator-based assessment. Global Environmental Change, 20(4):597–611. Guillaumont, P. (2001 n°2 ). On the economic vulnerability of low income countries, CERDI. Etudes et Documents. Guillaumont, P. (2006). Macro vulnerability in low income countries and aid responses, in F. Bourguignon, B. Pleskovic, and J. van der Gaag (eds) Securing Development in an Unstable Word. ABCDE Europe 2005. Washington, DC: World Bank, pp. 65–108.

Measuring Structural Economic Vulnerability in Africa    421 Guillaumont, P. (2007). “La vulnérabilité économique, défi persistant à la croissance africaine” in African Development Review, Vol. 19 no 1 p.123-162, April. Guillaumont, P. (2008). “Economic Vulnerability: Still A Challenge For African Growth”, in Kasekende, L. and Ajakaiye (eds), AFDB Book Accelerating Africa’s Development Five Years Into the 31 Century, p. 608–637. Guillaumont, P. (2009a). Caught in a trap. Identifying the least developed countries. Economica, 2009:386. Guillaumont, P. (2009b). An economic vulnerability index: its design and use for international development policy. Oxford Development Studies, 37(3):193–228. Guillaumont, P. (2010a). Assessing the economic vulnerability of small island developing states and the least developed countries. Journal of Development Studies, 46(5):828–854. Guillaumont, P. (2010b). Considering vulnerability as an aid allocation criterion, Parliamentary Network on the World Bank (PNoWB). Network Review, 6(September):14–15. Guillaumont, P. (2013). Measuring Structural Vulnerability to Allocate Development Assistance and Adaptation Resources. Ferdi Working paper p. 68. Guillaumont, P., and Guillaumont Jeanneney, S. (2009). State fragility and economic vulnerability. What is measured and why? Paper prepared for the European Report on Development. Ferdi Working paper p. 07. Guillaumont, P., Guillaumont Jeanneney, S., and Wagner, L. (2010). How to take into account vulnerability in aid allocation criteria—ABCDE Conference Stockholm. Ferdi Working Paper, p. 13, May. Guillaumont, P., McGillivray, M. and Wagner, L. (2013). “Performance Assessment: How it Depends on Structural Economic Vulnerability and Human Capital Implications for the Allocation of aid”, Ferdi Working paper, p. 71, June. Guillaumont, P., and Simonet, C. (2011a). Designing an index of structural vulnerability to climate change. Ferdi Policy brief, B. 18, March. Guillaumont, P., and Simonet, C. (2011b). To What Extent are African Countries Vulnerable to Climate Change? Lessons from a New Indicator of Physical Vulnerability to Climate Change. Ferdi Working paper, I.08, November. Ramey, G., and Ramey, V.A. (1995). Cross-country evidence on the link between volatility and growth. American Economic Review, 85(5):1138–1151. United Nations (2012). General Assembly Resolution (A/C.2/67/L.51) on Smooth transition for countries graduating from the list of least developed countries. Wheeler, D. (2011). Quantifying vulnerability to climate change: implications for adaptation assistance. Center for Global Development, Working Paper 240.

Appendix 1  Level of the Economic Vulnerability Index in 2011 Calculated at Ferdi with New Data According to  the UNCDP 2011 Definition EVI 2011 (2011 definition)

Country Name Algeria Angola Benin Botswana Burkina Faso Burundi Côte d’Ivoire Cameroon Cape Verde Central African Republic Chad Comoros Congo Dem. Rep. of the Congo Djibouti Egypt Equatorial Guinea Eritrea Ethiopia Gabon Gambia Ghana Guinea Guinea–Bissau Kenya Lesotho Liberia Libyan Arab Jamahiriya

LandLDCs locked FS x x x x

X x x x

X X

x

x

X

x x

x

X x

x

x

x x x x

x x

x x x x x

x x x x x

Exposure Index

Shock Index

EVI 2011

Value Rank

Value Rank

Value Rank

Change (EVI2011EVI2000)

13.2 38.1 37.9 47.1 35.0 42.2 28.4 27.8 43.1 43.4

2 27 25 46 21 37 10 9 39 41

33.1 61.3 31.7 34.5 38.8 65.6 43.3 18.4 27.2 19.1

19 46 16 21 26 48 35 3 10 4

23.1 49.7 34.8 40.8 36.9 53.9 22.3 23.1 35.1 31.3

6 44 18 31 25 48 4 5 19 13

–0.9 –1.5 –10.1 –4.3 –7.4 2.3 1.4 –2.7 –3.2 –1.4

37.7 57.3 38.4 29.7

24 52 30 14

74.3 38.5 30.1 45.7

50 25 15 38

56.0 47.9 34.3 37.7

50 42 17 27

15.9 –7.8 –2.4 –2.0

48.1 21.2 43.1

48 4 40

44.1 15.6 41.1

37 2 30

46.1 18.4 42.1

40 2 33

–11.3 –6.5 –9.4

29.3 30.1 43.7 49.8 29.8 34.6 58.0 24.8 44.6 49.3 26.4

12 17 43 50 15 20 53 6 44 49 7

88.7 32.6 33.7 84.9 29.2 20.3 61.6 28.8 39.8 53.6 41.3

53 18 20 52 13 5 47 12 27 43 31

59.0 31.4 38.7 67.3 29.5 27.4 59.8 26.8 42.2 51.5 33.8

51 14 29 53 12 10 52 8 34 46 16

2.5 17.4 –17.9 1.0 –1.8 –2.7 –2.1 –12.5 3.7

Country Name Madagascar Malawi Mali Mauritania Mauritius Morocco Mozambique Namibia Niger Nigeria Rwanda Sao Tome and Principe Senegal Seychelles Sierra Leone Somalia South Africa Sudan Swaziland Togo Tunisia Uganda United Rep. of Tanzania Zambia Zimbabwe

Exposure Index

LandLDCs locked x x x x

FS

x x

x

x

x

x x

x

x x x

x

x x x

x x

x

x x

x

x

x x

x

x

x

x x

x

Value

Rank

Value

Rank

Value

Rank

Change (EVI2011EVI2000)

33.6 41.6 38.1 43.6 42.1 12.5 39.6 37.9 34.1 29.9 38.8 55.8

18 34 28 42 36 1 32 26 19 16 31 51

40.1 54.5 32.5 47.8 14.7 20.8 51.1 37.1 41.7 42.7 51.2 30.0

28 44 17 39 1 6 40 24 32 33 41 14

36.8 48.0 35.3 45.7 28.4 16.7 45.4 37.5 37.9 36.3 45.0 42.9

24 43 20 39 11 1 38 26 28 22 37 36

6.2 –4.0 1.3 8.9 –10.5 –5.0 –2.0 0.0 –5.0 –6.4 –4.4 –13.7

35.8 47.2 41.0 42.8 23.5 29.5 44.7 38.3 16.5 28.6 26.8

23 47 33 38 5 13 45 29 3 11 8

37.0 35.8 59.1 51.2 24.5 75.2 40.5 28.8 22.7 42.8 26.9

23 22 45 42 8 51 29 11 7 34 9

36.4 41.5 50.1 47.0 24.0 52.4 42.6 33.5 19.6 35.7 26.9

23 32 45 41 7 47 35 15 3 21 9

–2.8 –0.7 0.5 –14.3 0.4 0.3 0.0 –5.4 –5.7 –12.2 –2.8

42.0 35.4

35 22

67.2 43.5

49 36

54.6 39.5

49 30

9.4 2.5

Shock Index

EVI 2011

Appendix 2  Level of the Quadratic Physical Vulnerability to Climate Change Index Calculated by Ferdi and Rank of Countries PVCCI Country name Algeria Angola Benin Botswana Burkina Faso Burundi Côte d’Ivoire Cameroon Cape Verde Central African Republic Chad Comoros Congo Dem. Rep. of the Congo Djibouti Egypt Equatorial Guinea Eritrea Ethiopia Gabon Gambia Ghana Guinea Guinea-Bissau Kenya Lesotho Liberia Libyan Arab Jamahiriya Madagascar Malawi

LDCs

Landlocked FS

x x x x

x x x x

x x

x

x

x

x x

x

x x

x

x

x x x x

x x

x x x x x

x x

x x x x x

x

x

Progressive shocks

Recurrent shocks PVCCI

Value

Rank

Value

Rank

Value

Rank

55.3 28.2 24.9 60.9 50.7 23.3 16.6 20.0 50.6 14.9

46 25 21 52 37 19 4 12 36 1

48.1 60.3 48.0 50.3 56.5 72.0 43.9 48.0 50.2 44.1

21 48 19 28 43 53 9 20 27 10

51.8 47.1 38.2 55.8 53.7 53.5 33.2 36.8 50.4 32.9

38 32 15 49 44 43 4 10 37 3

52.4 29.4 15.9 18.2

39 28 2 9

53.4 29.0 51.5 43.5

36 1 31 8

52.9 29.2 38.1 33.4

42 1 14 5

53.5 54.5 18.1

43 45 7

38.9 42.6 47.1

5 7 17

46.8 48.9 35.7

30 36 8

50.1 35.4 17.5 51.8 18.9 20.9 22.3 44.9 22.0 19.0 58.3

35 30 5 38 10 13 17 32 16 11 50

46.4 45.5 52.3 58.0 48.9 53.0 62.4 46.8 49.6 47.2 46.8

14 11 32 46 23 35 51 15 25 18 15

48.3 40.8 39.0 55.0 37.0 40.3 46.8 45.9 38.4 35.9 52.9

34 21 17 47 12 19 31 28 16 9 41

22.8 25.6

18 22

53.8 51.4

37 30

41.3 40.6

22 20

Country name Mali Mauritania Mauritius Morocco Mozambique Namibia Niger Nigeria Rwanda Sao Tome and Principe Senegal Seychelles Sierra Leone Somalia South Africa Sudan Swaziland Togo Tunisia Uganda United Rep. of Tanzania Zambia Zimbabwe

Progressive shocks

Recurrent shocks

PVCCI

LDCs

Landlocked FS

Value

Rank

Value

Rank

Value

Rank

x x

x

52.6 56.0 16.5 53.5 26.9 61.0 56.9 33.8 22.0 18.2

40 47 3 44 23 53 48 29 15 8

51.1 57.3 39.3 31.2 55.3 52.6 55.5 52.6 54.6 71.3

29 44 6 3 41 34 42 33 40 52

51.8 56.6 30.1 43.8 43.5 57.0 56.2 44.2 41.6 52.0

39 51 2 26 25 52 50 27 23 40

53.3 29.3 24.6 53.3 47.4 58.2 39.2 17.9 59.0 21.9 28.8

41 27 20 42 33 49 31 6 51 14 26

54.2 53.8 46.1 57.4 50.0 62.1 29.3 46.1 31.2 48.9 48.5

39 37 13 45 26 50 2 12 3 24 22

53.8 43.3 37.0 55.4 48.7 60.2 34.6 34.9 47.2 37.9 39.9

45 24 11 48 35 53 6 7 33 13 18

28.2 47.7

24 34

58.5 60.7

47 49

45.9 54.6

29 46

x x

x

x x

x

x x x

x x x

x x

x

x x

x

x

x x

x

x

x

x x

x

Appendix 3 EVI 2011 (06–09 definition) as a function of quadratic PVCCI 70

Gambia

Eritrea

60

Chad

EVI 2011 (06–09 definition)

Comoros

Guinea-Bissau Seychelles

50

40

Gabon Equatorial Guinea Swaziland Sierra Leone Malawi Democratic Republic of the Congo Congo

Burundi

Sao Tome and Principe Zambia

Lybia

Angola Burkina Faso Djibouti Madagascar Central African Republic Liberia Nigeria Mali Cape Verde Rwanda Lesotho Cote d'Ivoire Togo Mozambique Benin Senegal Mauritius Ghana

30

Uganda

Mauritania

Namibia

Zimbabwe Niger

Algeria

Cameroon Guinea

Sudan

Somalia Botswana

Ethiopia

United Republic of Tanzania

20

Tunisia Morocco

30

35

40

Kenya

45

PVCCI. quadratic

South Africa Egypt

50

55

60

Chapter 22

Measu ring Demo c rac y An Economic Approach Célestin Monga

22.1 Introduction Something puzzling happened on March 21, 2014. Nicolas Sarkozy, who had been President of France until only a couple of years earlier, published an article in the daily newspaper Le Figaro accusing the French government of acting like a “dictatorship” and trampling “human rights.” His article followed revelations in the French media that independent judges had tapped several of his telephones. The bugging had revealed an alleged attempt by the ex-President to derail investigations against him—which covered, amongst other things, an accusation that he took money from the former Libyan leader Muammar El Kadhafi. Insisting that the bugging could only have been inspired by his successor and rival François Hollande as part of a “political persecution,” Sarkozy wrote: “I learn in the press that all my telephones have been bugged for eight months. The police have every detail of my intimate conversations with my wife, children and friends … One can easily imagine who is reading the transcripts!.… This is not an extract from that marvelous film The Lives of Others on East Germany and the activities of the Stasi. This is not the actions of some dictator somewhere against his political opposition. This is France.” (Sarkozy 2014). Sarkozy’s suggestion that his country had become a communist-like police state and dictatorship provoked equally blistering responses from President Francois Hollande and members of his administration. The Head of State said that is was “intolerable” to compare France with the former East Germany. The interior minister accused Sarkozy of trying to “protect himself ” behind a screen of “rage.” Other members of the ruling coalition mocked the former President’s angry article and spoke of the “Berlusconisation” of Sarkozy … This was about France, the country known in history and political science books as “the mother of democracy.” And this was not 1814 or 1914 but 2014… About the same time, Russian President Vladimir Putin was being described by leaders in most of the Western world as a dictator, and his country was “suspended” from the Group of 8 after accepting the secession of Crimea from Ukraine and its entry into the Russian Federation. Yet in Russia, Putin’s approval ratings reached records high (around 85 percent),

428   Methodological Issues with his supporters arguing that the Western World had no moral grounds to criticize his actions after encouraging the overthrow of a legitimately elected Ukrainian President in 2009 and 2014. In Venezuela and Thailand, other legitimately-elected leaders were being pushed out of office only a few months into their terms by angry crowds, again with tacit (if not public) support from the so-called international community. In Africa Egypt, the first democratically elected President in the 4000-year history of the country was arrested by an army general and put on trial while hundreds of his supporters were being sentenced to death—again, with wide acceptance from the “leaders of the Free World.” These examples, among many other puzzling events in an era that Fukuyama (1992) optimistically dubbed “the End of History,” offer an idea of the difficulties of comparative political well-being exercises. They also raise some basic but fundamental questions: what is going on? What exactly is democracy and how should it be measured across boundaries and even within different regions of the same country? “I can’t imagine Igbos traveling four thousand miles to tell anybody their worship was wrong!,” once said Nigerian novelist Chinua Achebe, expressing his faith in the intrinsic humility and sense of tolerance of his people. He also contrasted such unpretentious philosophical attitude with the prescriptive, self-righteous approach of the Arab or Western explorers and religious pioneers who devoted much of their lives traveling Africa in military expeditions and conquests, and brutally replacing value systems in faraway societies with theirs. Achebe’s assertion, whether right or wrong, points to the fundamental questions of truth and interpersonal comparisons of welfare (Elster and Roemer 1991). There lies the source of the still ongoing debate over what should be the proper criteria for validation of political well-being—and democratization. Measuring democratization and democracy is in itself an almost utopian endeavours. For if one assumes that this notion has a definable content, does one know enough about it to grasp it, to define its outlines, to identify it precisely, to chart its path, and follow its movement? How is one to launch into a comparative analysis of the idea of freedom, which has divided philosophers from the very beginning? And the obsessive desire to measure everything—including the most fleeting, uncertain sensations-does it run parallel to our confident belief in self-knowledge, in cognitive mechanisms, in numbers, whether one sides with those who believe such forms of knowledge are innate or those who view them as acquired and thus strongly dependent on social conventions? It is a challenging task, especially when one is aware of how risky it is to refute the power of our impressions, of our instinctive faculties of perception, understanding, and interpretation of phenomena (cf. Dennett 1993). But it is a rather important one for developing countries in an era when a discourse of conditionality still dominates North-South monetary relationships: developmental aid, it is frequently said, should be distributed in function of “good governance” and “political and economic reforms”—though these last often remain at the level of mere rhetoric and promises and do not correspond to reality (Kahler 1992). There is, in fact, nothing new about the current pessimism concerning democracy, except that some would apply it to Africa alone. In all ages, scholars who have attempted to define the precise content of the democratic idea have soon enough become skeptical. In a widely acclaimed work, Dahl (1971) spoke of the eight pillars of democracy as laid out in the Universal Declaration of Human Rights. Studying the so—called “free” nations, he concluded that no nation in the world is truly democratic. He put forth the notion of polyarchic regimes to describe Western “democracies.” Concurring with this view, Crozier, Huntington,

Measuring Democracy   429 and Watanuki stated in 1975: “What are in doubt today are not just the economic and military policies but also the political institutions inherited from the past. Is political democracy, as it exists today, a viable form of government for the industrialized countries of Europe, North America, and Asia?” (1975: 2). To illustrate their cause for concern, they cited German Chancellor Willy Brandt’s catastrophic prediction: “Western Europe has only 20 or 30 more years of democracy left in it; after that it will slide, engineless and rudderless, under the surrounding sea of dictatorship, and whether the dictation comes from a politburo or a junta will not make that much difference” (Crozier, Huntington, and Watanuki 1975: 2). To varying degrees, skepticism as to the intrinsic durability of democracy has always been present in thought, pushing certain intellectuals to become professional doomsayers of democracy (the most famous case being Alexander Zinoviev). The implosion of capitalist economies symbolized by the discontinuance of the Bretton Woods system of fixed exchange rates, the increase in social problems in most industrialized democracies after the 1973 oil crisis, and the tremendous upsurge in speculation in the wake of financial deregulation during the 1980s have finally caused political scientists to question individualism, the foundation of liberal democracy since the Enlightenment. This chapter, which builds on Monga (1996a), briefly discusses the foundations of these contradictions, circumscribes the narrative fields to which they belong, and proposes a framework for the analysis of current political transformations (the Democracy coefficient). Though the proposed framework may be applied to any region of the world, Africa was chosen for the elaboration of the model in its present form. Section 2 addresses the primary difficulties in evaluating the democratization process, calling attention to the poor quality of the information available in the media on Africa and to the conflicting perspectives dividing researchers (universalism vs. relativism). Section 3 explores the sources of confusion, analyzing the epistemological, conceptual, and methodological problems that all paradigms and “measurements” of freedom inevitably raise. Using an economic approach, section 4 proposes the Democracy coefficient, an integrative, dynamic model that attempts to reconcile “universalists” and “relativists.” Section 5 offers concluding thoughts.

22.2  Challenges of Assessing Democratic Progress The fall of the Berlin Wall in November 1989 and the liberation of Nelson Mandela from prison in February 1990 led to a widely held belief among researchers and political analysts that the world was witnessing a third wave of democratization (Huntington 1991).1 1  Huntington contends that the world has witnessed a third wave of democratization that purportedly began with the Revolution of 25 April 1974 in Portugal. “The first ‘long’ wave of democratization began in the 1820s, with the widening of the suffrage to a large proportion of the male population in the United States, and continued for almost a century until 1926, bringing into being some 29 democracies. In 1922, however, the coming to power of Mussolini in Italy marked the beginning of a first ‘reverse wave’ that by 1942 had reduced the number of democratic states in the world to 12. The triumph of the Allies in World War II initiated a second wave of democratization that reached its zenith in 1962 with 36 countries governed democratically, only to be followed by a second reverse wave (1960-1975) that brought the number of democracies back down to 30.” (Huntington 1991: 12).

430   Methodological Issues The optimism was understandable: for several thousand years, the world had been a brutal and autocratic place. Even after the eighteenth century Enlightenment, only the American Revolution appeared to be a sustainable democracy. During the nineteenth-century monarchists fought a prolonged rearguard action against freedom movements in Europe, and even the Meiji Revolution in Japan could not be labeled a full-fledged democracy. In the first half of the twentieth-century nascent democracies collapsed in Germany, Spain, and Italy. There was a major turnaround in the second half of the twentieth century, as democracy suddenly took root in some of the most unlikely places: Germany emerged from Nazism to establish a credible political system that was consistent with the principles of the 1948 United Nations Declaration of Human Rights. India, which has the world’s largest population of poor people, freed itself from colonialism, and designed and implemented a political system viewed as democratic. Mandela’s South Africa, which had been traumatized by apartheid, was able to build a political system that appeared stable and representative of the country’s social groups. More generally, decolonization created many new democracies in Africa and Asia, and autocratic regimes gave way to democracy in Greece (1974), Spain (1975), Argentina (1983), Brazil (1985), and Chile (1989). The collapse of the Soviet Union opened the way to many democracies in central Europe. Still, in the first decade of the twenty first century, the global picture seemed less convincing and the most popular measurement instruments provided a mixed assessment of the pace of democratization around the world. Even though around 40 percent of the world’s population, more people than ever before, lives in countries that are committed to holding hold “free and fair” elections on a regular basis, democracy’s global advance seems to have come to a halt, and may even have gone into reverse in many places. Between 1980 and 2000 the cause of democracy experienced only a few setbacks, but since 2000 there have been many. And democracy’s problems run deeper than mere numbers suggest. Many nominal democracies have slid towards autocracy, maintaining the outward appearance of democracy through elections, but without the rights and institutions that are equally important aspects of a functioning democratic system. (The Economist 2014: 44)

According to Freedom House, perhaps the most influential institution in the field of democratic measurement, nearly 2.5 billion people still lived under oppressive rule in 2013 (Freedom House 2014). Its assessment was that the state of freedom declined for the eighth consecutive year in 2013—despite a positive long-term trend. Out of 195 independent nations, 88 were classified as “free,” 59 as “partially free,” and 48 as “not free” (Figure 22.1). The Polity IV, the most widely used data resource for studying regime change and the effects of regime authority, also offers a generally positive assessment of democratic consolidation around the world and in Africa in particular, especially since 1990 (Figure 22.2).2 In the aftermath of the 2008 global financial and economic crisis (the Great Recession), many people even in industrialized countries seen as “old democracies” became disillusioned with the workings of their political systems—particularly when conservative and progressive governments alike decided to bail out financial institutions with taxpayers’ 2  A similar trend is observed by The Economist, which has published every year since 2007 a Democracy Index.

Measuring Democracy   431 100% 90% 80% 70% 60% 50% 40% 30% 20% 10%

Partly Free

12

10

20

08

20

06

20

04

20

02

20

00

20

98

20

96

19

94

19

92

19

90

Free

19

88

19

86

19

84

19

82

19

80

19

78

19

76

19

74

19

19

19

72

0%

Not Free

Figure  22.1   Freedom in the World, 1972–2013. Source:  Author, from data by Freedom  House.

5

1960 sub-sample 1965 sub-sample

Average Polity2 Score

1975 sub-sample 1990 sub-sample 0

–5

1960

1970

1980

1990

2000

2010

Year

Figure  22.2   Africa:  Average Polity 2 Score, 1960–2010. money and then stood by impotently as rich, private sector bankers and managers continued to pay themselves large bonuses. The emergence of China on the global scene not only as a dominant and indispensable economic force but also as the proponent of a “new” political model has forced researchers to rethink the foundations, scope, and ethics of democracy. Many observers were shocked to learn from the 2013 Pew Survey of Global Attitudes that 85 percent of Chinese were “very satisfied” with their country’s direction, compared with only 31 percent of Americans. While China’s critics point to the perpetuatoin of a single-party communist system, the imprisonment of dissidents, and censorship of freedom of speech, they acknowledged that the Chinese government was able to lift some 600 million people out of poverty in just thirty

432   Methodological Issues

Upper level: Superstructure

years (1979–2009, “an accomplishment unparalleled in human history” (Obama 2009). The perplexing question is whether the Chinese model—tight control by the Communist Party, coupled with a relentless effort to recruit talented people into its upper ranks—is more efficient than the traditional Western form of democracy, and less susceptible to gridlock. Despite its rigid structure, that model provides for a change in political leadership every decade or so, with a constant supply of new talent (party cadres are promoted based on their ability to deliver on public policy goals). The challenges of assessing democratic progress are particularly difficult in Africa, which is often seen as “the most politically volatile region,” with major democratic breakthroughs in some countries, and coups, insurgencies, and authoritarian crackdowns in others. From Egypt to Somalia, from Libya to the Central African Republic, African political markets display many peculiarities, indeed. The structure of political supply and political demand there is made even more complex than elsewhere because of the extraordinary high level of external influence by powerful actors such as the international financial institutions, the global non-governmental organizations, and the always present former colonies. This creates overlapping visible and invisible political worlds where democratic rules and practices are difficult to assess (see Figure 22.3). Such a mosaic of players tends to complicate to the extreme the rules of the political game, their implementation, and the criteria for validation of what democracy means in the African context (Monga 2010). It is therefore not surprising that Cameroon, for instance, has basically had two Constitutions in force since 1996, without raising concern in the so-called

Transnational Political Lobbies

Transnational Financial Networks

Invisible part

Transnational Sects and NGOs

Lower level: Technostructure

Visible part Government Institutions

Political Parties

Civil Society

Figure  22.3   Africa’s visible and invisible political worlds. Source:  Monga (1996b).

Measuring Democracy   433 international community. By contrast, Ivory Coast underwent a period of political turmoil after the 2011 presidential elections that led to military intervention by France and the removal of a President who was designated the winner by the Constitutional Court. Sudan’s President Omar El Bechir has been indicted by the International Criminal Court and yet welcomed by the African Union as the legitimate leader of his country. Likewise, Zimbabwe President Robert Mugabe has been re-elected several times despite being labeled a “dictator” in much of Western world. The question must therefore be raised: how much does one really know about what is going on politically in the world in general and in Africa in particular? Is there a mystery of African governance? (Lin and Monga 2012). Even if one assumes that a crude, and necessarily arbitrary, slicing up of the time continuum to identify “waves of democratization” is possible, and that such a teleological approach to social change is defensible, what impact have they had on Africa? What path have they followed over the past decades? Has there been an “advance” or a “decline” of the democratic idea? And how does one define and measure such progress or regression? Are the courses of action taken in different countries always comparable? Such questions are discussed in the next sections.

22.3  Sources of Confusion: Epistemology and Methodology Why are there such diverging views over the validity of democratic progress, especially in developing regions of the world? At the practical level, the most obvious reason is the ability of authoritarian regimes to successfully rebrand themselves and ride the wave of change, often by adjusting cosmetically to demands for freedom, and cleverly exploiting the opposition’s weaknesses and vulnerability (Monga 1996b). But there are also three deeper, philosophical explanations. The first has to do with the implicit epistemological assumptions underlying theories of democratic consolidation and their measurement tools such as the Freedom House Index or the Polity Scores, and the old dispute between universalism and cultural relativism. The second involves the limits that each method of comparative analysis de facto sets for itself, whether it adopts an “objective” approach or a “subjective” one. The third is the myth of a normative ethics of political well-being that would be valid across time and places. Let’s review them briefly in turn.

22.3.1  Epistemological issues and hidden assumptions Most of the themes that fuel the debate—and upon which comparative political scientists base their democratic evaluation methods—are derived from Immanuel Kant’s philosophy, and the belief in the universal, hieratic, sovereign subject. Yet, ever since Plato affirmed that the truth is accessible to anyone of us, the definition of the subject appears less “centralized,” and philosophers less inclined to sublimate a unique perception of reason. Not surprisingly, Kant’s overweening confidence in the power of reason and in the autonomy of the rational subject (i.e. the “democratic citizen”) has elicited interesting critiques from those

434   Methodological Issues who believe that, on the contrary, the contingent, nature of rules and criteria, which alone determine what is considered rational in a given place and a given historical era, ought to be substituted for these assumptions. Baynes et al. have nicely summed up this critique of the purity and unity of rationalism: It is no longer possible to deny the influence of the unconscious on the conscious, the role of the preconceptual and nonconceptual in the conceptual, the presence of the irrational—the economy of desire, the will to power—at the very core of the rational. Nor is it possible to ignore the intrinsically social character of ‘structure of consciousness’, the historical and cultural variability of categories of thought and principles of action, their interdependence with the changing forms of social and material reproduction. (Baynes et al. 1987: 4)

Another problem that must be addressed when seeking to conceptualize each individual’s experience of freedom is that of the status of knowledge as representation. Numerous philosophers have rejected the premise that the human subject is confronted with a world of objects independent of the self, and about which s/he “freely” forms a more or less accurate idea. Many writers have underscored the illusory nature of a set of fixed, autonomous themes relative to the sovereignty of self-consciousness; of a set of “pure” truths destined to be consumed by the spirit; and of the immutable nature of objects of thought that may be inserted into rigid schemes. These false assumptions are the primary tenet of empiricism (Davidson 1977)—and, of course, empirical research lies at the heart of social scientific research and of the methodologies adopted by Africanist political scientists. Knowledge of a given object is necessarily limited insofar as it is inscribed in a mental structure that pre-interprets it; and every knowing subject is in reality actively engaged in the work of interpretation s/he undertakes. The dream of a thinking subject that is detached from the world it endeavors to analyze is completely utopian. As Taylor notes, “the condition of our forming disinterested representations of the world is that we be already engaged with it. And the kinds of representations we form will depend on the kinds of dealings we have with it” (1987: 461). For political scientists who attempt to identify and measure freedom and democracy, the implications of that epistemological challenge are important: how to rid oneself of the presumption of distance and detachment that inhabits those who analyze sociopolitical processes? How can one haughtily and serenely attempt to define the different components of the democratization process that one endeavors to observe, analyze, and evaluate? These questions are all the more haunting in the context of democratic measurement, as one should call into question the artificial dichotomies, long sustained by classical philosophy, between fact and commentary, logos and mythos, logic and rhetoric, the literal and the figurative. There is no such thing as a self-justifying logocentrism and pure reason is always an illusion. Clearly, it is necessary to question most of the conscious and unconscious assumptions governing the interpretive frameworks, analytic models, and evaluative techniques used to assess social change in Africa. The interpretive frameworks and instruments of measurement used by the Freedom House and others fail to take into account many of the lessons of post-Enlightenment thought. Concepts such as good governance, human rights, civil liberties, participation, etc. are neither elucidated nor legitimated by the actual experience of those to whom they are applied, and they do not necessarily have the same meaning across places and time. Similarly, the complexity of language games, one of the pillars of Foucault’s genealogy of knowledge, appears neither in the theoretical models nor in the empirical

Measuring Democracy   435 research devoted to the problem of freedom in Africa. In other words, in the choice of variables and parameters used to study democracy in Africa, postmodernism’s most important lesson remains totally ignored. The deconstruction of Western metaphysics has yet to occur in the field of Africanist political science. It must, however, be undertaken if one rejects the premise that there exists a single democratic truth, which the West may have carefully protected since the time of Ancient Greece.

22.3.2  Universalism versus relativism A major epistemological challenge is likely to remain: how to demonstrate intellectual broad-mindedness toward various conceptions of freedom and tools for its measurement without losing sight of the fundamentals whatever they are? How is it possible to share democratic truth with all the peoples of the world—whose unity would have to be assumed—without running the risk of altering it? It is indeed important not to trivialize truth and falsehood, good and evil, justice and injustice. Even as they dismantled, like Nietzsche before them, Plato’s idea of a single truth, underscoring its complicity with such notions as “power,” “knowledge,” “constraints,” and “privileges,” philosophers of hermeneutics (Ricoeur 1969, for instance) refute the equal validity and facile conflation of all “truths.” Yesterday’s communist dictators demanded the right to their own “truth,” as did the political opportunists who organized ethnic cleansing campaigns in former Yugoslavia and Rwanda, in the name of “defending” a particular ethnic identity. The preceding discussion explains why those who evaluate political progress in the Third World fall into two basic categories: the universalists and the relativists. The former believe in the existence of a universal model of liberal democracy, as laid out in the provisions of international law concerning human rights and communities; they heed the manner in which southern authoritarian regimes attempt to adopt this model. They contend that the conception of democratic values found in international covenants forms part of humanity’s common spiritual heritage. The latter group rejects the idea of a universal model of human rights and demand that each country have the right to develop an alternative model, independent of the Western vision of freedom and more in keeping with the local values and culture. They speak of the specificity of “local” or “traditional” identities and cultures to defend the sovereign right of every people to decide the democratic question. This perspective is above all rooted in structuralist and postmodern thought. The sharp dichotomy between these two paradigms has turned the divergent viewpoints on the way in which to evaluate the democratization process that began in the early 1990s in numerous African nations into a veritable ideological labyrinth. The pride and optimism of African leaders, who all claim to have undertaken democratic reforms in their countries, is generally met with skepticism on the part of institutional observers (monitoring and human rights organizations, chancelleries), and with pessimism on the part of scholars and the media. Of course these contrasting viewpoints are at least partially explained by the fact that the evaluation of democracy is bound to be a frustrating exercise. While both camps have some valid arguments it is difficult to dispute the observations that all criteria used to evaluate democratization derive from international law, whose principles date back to the Magna Carta, issued by King John of England in 1215. The French Déclaration des Droits de l’Homme et du Citoyen (1789) and the Bill or Rights attached to

436   Methodological Issues the United States Constitution (1791) relied on the same principles, which were subsequently woven into various treaties, conventions, and resolutions of the League of Nations and the United Nations. Political scientists refer exclusively to these texts, the most recent versions of which are the Universal Declaration of Human Rights, unanimously adopted by the United Nations in 1948, the International Covenant on Civil and Political Rights, and the International Covenant on Economic, Social and Cultural Rights (1976). But even if one assumes that each sovereign member-state of the United Nations has freely chosen to ratify these conventions, no one would seriously dispute their Eurocentric quality. And if one admits that democracy is not a mere abstraction but a process in which very different civilizations throughout the world are continually engaged, then it is evident that the current models are at the very least incomplete.

22.3.3  The myth of a normative ethics of political well-being All currently available instruments for the evaluation of democratization in Africa and elsewhere subscribe to a normative ethics of democratic well-being, applicable to all citizens and all nations. But how is one to compare the different ways in which democracy is experienced? How is one to believe that measurement instruments derived from the Universal Declaration of Human Rights alone are comprehensive enough to cover all the nuances of the ethics of political well-being (or of greater democratic welfare)? Here the problem as to the validity of normative concepts touches upon the issue not only of space but of time: beyond the fact that the experience of freedom—and hence the evaluation of democracy—is different in Timbuktu, Mali and Timbuktu, Illinois, there is the element of time: men’s preferences, even in the same country, do not remain constant throughout time. The Russians, Texans, and Zulus of today do not necessarily expect the same things out of life that their grandparents and ancestors did. Their ideas about politics (their integration into the political market, their demands on politicians, their conditions for signing onto the social contract) naturally evolve over time, if only as a result of demographic changes. There is, then, a spatiotemporal problem in the way nations are evaluated by monitoring organizations: though it is never explicitly stated, all current indicators offer slices of political reality (as they perceive and conceive of it), snapshots of sorts of the status of democracy. In so doing, they view political reality through the lens of their own criteria, and adopt what economists refer to as a transversal perspective (sociologists speak of a synchronic perspective). At the very least, a second dimension is called for, one that would follow, over time, the democratic performance of different social groups. Instead of merely commenting upon the change in a country’s overall rating from one year to the next, using a rather arbitrary scale, why not adopt a more focused perspective and measure the ratings of identical groups (the same actors, if you will) over time? Why not attempt to track individual or collective stories by way of longitudinal analyses (in sociology, the diachronic approach)? When Freedom House for instance lists Country A and Country B among those countries in the world registering a decline in freedom and a change in category from one year to another, what is it really telling us? Did these two countries really move in the same direction over this 12-month period? Tremendous political change may have occurred in Country A, yet it is lumped together with Country B. Is the situation really the same in all these countries as the information suggests? Probably not. This is just one example of the type of hasty generalizations and

Measuring Democracy   437 misleading simplifications the choice of an exclusively synchronic approach to democratic measurement can lead to. Here again, the conceptual incoherence of the measurement tool stems from the epistemological incoherence that drives democracy monitors to define the political ethics attached to democracy in normative terms. Of course, it is conceivable that every society wishes to establish a set of “definitive” ethical principles, given that such carefully preserved and protected political precepts (a list of Ten Commandments of sorts) allow it to set the parameters of acceptable behavior and to safeguard the ideals that are to be passed on to future generations. The mythologies surrounding the American “Founding Fathers” and the French revolutionaries belong to this type of collective imaginary, which, in all countries, is part of the political game. It is one way among others to give one’s society or nation a sense of “identity,” something that is especially important in an age when the globalization of the market and the telecommunications revolution have complicated the primary task of politicians, which is to give some sort of meaning to a particular geographical area and to sustain the illusion of the significance of borders and nationalities. But scholars who endeavor to compare different political experiences on the basis of ethical norms developed in a specific time and place run the risk of failing to grasp the nuances of the function of political utility, of which all individuals, wherever they may live, have an idea, expressed not by way of an explicit formulation but through direct experience with its implications. This leads us to a discussion of the notion of interpersonal utility comparisons. The debate over the possibility of comparing notions of utility across individuals is decades old (Robbins 1938). Economists and philosophers have long traded arguments back and forth, and some have questioned the very possibility of giving any meaning whatsoever to interpersonal comparisons—this at least partially pertains to the democratic idea as Dahl has defined it (1971). Mathematicians have developed sophisticated methods to resolve the issue, but as Hammond (1991) has shown, the results of a century’s worth of work are rather disappointing. In all the existing indicators, political welfare is viewed as the sum of individual “welfares”—as defined by international law, that is to say, in a rather arbitrary and rigid manner, and without taking into consideration either personal value scales or even geographical and temporal variations. The least one can say is that thought on the subject of values has yet to lead to the development of an analytical framework that passes social scientific muster.

22.4  A Comparative Theory of Democratization Having shown in sections 2 and 3 how philosophical presuppositions compromise the “neutrality” of currently available democratic measurement instruments, I shall begin this section with a brief discussion of the epistemological framework of a unifying model so as to justify its theoretical postulates. I go on to present the technical properties and mathematical structure of a different tool, the Democracy coefficient, whose conceptual scheme and construction technique distinguish it from other indices in the literature on democratic measurement. It proposes an economic approach to the measurement of democratization and democracy.

438   Methodological Issues

22.4.1  An economic approach to democratization: an alternative to traditional regressions The goal of any index in this field should be to formalize the political game and its mechanisms in such a way as to express, beyond various systems of representations, a quantitative value of political welfare. A good tool for assessing political change must take account of heterogeneity in country circumstances and among political agents. This would require that analyses of political development processes be carried out not only at the aggregate/universal level but also at a country/disaggregated level. The reason is straightforward: beyond some broad and vague general principles, the people of Bolivia do not necessarily envisage democratic consolidation as those in Swaziland or in Japan. Moreover, even in the same country, there is no valid philosophical justification that the conceptualization of democratic principles remains the same over decades and centuries. After all, as Cioran once said, “tradition is a right granted to the dead.” In an ever-changing world, people should indeed be free to set their priorities and redefine the legal and regulatory political frameworks to reflect their changing values and sociopolitical preferences. The Democracy coefficient is built on these dynamic principles and allows for a less easily contested understanding, measurement, and possible prediction of the direction of sociopolitical change. Architects of models generally base their preliminary hypotheses on past-observed phenomena and then test them against the data. The approach here is necessarily different insofar as the underlying definition of democracy is not exclusively normative and universalist. More inclusive and supple, it takes into account not only the issue of heterogeneity and the values that each community decides to endorse but also the body of universal texts that sovereign states ratify when they elect to become a member of the international community. The proposed index thus has three sources of legitimacy: it reflects an observation of the most important phenomena of African political life since independence; it also reflects not only international law on the subject of human dignity but also the democratization validation criteria used by the African political actors themselves, who must be surveyed about their preferences in the context of their particular environment. Theories of democratic development based on the notion of social capital (Putnam 1993) suggest that the actual participation of citizens in political and economic life-which may be seen as the best approximation of the democratic idea-is a function of the consensus of the principal political actors on the rules of the game as well as on the values public authority is called upon to preserve and protect. This Democracy coefficient therefore refutes both culturalist theories (which subordinate democratic consolidation to cultural, ethnic, or civilizational factors) and deterministic theories, which focus on the economic and social situation or on the balance of power between the major powers in the international state system. Analytical frameworks from economics can help. One obvious way to go would be to think of democracy as a function of a consensus observable over time, written as:

D = f (C , t ) + ε (1)

where C represents the level of consensus and t represents time. A set of behavioral relationships can then be introduced to link changes in the level of consensus to a set of variables which describe the initial sociopolitical conditions, policy variables, and the institutional

Measuring Democracy   439 environment. One can label these set of variables “Z,” which gives us a reduced form democratic model where aggregate democratic growth, D , is related to this set of “determinants of democracy”, Z (be taken to express the effect of the level and quality of political participation and t the effect of changes in the mores and behavior of political elites over time, etc.), through some function f and a set of parameters β:

 D = f (β, Z ) (2)

This methodological route leads to the search for the determinants of democratization, and to spell out possible variables that would qualify in the African context. The econometric literature on democracy and democratization that has ventured in this direction has relied on a function f that is essentially linear—the common “democracy regressions” have mostly investigated the linear additive effects of institutional quality, governance, level of economic development, etc., on democratic consolidation. Yet, anecdotal evidence has shown that countries vary with regard to the conditions under which they are able to generate and sustain democratization. The effect of “democracy determinants” is heterogeneous across countries and depends on complementary institutions and policies. In theory, adding sufficient complexity to the above model to, for example, to take account of policy and institutional complementarities, we should in principle be able to account for country specificity. In practice, accounting satisfactorily for country heterogeneity would require a high degree of complexity on the specification of the function f. In fact, the model would rapidly become very complex. Furthermore, digging deeper within concepts such as governance or institutional quality and taking into account actual policy levers such as the level of implementation of existing laws and regulations would multiply the number of dimensions of Z and lead to an even more complicated model. The problem with a highly complicated model of such nature at the aggregate level is that it cannot be estimated, because of the limited number of observations across countries and over time. There is simply not enough data at the aggregate level. Moreover, running such a regression would imply that one addresses the well-known econometric problems associated with it. First, there is multicollinearity, which means that two or more of the explanatory variables of democracy will be found to be very closely related and, as a result, one may not be able to distinguish the effect of one from the effect of the other. This suggests that one or more of the explanatory variables are linearly related in the sample, but that there is no real causal relationship between them. The consequences are not negligible: while the coefficient estimates will be unbiased even if there is multicollinearity, standard errors of estimates will increase/t-stats will decrease. This means that variables that play a significant role in explaining the value of the dependent variable (democracy) will not have estimated coefficients that are significantly different from zero. In addition, estimates will be very sensitive to changes in specification of the model. In such situations, dropping a variable from the regression or even excluding some observations may lead to large changes in estimated coefficients, leading to an important lack of confidence in the robustness of estimates. This effect is due to the high correlation between explanatory variables. Still, it would not be a problem to derive good predictions given a complete set of explanatory variables, even if their effects cannot be separated but it would be nearly impossible to separate the effects of changes in individual variables. But in the African context—where the polity

440   Methodological Issues is often fractionalized because of the pervasive history of poverty—it would be risky to try to predetermine a single, “universal,” generic and aggregate list of explanatory variables for political consensus and democratic consolidation. Such an exercise is probably best carried out at the level of each country or region, and through rigorous surveys of the citizens/political agents themselves—not some external experts. Second, there is endogeneity, which presumes some functional relationship between explanatory variables, the error term, and/or the dependent variable in addition to the relationship that is to be estimated. It may be due to reverse causality, sample selection, or some correlated missing regressors, and always induces bias. It is well known that correlation does not equal causation. Higher levels of political participation may drive higher levels of democratization. Alternatively, an unobserved variable may jointly determine both high levels of institutional quality and high levels of democratization. Or both might be true. But the main reason why a correlation between institutional quality and the level of democracy may not allow concluding that this correlation is causal, and running from institutional quality to democratization, is that observational data is usually not randomly assigned. This is certainly the case when it comes to institutions. If institutions were randomly assigned within and across countries, establishing causality would be as easy as ascertaining whether institutions and some measure of democratic development are correlated. While there are various ways of dealing with this problem—most notably through the use of instrumental variables thought to have no direct association with the outcome under study, that is, democratization—it is also unlikely that it can be resolved satisfactorily when the list of truly valid instruments is not obvious (Bound et al. 1995). Third, there is the omitted variable problem, which will cause coefficient estimates to be biased. The typical way of solving the omitted variables problem is to find instruments, or proxies, for the omitted variables, but this approach makes strong assumptions that are rarely met in practice (Leightner and Inoue 2012). Confronted with such methodological issues, the economist who aims at applying the tools of his discipline to the study of African political markets must search some alternative routes. One is to introduce stronger assumptions and priors about the underlying processes. In other words, use an individual country focus with priors on the specification of the function f. This would essentially amount to using a structural (rather than reduced-form) model with the hope of achieving country specificity. It is an equally risky route, as it would not solve the epistemological problems discussed in section 3, or provide good answers to the issue raised by the relativists, i.e. “whose conception of democracy is being measured?” The second and more promising alternative is to choose a completely different methodological route and to carry out the analysis at the “disaggregated level,” that of the political agents in each country. Rather than seeking to pinpoint the determinants of democracy from traditional regression analyses, it is advisable to go back to equation (1) and examine the constitutive elements of the model. The central concept of consensus can then be broken down into three important components: • First, the consensus on the rules of the political game and the values to be protected, recorded in writing (or understood by the principal actors), for it all begins here. This includes, most notably, the constitutional framework adopted by each nation. Let us call this the design of democracy (α).

Measuring Democracy   441 • Second, the consensus on the everyday application (β) of the agreed-upon rules; that is, the way the democratic framework is interpreted, applied, and actually experienced by the principal political actors. • Third, the consensus on the maintenance of democracy (γ), that is, the major players’ long-term commitment to mechanisms of democratic consolidation; e.g. important institutional changes (constitutional amendments, administrative divisions, matters of citizenship), methods governing the adoption and repeal of laws and regulations pertaining to sensitive matters (the voting calendar, the national economic agenda). Focusing on these three components of consensus, and without entering into an analysis of their respective determinants, the democratic equation may be written in the form of a simple identity:

D ≡ α + β + γ (3)

Variables underlying these three notions need not to be identified through regression analyses. To better grasps the complexity of political transformations while ensuring the legitimacy of the democratic experience from the people themselves, the model proposed here attempts to counterbalance the “subjectivity” that lurks behind current evaluative methods, thereby delegitimizing the findings of researchers. It consists in calculating, for each country, an index number that corresponds to the sum of the three components in equation (3), each of which is tied to a battery of indicators that capture the actual meaning of democracy, as it is at once defined in international law and by local political actors. The Coefficient brings to political analysis several quantifiable concepts that allow democracy to be legitimated by those who are most directly concerned, permit interpretative frameworks to be brought into uniformity so as to facilitate comparative judgments, and enable empirical criteria to be clarified. This approach may be graphically represented as in Figure 22.4. The notable advantage of such an approach is that it brings about a necessary synthesis of “universal” and “local” values, brings together the “truth” of international law and the “truth” of political relativism. It at once solves the problem of Western ethnocentrism, which has marked the conception of liberal democracy advocated by the UN and human rights organizations, and that of the corrupt use of the principle of political sovereignty, which certain African leaders have invoked to justify authoritarianism. Moreover, the distinction between the design, application, and maintenance of democracy highlights the need not to underestimate the capacity of sitting governments to deceive. It opens up the possibility of an in-depth examination, one that discerns the merely tactical changes in the tone of official discourse, the (theoretical) adoption of new rules of the political game in the context of the modernization of national political life, the actual application of the adopted rules, and the consolidation, over time, of the processes of democratization. It is essential to take such an approach, because analyses that rely on traditional evaluative criteria have revealed their limitations: the simple adoption of a constitution based on the separation of powers or the mere organization of “pluralist” elections conducted under the complacent eye of international observers is not sufficient for a country to be awarded the label of “democratic.” Other elements must be taken into account. Before discussing the different conceptualization stages and practical methodology of this two-pronged approach (the following sections) as well as the techniques used to assign

442   Methodological Issues

International Law (Universalism)

Local Values and Norms (Relativism)

Design

Democracy Coefficient

Ite

rat

ive

pro

ces

s

Implementation

Maintenance

Figure  22.4   A  dynamic model of comparison of political well-being. numerical values to the various factors under study, I shall first address the philosophical foundations and the method employed to aggregate the three main components of the Index.

22.4.2  Philosophical assumptions and semantic clarifications The Democracy coefficient attempts to temper the imperiousness of Western reason while doing justice to the humanist impulse of contemporary philosophical thought. For reason, despite its arrogance and limitations, nevertheless serves a regulative purpose, as Putnam (1982) has pointed out. It allows us to critique historically stratified political traditions and to retain a minimal amount of skepticism toward all cultural practices and all discourses. Nevertheless the pluralism of forms of political discourse does not present an insurmountable obstacle. Although the idea of a single truth has been rejected, one should not automatically embrace difference as such, because our interpretive methods do not in fact confirm with absolute certainty that the conceptual schemes underlying the development of diverse cultures necessarily differ from our own. In other words, it is theoretically conceivable that the foundations of the political and social organizations found among the Pygmies, for example, and among the Slavs are not as dissimilar as they appear. It is indeed important to move beyond the universalist–relativist dichotomy insofar as it leads to a conceptual impasse in the field of political analysis. The politicization of the debate on values inexorably gives rise to a sterile hierarchy of particular truths. As West observes, the aim is not giving up on universality, not giving up on threadbare notions of objectivity, not giving up on notions of balanced analysis, so [the debate] is not reduced simply to partisanship, it’s not reduced solely to power struggle. There is space for critical exchange, but only

Measuring Democracy   443 if we acknowledge that truth has only a negative function, if we never reduce truth claims to assertabilitly claims, if we recognize that all truths have a small ‘t’, if we recognize that a big ‘T’ is always a fish that stands outside our conceptual net. (1993: 123)

A more rewarding philosophical approach is to stop constantly pitting these two approaches against each other, stop clinging to the illusion of immutable, mutually exclusive truths (“ours” vs. “theirs”) and start trying to bring together the best parts of each approach. On the one hand, it is illusory to believe that the exalted humanism that underpins the so-called “international” covenants on democracy and freedom may be transformed into valid, practical tools for the measurement of democracy. On the other hand, the apostles of a certain type of universal rationalism are right to be wary of the “uses of diversity,” especially in the political realm. As Davidson rightly puts it “there is no chance that someone can take up a vantage point for comparing conceptual schemes by temporarily shedding his own” (1984: 185). The political evaluation model put forth here is based on a pluralist conception of truth and reflects a dynamic, unstable truth insofar as “a) truth is always to be made … b) it is not pure creation, as it only operates within known and acknowledged constraints and limits …  c) yet it always remains an invention, not a simple reproduction of archetypes … Veracity emerges in that triangle that includes the truth, the others and the self, inseparably, with neither party being the judge of last resort” (Eboussi Boulaga 1991: 260). The Democracy coefficient seeks above all to refute the idea of an a priori political knowledge and a self-justifying democratic philosophy that ascribes to itself a normative value simply because it is derived from the precepts of habeas corpus and is endorsed by international human rights organizations. But it also rejects the relativism that deconstructs the Western concept of freedom for the express purpose of constructing oppressive systems. It is easy enough to state such postulates as long as one remains at the level of abstract principles. Moving from the epistemological to the formulation of a comparative model of political evaluation is a more delicate exercise. The proposed Democracy coefficient attempts to avoid the primary conceptual and methodological pitfalls discussed above, which limit the legitimacy and acceptance of such popular democratic measurement instruments as the Freedom House and Polity IV indices. Rather than limiting itself to variables drawn from the context of international law (the Universal Declaration of Human Rights, international conventions on economic, social and cultural rights) as these indices do, the approach here is deliberately different: it defines democracy as the merger and constant aggiornamento of (i) the universal rules and values of human dignity as laid out in international law, and (ii) the rules of politics as defined and assessed by the political class of each nation. The Democracy coefficient reflects this two-pronged approach and endeavors to synthesize the following two elements: it uses the rating the main local political actors give their country in the area of respect for human rights, as defined in the texts adopted by the UN and ratified by almost every nation in the world;3 and the rating not only foreign analysts but local political actors give a country in the area of respect for the internally agreed-upon rules and values. 3  The most important among them is the Universal Declaration of Human Rights, adopted 10 December 1948, General Assembly Res. 217a, U.N. Doc. A/810.

444   Methodological Issues

22.4.3  Analytical structure of the democracy coefficient This section discusses the methodology for the Democracy coefficient. It presents the different steps leading to the construction of each of its components, the choice of measurement indicators, and the assignment of numerical values to these last. It begins with a short rationale for the adoption of a two-pronged approach. It then goes on to state the concepts that emerged from the method and conclude with a brief presentation of the techniques used to assign numerical values to the variables of interest. The ultimate goal here is to measure democratic consolidation (assumed to be a proxy for political well-being) across countries, despite the reality that citizens and political agents have different preferences and values. While they may adhere to the so-called “universal” norms and belong to countries that have signed and ratified all major international treaties and covenants on freedom and political rights, they also cherish their particular set of national or regional (often subjective and evolving) local values and cultures. Therefore, the proposed Democracy coefficient attempts to measure both processes, through two types of indicators: Category A and Category B. This integrative conception of democracy proceeds in two stages. First, it defines a number of measurable concepts which are rated to reflect the level of adherence to international law in the area of human rights—Category A indicators. Then it draws up from survey results a second list of measurable concepts, defined, this time, by the local political actors in each nation—these shall be referred to as category B indicators. Category A indicators are to be chosen for each country or region on the basis of Ranney and Kendall’s (1969) and Dahl’s conceptualizations of democracy, which best grasp the essence of the Universal Declaration of Human Rights. Outlining the basic principles for a model of democracy, Ranney and Kendall wrote: In order for a Government to be called a ‘democracy’…, it must exhibit the following minimum characteristics: (1) Those who hold office in it must stand ready, in some sense, to do whatever the people want them to do, and to refrain from doing anything the people oppose; (2) each member of the “community” for which it acts should have, in some sense, as good a chance as his fellows to participate in the community’s decision-making—no better and no worse; and (3) it must operate in terms of an understanding that when the enfranchised members of the community disagree as to what ought to be done, the last word lies, in some sense, with the larger number and never the smaller… (1969:  45–46).

Four imperatives follow from this: respect for the sovereignty of the people, political equality of all citizens, regularly held elections, and respect for the opinion of the majority. Dahl’s model takes up and refines these concepts. Enumerating the characteristics of an ideal—and utopian—polity Dahl defines democracy as follows: a political system one of the characteristics of which is the quality of being completely or almost completely responsive to all its citizens … In order for a government to continue over a period of time to be responsive to the preferences of its citizens, considered as political equals, all full citizens must have unimpaired opportunities: (1) to formulate their preferences (2) to signify their preferences to their fellow citizens and the government by individual and collective action (3) to have their preferences weighed equally in the conduct of the government, that is, weighted with no discrimination because of the content or source of the preference. (1971: 2)

Measuring Democracy   445 Based upon these principles, Dahl formulates a list of eight conditions for democracy: (i) Freedom of assembly; (ii) Freedom of expression; (iii) Right to vote; (iv) Right to hold elected office; (v) Free competition among political leaders for the votes of citizens; (vi) Pluralism of information; (vii) Free and fair elections; and (viii), Existence of institutions ensuring citizen input into public policy decisions. In the proposed Democracy coefficient, category A indicators take account of the principles stressed by Ranney and Kendall as well as by Dahl. Some of them are virtually unattainable anywhere in the world—Dahl himself speaks of democracy as a “hypothetical” system and prefers the concept of polyarchy. But they provide an appropriate measurement framework, one that reflects the essence of the universalist view of human rights, as laid out in international law. Category B indicators represent an attempt to take account of the democratic principles that are specific to a given “culture” (as they are defined by each nation’s political class). A sample of the most influential members of the political class—which here includes civil society4 as well as political parties—must be surveyed in this regard. The rules and “values” that are to be selected as category B indicators are then culled from the results of this survey, the methodology and organization of which draws upon marketing techniques Green and Wind (1975) developed to “measure consumers’ judgments.” These category B indicators are to be defined freely and assessed by the very political actors who select them—not by external, foreign observers. In fact, category B indicators are to be selected and measured the same way economists measure inflation through the consumer price index. The final index is the arithmetic mean of the two scores.

22.4.4  Aggregation, sampling, and the assignment of numerical values Quantifying country performance on the ideals and values derived from this two-pronged approach requires ratings by experts and observers, as well as survey results carried out at the level of each county or region under analysis. It must be done both at the level of public statements and legal documents (de jure) and at the level of actual policies and practices (de facto). The method proposed here to assign numerical values to phenomena attempts to temper, to the extent possible, sources of subjectivism. It identifies the major ingredients for the advent of democracy as prescribed by international law—this is done through ratings by experts—and also assesses how the democratic consolidation process is being perceived by political agents—this is done through field surveys, documental research, and interviews with local political actors. That combined approach yields a list of “legitimate” indicators that by and large reflect the conception leaders and people have of democracy—and the conditions under which they adhere to the rules of the political game. For each Category A indicator, a country or region is rated by at least two different analysts on a scale of 0–10, 0 being 4  The definition of civil society proposed in Monga (1996b) is used here; civil society is taken to include only those socio-professional groups, religious or cultural associations, and influential members of the community that subscribe to the democratic ideal (in the broadest sense) and attempt to construct institutional and informal constraints limiting the power of the state and political parties, which naturally tend toward totalitarianism and coercion.

446   Methodological Issues the lowest rating in each category. Category B indicators are rated by a representative sample of local political actors. The criteria used to rate each Category A indicator can vary according to the topic under consideration. For example, the ability to implement public policy, the indicator of the governability of societies that have developed a panoply of techniques to resist authoritarianism, can be picked as a proxy for administrative capacity and measured on the basis of the most recently available figures on the collection rate of tax revenue (generally those from the preceding year published in accordance with the Budget Bill or IMF estimates). For the purposes of aggregation, the statistics are then converted into a rating on a scale of 0-10. The evaluation of category B indicators is left to local political actors who are best acquainted with, and thereby best able to sense, analyze, and judge, their country’s political reality. The use of surveys for the evaluation of Category B indicators conforms to the standards of statistical theory. One of two methods may be employed to select a sample: quota or probability sampling. The first method would here consist in selecting a sample of individuals from each country’s political class that reflected the political class as a whole—for example, the proportion of men and women, the percentage of people from given age groups, regions, socio-professional backgrounds, etc. The names of the individuals surveyed need not be known, for the important thing is to interview people from each group proportionate to their numbers in the total population of political actors. I would reject this method in its pure form because its major advantages (economy, suppleness of use) are outweighed by its main disadvantages (too few restrictions on the investigator, risk of error), which would be magnified in the African context. The second possibility is probability sampling, which statisticians view as more reliable because the margin of error may be measured and controlled. It consists in selecting respondents at random; that is, each member of the political class has an equal chance of being chosen. This technique is derived from the law of large numbers. It is especially useful in surveys where samples are large. When the sample is quite small (several dozen or a few hundred people), the quota method is generally to be preferred. I would opt for a combination of the two. The high cost of conducting a survey where the respondents are dispersed over a very large geographical area may be avoided as may the possibility of an unrepresentative sample. This is known as the stratified-sample method. Starting from the general typology of actors in the political class, one can control certain characteristics and then select respondents at random from each stratum—for example, the fraction of randomly selected women will correspond to the ratio of women active in public life; each age group will be proportionately represented, etc. Albeit fastidious, this method lowers the non-response rate and eliminates the need for statistical adjustments. The sample established, the field survey may be carried out and the results tallied. The arithmetic mean of the sum total of points given by all respondents to a given category B indicator constitutes its score. Categories A and B indicators are then constructed for the three mentioned conceptual categories: • The Design of Democracy (α): This is the analysis of the appropriateness of the new rules of the political game in the country whose democratization process is being evaluated—the principal international conventions the government has ratified, the constitutional framework, the laws and regulations on political pluralism, etc. A methodological assumption is adopted for this component: given that the primary rules of the political game are established once and for all (countries do not normally change

Measuring Democracy   447 Constitutions too frequently), it would be unadvisable to give this component of the Index the same importance indefinitely. To avoid this difficulty, the weight of this component shall be taken to diminish over time. To compute its relative importance at a given moment in time, I have employed the mathematical formula Atkinson has proposed for the calculation of diminishing returns. The total number of points given a country in the design-of-democracy category is weighted according to the length of time the democratization process has been in place5 before being integrated into the computation of the country’s overall index. Thus its point total is analyzed using a coefficient of the elasticity of the marginal political utility variable ε (from 0 to 1), according to the amount of time the process has been in existence. This may be written:

1− ε PU (α ) = 1 (1 − ε ) × α ( ) (4)

If ε = 0, ⇒ the utility has scarcely diminished. This means that the democratization process is still relatively new and fragile, and that the design of democracy component α is fully weighted in the calculation of the Index. If ε = 1, ⇒ this indicates that the democratization process in this country is fairly old and well-entrenched, and that less weight should be given to α in the computation of a country’s overall index. Mathematically, UP(α) is then equal to logα. Within the framework of this model, three possible values are assigned to the coefficient of the elasticity of political utility, according to the amount of time the democratization process has been in place (d*): If the democratization process is less than six years old (d* < 6 years), which is the case in African countries where constitutional changes typically occur towards the end of the five-year presidential term, the value of coefficient ε is 0, which means that there are no diminishing utility returns for α. If the democratization process is between six and twelve years old (6 < d* < 12), the value of coefficient ε is 0.5, which indicates that there has been a decrease in the utility of α. This is the case in a country such as Benin, where the debate between political actors does not so much concern constitutional reforms as it does respect for the constitution and laws in place. If political reforms are more than twelve years old (d* > 12), the value of coefficient ε is 0.75, which means there has been a marked decrease in the political utility of α. This applies to such countries as Botswana and Mauritius, the two countries considered the most democratic. A concrete example sheds light on the computation of coefficient α. Monga (1996a) uses data from 1994 surveys to calculate the Democracy coefficient for Cameroon at that time. He identifies eighteen indicators under the design-of-democracy component. Each indicator is rated on a 0-10 scale, at 10 points each, thus the highest possible score is 180. Cameroon received 73 points from survey results (α). Political reforms were launched in 1991 (d* < 6 5  The inception date of political reforms is defined as the date upon which the first opposition party recognizes the democratization process to have begun or, in the case of a few countries such as Uganda that do not have political parties, the date upon which the first elections are held that international observers deem to be truly open to candidates who are not affiliated with the sitting government. So-called “pluralist” elections organized in the framework of a single party such as were held in the Ivory Coast and Cameroon during the 1980s do not mark the onset of a “valid” democratization process.

448   Methodological Issues years when the analysis was carried out), therefore the coefficient ε value was 0). The weighted diminishing utility was thus: PU(α) = 1/(1–0) × 73(1–0) = 73

(5)

This score was converted into a component of the final index using a well-known dynamic aggregation method:



(Maximum Value - Obtained Value) / (Maximum Value – Minimum Value)    α = (180–73) / (180–0) = 0.594

(6)

The numerical value assigned to the two other components in the Index is calculated more simply because the problem of diminishing marginal utility is less critical. • Implementation of the Democratic Concept (β), that is, the actual application of the texts that have been legally adopted, as subjectively assessed through surveys. This includes an evaluation of the prevailing rules governing the national political market, an analysis of the functioning of the agreed-upon institutional framework, and a monitoring of the country’s day-to-day respect for human rights and good political governance. The component β value is computed using the above-mentioned aggregation method:

(

)(

)

β = Maximum Value − Obtained Value Maximum Value − Minimum Value p (7) • Democratic Maintenance (γ). Democracy is never won once and for all. Rather, it is an ongoing pursuit of ever-evolving imperatives, a dynamic process the validity of which is never permanently established. Its maintenance therefore involves a constant renewal of the determinant values of well-being and a smooth functioning of both established institutions and institutions that may later be created to ensure respect for the new set of values. Using the same aggregation method, the component γ value is also:

(

)(

)

(8) γ = Maximum Value − Obtained Value Maximum Value − Minimum Value Knowing the values of α, β, and γ, one can calculate the Index. In the light of the approach taken (a comparison of the actual score to the highest, hoped-for score), these components cannot but provide a measurement of how much farther a country has to go before it approaches the democratic ideal. The arithmetic mean of these three “sub-index-numbers” expresses a nation’s democratic deficit. The country’s or region’s democracy coefficient is calculated as the difference to the unity. In Monga (1996a), Cameroon’s was determined as follows: (9) (α + β + γ ) 3 = (0.594 + 0.581 + 0.7 ) 3 = 0.625 whence: The Democracy coefficient on 30 September 1994: 1 − 0.625 = 0.375. Despite its simplicity, the proposed coefficient allows for a more precise and legitimate evaluation of political progress. The synthesis of several dozen indicators, analyzed from a

Measuring Democracy   449 simultaneously universalist and relativist perspective, provides a more accurate picture of the political landscape, a better assessment of the balance of power and social dynamics. The usual criticisms concerning the (dubious) reliability of the statistical data used for such indices or the unrepresentative nature of this composite index, which, like any arithmetic mean, tends to flatten out the extremes (and hence possible variations in numerical values obtained in different regions or social groups) cannot be leveled against the Democracy coefficient—at least not to the same extent—especially in the light of the fact that the survey samples can be refined, multiplied, or disaggregated, enabling one to discern various levels of judgment and to calculate with ever greater precision sub-index-numbers.6

22.5  Conclusion: Beyond Objectivism and Subjectivism For several centuries now, thinking on the conditions of human happiness has been mired down in the dispute between objectivism—which postulates our capacity to conceptualize the happiness of others—and subjectivism, which assumes the converse. Africa’s diverse and complex recent sociopolitical history provides the ideal base for devising a democratic measurement tool that uses economic methodology and enriches the existing literature. This chapter has critically reviewed the philosophical foundations for the most popular current indicators and proposed the Democracy coefficient as a more satisfying alternative. It has argued that this new tool helps avoid the conceptual impasses to which classical democratic measurement instruments have led. To Locke, who thought that the ultimate goal of those who governed was to secure happiness for the people, Leibniz responded:  “a certain uneasiness in longing for the good together with a continuous and uninterrupted progress to the greatest goods is even better than to possess the good.” (1959: Chapter 21, section 36). “Without perpetual progress and novelty there is neither thought nor pleasure.” (1967: 101). He thus introduced the idea that a certain amount of malaise is necessary if humans are to feel, understand, and measure their well-being. Arguing along the same lines, Kahneman and Varey (1991) go even further, maintaining, for example, that the master-slave dialectic may lead to a reversal of the situation: in adapting himself to his condition, the slave may disrupt the values attributed to a subjectively defined utility and modify thus the equation for well-being—rendering all comparative theories on the matter inoperative. The transient nature of democracy quickly brings us back to this fundamental debate. Can we make judgments for others? Can we interpret and measure with a fair amount of accuracy the different mental states and hedonistic methods of satisfaction that lie below the surface 6  Perhaps it would be preferable to use the geometric mean rather than the arithmetic mean, given that the choice of the latter de facto implies that the three components of the Index may offset one another. If, on the other hand, the geometric mean is utilized, one zero value would negate the total index value. But the arithmetic mean has the marked advantage of simplicity. Its use should be abandoned only if a sensitivity analysis—how the classification of various countries is affected by changes in the Index’s underlying assumptions, notably the relative importance of each of its three components—reveals significant shifts in countries’ rankings.

450   Methodological Issues of the democratic idea? Comparative theories on well-being do not offer precise answers to these questions. The main advantage of the approach I have proposed is that it allows one to circumvent the obsession with objectivity without falling into the trap of unbridled subjectivity. My model valorizes individual subjectivities in a manner that does not affect the nature and value of the final product—freedom. The idea that democracy ought to contain a component reflecting personal preferences satisfies the requisite of individual autonomy put forth by certain researchers (Geuss 1979; Elster 1983), and allows for the elaboration of an index meeting the criteria of appropriateness and practical utility. Though they are essential to the organization, functioning, and stability of any political system, the subtleties contained in the notion of political utility are never taken into account in the literature on democracy. Classical democratic measurement instruments (the Freedom House Index, the Polity Scores, Humana Guide, etc.) are too quick to presume the uniformity of personal levels of utility and individual criteria for democratic satisfaction. They presuppose, for instance, that people are happy to live in countries that have been designated as democratic even if they are unaware that they have been granted the privilege of freedom. They also assume that the way the absence of freedom is experienced is necessarily identical in all countries classified as non-democratic according to their indicators (which, insofar as they are never tested against local “values,” are illegitimate). The model I have put forth does not of course solve all these conceptual difficulties; but by aggregating the subjectivities of a large number of people, it lowers the risk of error through diversification (as financial theories on portfolio management have taught us) and relativizes the inconsistencies (in terms of levels of political well-being) within and across individuals. By placing the notion of freedom in the “here and now,” the proposed Coefficient allows one to capture the actual experience of liberty not as prescribed by some outside convention but as actually conceived and lived by the people in various places around the world. This is of the essence if one agrees with West that even in so-called politically advanced nations, democracy is tied to ever-evolving philosophical choices and sociological identities. “Democratic practices are themselves deeply rooted in precisely the nuanced historical sense, the subtle social analysis, and the self-correction and self-critical process of never blocking the road to inquiry” (West 1993: 123). Including the important dimension of the eternal nowness of the democratic idea and experience, the Coefficient exhibits a notable “philosophical” advantage, especially if one accepts the principle of multiple truths and the idea of a universal conception of human rights, which the followers of a Nietzschean, deconstructive tradition (Mudimbe 1988) as well as the theorists of an African-specific humanism (Obenga 1990) have embraced. Finally, unlike other existing indices (notably the Freedom House Index and the Polity Scores), the Democracy coefficient allows for an infinite variety of numerical comparisons and hence for a more precise measurement of the amplitude of political and social movements (advance/decline). Whatever the strengths of the liberal democratic model, strengths a large part of humanity has come to recognize and embrace, a reconceptualization of democracy is urgently called for if one is to restore to the concept its dynamism and intrinsic pluralism. Research institutions, international organizations, chancelleries, monetary institutions, and investors cannot continue to “measure” the degree or speed of democratization by simply consulting arbitrary compilations based on news bulletins sent over wire services.

Measuring Democracy   451 The myth of a normative ethics of political welfare is as problematic as its symmetrical counterpart of unbridled relativism. The U.N. member-states parties to the Universal Declaration of Human Rights have demonstrated their commitment to a set of universal values, but this should not be interpreted as an endorsement of philosophical absolutism, especially in the light of the fact that the notion of political utility cannot be reduced to principles of international law. Because of its integrative and dynamic approach as well as its flexibility and authenticity, the Democracy coefficient suggests a way to address some of the crucial questions that have always preoccupied humanity (freedom, well-being, happiness) but that we have never been able to conceptualize in an “acceptable” fashion.

References Baynes, K., Bohman, J., and McCarthy, J. (1987). Introduction, in Baynes et al. (eds), After Philosophy: End or Transformation? Cambridge, MA: MIT Press. Bound, J., Jaeger, D.A., and Baker, R.M. (1995). Problems with instrumental variables estimation when the correlation between the instruments and the endogeneous explanatory variable is weak. Journal of the American Statistical Association, 90(430):443–450. Crozier, M., Huntington, S.P., and Watanaki J. (1975). The Crisis of Democracy. New York: New York University Press. Dahl, R.A. (1971). Polyarchy: Participation and Opposition. New Haven: Yale University Press. Davidson, D. (1984). Inquiries into Truth and Interpretation. Oxford: Oxford University Press. Dennett, D.C. (1993). Quining Qualia, in Alvin I. Goldman (ed.) Readings in Philosophy and Cognitive Science. Cambridge, MA: MIT Press, pp. 381–414. Eboussi Boulaga, F. (1991). A contretemps—L’enjeu de Dieu en Afrique. Paris: Karthala. Economist (2014). What’s Gone Wrong with Democracy? March 1, pp. 43–48. Elster, J. (1983), Sour Grapes: Studies in the Subversion of Rationality. Cambridge: Cambridge University Press. Elster, J., and Roemer, J.E. (1991). Interpersonal Comparisons of Well-Being. Cambridge: Cambridge University Press. Freedom House (2014). Freedom in the World 2014, New York, http://www.freedomhouse.org/ sites/default/files/Freedom%20in%20the%20World%202014%20Booklet.pdf. Fukuyama, F. (1992). The End of History and the Last Man. New York: Free Press. Green, P.E., and Wind, Y. (1975). New way to measure consumers’ judgments. Harvard Business Review, 53(4):107–117. Geuss, R. (1979). The Idea of Critical Theory. Cambridge: Cambridge University Press. Hammond, P. J. (1991). “Interpersonal Comparisons of Utility: Why and How They Are and Should Be Made”, in Jon Elster and John E. Roemer(eds), Interpersonal Comparisons of Well Being. Cambridge: Cambridge University Press, pp. 200–254. Huntington, S.P. (1991). Democracy’s third wave. Journal of Democracy, 2(2):12–34. Kahler, M. (1992). External influence, conditionality, and the politics of adjustment, in S. Haggard and R. Kaufman (eds), The Politics of Economic Adjustment. Princeton: Princeton University Press, pp. 89–136. Kahneman, D., and Varey, C. (1991). Notes on the psychology of utility, in J. Elster and J.E. Roemer (eds), Interpersonal Comparisons of Well-Being. Cambridge: Cambridge University Press, pp. 127–163.

452   Methodological Issues Leibniz, G.W. (1959). Nouveaux essais sur l’entendement humain—Neue Abhandlungen über den menschlichen Verstand, in W. von Engelhardt and H. H. Holz (eds), Philosophische Schriften, Vol III/1. Darmstadt: Wissenschaftliche Buchgesellschaft. Leibniz, G.W. (1967). Confessio Philosophi—Ein Dialog, von Otto Saame (ed.). Frankfurt/Main: Klostermann. Leightner, J.E., and Inoue, T. (2012). Solving the omitted variables problem of regression analysis using the relative vertical position of observations. Advances in Decision Sciences, vol. 2012, Article ID 728980, http://dx.doi.org/10.1155/2012/728980. Lin, J.Y., and Monga, C. (2012). Solving the mystery of African governance. New Political Economy, 17(5):659–666. Monga, C. (1996a) Measuring Democracy:  A  Comparative Theory of Political Well-Being. Boston University African Studies Center, Working Paper no. 206, 2 volumes. Monga, C. (1996b), The Anthropology of Anger: Civil Society and Democracy in Africa. Boulder, and London: Lynne Rienner Publishers. Monga, C. (2010). Civil society and sociopolitical change in Africa: A brief theoretical commentary, in P. Soyinka-Airewele and R. Kiki Edozie (eds), Reframing Contemporary Africa: Politics, Economics, and Culture in the Global Era. Washington, DC: CQ Press, pp. 144–156. Mudimbe, V.Y. (1988). The Invention of Africa: Gnosis, Philosophy, and the Order of Knowledge. Bloomington and London: Indiana University Press and James Currey. Obama, B. (2009). Remarks by President Barack Obama at Town Hall Meeting with Future Chinese Leaders, Museum of Science and Technology, Fudan University, Shanghai, China, November 16. Obenga, T. (1990). La philosophie africaine de la période pharaonique. Paris: L’Harmattan. Putnam, H. (1982). Why reason can’t be naturalized. Synthese, 52:1–23. Putnam, R. (1993). Making Democracy Work:  Civic Traditions in Modern Italy. Princeton: Princeton University Press. Ranney, A., and Kendall, W. (1969). Basic principles for a model of democracy, in C.F. Cnudde and D.E. Neubauer (eds), Empirical Democratic Theory. Chicago: Markham Publishing Company, 1969, pp. 41–63. Ricoeur, P. (1969). Le conflit des interprétations—Essais d’heuméneutique. Paris: Seuil. Robbins, L. (1938). Interpersonal comparisons of utility:  a comment. Economic Journal, 48:635–641. Sarkozy, N. (2014). Ce que je veux dire aux Français. Le Figaro, March 21. Taylor, C. (1987). Overcoming epistemology, in K. Baynes, J. Bohman, and J. McCarthy (eds), After Philosophy: End or Transformation? Cambridge, MA: MIT Press, pp. 459–488. West, C. (1993). Beyond Eurocentrism and Multiculturalism, vol. 2, Prophetic Reflections: Notes on Race and Power in America. Monroe, ME, Common Courage Press.

Chapter 23

Measureme nt and Analysi s of C om petiti v e ne s s Olumide Taiwo and Julius A. Agbor

23.1 Introduction The concept of international competitiveness has gained prominence in both policy and academic circles specifically in relation to countries’ external balance positions.1 Policymakers are increasingly evaluating their economies in relation to global markets in a bid to identify the fundamental drivers of competitiveness as well as the associated constraints. This is particularly important for African economies, most of which lag behind the rest of the world in measures of economic, financial, and human development. Although the competitiveness discourse has come of age since its origin in macroeconomic theory and policy, issues relating to its definition and measurement are far from settled. Scott (1985) defined national competitiveness as “a nation state’s ability to produce, distribute, and service goods in the international economy in competition with goods and services produced in other countries, and to do so in a way that earns a rising standard of living.” In Fagerberg’s (1988) view, competitiveness refers to the ability of a country to achieve the twin goals of raising the living standards of its citizens by way of sustained growth in income and employment, and doing so without running into balance of payment difficulties. The OECD Program on Technology and the Economy (1992) defined competitiveness as “the degree to which, under open market conditions, a country can produce goods and services that meet the test of foreign competition while simultaneously maintaining and expanding domestic real income” (237). These definitions emphasize strategic competitive advantage achieved

1 

Measures and determinants of international competitiveness are becoming hot topics in academic and policy circles. As Ramirez and Tsangarides (2007) posit, competitiveness analysis is about identifying the elements necessary to ensure sustainable growth and improvement in living standards.

454   Methodological Issues through high value addition and economies of scale rather than comparative advantage based on resource endowments.2 We take this distinction seriously. In terms of measurement, economists generally lean towards the macroeconomic framework where competitiveness is assessed by a host of price-cost measures.3 Among these measures, the unit labor cost (ULC) and the Real Effective Exchange Rate (REER) have been prominent. In this framework, improvements in a country’s ULC relative to the rest of the world are thought to increase its volume of international trade, while a competitive (low) REER is thought to attract foreign demand thus increasing a country’s share of world market. Hence, overvaluation of a country’s currency (a manifestation of REER misalignment) results in loss of competitiveness. The same happens when the ULC exceeds or grows faster than the rest of the world. Therefore, explanations for gain or loss of competitiveness have always been sought from the underlying causes of movements in the REER and the ULC. Recent developments in global trade, notably the growing importance of global value chains, are challenging existing competitiveness frameworks, and in particular, the validity of conventional measures of REER. The models underlying the REER assume that the final goods traded in international markets are wholly produced by domestic factors (and therefore ignore trade in intermediate goods), which explains why competitiveness has been measured using domestic consumer prices and gross trade data. The increasing prominence of trade in intermediate goods arising from globalization of value chains imply that competitiveness needs to be evaluated both on the basis of value addition and on prices reflecting the cost of intermediate inputs. There have been two major efforts to revise the macroeconomic framework in the light of these developments. Bems and Johnson (2012) proposed a Value-Added Real Effective Exchange Rate (VAREER) that uses value-added trade and prices of factors of production with a view to tailoring the assessment toward competitiveness in the segment of the value chain (denoted as “tasks”) that drives a country’s trade in international markets. Further, Bayoumi et al. (2013) proposed a Goods Real Effective Exchange Rate (GOREER) that is a product of two components. One component measures competitiveness in domestic value-added content of goods traded while the other measures competitiveness in foreign value added content. Saito et al. (2013) examined the two measures and concluded that they both exhibit trade-offs in different contexts. In spite of these efforts, the macroeconomic framework of competitiveness faces three principal limitations when applied to African economies. First, the framework is more suitable for economies exporting manufactured goods rather than for exporters of raw materials, owing to the fact that prices of raw materials are determined in international commodity markets and therefore not significantly influenced by either REER or ULCs of the countries of origin. In other words, export demand for primary products is neither sensitive to the producer’s exchange rate nor to its domestic cost of production,4 but rather depends on international 2

  This also marks a distinction between the emphasis on increasing returns to scale in the new trade theory and comparative advantage underscored in the conventional trade theory. 3   Durand and Giorno (1987) provide an excellent overview of the different types of indices proposed and applied in the literature. 4  In several instances across Africa, domestic production costs seem to rise as a result of growing insecurity and social unrest. Examples include the crisis in the oil-rich Niger-Delta region of Nigeria, the Islamist hostage crisis that is affecting natural gas production in Algeria and the terror attacks on uranium mines in Niger, just to name these few. While these incidents do not necessarily affect export prices of the particular commodities, they nonetheless undermine competitiveness by raising domestic cost of production.

Measuring Democracy   455 market prices. Given that most African countries export mainly raw materials and are mostly price-takers in the markets, this framework is of little relevance. Second, improvements in non-price factors—which may raise the level of productivity in the economy and thus its overall competitiveness—may not lead to increases in the volume of international trade but might instead show up in improvements in the terms of trade. This may be more important for developing countries with large non-traded sectors and may cause the REER to miss important gains in competitiveness.5 Incidentally, most African economies are dominated by services sectors that are largely informal and non-traded. Third, movements in the REER of small open economies hardly reflect the state of the countries’ competitiveness owing to the preponderant influence of external shocks (favorable and unfavorable) arising from international goods and capital markets. To the extent that adjustments in these economies are sluggish, changes in competitiveness indices might well reflect those exogenous factors rather than actual changes in domestic conditions of production and value addition. This chapter proposes an alternative to the macroeconomic framework and evaluates the merits of an alternative index in assessing the competitiveness of African economies. In particular, it builds on the micro foundations of the business strategy literature to espouse the merits of the Trade Weighted Value added per capita (TWV) over the REER in the context of African economies. The remainder of the chapter is structured as follows. Section 2 discusses the conceptual framework, compares the macroeconomic and the business strategy perspectives, and sets the stage for the TWV. Section 3 compares the performance of REER and TWV in the African context and discusses the differences while Section 4 concludes.

23.2  Conceptual Framework Policymakers, scholars, and analysts consider competitiveness as an important goal. Be that as it may, there are many views about the appropriate approach to analyzing the concept. The three major perspectives identified in the literature are the macroeconomic, international competition, and business strategy perspectives. However, this contribution focuses on the macroeconomic and business strategy frameworks from where the REER and the TWV derive respectively. A discussion of the two frameworks follows below.

23.2.1  The macroeconomic framework The macroeconomic perspective originates from macroeconomic theory and policy and is influenced by the framework outlined in Corden (1994) and Boltho (1996). In the framework, competitiveness entails maintaining internal and external balance in the short-run (Wignaraja 2005). Internal balance is usually defined in terms of full employment (the lowest possible rate of unemployment that is consistent with an acceptable rate of inflation) while external balance is defined in terms of current account equilibrium (or some desirable 5 

This point is emphasized by Durand and Giorno (1987: 149).

456   Methodological Issues level of the current account). In this context, international competitiveness is defined as the level of the real exchange rate that, in combination with the requisite domestic economic policies, achieves internal and external balance (Boltho 1996). Thus, competitiveness policy is synonymous with exchange rate policy and competitiveness is assessed through the real exchange rate. This approach emphasizes the exchange rate as the strategic variable and hinges on the link between the real exchange rate, balance of payments, resource allocation across sectors and competitiveness. For example, large current account deficits are related to exchange rate appreciations which in turn hamper the development of tradables including manufactured exports (Wignaraja 2005). Economic theory (in particular trade theory) defines the real exchange rate as the ratio of domestic prices of non-tradables to tradables, e = pn / pt . An increase in the ratio denotes an appreciation of the exchange rate while a decrease denotes depreciation.6 However, this definition of the real exchange rate faces two empirical challenges. First, because the measure uses domestic prices, it lumps exports and imports into the same category as tradables. Boltho (1996) argued that the measure is only appropriate for small open economies where the terms of trade are set by the world market. Second, regular data on tradable and non-tradable prices are hardly available (Wignaraja 2005; Boltho 1996). These challenges have led scholars to rely on proxies for the real exchange rate. The first set of proxies are indicators of relative consumer prices such as the Consumer Price Index (CPI) and other indices of cost of living. While data is readily available in most countries, these measures suffer drawbacks such as inclusion of a range of goods and services that are not subject to international competition and variation of components and weights across countries. Relative indicators based on GDP deflators are sometimes used as remedies but these are also beset by similar limitations. The second set of proxies are those measuring relative producer prices of traded manufactured goods and are typically collected from declarations at the customs. Although these measures have some merits in the sense that the data is easy to collect and they relate to actual trade, they also suffer many setbacks. First, by focusing on actual trade, they ignore potential trade and therefore fail to cover all tradable goods and sectors. Such exclusion may be problematic by not taking into account possible loss of competitiveness of excluded goods as they become too highly priced to be traded. Second, there are variations in the quality of the measures across countries as well as lack of homogeneity in weighting and coverage. These shortcomings make international comparisons less meaningful. Third, changes in competitiveness tend to be heavily influenced by changes in prices of intermediate goods. Fourth, by focusing on relative price changes, these indices are only meaningful in markets with differentiated products. In perfectly competitive settings where prices are given, competitiveness manifests in terms of profits rather than prices (Boltho 1996). The third set of proxies are those measuring relative costs. The most commonly used is the index of Unit Labor Cost (ULC) in the manufacturing sector, defined as labor cost per 6  This is one of the ways the real exchange rate has been defined. Another interesting way of defining the real exchange rate (credited to Rudiger Dornbusch) is the ratio of domestic wages to the nominal exchange rate (measured as price of foreign currency in units of local currency). In this setting, the real exchange rate can be defined as the domestic wage in the foreign currency, typically the US dollar. If wages are high, then domestic producers of tradables will not be able to compete against imports and therefore will export less. Overvaluation then leads to high (cheap) imports and low (costly) exports.

Measuring Democracy   457 unit of manufactured output. The advantage of this measure is that improvements in competitiveness, through increase in labor productivity, fall in wages or nominal exchange rate depreciation, are associated with “either declines in tradable prices or with increases in profitability, or with a mixture of the two, depending on what strategies firms follow and on the nature of the markets in which they compete” (Boltho 1996: 3). The challenge with the ULC is that it is an aggregate concept that focuses on labor cost and productivity while ignoring other costs of production. This selective focus on labor is justified on two grounds. First, it is assumed that all non-price factors and other costs relevant to competitiveness are embodied in the production function and are therefore captured in labor productivity.7 However, Monga (2013) argued that if things were that simple and levels of transaction costs induced by non-price factors are unimportant, then manufacturing firms would have been moving from China, Brazil and other emerging economies where unit labor costs are already rising into low-wage labor intensive African countries such as the Democratic Republic of Congo, Ethiopia and Tanzania. Second, it is argued that labor costs are more important determinants of competitiveness than the cost of capital and other inputs that are assumed to be equalized across countries through international trade. For example, Boltho (1996: 3) argued that “cost of capital and other raw materials will be more similar across countries due to capital mobility and the existence of international commodity markets.” Meanwhile, while the assumption of capital mobility may hold among the group of industrialized countries, extending it to the global economy may be implausible given higher cost of capital in less developed countries than industrialized countries. The idea that increases in a country’s relative ULC leads to loss of competitiveness could be weaker than portrayed for many reasons. For example, differential changes in non-labor costs will affect competitiveness but might not be reflected in ULC. Also, higher capital-labor ratio, which entails higher capital costs and lower labor costs, could lead to relative ULCs that overstate competitiveness.8 Fagerberg (1988) noted that countries which achieved the fastest growth in terms of exports and GDP in the early post-war period were those that also experienced much faster growth in relative ULC. This phenomenon, referred to as “Kaldor Paradox” in reference to Kaldor (1978), implied that the focus on relative unit labor costs as an important determinant of competitiveness is rather too simplified, and could be sometimes misleading. In practice, most analysis use the REER obtained from the purchasing power parity (PPP) framework. The REER is obtained by deflating a trade-weighted average of nominal exchange rates between a country and its trading partners. The deflators being used in practice include relative consumer prices, relative producer prices, relative GDP deflators and unit labor costs in manufacturing. Researchers have noted the shortcomings of the REER, particularly in relation to producers of primary commodities. Cashin et al. (2002) pointed out that the usual behavior of REER in many developed and developing countries may not apply in the case of commodity-exporting countries. Harberger (2004: 4–5) argued that the common conceptual measure of real exchange rate as the ratio of the price of non-traded to traded goods runs into trouble when large inflow of earnings due to a rise in commodity

7  These include other costs of doing business notably those related to the state of infrastructure, institutions, access to capital, quality of human capital and governance institutions. 8  This insight is credited to Scott Rogers.

458   Methodological Issues prices induces an appreciation of the real exchange rate as a result of excess supply of foreign currency, a phenomenon commonly referred to as the Dutch disease. In a related literature, Fagerberg (1988, 1996) and Dosi et al. (1990) concluded that a competitive real exchange rate alone does not deliver international competitiveness if backward institutions, deficient technology, inefficient business environment, poor infrastructure and low human capital characterize an economy. Wignaraja (2005) argued further that these factors could be more important for competitiveness of developing countries, especially those in Africa. Thus, the utility of relative price-cost measures in assessing competitiveness in the presence of substantial structural and capacity constrains is questionable.

23.2.2  The business strategy framework This approach originates from the business studies literature and was pioneered by Porter (1990), who, in his study of eight developed and two newly industrializing countries, attempted to explain why some countries are more successful in particular industries than others (Moon et al. 1998; Smith 2010). In contrast to the macroeconomic approach, this approach considers a nation as an aggregation of industries and applies micro-level business strategy concepts in studying international competitiveness. In effect, the approach implies that continuous upgrade is the key to sustaining a competitive edge and competitive advantage of a country is the result of firm-level innovations and success in gaining large shares of world markets. Further development in the microeconomic literature on learning and innovation has given rise to extension of the initial framework to emphasize creation and adoption of technology as drivers of competitive advantage. The basic underlying model, referred to as the “Diamond Model,” classifies economies into four stages that are reminiscent of the Rostow stages of development. These are the factor-driven, investment-driven, innovation-driven and wealth driven stages.9 The model considers competitiveness as the outcome of interactions among the following four critical attributes (diamonds) of a nation: 1. Factor condition: this relates to the country’s position in factors of production such as skilled labor and infrastructure necessary to compete in industries. 2. Demand condition: this captures the nature of home-market demand for products or services. 3. Related and supporting industries condition: this refers to the presence of supplier industries and other related industries that are internationally competitive. 4. Firm strategy, structure and rivalry: this captures the conditions governing how firms are created, organized and managed as well as the intensity of domestic competition. In terms of usefulness in economic analysis, a major criticism of the diamond model has been that it fails to specify how we measure competitiveness—whether it is total factor productivity or something entirely different. Economists have attempted to distill the framework in order to proffer relevant measures of competitiveness. Gray (1991) suggested that Porter’s 9  Wignaraja (2005) notes that the diamond model was influential in the development of the Global Competitiveness Indicator (GCI) published regularly by the World Economic Forum, and that Professor Porter served as advisor in the process.

Measuring Democracy   459 definition of national competitiveness comes down to the rate of growth of GDP. Reinert (1995) disagreed with Gray and contended that the definition is hardly operational. In his view, competitiveness is divorced from issues of productivity and efficiency, and high productivity levels do not necessarily lead to competitiveness. He contended that “[a]‌lthough it is difficult to be competitive if you are not efficient and have a high productivity, it is by no means obvious that being the most efficient producer of an internationally traded product makes a country competitive—i.e. enables it to raise the standard of living” (26). The model has also been criticized in relation to its usefulness in analyzing trade patterns. Krugman (1980) objected to the idea that countries compete in international markets like corporations. He contended that international trade is not a zero sum game but one in which specialization and trade according to comparative advantages yield welfare gains to all nations. Waverman (1995) also described the model as too general in that it tries to explain every aspect of international trade and competition but eventually describes nothing. However, Grant (1991) contended that the model does better in understanding the patterns of trade and investment in the new world economy than the theories of trade and investment. On the international business side, most of the criticisms have focused on what is missing in the model. Critics point out that the model ignores the attributes of a country’s largest trading partners and is flawed if applied to small trading economies (Rugman 1991, 1992). Others point to omission of the role of multinational corporations (Dunning 1993). Following these criticisms, the model has been extended to account for the role of external diamonds, resulting in the double diamond model (Rugman and D’Cruz 1993; Rugman and Verbeke 1993), the generalized double diamond model (Moon et al 1998) as well as multiple diamond models (Bellak and Weiss 1993; Cartwright 1993). Porter’s approach has since been improved by incorporating the critiques from economists and business strategists. Indeed, the enriched model now serves as the underlying framework for the most widely referenced measure of competitiveness:  the Global Competitiveness Index (GCI) prepared by the World Economic Forum, as well as other competitiveness rankings produced by other institutions.10 The GCI defines competitiveness as: “the set of institutions, policies and factors that determine the level of productivity of a country,” and analyzes competitiveness based on both “microeconomic and macroeconomic foundations of national competitiveness.” The index is built on twelve pillars grouped into three categories that reflect the key drivers of competitiveness in economies at different stages of development:  the factor-driven, efficiency-driven and innovation-driven stages. The GCI is sensitive to these differences by varying the weights assigned to the sub-indexes in the computation of national competitiveness along the stages of development. The World Bank Doing Business Index (DBI) also examines some of the components of the GCI framework. Although the DBI ranking of countries is not in lockstep with the GCI’s, a correlation coefficient of 0.83 was established between the 10  Other competitiveness ratings include World Competitiveness Yearbook (WCY), Irish National Competitiveness Council (NCC), Doing Business Index (DBI) and Africa Competitiveness Report (ACR). The WCY framework identifies four main aspects of competitiveness: economic performance, government efficiency, business efficiency and infrastructure, and produces a ranking of countries along those lines. The Irish NCC distinguishes between inputs to national competitiveness (over which policy-makers have considerable leverage) and so-called “essential conditions” that must be present.

460   Methodological Issues rankings in GCI 2012/13 and DBI 2013.11 The African Competitiveness Report (ACR) complements these efforts by underscoring the continent’s competitiveness challenges while highlighting “areas requiring policy action and investment to ensure Africa lay the foundation for inclusive and sustained growth.” The generalized double diamond model (Moon et al. 1998) provides a simple analytical framework that incorporates the basic features of the competitiveness index, and is consistent with recent developments in international trade, and in particular, with globalized supply chains. The model thus offers the basis for an analytic measure of competitiveness in value addition in the following ways. First, production sharing through global value chains is implied in the model through the role of trading partners’ diamonds. In this setting, the final goods traded by a country in the international market are not produced exclusively by means of domestic factors but could also have trading partners’ factors embedded in them. Therefore, as Bayoumi et al. (2013) demonstrates, a country could experience loss of competitiveness in domestic factor costs (due to weakness of domestic diamond) but may not lose competitiveness in the pricing of traded goods due to the moderating influence of trading partners’ factors (due to strength of foreign diamonds). A particular implication of the foregoing is that factor prices do not have to equalize across countries for trade to take place, thus violating the neo-classical law of factor-price equalization and perfect competition. Reinert (1995) concluded that it is this violation, consistent with imperfect competition and economies of scale in international trade that underlies international competitiveness. Second, the business strategy framework, by design, emphasizes factors that are pertinent to the creation of high value sectors and industries through economies of scale. Under imperfect competition, the entire benefits of reduced costs associated with economies of scale are not passed onto consumers in the form of lower prices through international trade as would be the case under perfect competition. Rather some of the benefits (taken as “rents”) are kept within the producer countries and distributed in the form of profits, wages and, ultimately, government income through taxation, all of which are important determinants of living standards. Therefore, a country becomes competitive by reallocating resources to the “high-value sectors or industries” that in effect lead to rising national living standards while simultaneously producing goods that meet the test of international markets. That is, a country becomes competitive as it strategically accumulates higher rents (or value) compared to the rest of the world (importantly its trading partners), and does so by reallocating more of its human and physical factors into value-creating activities. Third, a reasonable measure of competitiveness in the framework must rely on the rate of accumulation of distributable rents per unit of tradable output, whether the output is intermediate or final good. From first principles, the simplest measure is of the form of C = (A–B)/Y, where C is rent-per-unit of output, A is current-year rent distributed in the form of wages and profits in those sectors, B is what wages and profits would be in the absence of current year rents (proxied perhaps by previous year’s wages and profits) and Y is tradable output. However, by focusing on the tradable sectors alone, this measure ignores the Balassa–Samuelson effect, that is, diffusion of gross rent into non-tradable sectors of the 11  This is taken from a presentation by Scott Rogers, at the Centre for the Study of the Economies of Africa on June 26, 2013.

Measuring Democracy   461 economy that may induce increases in wages, prices and profits there. In addition, other factors of production, including capital, may capture part of the rent through deliberate pricing mechanisms. For example, financial intermediaries in the services sector will capture part of gross rents through service fees and charges to tradable sectors. In the final analysis, a robust index of competitiveness would cover both tradable and non-tradable sectors, in order to capture gross distributable rents. In this framework, competitive economies are those able to accumulate aggregate rents faster than others. This view is consistent with Reinert’s (1995) suggestion that rapid changes in the level of productivity tend to raise competitiveness. The merits of the value-addition approach to assessing competitiveness of African countries are as follows. First, it does not require that countries produce final manufactured goods since the measure applies well to economies that produce intermediate goods. Consequently, the measure of competitiveness derived from this framework is applicable to economies at different stages of development irrespective of whether they are in steady state or not. Second, the approach emphasizes strategic competitive advantage that countries deliberately pursue through policies rather than comparative advantage in resource endowments. For instance, countries that simply dig raw materials and ship them overseas without capturing portions of global value chains would not make gains in competitiveness under this framework. Third, data on aggregate rents (value-added) for all countries are readily available in the National Income and Product Accounts (NIPA). These particular merits render the framework suitable for analyzing competitiveness of African economies.

23.3  Measuring Competitiveness We propose an alternative measure of competitiveness: the Trade-Weighted Value-added per capita (TWV) of a country relative to its main trading partners.12 In effect, by comparing accumulation of value-added induced by rents in a country relative to its main trading partners, and in essence, comparing rate of value addition at home and abroad, the proposed measure is consistent with the interpretation of competitiveness under the new trade theory. Notably, variants of TWV have been used in various forms to reflect phenomena that are tangential to the objective pursued here. The denominator, Trade-Weighted Real Value-added of a country’s trading partners (see Dos Santos et al. 2003) is a well-established determinant of price and demand for a country’s exports in the international trade literature (Cronovich and Gazel 1998; Vieira and Haddad 2011).13 Movements in the ratio of domestic value-added to foreign value-added is a plausible indicator of the extent to which a country is making technological progress and gaining larger shares of global value chains relative to the rest of the world in the double-diamond framework. In line with this interpretation, Abdih and Tsangarides (2006) referred the measure as “productivity index” and used it as proxy for technological progress in their analysis of equilibrium real effective exchange rate. 12  The measure is in real terms and is obtained using statistically determined weights of trading partners. 13  In addition, Adolfson et al. (2007) demonstrate its usefulness in Central bank forecasting models.

462   Methodological Issues Wagner and Zeckhauser (2006) used a similar measure to demonstrate the differential rates of competitive progress among 157 countries around the world over the period 1960-2000. A  graph of trade-weighted relative GDP per-capita against relative GDP per-capita14 shows Singapore, South Korea and Ireland as three countries that achieved substantial progress during the period covered, with Singapore being the most successful in moving from the group of least competitive countries to the group of highly competitive countries during the period. Coincidentally, Singapore and South Korea were the two Newly Industrializing Countries (NICs) included in the study by Porter (1990) that led to the development of the diamond model. In particular, the outcome which shows that Singapore was more successful than South Korea is consistent with the finding by Moon et al. (1998) in the context of the generalized double diamond model.

23.3.1  TWV for African countries 1980–2011 Figure 23.1 presents a graph showing the evolution of TWV for a sample of African countries and the position of their GDP per capita relative to the Sub-Saharan African (SSA) average for three ten-year periods covering 1981–2010. The empirical range of values on the vertical axis is reasonable given that all the economies are smaller than their trading partners. On the horizontal axis, a line drawn at the value of 1 indicates the position of the average SSA economy during the respective period. A movement northward (along the vertical axis) reflects an increasing rate of value addition per capita relative to the country’s main trading partners (driven by gains in global value chains) while an eastward movement (along the horizontal axis) reflects expansion of the economy relative to SSA average. An economy that is simultaneously gaining increasing share of global value chains and expanding faster than the average SSA rates would progress in the north-east direction. Economies that basically extract and sell primary commodities in response to world demand would remain roughly in the same spot for the entire period. Because many SSA countries are doing similar things, each country can only expand at the average SSA rates. As the figure shows, this is the experience of a vast majority of the countries that remained in the bottom-left corner of the graph through the entire period. The trajectories of two countries, South Africa and Gabon, provide important lessons. Although not located in the bottom-left quadrant, South Africa remained roughly at the same spot through the entire period. The country started out as roughly six times the average SSA country in terms of per-capita GDP but its growth has been driven by the extractive industries since the 1990s. Because its growth driver is similar to many African countries, the rate of expansion of its economy (on per-capita basis) is not faster than the average SSA. In addition, its TWV stayed roughly in the same spot.15 On its part, Gabon started out in the early 1980s with per-capita GDP that was about nine times the average SSA and a TWV of roughly 33 percent. By the 2000s, these positions have fallen to seven times and about

14 

The former is relative to trading partners while the latter is relative to the world. Indeed, South Africa’s exports transited from a fairly diversified structure in the nineties to a mineral and resource dominated structure in the mid and late 2000s in ways that have been demonstrated to be consistent with China’s demand for resources (Onyekwena and Taiwo 2013). 15 

Trade-Weighted Value-added per capita 0 .05 .1 .15 .2 .25 .3 .35 .4 .45 .5 .55 .6

TWV and relative GDP per capita 1981-1990

Gabon Seychelles South Africa Algeria MoroccoTunisiaMauritius Egypt, Arab Rep. Cape Verde Congo, Rep. Cameroon Cote d'Ivoire Senegal Zambia Equatorial Guinea Zimbabwe Angola Kenya Liberia Gambia, The Sudan Benin Mauritania Tanzania Guinea Togo Mali Comoros Central Nigeria African Republic Rwanda Madagascar Sierra Leone Guinea-Bissau Burkina Faso Chad Congo, Dem. Rep. Uganda Mozambique Burundi Ghana Niger Malawi Ethiopia

0

1

2

3

4

5

6 7 8 9 10 GDP per capita / SSA Average

11

12

13

14

15

Trade-Weighted Value-added per capita 0 .05 .1 .15 .2 .25 .3 .35 .4 .45 .5 .55 .6

TWV and relative GDP per capita 1991-2000

LibyaSeychelles

Djibouti

South Africa Mauritius

Namibia

Swaziland Benin Tunisia Egypt, Arab Rep. Algeria Morocco Cape Equatorial Verde Guinea Congo, Rep. Senegal Zimbabwe Togo Mali Cameroon Cote d'Ivoire Kenya Sudan Zambia Gambia, The Tanzania Guinea-Bissau Angola Guinea Mauritania Nigeria Mozambique Uganda Lesotho Burkina Faso Chad Rwanda Comoros Ghana Madagascar Central African Eritrea Malawi Niger Sierra Leone Liberia Ethiopia Congo, Burundi Dem. Rep.Republic

0

1

2

3

4

Gabon

Botswana

5

6 7 8 9 10 GDP per capita / SSA Average

11

12

13

14

15

Trade-Weighted Value-added per capita 0 .05 .1 .15 .2 .25 .3 .35 .4 .45 .5 .55 .6

TWV and relative GDP per capita 2001-2010

Libya Equatorial Guinea Seychelles Benin

Djibouti

Namibia Swaziland Egypt, Arab Rep.Tunisia Cape Verde Morocco Algeria

South Africa Mauritius Gabon Botswana

Togo Guinea Congo, Sudan Rwanda Senegal Gambia, The Rep. Cameroon Angola Tanzania Mali Mauritania Zimbabwe Cote d'Ivoire Guinea-Bissau Uganda Kenya Zambia Mozambique Nigeria Ghana Eritrea Comoros Lesotho Central African Liberia Burkina Faso Ethiopia Madagascar Chad Malawi Niger Sierra Leone Burundi Congo, Dem. Rep.Republic

0

1

2

3

4

5

6 7 8 9 10 GDP per capita / SSA Average

11

12

13

14

15

Figure  23.1   TWV and relative GDP per capita 1981–1990. TWV and relative GDP per capita 1991–2000. TWV and relative GDP per capita 2001–2010. Source:  Olumide Taiwo and Julius A.  Agbor.

464   Methodological Issues 18 percent respectively. Gabon thus presents the case of a country that went backward. While this loss of advantage might reflect the growing diversification of trading partners16 it is not a sufficient explanation for the decline. A group of countries led by Benin, which includes Djibouti and Swaziland to lesser degrees, exhibited a particularly different type of progress. In the case of Benin, there was a gradual upward movement on the vertical axis beginning in the 1990s although there was no movement on the horizontal axis through the entire period. In essence, the country seems to be doing well at capturing larger components of value chains but only expanding at the average SSA rate.17 A few countries emerged during the period with impressive north-east trajectories, both gaining in terms of value chains as well as expanding faster than the SSA average rate. The most notable success in this sense is Equatorial Guinea which moved from just about the average SSA economy and TWV of 4 percent in 1986 to nearly 14 times and 58 percent respectively in 2011. Other notable successes are Libya and Mauritius to a large extent, and Botswana and Cape Verde to lesser degrees. While Equatorial Guinea and Libya have not been included in the GCI, Mauritius and Botswana which progressed north-eastward, and Seychelles which remained in the north-east through the period, have received high rankings.

23.3.2  TWV versus REER In principle, export price competitiveness (low REER) leads to increased shares in international markets and induces re-allocation of domestic resources from non-traded to traded sectors. The movement of (human and physical) resources into the value-addition sectors enables the country to achieve two things: (i) accumulate value at a higher rate than the rest of the world and, (ii) do so in a more inclusive manner. These ultimately lead to increase in domestic value-added per capita relative to trading partners, leading to competitiveness in value addition (high TWV). Therefore, for consistency, the TWV and the REER should be negatively correlated. Table 23.1 presents correlations between the TWV and REER for a sample of African countries. Because the REER series are only available for 1998–201218 and the TWV is available for 1980–2010, the table presents correlations for the overlapping 13-year period 1998–2010. The indices are substantially positively correlated in the top 13 countries (marked in red) and negatively correlated in the bottom 19 countries (marked in blue). Interestingly, out of the nine countries marked in red, five of them are major oil exporters (Equatorial Guinea, Sudan, Nigeria, Cameroon, and Gabon) demonstrating that the REER 16  In the earliest period, 1980-1985, France, USA, Spain, UK, and Korea were the top five trading partners with trading weights in descending order. During 2006–2011, the list changed to USA, China, France, Trinidad and Tobago, and Thailand. 17  There is anecdotal evidence to support this implication. The country is a notable “passage” especially for importation of goods prohibited at Nigerian ports. In addition, its agro processing industry is prominently involved in packaging of finished fruit juices, a task that is located near the end of the fruit processing value chain. 18  The data is obtained from UNCTAD.

Measuring Democracy   465 Table 23.1  TWV and REER 1998–2010. Country 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

Equatorial Guinea Sudan Mali Nigeria Guinea Bissau Ethiopia Cameroon Central Africa Republic Gabon Botswana Comoros Mauritius Senegal Mozambique Madagascar Congo republic Cape Verde Mauritania Niger Morocco Ghana Burundi South Africa Swaziland Congo Democratic Rep Cote d’Ivoire Kenya Rwanda Chad Malawi Burkina Faso Gambia Zambia Lesotho Uganda Tunisia Djibouti Tanzania Sierra Leone

Corr Coef. 0.89 0.83 0.76 0.72 0.69 0.64 0.54 0.39 0.27 0.26 0.26 0.25 0.21 0.07 0.05 0.03 –0.01 –0.14 –0.16 –0.17 –0.22 –0.25 –0.26 –0.26 –0.44 –0.49 –0.54 –0.60 –0.62 –0.67 –0.70 –0.72 –0.75 –0.78 –0.78 –0.88 –0.90 –0.91 –0.92

is vulnerable to the Dutch disease. Inclusion of Botswana and Mauritius in this group also implies that the REER is moving in non-competitive direction in economies that have made substantial gains in competitiveness during the period. This is arguably a reflection of the Balassa–Samuelson effect. Similarly, it is not certain whether the REER reflects technological change, realization of economies of scale or gains in market shares in some of the cases where it is negatively

466   Methodological Issues correlated with the TWV. Its movements in some cases could simply reflect changes in foreign monetary policies that are not remotely connected with technological progress and value addition in the domestic economy. For example, Ramirez and Tsangarides (2007) show that during 2001–2006, the REER appreciated at the same time as the economies of the CFA franc zone19 lost competitiveness in value addition. However, the appreciation of REER had nothing to do with technological downturn or any fundamentals of the economies. Instead, it was the consequence of strengthening of the Euro, the currency to which the CFA franc is pegged. In summary, the TWV is potentially more relevant than the REER for measuring competitiveness in the African context for the following additional reasons. First, as noted, the TWV is a good proxy for technological progress in an economy (Abdih and Tsangarides 2006). It is well known that exporters of raw materials capture very trivial portions of product value chains. Banga (2013) shows that only 8 percent of total value added in global value chains accrue to less-developed and developing countries whose exports are typically dominated by raw materials. In the absence of improvement in domestic technology to enable primary commodity exporters capture increasing portions of value chains while their trading partners continue to advance technologically, the process of normalizing domestic value addition rates by the weighted average of trading partners’ would lead to a loss of competitiveness. Second, the TWV captures changes in both price factors and non-price factors such as structural and capacity constraints that affect competitiveness. Movements in the measure therefore reflect changes in labor productivity, capacity, other costs of doing business related to the state of infrastructure, governance, rent-seeking behavior and other non-price factors in a country relative to its trading partners. Third, a good measure of competitiveness must be relevant to countries with different economic structures at different stages of development and enable international comparison despite these differences. International comparability of this measure has been demonstrated by Wagner and Zeckhauser (2006).

23.4 Conclusion After a period of slow growth from the late 1970s into the early 1980s, African countries began implementing sound macroeconomic policies during mid-1980s to early 1990s with the view that a stable macro-economy would create conditions for a competitive economy. Despite the stable macroeconomic environment achieved through these reforms, the expected investment and growth outcomes did not manifest. Indeed African policymakers refer to the 1990s as the “lost decade.” On the other hand, the next decade ushered in impressive macroeconomic performance evident in low inflation rates, improved current account positions and relatively high rates of economic growth, but these were accompanied by lack of productivity growth in most sectors and rising rates of unemployment and poverty. These two regimes present a paradox that brings into question the role of macroeconomic

19  The countries are Cameroon, Gabon, the Central African Republic, the Republic of Congo, Equatorial, Guinea and Chad, Benin, Burkina Faso, Côte d’Ivoire, Senegal, Togo, Mali, Niger, and Guinea-Bissau.

Measuring Democracy   467 performance in Africa’s competitiveness as well as the relevance of the macroeconomic approach to analyzing competitiveness. It is evident that this paradigm is yet to address the fundamental determinants of competitiveness on the continent. The prevailing paradigm of price–cost competitiveness rests on, among others, the fundamental assumption of harmony between external and internal balance. The idea being that once the external is achieved, the internal would follow. Thus, public policy has a very narrow but direct role in competitiveness: government simply intervenes in the currency exchange market when necessary. Policymakers in African countries have generally followed this blueprint under the surveillance of the International Monetary Fund. The quality of skills, education and health of the population, infrastructure, internal mobility, strategic urbanization and most other conditions necessary for achieving internal balance are under-emphasized. Consequently, programs which focus on real domestic issues are half-heartedly pursued only to the extent that they are believed to be good for social development. For instance, education expansion programs in many African countries are being encouraged merely as a strategy of enhancing primary or secondary enrolments, rather than as strategies for developing competitive workforce. It is true that African economies are growing rapidly and increasingly undergoing structural transformation, but its pattern of transformation is strikingly different from that observed in the advanced economies and successfully replicated by emerging economies of Asia and Latin America. Despite enormous potential for labor-intensive industrialization, the African continent is in fact less industrialized currently than it was in the 1980s. At the moment, only a few African countries are capable of generating up to 10 percent of GDP from manufacturing. Rather than moving from the farm into the tradable manufacturing sector, workers are instead migrating from rural sector into urban non-tradable and largely informal commerce and distribution sector. Given that African economies are substantially lacking in manufacturing output, and much less exports, the relevance of a price-cost framework in analyzing their competitiveness is questionable. The fact that many African countries are increasingly being ranked lower in global competitiveness rankings; in spite of their remarkable growth performance should provoke policy makers to ponder on where they might be missing the mark. We hope that the business strategy competitiveness framework proposed in this chapter would inspire policy makers on the continent to be more adventurous in pursuing such broader policies as population health, skills and infrastructure development, industrial clusters and other relevant industrial policies.

References Abdih, Y., and Tsangarides, C. (2006). FEER for the CFA Franc. IMF Working Paper No. 06/236, Washington, DC: IMF. Adolfson, M., Andersson, M.K., Linde, J. et al. (2007). Modern forecasting models in action: improving macroeconomic analyses at central banks. International Journal of Central Banking, 3(4):111–144. Banga, R. (2013). Measuring Value in Global Value Chains, Background Paper for Regional Value Chains, Unit of Economic Cooperation and Integration amongst Developing Countries (ECIDC). New York: UNCTAD.

468   Methodological Issues Bayoumi, T., Saito, M., and Turunen, J. (2013). Measuring Competitiveness: Trade in Goods or in Tasks? IMF Working Paper 13/100, Washington DC. Bellak, C.J., and Weiss, A. (1993). A note on the Austrian diamond. Management International Review, 33(2):109. Bems, R. and Johnson, R. (2012). Value-Added Exchange Rates. NBER Working Paper No. 18498. Boltho, A. (1996). The assessment: international competitiveness. Oxford Review of Economic Policy, 12(3):1–16. Cashin, P., Cespedes, L., and Sahay, R. (2002). Developing Country Real Exchange Rates: How Many Are Commodity Countries? IMF Working Paper WP/02/223, Washington, DC: International Monetary Fund. Cartwright, W.R. (1993). Multiple linked diamonds and the international competitiveness of export-dependent industries:  the New Zealand experience. Management International Review, Special Issue, 33(2):55–70. Corden, W.M. (1994). Economic Policy, Exchange Rates and the International System. Oxford: Oxford University Press. Cronovich, R., and Gazel, R. (1998). Do exchange rates and foreign incomes matter for exports at the state level? Journal of Regional Science, 38(4) 639–657. Dos Santos, C.H., Shaikh, A., and Zezza, G. (2003). Measures of the Real GDP of U.S. Trading Partners: Methodology and Results. Working Paper No. 387, The Levy Economics Institute of Bard College. Dosi, G., Pavitt, K., and Soete, L. (1990). Economics of Technical Change and International Trade. Brighton: Wheatsheaf. Dunning, J.H. (1993). Internationalizing Porter’s diamond. Management International Review, Special Issue, 33(2):7. Durand, M., and Giorno, C. (1987). Indicators of international competitiveness: conceptual aspects and evaluation. OECD Economic Studies, 9:147-181. Fagerberg, J. (1988). International competitiveness. Economic Journal, 98(2):355–374. Fagerberg, J. (1996). Technology and competitiveness. Oxford Review of Economic Policy, 12(3):39–51. Grant, R.M. (1991). Porter’s competitive advantage of nations:  an assessment. Strategic Management Journal, 12(7):535–549. Gray, H.P. (1991). International competitiveness: a review article. International Trade Journal, 1(2):503–517. Harberger, A.C. (2004). The Real Exchange Rate: Issues of Concept and Measurement, A Paper Prepared for a Conference In Honor Of Michael Mussa, University of California, Los Angeles. Kaldor, N. (1978). The effect of devaluations on trade in manufactures, in Further Essays on Applied Economics. London: Duckworth, pp. 99–118. Krugman, P.R. (1980). Scale economies, product differentiation, and the pattern of trade. American Economic Review, 70(5):950–959. Monga, C. (2013). Winning the jackpot: jobs dividends in a multipolar world, in J. Stiglitz, J. Yifu, and E. Patel (eds), The Industrial Policy Revolution II: Africa in the Twenty-First Century. Palgrave Macmillan, pp. 135–173. Moon, H.C., Rugman, A.M., and Verbeke, A. (1998). A generalised double diamond approach to the globalized competitiveness of Korea and Singapore. International Business Review, 7:135–150.

Measuring Democracy   469 OECD, (1992). Technology and the Economy: The Key Relationships. Paris: OECD. Onyekwena, C., and Taiwo, O. (2013). An examination of South Africa’s trade with the BRICs. Working Paper. Centre for the Study of Economies of Africa, Abuja Nigeria. Porter, M.E. (1990). The Competitive Advantage of Nations. London: Macmillan Press. Ramirez, G., and Tsangarides, C.G. (2007). Competitiveness in the CFA Franc Zone. IMF Working Paper WP/07/212, Washington, DC: International Monetary Fund. Reinert, E.S. (1995). Competitiveness and its predecessors: a 500 year cross national perspective. Structural Change and Economic Dynamics, 6:23–42. Rugman, A.M. (1991). Diamond in the rough. Business Quarterly, 55(3):61–64. Rugman, A.M. (1992). Porter takes the wrong turn. Business Quarterly, 56(3):59–64. Rugman, A.M., and Verbeke, A. (1993). The Double Diamond Model of international competitiveness:  the Canadian experience. Management International Review, Special Issue, 33(2):17–39. Saito, M., Ruta, M., and Turunen, J. (2013). Trade Interconnectedness: The World with Global Value Chains. Washington, DC: International Monetary Fund. Scott, B. (1985). U.S. competitiveness concepts, performance, and implications, in S. Bruce and G. Lodge (eds), US Competitiveness and the World Economy. Boston: Harvard Business School Press. Smith, A.J. (2010). The competitive advantage of nations: is Porter’s diamond framework a new theory that explains the international competitiveness of countries? Southern African Business Review, 14(1):105–130. Vieira, F.V., and Haddad, E.A. (2011). A Panel Data Investigation on the Brazilian State Level Export Performance. The University of Sao Paulo Regional and Urban Economics Lab Working Paper. Wagner, G., and Zeckhauser, R. J. (2006). “Trading Up”. Unpublished Manuscript, Kennedy School of Government, Harvard University. Available at: http://www.hks.harvard. edu/m-rcbg/research/wagner_trading_up.pdf. Last accessed on September 11, 2014. Waverman, L. (1995). A critical analysis of porter’s framework on the competitive advantage of nations, in A. Rugman, J. Van den Broeck, and A. Verbeke (eds), Research in Global Strategic Management: Beyond the Diamond. Volume V, Greenwich, CT: JAI Press. Wignaraja, G. (2005). Competitiveness analysis and strategy, in Competitiveness Strategy in Developing Countries: A Manual for Policy Analysis. Routledge Studies in Development Economics. Routledge: London and New York, pp. 15–60.

Pa rt  I I I

H I STOR IC A L T R AJ E C TOR I E S A N D E C ON OM IC LANDSCAPE

Chapter 24

Af ri ca’s New E c onomi c Opp ortu ni t i e s Paul Collier

24.1 Introduction Investor sentiment about Africa has been subject to extraordinarily wide swings. In 2002 The Economist magazine aptly encapsulated the then-current perspective on the region with its cover story of “The Hopeless Continent.” A decade later Africa had become “The Rising Continent,” commonly supported by the statistic that the region had “six of the ten fastest-growing economies in the world.” Yet in the intervening decade Africa had been through neither fundamental changes in policy nor structural transformation. The instability of investor sentiment did not reflect an unstable African reality, but rather the extrapolation of recent experience. Extrapolation has been the dominant option because there has been little understanding of the region. Africa has simply been categorized as “different.” I will argue that Africa is distinctive but not different. Africa is not different because its key characteristics have similar consequences globally wherever they are found. But it is distinctive in that its characteristics, while not unique, are atypical. An implication is that Africa is not a mystery: we can do better than simply extrapolate whatever is happening currently. Global economic events have distinctive but reasonably predictable consequences for the region. The key global economic event of the past decade has been the growth of China. This is having a range of important consequences due both to competition in supply and to increases in demand. Taking first the impact of Chinese supply, as Chinese manufacturing clusters grew they became hyper-competitive. By this I mean that they were able to reap the productivity gains that accrue from cluster scale economies, while benefiting from wage levels comparable to those in low-income societies. Only by the end of the first decade of the twenty-first century did the growth of Chinese production lead to significant tightening of the labor market and so begin to erode this winning advantage. The phase of hyper-competition had consequences both for middle-income countries globally, and for low-income Africa. In many middle-income countries industrialization stalled; a phenomenon that became known as

474    Historical Trajectories and Economic Landscape the “middle-income trap.” For low-income Africa it closed off the possibility of breaking into global manufacturing. However, for Africa the key consequence of Chinese growth was the impact on demand. The rising demand for material inputs dramatically increased the prices of natural resources that mattered for Africa because of its distinctive geography. Being a huge landmass with limited rainfall it has been thinly populated. The consequential high ratio of land to people implies that natural resources endowments are likely to be important relative to income. Directly, higher prices substantially raised Africa’s revenues from its existing resource exports. But even more important, they triggered a wave of prospecting for new resources. As of the Millennium, Africa was the least-prospected region. Its known resource endowment per square mile was only one-fifth of that for the OECD. Obviously, this was unlikely to be because Africa had actually got fewer resources: over the two huge landmasses of Africa and the OECD, the random processes of geology are likely to have produced average endowments that are similar. The markedly lower known endowment of Africa simply reflected less past investment in search: Africa did not know what it had. Since the resource extraction companies understood this, they concentrated their search in Africa. As a result, so much was discovered that many African countries switched from being resource-scarce to resource rich. Whereas the first decade of the twenty-first century was one of discovery, the second will be a decade of extraction as discoveries are converted through further investment into mines and wells. This immediately yields a powerful prediction: Africa will continue to grow rapidly during this second decade. But the past growth underpinned by rising prices for exports will be replaced by growth underpinned by rising quantities. Rising exports of natural resources commonly cause Dutch disease: non-resource exports become uncompetitive. Historically, in Africa the clearest example of this process was in Nigeria during the phase of rising oil exports, 1966–1986. Nigeria’s initially massive exports of agricultural produce were effectively wiped out by real exchange rate appreciation. In the current content, the prediction of Dutch disease is important for understanding whether low-income Africa will be able to take advantage of the rise in Chinese wages to break into the global market for manufactures. Whereas until recently this was not possible because Chinese wages were too low, it may now be impossible because African resource exports, and hence real exchange rates, become too high. However, while this pattern describes Africa on average, there will be considerable variation. Because natural resources are randomly distributed they are highly uneven, so that some countries are well endowed and others resource scarce. For example, as of the Millennium Kenya had been unlucky in its known resource endowment. As a result, during 2000–2010 rising resource prices were a problem not a benefit: its terms of trade deteriorated by 10 percent. The high prices triggered prospecting and Kenya discovered oil, which will come on-stream during the coming decade. Africa’s distinctive geography has other important consequences. Because historic population density was very low, Africa’s population became ethnically and linguistically fragmented. As a result, during decolonization the continent became divided into many small countries. Even with this high degree of political fragmentation, most countries were internally ethnically fragmented. Given that Africa was a huge landmass, the proliferation of countries resulted in many being landlocked. These two features—resource endowment and location—generate a simple but useful two-by-two categorization of Africa’s countries. I will consider these four categories in turn.

Africa’s New Economic Opportunities    475

24.2  Coastal Resource-scarce Countries One of Africa’s most successful economies, Mauritius, epitomizes coastal resource scarcity. Initially an extremely poor country, in the mid-1980s it succeeded in breaking into unskilled, labor-intensive manufacturing. As China became hyper-competitive in this market, Mauritius was able to ascend the skill escalator, shifting to progressively higher quality niches in the garments market. However, no other African country was able to replicate the successful entry to manufacturing before China became hyper-competitive. Only in the current decade has the opportunity reappeared. Countries such as Eritrea and Liberia have the economic fundamentals necessary for breaking into global manufactures. However, the economic fundamentals of location and income levels are at most necessary, rather than sufficient, conditions: they create a shortlist of possible cities. Because manufacturing benefits from cluster scale economies it will never be spread thinly over all the cities on this shortlist:  firms will choose between them. In their choice, the remaining variations in labor costs will not be decisive: all countries on the shortlist have cheap labor, but labor costs account for only about 15–20 percent of value-added. Firms will therefore base their selection predominantly on variations in other costs. Many of these costs are affected by government policies. Hence, a government that wants to encourage entry to global manufacturing will need to adopt policies that reduce these costs. One such policy is to remove barriers to international trade. Modern manufacturing uses global supply chains in which the finished product gradually accumulates as small amounts of value are added to imported components at many different stages. For example, buttons are made in one location, zips in another, and the buttons and zips are sewn onto a garment in a third. Hence, high import barriers kill the possibility of participating in such global supply chains. Mauritius implemented a radical liberalization in trade policy without upsetting the protected import-substitute sector by the simple and effective expedient of creating an encompassing free trade zone: all exporting firms were able to import their inputs free of restrictions. A second important policy is the provision of basic infrastructure. Manufacturing supply chains depend upon swift logistics and reliable energy, so well-functioning ports and electricity generation are critical. Because manufacturing clusters together, this infrastructure does not need to be provided nationally. In infrastructure-scarce countries it is better swiftly to concentrate improved provision in one or two locations rather than gradually improve it everywhere. While infrastructure is important, manufacturing is more than just a mechanical process of production and distribution, transporting and converting intermediate inputs and energy into outputs taken to markets. Manufacturing is a business: firms enter into and enforce a set of contractual relationships. The functioning of the regulatory environment and the legal system are therefore also important. In breaking into global manufacturing, countries face a coordination problem. Because cluster scale economies are so important the first firms to enter a potential cluster are liable to lose money. Even though, once established, a new cluster might be viable, there is no market process that coordinates the entry decisions of many firms so as to jump rapidly to a critical mass. Mauritius succeeded in breaking in to garments manufacturing because by

476    Historical Trajectories and Economic Landscape chance changes in international trade policies provided a strong and coordinated incentive for firms to relocate. In 1984 the Multi-Fibre Agreement introduced quotas designed to curtail Asian exports, and Mauritius was granted a quota that it could not initially use. By relocating from Hong Kong to Mauritius, garments firms could maintain their access to European markets and many did so, enabling Mauritius to jump straight to a viable cluster of producers. Currently, there is no equivalent of the Multi-Fibre Agreement, but some such pump-priming arrangement is highly desirable. An obvious policy vehicle would be the Economic Partnership Agreements of the European Union, which, having long been gridlocked in a negotiating impasse, are ripe for revision.

24.3  Landlocked, Resource-scarce Countries The most striking difference between Africa and other developing regions is in the proportion of the population in landlocked, resource-scarce countries. While this has diminished due to resource discoveries, there are still important examples such as Ethiopia, Rwanda, Burundi, the Central African Republic, and Malawi. Outside Africa, areas with these poor endowments seldom became independent countries: rather, they became the hinterlands of countries that are overall more fortunately endowed. With hindsight, the creation of so many such countries in Africa may have been a mistake, but recent political secessions are actually adding to the number of such countries. The secession of Eritrea turned Ethiopia into a landlocked, resource-scarce country, and the secession of South Sudan created a new landlocked country, albeit with some oil. Small, poor landlocked countries face acute and distinctive problems of economic development. With poor transport connections to global markets most of their production will necessarily be orientated to the domestic market. But since the domestic market for most goods and services is tiny, it is not possible to reap scale economies and so costs are high and sometimes prohibitive. As the economy grows, markets periodically expand so that new activities become viable, but this process depends upon pioneer investors who take the risk of discovering whether new activities are profitable. This is a crucial distinguishing feature of growth in small isolated economies: they are unusually dependent upon pioneer invest­ ors. If pioneer investors fail they bear the full cost, but if they succeed other investors copy them. Hence, they are not able to appropriate all the benefits of the risk that they have taken. Because pioneering investment generates this beneficial externality and is disproportionately important in poor, landlocked economies, the rate of private investment in these economies will be too low from the perspective of the social optimum (Collier 2013). To date, this problem has not even been recognized, let alone rectified, either by governments or donors, yet it differentially affects the poorest countries in the world. Globally, there are some obvious examples of successful landlocked, resource-scarce economies such as Switzerland and Austria. However, these countries have benefited enormously from their neighborhood. In effect, being landlocked has not cut them off from international markets but rather placed them at the heart of a regional market. More generally, historically the most promising strategy for such countries has been to orient their economies towards trade with their more fortunately endowed neighbors. This shows up in the growth spillovers. Globally, on average if neighbors grow at an additional 1 percentage

Africa’s New Economic Opportunities    477 point, that raises the growth of the country itself by 0.4  percent (Collier and O’Connell 2008). Outside Africa the landlocked, resource-scarce economies on average gain larger spillovers, at 0.7 percent. Thus, they are consciously orienting their economies towards making the most of these growth spillovers. In Africa, the growth spillover for the landlocked, resource-scarce economies from the growth of their neighbors is a mere 0.2 percent. In other words, their economies are not oriented towards their neighbors. Regional integration thus matters a lot for the landlocked, resource-scarce countries. However, it is not a policy that their governments control: whether integration is feasible depends upon the policies of their more fortunate neighbors. The integration agenda is partly a matter of formal trade policies such as tariff barriers. However, at most African borders practical trade policy such as the removal of roadblocks and harassment by customs officials is more important. It is also a matter of infrastructure: roads need to be built and maintained. Limao and Venables (2001) show that the transport costs faced by landlocked countries depend upon the investments of their coastal neighbors in transport infrastructure. Africa’s regional organizations, and its international donors, might do more to promote these agendas. It is possible that developments in connectivity, notably e-trade and air-freight, might offer new routes to global integration for landlocked countries. In 2012 a Chinese footwear company shifted some production to Ethiopia, and there are plans for a similar shift of garments production to Rwanda. Clearly, the landlocked countries should push these opportunities to the hilt. Whereas being landlocked is not a choice, being “airlocked” is largely a matter of policy. The policies that produced high-cost monopolies such as Air Afrique were evidently mistaken. Similarly, the twin pillars of e-trade are international telecoms and post-primary education. Policies that raise the cost of international telecoms, or make access unreliable, and the neglect of post-primary education, would thus be costly for landlocked, resource-scarce Africa.

24.4  Landlocked, Resource-rich Countries Botswana is landlocked and resource rich: it is as successful as Mauritius. Indeed, for many years Botswana was the fastest-growing economy not just in Africa, but in the world. Botswana had no realistic opportunity of following the same path as Mauritius: it could not break into global markets for manufactures. The combination of high transport costs, a small domestic market, and an equilibrium real exchange rate that appreciated as a result of substantial diamond exports, made most manufactures uncompetitive. Botswana pioneered a radically different strategy. Botswana is not alone in being landlocked and resource rich. Zambia, Chad, South Sudan, and Mali have the same characteristics and could benefit from following the Botswana example. Even some resource-rich countries that are not landlocked turn out to have the same economic fundamentals despite their apparently superior geography. Angola, Equatorial Guinea, and Gabon have so much oil that, like Botswana though for a different reason, their equilibrium real exchange rates will be too high to make non-resource exports viable for the foreseeable future.

478    Historical Trajectories and Economic Landscape The key policy challenge for Botswana, and for other countries with the same economic fundamentals, has been to convert resource revenues into higher living standards for its population. The first step in this is to capture the rents on resource extraction for society by means of an effective tax regime while not killing off the process of extraction. Botswana approached this by establishing a partnership with the diamonds company De Beers. The joint company, Debswana, has been highly successful in meeting the interests of the country while being commercially viable. The next step is to ensure that revenues accrue to the national budget so that they can be spent through a coherent and transparent decision process. Botswana has achieved this, but some African governments have not. For example, in previous decades Cameroonian oil revenues were handled in an opaque manner by the president, and have never been adequately accounted for. Membership of the Extractive Industries Transparency Initiative (EITI) now provides a useful global litmus test of whether a government is achieving reasonable standards. EITI is no longer just for developing countries: five of the G8 countries have committed to implementing it. Having got resource rents into a budget process, the remaining task is to spend them effectively. One key issue is the balance of spending between assets and consumption. If the resource rents are unlikely to last for more than two or three decades it is important to use much of the revenues to accumulate assets: natural assets are being depleted and other assets should be accumulated in their place. Botswana has been very careful to do this. As to which assets should be acquired, both domestic investment and foreign financial assets are appropriate. As discussed, poor connectivity with the outside world poses severe problems for landlocked countries. Hence, a priority for public investment is the infrastructure needed for efficient transport and telecommunications. However, foreign assets are also useful for two distinct reasons. Botswana found that as a small, landlocked market, opportunities for domestic investment were not abundant. Rather than invest wastefully domestically, it accumulated foreign assets as well as domestic ones. As a result of this wise decision Botswana now has the multi-billion dollar Pula Fund. The Fund is particularly useful because, as with most natural resource revenues, diamond income is subject to periodic shocks. The Fund has been used to smooth public spending during periods in which export revenues have temporarily crashed, thereby avoiding acute recessions. While most of Africa’s resource-rich countries would be prudent to have substantial savings rates out of resource revenues, for those countries in which extraction of the resource endowment can continue for many decades and poverty is initially severe, the urgent priority is to augment consumption. This has implications both for supply and demand. Considering first the supply side, resource revenues accrue in foreign exchange and so can ultimately only be used to augment the supply of imports. But people consume many goods and services that cannot readily be imported: the category which economists term “non-tradable goods.” In poor countries the key examples of such goods and services are housing, healthcare, education, and transport. In order for the consumption of these goods to be increased, their domestic production must be increased and this will require investment in these sectors. In turn, for investment to be productive the process by which it is undertaken is critical. A range of possible investments must be prepared; a selection process is needed which prioritizes among them; the implementation of selected projects should be well managed; and the results evaluated to enable learning. The quality of public investment management, with each of these four stages distinguished, has now been benchmarked by the IMF in the Public

Africa’s New Economic Opportunities    479 Investment Management Index. This is publicly available online for most African countries and for over 70 countries worldwide. The process of building these capacities can be thought of as “investing-in-investing” and is a crucial step in being able to harness revenues for growth (Collier 2010). Botswana prioritized investing-in-investing, putting in place an effective system for deciding which public investment projects were worth undertaking. Turning to the demand side of raising living standards, households can only benefit from augmented supply if they can afford to acquire it. Some important services are sufficiently essential to all households that they can be directly financed by the government and provided for free without significant misallocation, the key examples being education and healthcare. But for most other goods and services the needs and preferences of households vary substantially and so it is more efficient for them to purchase what they want. Evidently, for households to be able to purchase the goods and services that raise their consumption level to a decent standard they need to get a decent income. In extreme situations the best way of doing this is simply to distribute some of the income that the government gets from natural resources directly to households. For example, Equatorial Guinea has so much oil that average per capita income is around $20 000. No matter how many public goods the government provides for free, people will not be able to increase their incomes from the present level of a few hundred dollars to $20 000 for many decades unless they get money directly from the state. But this is unusual. More commonly, the best way of raising incomes will not by be direct transfers but indirectly, through public investments that increase the capacity of ordinary people to be productive. For example, Angola currently contracts with Chinese construction companies to build housing, and then provides it at subsidized prices to middle-income households. This raises the living standards of some Angolan households but does not raise their income-earning capacity. Instead, the government could encourage the growth of indigenous construction companies that used indigenous labor. The houses would initially cost somewhat more to build, but in the process a new domestic construction sector would emerge that would provide decent incomes for many people. Such encouragement of a domestic construction industry was one of Botswana’s successful strategies.

24.5  Coastal Resource-rich Countries Coastal resource-rich countries face many of the same issues as their landlocked counterparts. However, if their resource revenues are likely to last only for two or three decades, it is important to plan for the time beyond their exhaustion. This is where having a coast is a major potential advantage because, like other coastal economies, these countries will be able to insert themselves into global markets for manufactures. Once resource revenues dwindle the equilibrium real exchange rate will depreciate, and so they will potentially be competitive. Superficially, the challenge is daunting. It might appear to be necessary to anticipate what niches the economy will be able to occupy in two or three decades time, and use some of the resource revenues to undertake the investments which will be needed to succeed in these niches. So stated, it is at once evident that it is impossible to forecast the development of market opportunities: product niches cannot be anticipated because products cannot be anticipated. In three decades there will be no I-phones, and though cars will still have wheels

480    Historical Trajectories and Economic Landscape they will probably use completely different power and navigation technologies. Given such radical uncertainty, what is it sensible for a government to do? While it is not possible to predict products three decades ahead, it is possible to predict with reasonable confidence the conditions that will be conducive for the production of whatever products are being made. That is, rather than trying to anticipate future products, governments should invest in the production “platforms” which will be needed for typical twenty-first-century products. The generic characteristic of these platforms will be human capital clustered in well-functioning cities. Currently Africa lacks such platforms: its workforce is less educated than that or other regions and its cities are neither productive locations to work, nor attractive places to live. If Africa’s coastal cities such as Lagos and Accra are to become places where ordinary workers can achieve twenty-first-century global incomes, they and their workforces will need to be transformed through sustained investment. Africa’s investment in human capital has to date focused predominantly on the expansion of primary education: the priority is now to raise both the quantity and quality of post-primary education. Africa has urbanized through the movement of people to cities: the priority is now the infrastructure, housing, offices, and factories necessary for productive jobs and decent homes.

24.6  Resource-rich and Ethnically Fragmented Societies The economic policies that a society adopts are the result of a political process. As noted above, one legacy of Africa’s historically low population density is its ethnic fragmentation. As a result, even though the region is fragmented into many countries, most countries are internally fragmented. Political allegiance has become organized according to ethnic identities. Such ethnic politics is often dysfunctional. Since voters are more responsive to identity than to government performance, the discipline normally provided by elections is weakened. Natural resources also generate political stresses. The revenues from resources reduce the need to tax the wider economy. This can weaken the incentives for citizens to scrutinize government, and for government to grow the economy. Not only is scrutiny weakened, but political power becomes more valuable, attracting the corrupt into politics. Not only do natural resource abundance and ethic fragmentation each generate political stresses, they interact adversely so that the stresses typical of societies that are so characterized are worse than simply being the sum of the two individual problems. They interact because both natural resources and ethnic identities are specific to location. The resources that are valuable to a nation are not spread equally across the population but are clustered in the territory of a particular ethnic group. Politics easily degenerates from being a cooperative game to supply public goods, into an ethnically organized zero-sum contest for the control of resource revenues. Typically, the well-endowed group regards the resource as its own, whereas other groups regard it as belonging to the nation. In the extreme this leads to violent conflict. In 2013 war flared up in South Sudan between the Dinka, who held political power, and the Nuer, who inhabited the region with oil. This mirrored an earlier attempted secession of oil-rich Biafra from Nigeria.

Africa’s New Economic Opportunities    481 Since most African countries are ethnically fragmented, and many are now becoming resource abundant due to prospecting, this juxtaposition will increasingly characterize the region. The crucial test of political leadership and voter sophistication will be whether these stresses can be surmounted through the design of appropriate political institutions. For example, the first President of Botswana, Sir Seretse Khama, had the foresight to persuade the clan leaders of his newly independent country that should diamonds be found they would belong to everyone. Behind the “veil of ignorance” that preceded geological search the clan leaders saw the advantage of such a policy. What can be said concerning the political institutions that are appropriate for societies in which political allegiances are aligned according to ethnicity? Autocracy in such a context is dangerous because it almost inevitably empowers the ethnic group of the autocrat over all others. But even democracy can have a similar effect, empowering only the majority group, unless it is complemented by strong checks and balances that protect the rights of minorities. The political institutions appropriate for resource-rich countries are checks and balances on power. Collier and Hoeffler (2009) find that checks and balances are differentially beneficial for growth in resource-rich countries: unlike other economies, the stronger are their checks and balances the faster their growth. In combination this suggests that Africa’s resource-rich, fragmented societies need democracies with distinctively strong checks and balances. Consistent with this prognosis, the institutions of Botswana have been of this form from Independence. Most African societies now have competitive elections. But few have yet built the checks and balances that effectively protect minority ethnic groups, or curtail the power of a government to squander or divert resource revenues should it wish to do so.

24.7 Conclusion Africa now has two economic prospects that are better than at any time since Independence. During the coming decade the exploitation of recent resource discoveries will provide major revenues for many countries under the direct control of its own governments. As China shifts its more labor-intensive manufacturing offshore, Africa will at last have a realistic chance of industrializing. For these opportunities to be seized, political institutions and economic policies will need to be improved. Improvement will not be easy: globally, institutions and policies tend to be slow changing. They reflect a political equilibrium and so the default option is for them to be persistent. I have suggested that a core objective of institutional change should be to strengthen checks and balances on the power of government. Such checks and balances are not usually built voluntarily by governments; they are introduced in response to popular pressure. Hence, rather than looking to political leadership, Africans may need to take widely-shared responsibility for institutional change. Informed and active citizens, rather than great leaders, may be the key to seizing the new economic opportunities. Urbanization and social media are increasing the power of citizens to bring pressure on governments, as demonstrated by the Arab Spring. But citizen pressure can be a force either for improvement or for deterioration in institutions and policies depending upon how well citizens understand the opportunities and challenges that their society faces. Empowered

482    Historical Trajectories and Economic Landscape but badly informed citizens can drive governments into implementing the temporary palliatives of populism. Fortunately, the expansion of African media, transformed access to global media, and the spread of education across the region are all raising the economic literacy of citizens. Citizen understanding has become the new battleground.

References Collier, P. (2010). The Plundered Planet. Oxford: Oxford University Press. Collier, P. (2013). Aid as a catalyst for pioneer investment. WIDER WP-2013-04. Collier, P., and Hoeffler, A. (2009). Testing the neo-con agenda: Democracy in resource-rich societies. European Economic Review, 53(3):293–308. Collier, P., and O’Connell, S. (2008). African growth: opportunities and choices, in B. Ndulu, S. O’Connell, R. Bates, P. Collier, and C. Soludo (eds), The Political Economy of Economic Growth in Africa, 1960-2000. Cambridge: Cambridge University Press, pp. 76–136. Limao, N., and Venables, A.L. (2001). Infrastructure, geographical disadvantage, transport costs, and trade. The World Bank Economic Review, 15(3):451–479.

Chapter 25

Tigers or Tiger Praw ns ? The African Growth “Tragedy” and “Renaissance” in Perspective

Christopher Cramer and Ha-Joon Chang 1

25.1 Introduction Between the mid-1990s and the middle years of the 2000s, a thriving cottage industry existed that purported to explain Africa’s “growth tragedy,” or its “chronic growth failure” (Collier and Gunning 1999a, b; Easterly and Levine 1997; Bloom and Sachs 1998). Africa is “beyond doubt,” one commentator argued, “the most tragic example of a growth disaster” (Smits 2006). In the last few years, the discourse on African growth has changed 180 degrees. In the last decade or so, the continent has grown faster than ever in its history and faster than many regions in the world (except Asia). Now the talk is of “African renaissance,” “Lions on the move” (McKinsey Global Institute 2010), “Tigers in Africa,” or the widespread “Africa Rising” narrative. Having improved their governance and established macroeconomic stability, it is argued, African economies are finally reaping the benefits of economic liberalization implemented throughout the 1980s and the 1990s. The only way for Africa is up, according to this narrative. This chapter aims to put the African growth experiences of the last half-century into perspective by critically examining these two simplistic discourses that have dominated discussion on African economies and produced misguided policy actions (and inactions), undue pessimism, complacency, and many other negative consequences.

1 

The authors would like to thank Ming Leong Kuan for thoughtful and extremely efficient assistance in marshaling data used in this chapter.

484    Historical Trajectories and Economic Landscape

25.2  The Growth Tragedy and Its Explanations 25.2.1  African growth tragedy The African growth experience had not always been described as “tragic.” Africa’s economic growth in the early days of decolonization (stretching between the late 1950s and the mid-1970s, but concentrated in the 1960s) had been relatively poor. But no one described it as a tragedy, even at the start of the structural adjustment programs (SAPs) in the late 1970s and the early 1980s, when many African countries were paraded as examples of failing state-led development strategy (e.g. in the Berg report, World Bank 1991). The talk of an African growth tragedy started in the mid-1990s, when a decade of “structural reform”—liberalization of international trade and investment, retrenchment of government spending, privatization or liquidation of state-owned enterprises and many government agencies, and deregulation of domestic financial and labor markets—had failed to produce the expected acceleration in growth. Indeed, in most countries, the result was a fall, rather than an improvement, in growth (Mosley, Subasat, and Weeks 1995). Not being willing to admit that SAPs may have actually reduced, rather than improved, economic performance, the search was on for factors that could explain why the African countries had very poor growth performance despite adopting “good” policies. All sorts of “meta-structural” factors—geography (especially being landlocked), climate, natural resource endowments, colonial history, culture, ethnic diversity, demography (especially low population density), governance (politics), and what not—were ventured as explanations of Africa’s unresponsiveness to the “right” medicine and often fed into cross-country growth regressions of the sort that Solow (2007: 51) found “statistically unprepossessing.” Many of these regressions sought to discover the source of the so-called “Africa dummy,” the residual in growth accounting that was not attributable to the main factors of production and that might capture the apparently distinct “African” growth experience. With meta-structures like these, it is not surprising that the African countries do not do well despite having gone through “structural” adjustment programs for a decade or longer. Constrained by these factors that can be changed only very slowly, if at all, Africa was doomed to stagnation, if not retrogression.

25.2.2  Tragedy what tragedy? Before we even start discussing the merits and the shortcomings of explanations of an African growth tragedy based on “meta-structural” factors, we should point out that there was really no African growth tragedy. This is in two senses. First, there can be no “African” growth story—tragedy or otherwise. We are not simply repeating the obvious—if frequently ignored—point that Africa is not homogeneous and is made up of dozens of countries with very diverse conditions. The point is that many similarities between African countries are more a function of their underdevelopment and history of colonialism, rather than of their being in “Africa.” And as we show below, in a comparison of recent economic history in Ethiopia and Mozambique, even countries with many

Tigers or Tiger Prawns?    485 apparent similarities in the same period can be fundamentally different in ways that matter for policy and growth. Second, there is no African growth “tragedy.” Low—and often negative—growth that prompted the talk of “tragedy” has not been a permanent feature of the continent but a phenomenon that was unique to the roughly 25-year period from the late 1970s to the early 2000s. Between 1960 and 1980, Africa grew at decent rates and as Jerven (2010) shows between 15 and 26 African countries were experiencing sustained growth episodes from the late 1950s to the mid-1970s in any one year. During this period, per capita income in sub-Saharan Africa (SSA) grew at 1.6% per year. While it was lower than in other developing parts of the world (East Asia 5.3%; Latin America 3.1%; the Middle East and North Africa 2%), it is a rate that compares favorably with those experienced in today’s rich countries during the Industrial Revolution, when they grew at around 1–1.5%. In the first decade of the 2000s (2000–2010), at 2.6% per annum, gross national income (GNI) per capita in SSA has grown faster than ever.

25.2.3  Can “meta-structural” factors explain poor African growth? If this is the case, explanations of Africa’s low growth during 1980–2000 relying on “meta-structural” factors—such as climate, geography, and history—are not convincing. Most of these factors have been present all the time but growth has gone up and down quite a lot in most African countries. This suggests, as we have pointed out above, that the “tragedy” narrative came into being in order to “explain away” the failures of the SAPs and their successor programs during the period. Moreover, the fact that a factor is given by nature or history does not mean that the outcome is predetermined. Indeed, the fact that most of today’s rich countries have also suffered from similar “meta-structural” handicaps suggests that all those structural factors are not insurmountable. For example, many of today’s rich countries used to have malaria and other tropical diseases, at least during the summer—not just Singapore, which is bang in the middle of the tropics, but also Southern Italy, the Southern USA, South Korea, and Japan. Malaria and other diseases have largely (although not entirely) disappeared in those countries not because their climates have somehow changed but because they have better sanitation (which has vastly reduced their incidence) and better medical facilities (which allow them to deal effectively with the few cases that still occur), thanks to economic development. Therefore, to blame Africa’s underdevelopment on climate is to confuse the cause of underdevelopment with its symptoms—poor climate does not cause underdevelopment; a country’s inability to overcome the constraints imposed by its poor climate is a symptom of underdevelopment. Likewise, much has been made of the landlocked status of many African countries. Landlockedness does impose economic burdens, but then how do we explain the economic successes of Switzerland and Austria? These are two of the richest economies in the world, but they are both landlocked. Some people may respond to this point by saying that those countries could develop because they had good river transport, but many landlocked

486    Historical Trajectories and Economic Landscape African countries are potentially in the same position; for example, Burkina Faso (the Volta), Mali and Niger (the Niger), Zimbabwe (the Limpopo), and Zambia (the Zambezi). So, once again, the argument is based on confusion between the cause and the symptom—it is the lack of investment in the river transport system, rather than the geography itself, that is the problem. One of the fastest growing non-resource-rich countries in Africa in the past 15 or so years, Ethiopia, is landlocked (not to mention being in a “rough neighborhood”), and lost access to the port of Massawa in Eritrea through political conflict, but has been investing heavily in upgrading road and rail links to coastal ports in Djibouti (and, through the ongoing LAPSSET project, in Kenya). Similar points can be made in relation to many of the “meta-structural” factors, not to speak of more malleable things like institutions (governance) but also things like ethnicity (Chang 2009a, b, 2010; Cramer 2006; Lonsdale 2012). The key point is that the growth tragedy literature has got the causality wrong—if anything, it is underdevelopment that is making “meta-structural” factors appear to be obstacles to growth, rather than those factors causing underdevelopment.

25.2.4  Questioning the method The African growth tragedy literature mainly identifies the causes and outcomes of the tragedy through cross-section growth regressions. There is no shortage of critiques of the growth regression literature, with implications for the interpretation and significance of the “African dummy” (see Rodriguez 2007; Kenny and Williams 2001; Deaton 2010). An important point raised by critics of growth regressions is that many of their results are not statistically robust, particularly when samples of countries change. Most contributions to the literature of the 1990s used a limited sample of African countries.2 Country samples varied, often because it proved too difficult to include data stretching back several decades for some countries. Which African countries are included makes a massive difference to results. Econometric specifications also vary and that also affects the validity of results and their interpretation. The number of regressors (explanatory variables) used in the vast growth literature is astonishing and exceeds the number of countries in the standard growth datasets (Durlauf and Quah 1999). As Deaton (2010) argues, a lot of growth studies barely relate to an economic model of growth and mostly rely on the inclusion/exclusion of variables to minimize estimation residuals. Much of the growth literature is dominated by linear specifications yet the historical evidence suggests that growth is a very complex outcome of often non-linear processes and relationships among factors. Rodriguez (2007) showed how different the results are from much less used, non-linear regressions compared with standard linear regressions. Overall, technique, ideological convenience, and the speed of computer analyses, rather than careful thought, have driven findings and recommendations. The problem of distinguishing correlation from causality and the possibility of “reverse causality” or “endogeneity” (i.e. when independent variables may be caused by the dependent variable as well) is one of the main headaches for econometricians and growth researchers. 2 

See Table 1 in Collier and Gunning (1999b: 65).

Tigers or Tiger Prawns?    487 Despite the lack of robustness in results and the questionable quality of the data, much of the literature draws strong policy implications. A growing number of researchers are trying to solve the correlation–causation conundrum by using instrumental variables, but instrumental variables are a misleading tool for a wide range of applications. Even when its use is appropriate, it is very hard to find an appropriate instrumental variable (IV) and even harder to draw a meaningful interpretation, especially in cross-country regression analysis (Deaton 2010).3 Also, statistically a lot of useful information is lost by excessive aggregation and using arbitrary regional dummies. A regional average masks significant variation that gets lost in the regression process. Some alternative econometric clustering techniques have shown that one out of four African countries would be classified as medium–high growth comparable to many Asian and Latin American countries. They conclude that there are no clear-cut continent-specific clusters, which means that, statistically, an African dummy is not very useful (Paap et al. 2005). As Jerven (2011: 3) concludes: “[T]‌he quest for the African dummy was an outcome of the specific methodology used, which again determined the handling of the growth evidence.” That evidence, in turn, is a problem. The shortcomings of much of the official data on African economies should cause widespread humility among those laying claim to grand explanations of trends, let alone their causes. As base years change (or become obsolete) unevenly across countries, as different countries adopt new approaches to the coverage of national accounts estimates at various points in time, as statistical agencies suffer variously the vagaries of resource allocation, and so on, the individual reliability and above all the comparability of data across African countries becomes deeply problematic (Jerven 2013; Devarajan 2013).4 Still far too few analyses of economic growth in Africa acknowledge this problem.

25.3  Africa Rising? 25.3.1  African growth revival As the African growth tragedy discourse was reaching its fever pitch in the early years of the 2000s, many African economies started growing faster. The result has been more than 3  The most widely cited example in the African growth literature is Acemoglu et al. (2001). In their attempt to avoid reverse causality and circular reasoning, these authors use patchy records on pre-quinine mortality among European settlers, soldiers and missionaries to assess the extent to which initial geographical conditions determined particular institutional paths. They do so because they find a correlation between this instrument and private property rights and output per head in the 1990s. The pre-quinine mortality rates are supposed to be “exogenous” (Deaton 2010 would say “external”) to post-colonial economic growth per capita, as there could not be a direct connection, thus making them an ideal instrumental variable (Austin 2008). Intuitively if an explorer in the 1600s met a hostile environment, this would have induced an extractive and predatory set of institutions and discouraged settlements, contrary to areas where settlements were viable and the need to reproduce “European” institutions became pressing early on. In this contrast, Australasia and North America, as opposed to other colonized regions, “adopted the kinds of institutions—primarily, private property rights—that already existed or were emerging in Western Europe” (Austin 2008: 999). 4  “African data exhibit deficiencies in accuracy, periodicity, and timeliness” (IMF 2013: 26).

488    Historical Trajectories and Economic Landscape a decade of unprecedented growth. Growing at 2.2% per annum, per capita GDP growth in SSA between 2000 and 2010 was the fastest ever for the region. It was also faster than that in Latin America and the Caribbean (1.8%) and not much behind that of the Middle East and North Africa (MENA) region (2.6%). Now, the talk is of an African renaissance or “uncaging the lions” (The Economist, June 2010). Having got rid of the bad policies through the SAPs in the 1980s and the 1990s, Africa is finally ready to take off; with plentiful natural resources, diasporas with business experiences in the advanced economies, and governments finally ready to “do business” many of them could be the next tiger economies (on “Africa’s $2.6 trillion business opportunity,” e.g. see McKinsey Global Institute 2010). The demographic composition that had commentators in a panic about the threat of a “youth bulge” (Urdal 2004; Cincotta 2008) is now more commonly cast as a massive market-in-waiting and a source of competitively cheap global labor (McKinsey Global Institute 2010). Natural resources are seen less often as curse and more as blessing.

25.3.2  Factors behind the African Renaissance One thing that the growth acceleration in the last decade has done is to discredit the African growth tragedy discourse. The growth acceleration has occurred without any significant changes in those “meta-structural” factors that were supposed to be responsible for the tragedy; it is not as if ethnic homogeneity suddenly increased through rearrangement of borders or landlocked countries miraculously got access to the sea due to fortuitous seismic activity. If anything, growth has quickened despite the deterioration in those “meta-structural” variables. For the most important example, as we mentioned earlier, Ethiopia, one of the fastest growing non-oil African economies, actually became more landlocked since the late 1990s!5 While the recent growth experience of Africa shows that the continent is not destined for growth failure, it is still premature to declare the continent to be on a new, sustainable growth path. First of all, in a number of countries (Chad, Sudan), the revival of growth is mainly the result of the end to their civil wars. Of the six rapidly growing non-resource-rich African countries highlighted by the IMF (2013) over the 1995–2010 period, four were post-conflict countries (Ethiopia, Mozambique, Rwanda, and Uganda). The importance of civil wars in explaining African countries’ (or for that matter any country’s) growth performance is illustrated by the fact that Cote d’Ivoire, a growth “star” in the earlier era, has seen its growth disappear after the civil wars in the twenty-first century; it has experienced the fourth biggest drop in the rate of growth of GDP per capita and the largest fall in the overall rate of growth of GDP (among African economies) between 1961–1980 and 1980–2012.6 Second, in quite a few other countries, growth acceleration was sparked by the discovery of new mineral resources. Since it found oil in the mid-1990s, Equatorial Guinea has been one of the fastest growing economies in the world, growing at double the speed of China. 5  Despite this, Ethiopia was nonetheless deemed to have more accessible port infrastructure than Mozambique (IMF 2013: 47). 6  Yet it is also intriguing to consider the role of wartime in potentially generating contributors to post-war growth and structural change, in terms of political organization, development of strategic vision, etc. (Cramer 2006).

Tigers or Tiger Prawns?    489 Between 1995 and 2010, its per capita GDP grew at the rate of 18.6% per year—more than double the rate in China, the international growth superstar, which grew at “only” 9.1% per year. Angola is another example of oil-based success, while growth in Ghana is picking up partly thanks to its recent oil discovery. Third, many countries have benefited from the rapidly growing demand for natural resources from China. By the early 2000s, the Chinese economy had become very significant and was growing at breakneck speed, sucking in natural resources from Africa and Latin America. Given that China is unlikely to repeat this rate of growth in the coming decades, the growth-boosting effect that Africa gets from China may be more limited. What is even more worrying than the growth being mainly due to one or more of these one-off factors is that economic growth in most African economies during the last decade has not been accompanied by an increase in productive capabilities or structural change. Manufacturing productivity in 15 major SSA economies studied by ACET (2013:  4)  “is low and stagnant” and the shares of medium and high level technology in production and exports are not only much lower than in ACET’s group of non-African comparator countries but these shares have been declining.7 Indeed, Rodrik (2011: 29) argues that Africa—and Latin America—have experienced “growth-reducing structural change during 1990–2005,” meaning a shift of labor from higher productivity activities such as manufacturing and towards lower productivity activities. Growth episodes, including in Africa, are relatively common; the problem is that many growth spurts are followed by severe growth collapses and the challenge is to convert spells of rapid growth into fundamental structural change that enables a more resilient, sustained growth (and employment) trajectory (North et al. 2009; Ocampo et al. 2009). The historical record (Ocampo et al. 2009) shows that in developing countries growth episodes are especially common in particular international conditions: cheap borrowing, high inflows of foreign investment, and high commodity prices. When these conditions fade, growth becomes far more vulnerable. Hence the need to convert recent African growth—in most cases sustained precisely by debt relief (HIPC), low interest rates, a commodity boom (in particular sustained by Chinese demand), and eager (but unreliable) capital flows—into structurally stronger development. Unfortunately, even since the growth revival, many African economies have experienced de-industrialization and the dependence on natural resource exports has gone up, rather than down, in many countries. Few African countries are on their way to becoming “tigers”; rather, they remain “tiger prawns,” dependent on international demand for their natural resource-based exports, like minerals, coffee, cocoa, seafood, and so on. Unless they become serious about diversifying and upgrading their productive capabilities, growth in African countries will remain subject to the vagaries of external demand and vulnerable to technological innovations for synthetic substitutes, as well as at the mercy of foreign aid, short-term global capital flow fashions, and largely ineffective “ethical trade” initiatives. Arguably, in the recent literature on African growth there has been too little focus on the significance of industrialization and the promotion of productive capabilities in cementing

7  ACET (2013) includes a group of 15 countries that account for 70% of the SSA population, 76% of its GDP, 80% of exports, and 85% of manufacturing value added. The ACET comparator group includes Brazil, Chile, Indonesia, Malaysia, Singapore, Thailand, South Korea, and Vietnam.

490    Historical Trajectories and Economic Landscape the gains of short-term growth spurts. While the hubris of cross-country growth regressions of the kind discussed above has faded from growth economics, it has been replaced by what often appears not merely an appropriate humility (Easterly 2009; Solow 2007) but rather a muddle of eclecticism—“sound” macroeconomic policy matters, sectoral promotion seems to matter but governments should be wary of trying to copy East Asian economies too much, states should be “developmental” but only in a “facilitative” way and not messing with comparative advantage signals, and so on (The Barcelona Development Agenda 2004; The Spence Commission 2008; Lin and Chang 2009). An alternative tradition of development economics drawing on Kaldor, Thirlwall, and others (Amsden 2001; Chang 2002; Ocampo et al. 2009) may offer a more coherent foundation—and one rooted more clearly in economic history—for considering the performance and prospects for sustained growth and development in Africa. To illustrate some of these arguments, we briefly discuss two diverging “success stories,” Ethiopia and Mozambique. The comparison suggests a need for nuance and restraint in waving about fashionable monikers like “Africa rising” and it suggests the significance of more detailed disaggregation than is the norm. It would be foolhardy to make grandiose predictions in either case. However, Mozambique arguably offers an example of a more natural resource-based growth strategy with little coherent strategy for structural change and very little “ownership” of policy. In contrast, Ethiopia, replete with risk and uneven success, demonstrates the possibilities of an unusually coherent development strategy, partly adhered to for reasons of political survival, and geared far more to structural change and broader (socially and spatially) growth.

25.3.3  A comparative anatomy of recent African growth: Mozambique and Ethiopia A comparison of recent economic history in Ethiopia and Mozambique helps to illustrate some of the themes addressed in this chapter. The two economies have been among the fastest growing SSA economies in the last two decades or so, with real GDP per capita growth growing at 4.2% and 4.4% between 1992 and 2012. The comparison is useful because it demonstrates the diversity of SSA, even where there are some common features.

25.3.3.1.  Similarities and differences in “meta-structural” factors Both Ethiopia and Mozambique suffered prolonged warfare and have been governed, since the end of civil wars in the early 1990s, by a single party under a formally democratic system. Neither of them record highly on “governance” scores (Ethiopia ranking 33rd and Mozambique 20th, out of 52, in the Ibrahim Index of African Governance 2013).8 Both have existed in “rough neighborhoods,” characterized by war, political violence, and poverty, although Mozambique’s neighborhood has become more settled since the end of apartheid, while conflicts in Ethiopia’s neighborhood have intensified, if anything. In each case, the ruling party has a history of commitment to socialist ideals but has managed deregulation and 8 

http://www.moibrahimfoundation.org/downloads/2013/2013-IIAG-summary-report.pdf.

Tigers or Tiger Prawns?    491 a “market-oriented” economic strategy. They have both experienced remarkable economic growth over fairly prolonged spans of time. Though the classification is arguably misleading for Mozambique, they both have been defined as falling into a group of rapidly growing non-resource-rich African economies (IMF 2013).9 However, there are also important differences in “meta-structures.” In terms of geography, Mozambique has a long coastline while Ethiopia has had unstable access to the sea via Eritrea during protracted (then) civil war and more recently has become officially landlocked since the secession of Eritrea in 1993; this route has been cut since war between the two countries in 1998. Ethiopia is a relatively densely populated country by African standards, with 83 people per km2, while Mozambique is rather sparsely populated, with 29 people per km2. Ethiopia was not colonized (and as Markakis 2011 argues, it took part in the nineteenth century scramble for Africa, expanding its own empire), while Mozambique has a bitter history of colonization, being one of the last SSA countries to become independent (in 1975 from Portugal).

25.3.3.2  Similarities and differences in growth strategies Mozambique and Ethiopia have pursued rather different growth strategies, based on different relationships with the outside world. Mozambique, often considered one of the great successes of internationally brokered war-to-peace transition, has remained for more than 20 years a profoundly aid-dependent country (Castel-Branco 2008). Ethiopia has also depended on large inflows of concessional funding but it has managed to make a better fist of protecting “policy space” (Whitfield et al. 2008; JICA 2011). This difference may owe something to political history. In Mozambique, the ruling party, Frelimo, which has been in power since independence in 1975, has not faced a serious political challenge since the end of the war in 1992. Since the end of apartheid in South Africa in 1994, Frelimo has ruled in an essentially stable and “friendly” regional neighborhood. Ethiopia, on the other hand, has been and remains surrounded by instability and in some cases hostility, with borders with Eritrea and Somalia as well as with Sudan and South Sudan. The ruling EPRDF coalition is still engaged in the project of creating a coherent Ethiopian nation (Markakis 2011). In terms, therefore, of the internal and external threats, which Doner et al. (2005) argue were important to driving developmental commitments in East Asia, it may be that the Ethiopian government is under more pressure to forge a development strategy that ensures its survival. Certainly, the EPRDF has hoist its flag on the mast of economic development and the shock of an internal democratic challenge in the 2005 elections appears to have provoked a serious reappraisal of strategy and a renewed, and adjusted, commitment to rapid economic growth. Growth in Mozambique has been sustained in a context of market liberalization and a fairly strong commitment to working within the parameters, outwardly at least, of an orthodox policy framework shaped by IFIs and bilateral donors. As Castel-Branco (2008) showed,

9 

It is acknowledged in IMF (2013) that, even when formally Mozambique did not qualify as a resource rich country, something changing now, indirectly it did benefit earlier from the resource boom thanks to “large external inflows to finance investment in natural resource production” (p. 32). Unfortunately the IMF does not provide the definition of “resource richness”.

492    Historical Trajectories and Economic Landscape there have been examples of a more coherent national interest shaping economic strategy, as in the sugar sector, but these are rare, and they owe more to “winners picking the state” than the other way round. In Ethiopia, by contrast, a much clearer strategic commitment to structural change and long-run national development has emerged. The government itself promotes a strategy of “democratic developmentalism” and the late Prime Minister Meles Zenawi advocated the case for developmental states in Africa (Zenawi 2012). But discussions of whether Ethiopia (or Rwanda, or possibly other African states) is a “developmental state” are not especially helpful:  they tend to be premature, since much of what is regarded as the content of a developmental state is defined after the fact, when the unevenness, experiment, and rough edges of policy in different sectors and at different times acquires a coherence conferred by outcomes. The reality, at least for now, is messier. In Ethiopia, there is evidence of policy learning-bydoing and of a combination of winners picking the state and of the state picking—and creating—winners. Recent research (Arkebe 2013) shows that policy in Ethiopia has not stuck to the dictates of current comparative advantage. If anything, policy has been driven by the powerful nudge of the balance of payments constraint, geo-political insecurity, and internal challenges, together amounting to a form of compulsions not unlike Hirschman’s (1958) unbalanced growth prompts. At the same time, the internal coherence and ideas of a group of leaders who had fought a 17-year rebellion appears to have forged a policymaking environment within which these pressures find nationalist responses. As JICA’s (2011) report puts it, the Growth and Transformation Plan (GTP) in Ethiopia is unusual in its brevity, coherence, and strategic direction.

25.3.3.3  Similarities and differences in economic performance Comparing the data on these two countries does not fully capture the differences in policymaking but begins to suggest divergent paths of growth and transformation. For example, while the data on manufacturing value-added (Figure 25.1) show that MVA as a share of GDP is higher in Mozambique and that MVA has grown rapidly in both countries in recent years (Figure 25.2), they do not fully capture the dynamics of manufacturing. Nor do they capture the different policy environments for industrialization Overall, growth in Mozambique has been driven heavily by aid and by FDI, which have sustained growth in a country whose current account deficit has remained at more than 13% of GDP from the mid-1990s till 2010 (IMF 2013a, Table 2.4). Aid flows to Mozambique were on average 14.5% of GDP between 1995 and 2010, the highest share in the group of rapidly growing non-resource-rich African countries during this period. Growth has been concentrated chiefly in services and increasingly in minerals and energy production, which have—despite the IMF’s classification of the country as a non-resource rich economy—taken a rising share of output and of exports (Figure 25.3). That growth is also highly unequal, both vertically and spatially. It has become less pro-poor, as Jones and Tarp (2013) and Ehrenpreis and Virtanen (2007) put it. It is largely concentrated in and around Maputo and in spots of mineral production and there is evidence that it is very much an urban phenomenon (Jones and Tarp 2013). Rural poverty has stayed at astonishingly high levels (some 70% according to some estimates). There has been

Tigers or Tiger Prawns?    493 18%

Ethiopia

16%

Mozambique

14% 12% 10% 8% 6% 4% 2%

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

0%

Figure 25.1   Manufacturing value added (% of GDP). Source:  World Bank’s World Development Indicators  2013.

1,200

Millions (US$)

1,000

Ethiopia Mozambique

800 600 400 200

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

0

Figure 25.2   Manufacturing value added (constant 2005 US$). Source:  World Bank’s World Development Indicators  2013.

very weak productivity growth in so-called smallholder agriculture (Arndt et  al. 2012). Survey data reviewed by Jones and Tarp (2013) suggest a decline, between 2002 and 2008, in access to extension information, use of irrigation, and use of pesticides, and no change in the use of chemical fertilizers. Ethiopia has actively promoted the expansion of agriculture and a diversification of exports, including both inter- and intra-sectoral export diversification, securing rapid growth in foreign exchange earnings from air travel (through the state-owned Ethiopian Airlines, in its more traditional passenger business and increasingly in the freight facilities developed to support the floriculture and broader horticulture businesses), from tourism, from cut flowers, and from sesame, and other commodities. Figure 25.4 shows that a declining share of export revenue has come from minerals and energy (unlike in Mozambique) and a growing, albeit still very small, share from manufactured exports. Export revenues more than quadrupled from 2004 to 2011, to more than $3 billion. As the

494    Historical Trajectories and Economic Landscape 100% 80% 60% 40% 20%

Agricultural products

Fuels and mining products

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1996

1995

1994

0%

Manufactures

Figure  25.3   Diversification of exports, Mozambique. Source:  WTO International Trade Statistical Database. Note:  The time series was not complete (missing data for 1997–1998).

100%

95%

90%

85%

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1995 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

80%

Agricultural products

Fuels and mining products

Manufactures

Figure  25.4   Diversification of exports, Ethiopia. Source:  WTO International Trade Statistical Database Notes:  1.  Agricultural products constituted the bulk of the export basket of Ethiopia (83–97%). The y-axis shows a narrower range (80–100%) to reflect changes more clearly. 2.  The time series was not complete (missing data for 1994, 1996–1998).

EIU (2012) put it: “Ethiopia has grown more quickly than almost any country on the continent while rejecting the advice from the IMF and others to open up the economy quicker than it would like.”10 In agriculture, the IMF argued that there was a relatively effective state strategy, the National Agriculture and Extension Intervention Programme (NAEIP).11 The NAEIP 10  http://country.eiu.com/article.aspx?articleid=659462850&Country=Ethiopia&topic=Economy&oid= 659462850&aid=1. 11  There is some reason to doubt official statistical claims about agricultural growth in Ethiopia (Dercon and Hill 2009).

Tigers or Tiger Prawns?    495 100000 90000

Ethiopia Mozambique

80000 70000 60000 50000 40000 30000 20000 10000

19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 20 06 20 08 20 10 20 12

0

Figure  25.5   Roads, total network  (km). Sources:  International Road Federation (World Road Statistics), Ethiopian Roads Authority.

provided seeds, fertilizers, and credit to 40% of farming households over ten years and produced a 50% increase in improved seed use and 30% increase in fertilizer use between 1998 and 2008 (IMF 2013). Also supporting a more broad-based economic growth, and acting as a stimulus to both agricultural trade and the geographical spread of manufacturing, the state has engaged in a push for road improvement. During the first phase of the Road Sector Development Programme (RSDP I) from 1997 in Ethiopia, 8709 km of roads were built or rehabilitated; during RSDP II 988 km were rehabilitated, 1758 km were upgraded and 628 km of new gravel roads were constructed. By 2008 more than 70 000 km of community roads had been constructed. Between 1997 and 2011 the road network, according to the Ethiopian Roads Authority, expanded from 26 550 km to 53 997 km, with a significant effect on the spread and growth of manufacturing enterprises (Shiferaw et al. 2013; Figure 25.5). The EPRDF leadership has been fairly effective in crafting a “rhetorical commonplace” (Jackson 2006) of growth and transformation as a national project, helping to shape and secure the legitimacy of resource allocations.12 This is reflected in the mobilization of domestic finance for the Grand Renaissance Dam on the Nile, the biggest hydroelectric power project in Africa and one that—largely thanks to hydro-political sensitivities with respect to Egypt—could not attract funding from the World Bank or even the Chinese government. Other dramatic efforts to overcome the infrastructural constraints on structural change include the largest electric run rail network in Africa, under construction since 2010, and Ethiopia’s commitment to the Lamu Port and South Sudan Ethiopia Transport (LAPSSET) project. As yet, this commitment to development, expressed officially in the Growth and Transformation Plan (GTP), has not led to significant structural change. At any rate, such change is not wholly obvious yet in all the statistics. Manufacturing has grown, but it remains a very small share of total economic activity. The country has secured an improvement in 12  The late Prime Minister, Meles Zenawi, was not alone among Ethiopian leaders in taking a serious interest in thinking through the dilemmas of development but his approach is captured relatively clearly in publications including Zenawi (2012) and de Waal (2012).

496    Historical Trajectories and Economic Landscape UNIDO’s Competitive Industrial Performance Index (CIP). Despite the importance in structural change of shifts in employment, labor market data are very poor. It is, however, notable that ILO indicators suggest that 3.4% of total employment in Mozambique was in industry in 2003, while the indicators suggest a rising trend in Ethiopia, from 2.3% in 1994 to 6.6% by 2005 (ILO Key Indicators of the Labour Market). There is a clear industrial policy but its effects have been uneven: for example, one recent study highlighted the successes in floriculture (effectively very like a manufacturing activity) and cement and contrasted these with the sluggish response to policy of the leather and leather products sector, hampered by entrenched interests and by the failure to secure a reliable supply of high-quality inputs from the massive Ethiopian livestock population (Arkebe 2013).13 From producing around half a million tons of cement in 1992, Ethiopia has grown to be one of the top three producers in Africa, producing more than 10 million tons in 2010. It has done this partly by nurturing domestically owned firms rather than handing over to the conglomerates that dominate global production. In the process, both policy and firm level capabilities have developed. Figure 25.6 shows some of the impact of a large infrastructure program, a remarkable social housing program (UN Habitat 2010), and the nurturing of a domestic (and foreign owned) cement production industry, in the rapid increase in gross value added in construction in Ethiopia, by contrast with Mozambique. What is evident from the study of specific sub-sectors like these is that, despite the limited overall structural change thus far, in Ethiopia there is an ongoing and strategically promoted process of acquiring and developing both firm level and policy level productive capabilities. Arguably, this contrasts with Mozambique. The aggregate data might not quite capture the differences—MVA is a higher share of GDP in Mozambique and MVA has also grown in recent years, chiefly propelled by the massive capital intensive so-called mega-projects such as the Mozal aluminum smelter that is effectively integrated into South Africa’s minerals–energy complex (Ehrenpreis and Virtanen 2007). Manufactures make up a tiny, and by contrast with Ethiopia stagnant, proportion of total export revenue. Government investment has largely stabilized in Mozambique since the mid-1990s by contrast with the increase in the same period in Ethiopia (Figure 25.7). Given that even IMF economists, (Ghazanchyan and Stotsky 2013) argue that public investment is more important than private investment in securing sustained growth, this is a worrying trend for Mozambique. Ethiopia’s investment in human capital as a base for structural change is partly captured by evidence (Figure 25.8A–C) that it has committed significantly more to secondary and tertiary education than has Mozambique. Interestingly, the heavy emphasis on secondary school education is what really distinguished the successful East Asian countries in the 1960s and the 1970s from the SSA economies (see the “initial conditions” chapter in Chang 2002). Figure 25.8D shows a dramatically higher number, and faster growth, of tertiary graduates in engineering, science, manufacturing and construction in Ethiopia than Mozambique. More detailed research suggests a lack of clear strategic commitment in Mozambique to the kind of fundamental structural change—especially led by industrialization and

13  The same is very much true of floriculture, which grew from virtually nothing in early 2000s to be one of the top five global exporters.

Tigers or Tiger Prawns?    497 1,400

Millions (US$)

1,200

Ethiopia Mozambique

1,000 800 600 400 200 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

0

Figure  25.6   Gross value added in construction (constant 2005  US$). Source:  UN Statistical Database.

20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

0%

Ethiopia

Mozambique

Figure  25.7  Government investment (Public Gross Fixed Capital Formation as a % of  GDP). Source:  World Bank’s World Development Indicators  2013.

manufactured exports—that this chapter has suggested is crucial to the prospects of converting the growth episode labeled “Africa Rising” into long-run sustained development. There was a renowned policy debate in the early post-war years in the 1990s about the appropriate policy towards the cashew nut—and cashew processing—sector. The government had no coherent strategy or way to respond to heavy external pressure rapidly and radically to liberalize the sector and, as a result, some argued that it lost the chance to revive and develop its once significant cashew processing industry (Cramer 1999; Macmillan and Rodrik 2003). Survey research in the early 2000s showed a narrowing of the manufacturing base in Mozambique, a reduction to more simple processes and a reliance on old machinery, and an emptying out of policy experience and know-how vis-à-vis manufacturing (Warren-Rodriguez 2010). The evidence suggests precisely the kind of anti-developmental structural change suggested in much of Africa by Rodrik (2011).

(a) 120% 100% 80% 60% 40% 20%

2003

2005

2007

2009

2011

2005

2007

2009

2011

2001

1999

1997

2003

Ethiopia

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

0%

Mozambique

(b) 40% 35% 30% 25% 20% 15% 10% 5%

Ethiopia

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

0%

Mozambique

Figure  25.8   (A)  Gross enrolment ratio (Primary)  (%). Source:  UNESCO Institute for Statistics.

(B) Gross enrolment ratio (Secondary)  (%). Source:  UNESCO Institute for Statistics.

(c) 9% 8% 7% 6% 5% 4% 3% 2% 1%

Ethiopia

2011

2009

2007

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

0%

Mozambique

(d) 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000

Ethiopia

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

0

Mozambique

Figure  25.8  (C) Gross enrolment ratio (Tertiary)  (%). Source:  UNESCO Institute for Statistics. (D) Tertiary graduates in science, engineering, manufacturing and construction. Source:  UNESCO Institute for Statistics.

500    Historical Trajectories and Economic Landscape

25.4 Conclusion In this chapter we have argued that both the older narrative of an African growth tragedy and the more recent narrative of Africa rising are both unwarranted, preventing a more realistic approach to understanding the challenges and prospects of long-term economic development in the continent. Having revealed the critical shortcomings of the “African growth tragedy” discourse based on “meta-structural” explanations, we devoted most of the chapter to discussing the new—and equally problematic—discourse of “Africa rising.” The fact that the recent growth revival in the continent largely derives from one-off factors that are unlikely to be repeated—the end of civil wars, a resource bonanza, or Chinese resource demands—is actually less of a problem; countries will always get these kinds of lucky breaks (as well as unfortunate events). In the long run, the more worrying problem is that few African countries have used the proceeds of export growth based on natural resources, cheap external borrowing, and an influx of foreign capital in building capabilities in higher productivity activities, whether through direct subsidies or through inducements to foreign direct investments that can, when supported by appropriate policy intervention, raise the capabilities of domestic producers. We then illustrated this general point by comparing and contrasting two “success” stories of Mozambique and Ethiopia. While recognizing the complexity of the picture, we identified that Ethiopia has been more conscious of the need to go against current comparative advantage and build long-term productive capabilities, not least through the increase in manufactured exports. The palpable, if not yet dramatic, shift in the country’s economic structure makes us cautiously more hopeful of Ethiopia’s prospect of long-term economic transformation, compared to Mozambique’s, despite the fact that the latter has also made some important progress. Our chapter shows that there is nothing inevitable and automatic about sustained long-run economic growth—or the lack of it. It is in large part driven by a country’s willingness to create a national vision for long-term economic transformation based on an increase in productive capabilities and to translate that vision into pragmatic policies. Those that have arguably made unusual efforts to support structural change, like Ethiopia, have done so by resisting the relentless and often simplistic narrative and mantra of external advisers, showing a wariness of Hirschman’s foreign economic expert syndrome.

References Acemoglu, D., Johnson, S., and Robinson, J.A. (2001). The colonial origins of comparative development: an empirical investigation. American Economic Review, 91(5):1369–401. ACET (African Centre for Economic Transformation) (2013). Growing Rapidly, Transforming Slowly. Accra: ACET. Amsden, A. (2001). The Rise of the Rest:  Challenges to the West from Late Industrializing Economies. New York: OUP. Arkebe, O.M. (2013). Industrial Policy in Ethiopia:  Accounting for Variation in Policy and Performance in Cement, Floriculture and Leather Goods Sectors, PhD Thesis, Department of Development Studies, SOAS, University of London.

Tigers or Tiger Prawns?    501 Austin, G. (2008). The “Reversal of Fortune” thesis and the compression of history: perspectives from African and comparative economic history. Journal of International Development, 20:996–1027. Barcelona Development Agenda (2004). The Barcelona Development Agenda. http://www. bcn.cat/forum2004/english/desenvolupament.htm. Bloom, D., and Sachs, J.D. (1998), Geography, demography and economic growth in Africa. Brookings Papers on Economic Activity, 1998(2):207–295. Castel-Branco, C. (2008). Aid Dependency and Development:  A  question of ownership? A critical view, Working Paper 01/2008, Maputo: IESE. Chang, H.-J. (2002). Kicking Away the Ladder: Development Strategy in Historical Perspective. London: Anthem Press. Chang, H-J. (2009a). Under-explored Treasure Troves of Development Lessons – Lessons from the Histories of Small Rich European Countries (SRECs), in M. Kremer, P. van Lieshoust, and R. Went (eds), Doing Good or Doing Better – Development Policies in a Globalising World. Amsterdam: Amsterdam University Press. Chang, H-J. (2009b). Economic History of the Developed World: Lessons for Africa, in S. Tapsoba and G. Oluremi Archer-Davies (eds), Eminent Speakers Series Volume II – Sharing Visions of Africa’s Development, Tunis: African Development Bank (can be downloaded from: http://www.econ.cam.ac.uk/faculty/chang/pubs/ChangAfDBlecturetext.pdf) Chang, H-J. (2010). 23 Things They Don’t Tell You About Capitalism. London: Allen Lane. Cincotta, R. (2008). How democracies grow up: countries with too many young people may not have a fighting chance at freedom. Foreign Policy, March/April:80–82. Collier, P., and Gunning, J.W. (1999a). Why has Africa grown slowly? Journal of Economic Perspectives, 13(3):3–22. Collier, P., and Gunning, J.W. (1999b) Explaining Africa’s economic performance. Journal of Economic Literature, XXXVII:64–111. Cramer, C. (1999), Can Africa industrialize by processing primary commodities: the case of Mozambican cashew nuts. World Development, 27(7):1247–1266. Cramer, C. (2006). Civil War is not a Stupid Thing:  Accounting for Violence in Developing Countries. London: C. Hurst. De Waal, A. (2012). The theory and practice of Meles Zenawi. African Affairs, December. Deaton, A. (2010). Instruments, randomization, and learning about development. Journal of Economic Literature, 48(2):424–455. Dercon, S., and Vargas Hill, R. (2009). Growth from Agriculture in Ethiopia: Identifying Key Constraints, Report prepared as part of study of agriculture and growth in Ethiopia. http:// www.ethiopianreview.com/2010/Growth%20from%20agriculture%20in%20Ethiopia%20 %20Stefan%20Dercon%20and%20Ruth%20Vargas%20Hill%20May%202009.pdf. Devarajan, S. (2013). Africa’s statistical tragedy. Review of Income and Wealth, 59(Suppl S1):S9–S15. Doner, R., Ritchie, B., and Slater, D. (2005). Systemic vulnerability and the origins of developmental states:  Northeast and Southeast Asia in comparative perspective. International Organization, 59(Spring):327–361. Durlauf, S., and Qhah, D. (1999). The New Empirics of Economic Growth. NBER Working Paper No. 6422. Easterly, W. (2009). The anarchy of success. New York Review of Books, November 19. Easterly, W., and Levine, R. (1997). Africa’s growth tragedy:  policies and ethnic divisions. Quarterly Journal of Economics, 112(4):1203–1250.

502    Historical Trajectories and Economic Landscape Ehrenpreis, D. and Virtanen, P. (2007). Growth, Poverty and Inequality in Mozambique. Country Study No. 10. Brasilia: International Poverty Centre. EIU (2012). Ethiopia:  at an economic crossroads. http://country.eiu.com/article.aspx?articl eid=659462850andCountry=Ethiopiaandtopic=Economyandoid=659462850andaid=1#. UtlNStCozdY.gmail. Ghazanchyan, M., and Stotsky, J. (2013). Drivers of Growth:  Evidence from Sub-Saharan African Countries. IMF Working Papers 13/236, Washington: IMF. Hirschman, A.O. (1958). The Strategy of Economic Development. New Haven:  Yale University Press. IMF (2013). Regional Economic Outlook: Sub-Saharan Africa—Keeping the Pace. Washington: IMF. Jackson, P.T. (2006). Civilizing the Enemy:  German Reconstruction and the Invention of the West. Ann Arbor: University of Michigan Press. Jerven, M. (2011). The Quest for the African Dummy:  Explaining African Post-Colonial Economic Performance Re-Visited. Journal of International Development, 23: 288-307. Jerven, M. (2010). African Growth Recurring: an economic history perspective on African growth episodes. Economic History of Developing Regions, 25(2):127–154. Jerven, M. (2013). Poor Numbers: How We Are Misled by African Development Statistics and What to Do About It. Ithaca: Cornell University Press. JICA (2011). Ethiopia’s Industrialization Drive under the Growth and Transformation Plan. Ch. 3 in Intellectual Partnership for Africa: Industrial Policy Dialogue between Japan and Ethiopia. JICA and GRIPS Development Forum, December 2011. http://www.grips.ac.jp/ forum/newpage2008/I&P_for_Africa.htm. Jones, S., and Tarp, F. (2013). Jobs and Welfare in Mozambique. WIDER Working Paper 2013/045. Helsinki: UNU/WIDER. Kenny, C., and Williams, D. (2001). What do we know about economic growth? Or, why don’t we know very much? World Development, 29(1). Lin, J.Y., and Chang, H.-J. (2009). Should industrial policy in developing countries conform to comparative advantage or defy it? A  debate between Justin Lin and Ha-Joon Chang. Development Policy Review, 27(5):483–502. Lonsdale, J. (2012). Ethnic patriotism and markets in African history, in H. Hino, J. Lonsdale, G. Ranis, and F. Stewart (eds), Ethnic Diversity and Economic Instability in Africa: Interdisciplinary Perspectives. Cambridge: Cambridge University Press, pp. 19–55. Macmillan, M., Rodrik, D., and Welch, K. (2003). When economic reform goes wrong: cashew in Mozambique. Brookings Trade Forum, 97–151. Markakis, J. (2011). Ethiopia: the Last Two Frontiers. Woodbridge: James Currey. McKinsey Global Institute (2010). Lions on the Move:  the progress and potential of African Economies. New York: McKinsey Global Institute. Mosley, P., Subasat, T., and Weeks, J. (1995). Assessing Adjustment in Africa, World Development. 23(9):1459–1473. North, D.C., Wallis J.J., and Weingast B.R. (2009). Violence and Social Orders: a conceptual framework for interpreting recorded human history. Cambridge: Cambridge University Press. Ocampo, J.A., Rada, C., and Taylor, L. (2009). Growth and Policy in Developing Countries: A Structuralist Approach. New York: Columbia University Press. Ohno, K. et al. (2008). Intellectual Partnership for Africa: Industrial Policy Dialogue between Japan and Ethiopia (Dec. 2011), Japan International Cooperation Agency (JICA) and GRIPS Development Forum. http://www.grips.ac.jp/forum/newpage2008/IandP_for_Africa.htm.

Tigers or Tiger Prawns?    503 Paap, R., Franses, P.H., and van Dijk, D. (2005). Does Africa grow slower than Asia, Latin America and the Middle East? Evidence from a new data-based classification method. Journal of Development Economics 77:553–570. Rodriguez, F. (2007). Policy makers beware: the use and misuse of regressions in explaining economic growth, Policy Research Brief No. 5, International Poverty Centre. Rodrik, D. (2011). The Future of Economic Convergence, paper prepared for the 2011 Jackson Hole Symposium of the Federal Reserve Bank of Kansas City, August 25–27th. Shiferaw, A., Soderbom, M., Siba, E., Alemu, G. (2013). Road Infrastructure and Enterprise Dynamics in Ethiopia. Working Paper no.  128, March 2013. College of William and Mary: Department of Economics. Smits, J.-P. (2006). Economic Growth and Structural Change in Sub-Saharan Africa during the Twentieth Century: New Empirical Evidence, Helsinki, International Economic History Association Conference. Solow, R. (2007). The last 50 years in growth theory and the next 10. Oxford Review of Economic Policy, 23(1):3–14. Spence Commission (2008). The Growth Report: Strategies for Sustained Growth and Inclusive Development. Commission on Growth and Development, Washington: World Bank. UN-HABITAT (2010). The Ethiopia Case of Condominium Housing: The Integrated Housing Development Programme. Nairobi: United Nations Human Settlements Programme. Urdal, H. (2004). The Devil in the Demographics:  the effect of youth bulges on domestic armed conflict, 1950-2000. Conflict Prevention and Reconstruction Paper 14. Washington, DC: World Bank. Warren-Rodriguez, A. (2010). Industrialization, state intervention, and the demise of manufacturing development in Mozambique. Ch. 3 in Padayachee, V. (ed.), The Political Economy of Africa. London and New York: Routledge, pp. 266–285. Whitfield, L. (ed.) (2008). The Politics of Aid:  African Strategies for Dealing with Donors. Oxford: Oxford University Press. World Bank (1991). Accelerated Development in Sub-Saharan Africa. Washington, DC: World Bank. Zenawi, M. (2012). States and markets: neoliberal limitations and the case for a developmental state, in A. Noman, K. Botchwey, H. Stein, and J. Stiglitz (eds), Good Growth and Governance in Africa: Rethinking Development Strategies. Oxford: Oxford University Press.

Chapter 26

T he Ec onomic L e g ac i e s of t he African Sl av e  T ra de s Warren C. Whatley

26.1 Introduction The early twenty-first century has witnessed a resurgence of interest in the economic history of Africa, partly because of renewed interest in the persistence of world inequality, partly because of new data on Africa’s past. Price theory predicts convergence in the incomes of trading partners, but many economists now believe the real world is not so ergodic—that to understand how complex systems like economies perform one must investigate how they got that way. The new institutional economics, endogenous growth theory, notions of historical development, path dependence, and even the renewed interest in genetics all hark back to the past as a source of information about the present. The renewed interest in the economic history of Africa fits squarely within this research agenda. It is the site of several recent discoveries about social responses to economic shocks and how they can be transmitted across time in ways that influence subsequent economic development. As such, the modern economic history of Africa has important lessons for economists, social scientist, students and policymakers in Africa and around the world. This chapter reviews this literature as it pertains to the African slave trades. There were four African slave trades of varying lengths and intensities. The transatlantic slave trade is the slave trade best understood by Western social science, primarily because of its contributions to the emergence of the Atlantic economy and to the persistence of inequality within it. There were three other slave trades—the trans-Sahara slave trade, the Indian Ocean slave trade, and the Red Sea slave trade—each part of regional systems in the Muslim world. Sub-Saharan Africa was the origin of almost all of these slaves, and while we have quantitative counts for all four slave trades, the data on the transatlantic trade are by far the most comprehensive. So is the empirical work. This review, therefore, focuses when necessary on the transatlantic slave trade, but reviews findings for all of Africa whenever possible. What was the legacy of these slave trades for Africa? Analysis begins by considering them as shocks to Africa—shocks that changed people’s behavior in ways that influenced subsequent economic structure and performance (Acemoglu and Robinson 2010). Were

The Economic Legacies of the African Slave Trades    505 the shocks large or small? Was the trade welfare enhancing or overall unequal for Africa? What about slave “production,” technological change and externalities? Did imported firearms expand production? What about the social costs? Where there negative externalities associated with the process of enslaving others and marching them to the coast, and if so where these social costs large or small? Lastly, did participation in the slave trade have lasting consequences for the structure and performance of African economies over the long term? These are the questions I address in this essay. To summarize up front, there is a growing body of evidence that the slave trades were substantial shocks to African economies with long-term negative consequences for economic growth. Even along the Gold Coast of West Africa, one of the most-developed regions of sub-Saharan Africa, the slave trade by the eighteenth century had become the dominant economic opportunity, and by a wide margin. Still, the international slave trade was a welfare loss for Africa, which begs the question “why were so many slaves exported?” The main reason is that slave production is organized theft, so by its very nature it generates negative externalities and overproduction. Overproduction was further encouraged by a major technological change in the seventeenth and eighteenth centuries—the imported gunpowder technology. Firearms imports reduced supply-cost by instigating regional arms races that compelled communities to “raid or be raided.” In the ensuing conflicts, large numbers of people were enslaved and shipped out of Africa. In Africa, however, people fought back, reconfiguring the ways they interacted with each other—in and between families, communities, ethnicities, and states. The general tendency was towards political decentralization as people and communities found ways to escape capture. In the process, the slave trade encouraged the development of local chiefly capacities that exercised absolutist rule over their people. The slave trade also encouraged the spread of polygyny and a culture of mistrust—all of which reduced long-term growth. These legacies remain developmental challenges for much of sub-Saharan Africa today. In the next section I review the literature on the slave trade as a shock to Africa, its magnitude and its short-term consequences. Next, I review the literature on long-term effects, beginning with a discussion of data and methodological issues, and then moving on to discuss the literature’s findings. I conclude with a brief discussion of future directions.

26.2  The Short-Term Effects of  the International Slave Trades 26.2.1 Quantity Sub-Saharan Africa was the southern frontier of the medieval Mediterranean and Eurasia worlds. Italian city-states, Arab caliphates, and European kingdoms all used conquest and trade to obtain treasure, land, and slaves. Muslim and Christian armies fought “just” wars that justified the enslavement of non-believers, and Roman and Islamic law contained well-developed civil codes that regulated the legal status of slaves (Patterson 1982). The trans-Saharan trade, the Indian Ocean trade, and the Red Sea slave trade were all part of these systems. By the middle of the fifteenth century the development of the lateen sails

506    Historical Trajectories and Economic Landscape enabled Portuguese caravels to sail down the west coast of Africa as well, initiating the transatlantic slave trade. The best records are for the trans-Atlantic trade.1 Modern empirical research began with the publication of Philip Curtin’s book The Transatlantic Slave Trade: A Census. It offered the first rigorous estimate of the numbers and general characteristics of the trade. Since that time a cooperative global effort has digitized the historical records of more than 35 000 transatlantic slave trade voyages. The Transatlantic Slave Trade Database (TSTD) will be discussed in detail in the next section. It estimates that more than 13 million people were taken from Africa and transported across the Atlantic to the Americas. Is this a large or small number? One way to answer this question is to estimate the share of population this represents. A widely used estimate of African population during this period is found in the Atlas of World Population History (1978). Of the 13 million slaves exported, 77  percent (10.1  million) originated along the west coast of Africa during the 150  years between 1701 and 1850. In 1700 the estimated population of this region was 28 million people (McEvedy and Jones 1978: 241–256). If the average life span was 30 years, then the 10.1 million slaves were produced over five lifetimes. That yields 2.6 million slaves produced per lifetime, or 9.3 percent of the total population. Collateral damage during conflicts could double this figure.2 According to these numbers an individual faced a 10–20 percent chance of being a casualty of the slave trade over his lifetime. Depopulation or a reduced rate of population growth does not necessarily lead to long-term underdevelopment. Ever since Domar (1970) we have understood that the developmental effect of a negative population shock (like the African slave trade or the European Plague) depends on the balance of political power. Increased labor scarcity in a competitive environment will unambiguously raise the opportunity cost of labor. If labor holds the balance of power vis-à-vis landlords and employers, then labor will benefit from higher wages, lower rents, and possibly greater political clout. However, if landlords hold the balance of power then they can cooperate politically to restrict labor mobility and depress the wage below its competitive level. Possible political outcomes include slavery and serfdom. Absolutist politics can choke off market pressures that could improve the standard of living of middle classes.3 The slave trade in Africa did not strengthen African middle classes. If anything, it increased inequality within societies and across societies. Some societies became successful slave raiders at the expense of the raided. The entire literature on the “transformation of slavery” in Africa is evidence that slave exports did not improve the welfare and political power of the laboring classes. While it is controversial just how widespread slavery in Africa was prior to the international slave trades, it is beyond dispute that it exploded during the era of the transatlantic slave trade.4 Some scholars estimate that by the time the international slave trade ended in the middle of the nineteenth century, 50 percent of West Africa’s population was enslaved. European nations justified their imperialist ambitions in Africa with a moral 1 

The best estimates of the trans-Saharan trade are found in Austen (1979). See Manning (1990) for regional estimates of the impact of the slave trade on population growth rates. 3  The Domar Hypothesis figures prominently in the economic history of Africa. For example, see Hopkins (1973); Ilifffe (2007); Austin (2008); and Fenske (2013). 4  The definitive work is Lovejoy (1983). 2 

The Economic Legacies of the African Slave Trades    507 obligation to eradicate slavery in Africa. Whatley (2014) presents evidence that the slave trade increased absolutist politics and reduced democracy.

26.2.2 Slave prices A higher international slave price also increases social inequality. Slaves originated in capture, so the subsequent prices paid measure a transfer of income and wealth from one society to another or one family to another or one class to another, without compensation. We are interested in slave prices paid on the coast of Africa because we want to understand how the international slave trade influenced the economic and political decisions of Africans. To address issues of resource reallocation within Africa we also need a domestic commodity price, one that indicates the relative profitability of slaving in Africa. African price data are scarce and require the imaginative use of available price data. Figure 26.1 graphs an estimate of real slave prices paid by British merchants operating on the coast of Africa, along with the real value of an ounce of gold in the city of London.5 Together these two prices are good indicators of the relative profitability of slaving or gold mining along the Gold Coast of Africa. The Gold Coast was one of the most developed regions in West Africa, exporting gold northward across the Sahara desert long before Europeans arrived by sea (Wilks 1982). The Portuguese sailed down the west coast of Africa in the mid fifteenth century looking for the source of the trans-Saharan gold. According to Figure 26.1, the fifteenth and sixteenth centuries were prosperous years for African gold producers. The trade was so regular that by the seventeenth century a rule of thumb had emerged among British gold traders operating on the Gold Coast. Goods worth £2 purchased in London would barter for an ounce of gold on the Gold Coast of Africa, or about half the £4.25 official price of gold set by the Master of the Mint in 1717 (Johnson 1966). By the beginning of the eighteenth century the real price of slaves had surpassed the real value of gold. By the end of the eighteenth century the real price of a slave was 12.5 times higher than the real price of gold. The relative international price of slave must have been even higher in other, less prosperous, regions of Africa.

26.2.3  The supply response What was the supply response to the increasing profitability of the international slave trade? Did Africans respond to the price increases in a profit-seeking manner, or were slaves produced as a byproduct of political conflicts internal to Africa? This debate has a long and contentious history, dating as far back as the British Abolition Movement of the eighteenth century.6 Whatley and Gillezeau (2011a) conducted a test. Following Curtin (1975) and 5  British slave prices (gross value of British exports to Africa—gross value of British imports from Africa)/(slave exports on British ships) in 1699 prices. Import and export values come from Johnson (1990). Real gold prices in London = the London mint price of gold/consumer price index, in 1699 prices. The figure shows the gold time series with two different deflators—Clark (2013); and Brown and Hopkins (1956). Gold prices are ten-year moving averages. 6  Abolitionists like Thomas Clarkson (1839) and Alexander Falconbridge (1788) argued that the British demand for slaves incited Africans to make war for captives. Proponents argued that slavery was an old established institution in Africa and that Africans were a war-like people before British meddling.

508    Historical Trajectories and Economic Landscape Real British Gold Prices and Slave Prices (in 1699 prices) 40

British Gold Prices (Brown-Hopkins deflator) British Gold Prices (Clark deflator) British Slave Prices

35

Pounds sterling

30 25 20 15 10 5 0 Year

1306

1356

1406

1456

1506

1556

1606

1656

1706

1756

1806

1856

1906

Year

Figure  26.1   Real British gold prices and slave prices (in 1699 prices). LeVeen (1975), if the supply of slaves were independent of international demand then the elasticity of slave exports with respect to the level of international demand would be zero. African conflicts would bring forth a politically generated ex ante supply. Price would simply allocate that exogenous supply among the competing slave ships docked off the coast of Africa at any point in time. This is the political model of slave supply where the supply curve is perfectly inelastic with respect to price. On the other hand, if international demand incited African societies to make war and kidnap, then increases in the level of international demand would bring forth additional supply. The supply curve will have some elasticity with respect to the level of international demand. Revenues per captive will increase, attracting more kidnappers, reducing the cost of military campaigns and covering additional costs of searching further inland for captives. Using a time series of British cargo shipments to Africa and slave exports on British ships, Whatley and Gillezeau (2011a) estimated a short-run elasticity of slave exports with respect to the real value of imports of 0.43. This is lower than existing estimates of long-run price elasticity, which fluctuate around 1.0.7

26.2.4  Trade and welfare Gemery and Hogendorn (1979) estimated the welfare gains and losses for Africa. They do this for the eighteenth century transatlantic slave trade. Their prior is that the trade reduced welfare, so they bias their estimates against their prior by overestimating gains and

7 

For other elasticity estimates, see Gemery and Hogendorn (1977); LeVeen (1975); and Curtin (1975).

The Economic Legacies of the African Slave Trades    509 underestimating losses. Gains emanate from new opportunities to trade slaves on the lucrative international market in exchange for imported goods from Europe and the America. They estimated the welfare gain to be equal to the total value of imported goods, which overestimates the actual increase to African GDP by the value of the prior stock of slave. Using slave quantity from Curtin (1969) and slave export prices from Bean (1974), they arrived at a maximum gain of £79.8 million for the eighteenth century Atlantic slave trade. They estimated losses as the subsistence incomes that the exported slaves would have produced, ignoring all other goods they would have produced like foods for urban populations and any other production above subsistence. Their review of the literature finds subsistence figures of £.8 to £1.2 per person per year for the eighteenth century. They estimate the years of production lost per slave export to range between 15 and 20 years. Estimates of fatalities during conflicts range between 20–50 percent of exports. Using Curtin’s estimate of 4.25 million slaves exported in the eighteenth century and the lower fatality rate of 20 percent, they estimated a minimum loss of £56–£84 million. The resulting maximum net gain from the slave trade was £23.8 million and was probably much lower. In addition, some of the imported goods were used to produce slaves and so should be counted as a cost of production not a gain from trade. Eltis and Jennings (1988) estimated firearms imports at 8  percent of import value for the eighteenth century. The Anglo-African Trade Statistics collected by the British Customs House estimated eighteenth century British gunpowder shipments to be another 8 percent of import value (Johnson, Lindblad, and Ross 1990). Together, eighteenth-century firearms and gunpowder came to approximately £13.5 million, reducing the maximum gain to £10 million for the entire eighteenth century slave trade, or £100 000 per year for all of Africa. This is an overestimate by construction. The slave trade reallocated income from victims to victors and from commoners to elites, but it was not an income bonanza for Africa.

26.2.5  Externalities and overproduction The obvious question then becomes “why were so many slaves exported?” The equally obvious answer is “because producers did not internalize the total cost of production.” Slaves were not produced, they were taken not from one’s own people (until late in the game) but from other peoples. Gains went to the victors, losses to the victims. Thomas and Bean (1974) likened this to a fishery where no fisherman has a property right to the fish, and so no one fisherman internalizes the full cost of catching an additional fish. Overproduction and depletion of the fishery is the expected outcome. Gemery and Hogendorn (1974) identify another source of overproduction: a prisoner’s dilemma arms race of “raid or be raided.” They put it this way: States playing no role in the slave trade, and therefore not receiving muskets in payment for slaves, found themselves on the losing side of an arms race. Their dilemma: without firearms defense was precarious. To get muskets, there must be something to export. The only item in great demand was slaves. Thus, it is not surprising that slave trading spread rapidly, especially in the eighteenth century when flintlock replaced the cumbersome matchlock. (Gemery and Hogendorn 1974: 242)

510    Historical Trajectories and Economic Landscape The following quote, from the Dutch Director General at Elmina Castle in 1730, describes what was happening: The great quantity of guns and gunpowder which the Europeans have brought have caused terrible wars between the Kings and Princes and Caboceers of these lands, who made their prisoners of war slaves; these slaves were immediately brought up by Europeans at steadily increasing prices, which in its turn, animates again and again these people to renew their hostilities, and their hope of big and easy profits makes them forget all labor, using all sorts of pretexts to attack each other or revive old disputes. (quoted in Richards 1980: 46)

Clear winner and losers emerged in the eighteenth century, especially along the lower Guinea coast where the new firearms technology was most effective (Kea 1971; Thornton 1999): Ashanti along the Gold Coast (Daaku 1970; Wilks 1975), Dahomey in the Bight of Benin (Law 2004, 1989, 1991), and the Aro trading network in the Bight of Biafra (Dike 1956; Olaudah and Allison 1995; Northrup 1978). Fenske (2013) presents evidence that negative climate shocks after 1830 reduced economic capacity to capture and export slaves. By the late eighteenth century Asante and Dahomey had become pass-through states that were able to extract tribute or toll as output flowed from the interior to coastal ports (Evans and Richardson 1995). The Aro trading network in the Bight of Biafra mediated the contradiction between violent production and orderly trade, using force, kinship, and religion to instigate small-scale conflicts and the flow of slaves into an orderly trade to the coast.

26.3  Long-Term Effects of the International Slave Trades The international slave trades were a significant shock to sub-Saharan Africa. At the height of the trans-Atlantic trade an individual living on the west coast of Africa faced a 10–20 percent chance of being a casualty of the trade. At the family level, Dalton and Leung (2014) and Fenske (2013) show that gender selection in the slave trades (being male-dominant in West Africa and female-dominant in East Africa) increased long-term polygyny rates in West Africa. The simulated relationship between polygyny, fertility, and savings presented in Tertilt (2005) shows that a ban on polygyny in sub-Saharan Africa would decrease fertility by 40 percent, increase savings by 70 percent and increase output per capital by 170 percent. At the macro-level the international trade raised the level of violence, reallocated resources toward slaving and changed the dynamics of politics and state formation. At the local level, it forced individuals to create new forms of protection (Diouf 2003). In the past ten years, economists have made great strides in identifying the long-term legacies of these macro- and micro-adjustments. In the remainder of this section I discuss this literature. First I discuss the data and methods used to trace the long-term effects of the slave trade over time and then I discuss the findings that are based on these data. Much of this discussion is preliminary because much of this research is new and still in working papers, but the empirical results across a growing number of studies are similar enough to warrant generalizations.

The Economic Legacies of the African Slave Trades    511

26.3.1 The data 26.3.1.1  The Transatlantic Slave Trade Database (TSTD), c. 1500–1860 As early as the 1960s, historians began collecting archival records on slave voyages across the Atlantic. By the late 1980s several independent efforts had collected records on approximately 11 000 voyages in 16 separate datasets. Starting in 1991 efforts to coordinate collection benefitted from a series of significant grants from a variety of sources. The resulting dataset is now generally available for easy query and download at www.slavevoyages.org. The final 2010 version of the database contains data on 34 948 transatlantic slave voyages with 276 variables for each voyage. Variables include the dates of travel, port of origin, national flag, African port of embarkation, the numbers of slaves embarked, port of disembarkation in the Americas, and much, much more. All of the studies reviewed in this section use the TSTD to measure the intensity of the slave trade shock.

26.3.1.2  The Ethnographic Atlas (EA), c. 1900–1920 The Ethnographic Atlas was initially published in the journal Ethnology in 29 installments from January, 1962, to July, 1980, by George P. Murdock. The goal was to create a quantitative database from the existing body of ethnographic monographs. Murdock used both primary sources—ethnographic and anthropological studies based on direct fieldwork—and secondary sources, which were themselves based on the archives of missionary societies, as well as field notes and special communications by anthropologists and others. The final dataset contained data on 1267 societies worldwide and was digitized and published in 1999 by J. Patrick Gray. There are over 100 variables in the Ethnographic Atlas. Most are categorical codes created from descriptions in the ethnographies—like whether or not the observed group practiced slavery, whether or not the dominant form of marriage was monogamy or polygyny, the degree of political hierarchy, inheritance patterns, political structure, sex taboos, and more. Murdock also mapped the locations of these societies Most studies of the long-term effects of the slave trade use this map to merge the transatlantic slave trade data (c. 1500–1865) with chronologically later sociological date like that in the EA (c. 1910). GIS software is used to link TSTD data and EA data using the spatial markers common to each. Once linked, the goal is to look for correlations between spatial variations in the intensity of slaving and spatial variations in subsequent social and economic outcomes recorded in the EA.

26.3.1.3  Late twentieth-century data The legacy of the slave trade can be traced beyond the EA by linking the TSTD to any contemporary data with spatial markers. The TSTD and the EA have been linked to late twentieth-century data on post-colonial nation-states. This is done by aggregating the EA homelands into areas that match the boundaries of post-colonial states. The TSTD and the EA have also been linked to contemporary survey data. The most common is the AfroBarometer Survey (http://www.afrobarometer.org/). They have also been linked to data from satellite images of light density at night, which has recently become a useful proxy for income in parts of the world that lack better income estimates. Together these linked data

512    Historical Trajectories and Economic Landscape form a series of spatially defined “snapshots” over time. Research on the legacies of the slave trade typically looks for correlations and causal relationships between spatial variations in the intensity of the slave trade shock in the past and spatial variations in subsequent social and economic outcomes.

26.3.2  Income and growth The modern line of research begins with Nathan Nunn (2008), who looks for cross-sectional correlations between state-level variations in slave exports per acre and state-level variations in income per capita in the year 2000. He collects available estimates of slave exports for the trans-Atlantic, trans-Saharan, Indian Ocean, and Red Sea slave trades. The spatial unit of analysis is the post-colonial nation-state. The only spatial identifier in the TSTD is the slave port of embarkation. The challenge is to allocate some of the coastal exports to interior state. Nunn accomplishes this by combining the slave export data with two additional sources of information. The first is the Murdock map. The second is a collection of 54 samples of slaves in the Americas that recorded ethnicities, representing 84 656 slaves with 229 distinct ethnic designations. Using the Murdock map, Nunn then determines the percentages of slaves in the Americas that came from coastal ethnicities and interior ethnicities. The ratio of coastal to interior slaves found in the American samples is used to allocate the slaves exported from TSTD ports to coastal and interior ethnicities in the Murdock map. These ethnic homelands and their slaves are then summed into territories that match the boundaries of post-colonial African states. Nunn does the same for the trans-Saharan, Indian Ocean, and Red Sea trades. State-level estimates of average per capita income in 2000 are then regressed on state-level estimates of slave exports per acre. These regressions reveal a robust negative correlation between slave exports per acre and income per capita in the year 2000, even after controlling for a variety of other factors like latitude and longitude, climate, coastline, religion, legal origin, colonizer, and natural resources. Reverse causality is possible, so identification strategy is important (Angrist and Pischke 2009). Nunn cannot conclude from ordinary least squares (OLS) regressions that the slave trade caused the lower incomes in 2000 because the intensity of slaving could have selected poorer regions first and foremost. To identify a causal effect running from past slaving to current income we need estimates of the unobserved counterfactual incomes for 2000—what incomes would have been had there been no slave trade. A first approximation would be incomes before the coming of the slave trade, or better still trends in prior income that could then be extrapolated to the present. These are not available. Population in 1500 is often used to measure prior development, but the available population estimates are extrapolations of twentieth-century counts back in time and contain endogenous information on later outcomes.8 Another approach is to use early travelers’ accounts and their observations on the material and political conditions of the empires that selected into the slave trade early. If more-developed regions entered first, then at least the temporal selection bias worked

8  The 1500 population estimate for Africa reported in McEvedy and Jones (1978) are primarily back-projections of colonial population estimates found in Kuczynski (1948, 1949). McEvedy and Jones do not discuss how they back-project.

The Economic Legacies of the African Slave Trades    513 against poverty selection. If the poorer areas entered the slave trade first, then OLS estimates contain both pre-trade poverty and trade-induced poverty. One of the earliest comparative observations comes from the diary of the travels of Vasco da Gama (1998), the first Portuguese explorer to successfully round the Cape of Good Hope in 1497–1498. The diarist, most likely Alvaro Velho (p. 18)9 notes that the West African coast was noticeably less developed than the East African coast, primarily because East Africa had been trading extensively in the Indian Ocean and Red Sea. Likewise, the trans-Saharan trade in the Western Sudan, which supported a series of empires from Ghana to Mali to Songhai, traded towards the Muslim world not the Atlantic. In 1500 the Atlantic coast was a backwater. In a global sense, the transatlantic slave trade touched poorest Africa first.10 Within West Africa, however, the selection bias was different. West Africa may have been poorer than East Africa but several well-developed empires did exist on the west coast, and these kingdoms were the first to trade slaves in large numbers—Wolof in the Senegambia region, Benin along the Guinea Coast, and Kongo in south-west Africa (Thomas 1997). These were empires based on inland river trades, and were anchored on the Atlantic outlets of the great rivers of Africa—Wolof and Benin at opposite ends of the Niger River and Kongo on the Congo River. The historical record shows that these were the West African kingdoms first visited by early trans-Atlantic slave traders and are often taken as evidence that the richest areas selected into the slave trade first. This is true, so long as the discussion is confined to the Atlantic coast of Africa. While it is impossible to reconstruct the pre-shock development trajectories of these early West African kingdoms, their responses to the shock are the first clear indications of how the slave trade altered African paths of development. The elites of these kingdoms initially participated in the slave trade, selling war captives and other victims to Europeans. The trade proved lucrative but also disruptive to social order. Benin and Kongo fought to withdraw from the trade (Egharavba 1968; Ryder 1969; Thornton 1983) and Wolof simply fell apart under its weight (Barry 1998). This does not mean that the richest areas of Africa were hit hardest by the slave trade, but as case studies they do suggest that the slave trade had a destabilizing effect on pre-existing paths of development. Case studies are informative but rare, so instrumental variables estimation has become the most commonly used econometric technique for identifying the causes of general trends. To continue with Nunn (2008), his instruments are the proximity of African regions to the major destinations for slaves around the world. Correlations between these distances and slave exports pick up demand-side influences on slave exports and are independent of the economic and political characteristics of the African suppliers. The closer an African society is to one of these destinations, the greater the demand intensity, regardless of the society’s level of development. Nunn then looks for correlations between these demand-generated

9  Page 18 reads “… the ‘diary’ is not the original document made by the diarist, but by a copyist in the 16th century. This copy came to light in the convent of Santa Cruz in Coimbra … and published in 1838 …” 10  One might conclude from this that since East Africa was richer and was trading slaves before Vasco Da Gama arrived then even globally the slave trade touched the richest parts of Africa first. This would be true if and only if the accounts of early Arab traders confirmed it. The Vasco Da Gama diarist’s observations are too late for East Africa.

514    Historical Trajectories and Economic Landscape estimates of slave exports and income per capital in 2000. He finds that this causal relationship is still negative and statistically significant.11

26.3.3  Domestic institutions How could the slave trade depress income growth over such a long period of time? According to Acemoglu, Johnson, and Robinson (2002) the most important channel would be domestic institutions (Acemoglu et al. 2001; Acemoglu and Robinson 2012). Their view is that during the age of discovery, the high mortality rate of Europeans settlers in Africa (due to diseases like malaria and yellow fever) discouraged settlement and encourage the establishment of extractive institutions like the slave trade.12 If Europeans had settled in Africa then they would have fought to establish domestic institutions that would have protect their property rights and their trade. Extractive trade, by contrast, seizes upon and transforms existing social relations to facilitate the export of goods in high international demand. Once established, indigenous elites who benefit from extraction have every incentive to block the development of more inclusive political and economic institutions, institutions that increase growth but also increase competing claimants to political power. The slave trade was certainly an extractive institution, but African peasants were not passive reactors. Unable to purchase insurance against capture they often devised other forms of protection. The most important was a common defense. Here the current literature on state capacity is instructive. Besley and Persson (2010) model the relationship between conflict, state capacity and development. In their model state capacity (the capacity to tax and spend on public goods) depends crucially on the demand for common-interest public goods like defense. If the demand for defense is high, then elites are free to tax all citizens maximally. There is little incentive to redistribute revenues to constituents because in times of war political opposition is weak. In times of peace, on the other hand, opposition can be substantial, forcing parties to compete for political supporters by redistribute revenues to constituents. Over time, a tradition of political competition can lead to political openness and accelerated economic growth.13 Research on the long-term effects of the slave trades tends to supports these views, but with an important caveat. The slave trade strengthened extractive institutions in Africa and the slave trade increase absolutist state capacity but primarily at the local level. By raising the price of slaves relative to the price of land the international slave trade reduced the 11 

There are major critiques of this study, as expected with a seminal paper. The TSTD records the vast majority of slave exports, but the data used to allocate the coastal exports to interior territories are thin. Errors in the Murdock map probably increase as one travels inland and the American samples are not a random sample of Africans in the Americas. The F-statistics for the first-stage could be larger. For criticisms, see Bhattacharyya (2009) and Bottero and Wallace (2013). For the most part, the errors in the data should contribute to attenuation bias which should bias the relationship with income towards zero. More importantly, many subsequent studies have used Nunn’s estimates of slaving intensity and found them to be correlated with many other long-term outcomes like political institutions, family structure and culture. Many of these studies are discussed below. 12  Again, as with seminal articles, this body of research and in particular the estimates of settler mortality have come under criticism. See Albouy (2012) and Acemoglu et al. (2012). 13  North, Wallis, and Weingast (2009) and Acemoglu and Johnson (2012) make similar arguments.

The Economic Legacies of the African Slave Trades    515 optimal territorial reach of political authority (Whatley and Gillezeau 2011a). Higher slave prices increased the return to slaving and decreased the return to growing a peasantry to be taxed. The rising level of conflict also increased the cost of protecting citizens, further reducing the incentive to grow a taxable citizenry. Political development took a decidedly local character. Several recent studies report evidence in support of this view. Osafo-Kwaako and Robinson (2013) use data from the Standard Cross Cultural Sample, another worldwide sample of ethnicities developed by Murdock, to test standard models of Eurasian state formation, which emphasize the importance of population density, inter-state warfare and trade. They find that these factors are positively correlated with political centralization outside of Africa but not in the African sub-sample. Whatley and Gillizeau (2011b) found that the slave trade reduced the territorial size of the societies found in the Ethnographic Atlas and in the Peoples Atlas. Nunn and Puga (2012) found that in Africa today, economic activity increases with the ruggedness of terrain whereas outside of Africa the opposite is the case. They interpret this as evidence that slaving in Africa dispersed populations and turned ruggedness into a defensive advantage. Obikili (2014) found that the slave trade increased political hierarchy in the early twentieth century but only at the local level. Whatley (2014) presents evidence that the slave trade strengthened absolutist chiefs at the local level, as measured by the rules governing political ascension that are recorded in the Ethnographic Atlas. Together, this body of research suggests that the slave trade—in particular the violence and conflict involved in capturing people—set in motion a process of political decentralization and the clustering of people into relatively isolated communities. It is too early to know if the slave trade actually decentralized societies or simply slowed the rate of political centralization. There are examples of decentralization—like Wolof, Kongo, and Benin—but the most likely general tendency is exemplified by the societies inland of the Bight of Biafra. Here is a region of Africa that remained politically decentralized throughout the slave trade era despite being one of the most densely populated, war-like, trading areas in West Africa. Instead of coalescing into a centralized kingdom, one finds villages continually making war on each other, a religious Oracle condemning people to slavery for the slightest infraction, and widespread kidnapping and slave trading to the coast. The transfer of captives to the coast is ordered by a tightly knit network of traders called the Aro, who ruled the region by force of religion, kinship connections, and violence. So long as neighbors were “others” then they were subject to capture. Political decentralization facilitated the “production” of slaves. The Aro network had every incentive to monopolize trade but no incentive to centralize the region politically. Political centralization would have minimized “otherness” and removed much of the pretext for the kinds of conflicts that produced large numbers of slave for exports. There is also evidence that this kind of political decentralization impeded economic development well into the late twentieth century. Gennaioli and Rainer (2007) reported OLS regressions showing a negative correlation between political decentralization, as measured in the Ethnographic Atlas, and public goods today like roads and infrastructure. Osafo-Kwaako and Robinson (2013) found the same negative correlation between decentralization in the past and poorer public goods and development outcomes today. Michalopoulos and Papaioannou (2013) showed that the same measure of political decentralization is the only variable in the Ethnographic that is statistically correlated with underdevelopment today, income being measured by satellite data on night-light density.

516    Historical Trajectories and Economic Landscape Obikili (2014) finds that politically decentralization areas in the EA exhibit higher levels of corruption today. Nunn and Wantchekon (2011) identified another long-term effect of slaving—mistrust. They link data from the TSTD to respondents’ answers to questions about trust in the AfroBarometer survey of 2008. They find that individuals from ethnicities and regions heavily shocked by the slave trade are less trusting of their neighbors today and less trusting of their political institutions as well. The size of the effect is small, probably because trust is also a fundamental ingredient in the making of domestic institutions. In a recent paper, Nunn (2014) presents evidence that slave-induced mistrust is weaker in states with strong rule-of-law and democratic institutions. He argues that the causality probably runs both ways—high trust societies tend to choose open political systems and open political systems tend to encourage and enforce trusting behavior.14

26.3.4 Colonialism What about colonialism? Long-term effects of the slave trade must pass through colonial occupation (approximately 1885–1960). Until recently, colonialism was considered to be the shock that fundamentally altered African economies, with the slave trade lumped together with a sort of primordial precondition called “precolonial Africa.” In the new slave trade literature a similar oversight is sometimes applied to colonialism.15 Future research will have to address this issue. Did the slave trade’s effect on decentralization influence the structure of colonial institutions? Did the institutions of colonial rule exacerbate or ameliorate the effects of the slave trade? These are frontier issues in the economic history of Africa. Early research suggests that the slave trade influenced the shape of colonial institutions, encouraged indirect colonial rule, and strengthened the authority of local absolutist chiefs. At the national level, Whatley (2014) reports a positive correlation between participation in the slave trade and the degree of indirect colonial rule. The measure of indirect rule is the percentage of court cases adjudicated in customary courts. Lange (2009) showed that this same measure of indirect rule is also negatively correlated with weak rule-of-law in postcolonial African states. Nunn (2014) presented evidence that weak rule-of-law in post-colonial African states allows the persistence of slave-induced mistrust. At the sub-national level, Obikili (2014) reported evidence that the slave trade fragmented political authority and strengthened the authority of local chiefs. Whatley (2014) presents evidence that the slave trade encouraged the spread of inherited local chieftancy. Acemoglu, Reed, and Robinson (2014) presented evidence that in colonial Sierra Leone the British colonial administration strengthen the local authority of chiefs, with detrimental long-term effect for a variety of local outcomes today, including literacy, school attainment, and health.

14 

On self-enforcing institutions and culture, see Grief (2006). Perhaps this is because of confusion about the dates of observations in the Ethnographic Atlas. The tendency is to interpret the data as a snapshot of precolonial Africa, but the modal date of observation is the 1920s, with 25 percent of the observations. 10 percent of the observations are as late as 1960. 15 

The Economic Legacies of the African Slave Trades    517 Finally, several papers have shown that the long-term effect of the slave trade on national income lay dormant until after colonial rule. Nunn (2008) established the negative relationship between past slaving and income per capita in the year 2000. In the same article he reports post-1950 income series high and low slave-exporting countries. High slave-exporting countries are always poorer than low slave-exporting countries, but the income gap between the two widens considerably after the late 1960s. Bottero and Wallace (2013) found a similar latent effect. When they replicate Nunn’s regressions using 1960 and 1950 income data they do not find a depressing effect of past slaving on income. They do find it for the years 1970 and 2000. Together, the macro-, micro-, and time-series evidence suggests that the slave trade’s effects on pre-colonial institutions may have mattered during colonialism but they mattered much more after independence, when local power bases were called upon to form a federal compromise. Michalopoulous and Papaioannou (2010) found that the pre-colonial boundaries of EA ethnicities explain more cross-sectional variation in post-colonial income than do post-colonial state boundaries, suggesting that local policies mattered more than national ones. Van de Walle (2001) argued that the fiscal crises of the post-colonial African nation-state is not due to excessive government spending but to nation-states’ inability to collect revenues from local authorities.

26.4 Conclusion There is a growing body of evidence that the slave trade had a lasting negative effect on the economic and political development of sub-Saharan Africa. The trade itself was not welfare enhancing. The social costs of production far outweighed the gain. The increased violence associated with stealing people polluted social relations. The major technological change of the time (the gunpowder technology) polluted them even more. People used a variety of mechanisms to fight back. Some moved further inland. Others moved to higher altitudes or more rugged terrain. The general tendency was towards political decentralization and the strengthening of local chiefly authority. A frontier question is how these institutions interacted with the colonial and post-colonial experience to shape the political and economic institutions and outcomes that we see in sub-Saharan Africa today? Customary authority, the salience of ethnic identity and the prevalence of polygyny are the major legacies of the slave trade. Colonialism reduced polygyny, but indirect colonial rule intensified local customary authority and ethnic identity. In fact, the maps used to draw the Murdock map of ethnic homelands, and the ethnographies that populate the EA were developed in the service of indirect colonial rule. Research on the slave trades has shown how these colonial ethnographies contain legacies of the slave trade. While many of the political customs described in the EA may predate the slave trade and colonialism, their current manifestations as local customary authority is an authority granted by the peoples to the leaders they chose to defend them from slavery. Indirect colonial rule legitimized these grants of authority and institutionalized them as part of the nation state. Research on the long-term effects of the slave trades suggests that these slave-induced local authorities have impeded economic

518    Historical Trajectories and Economic Landscape growth. 16 This remains a major challenge for African federalism and for the building of competitive state capacity at the local and federal levels.17

References Acemoglu, D., Johnson, S., and Robinson, J.A. (2002). Reversal of fortune: geography and institutions in making of modern world income distribution. Quarterly Journal of Economics, 117:1231–1294. Acemoglu, D., Johnson, S., and Robinson, J.A. (2001). The Colonial Origins of Comparative Development: An Empirical Investigation. American Economic Review, 91(5):1369–1401. Acemoglu, D., Johnson, S., and Robinson, J.A. (2012). The colonial origins of comparative development: an empirical investigation: reply. American Economic Review, 102(6):3077–3110. Acemoglu, D., Reed, T., and Robinson, J.A. (2014). Chiefs: Economic Development and Elite Control of Civil Society in Sierra Leone. Journal of Political Economy, 122(2):319–368. Acemoglu, D., and Robinson, J.A. (2010). Why is Africa poor? Economic History of Developing Regions, 25(1):21-50. Acemoglu, D., and Robinson, J.A. (2012). Why Nations Fail: The Origins of Power, Prosperity and Poverty. New York: Crown Business. Albouy, D. (2012). The colonial origins of comparative development: an empirical investigation: a comment. American Economic Review, 102(6):3059–3076. Angrist, J.D., and Pischke, J.-S. (2009). Mostly Harmless Econometrics:  An Empiricist’s Companion. Princeton: Princeton University Press. Austen, R.A. (1979). The trans-Saharan slave trade: a tentative census, in H.A. Gemery and J.S. Hogendorn (eds), The Uncommon Market: Essays in the Economic History of the Atlantic Slave Trade. New York: Academic Press. Austin, G. (2008). Resources, techniques, and strategies south of the Sahara: revising the factor endowments perspective on African economic development, 1500–2000. Economic History Review, 61:587–624. Barry, B. (1998). Senegambia and the Atlantic Slave Trade. Cambridge:  Cambridge University Press. Bean, R. (1974). A note on the relative importance of slaves and gold in West African exports. Journal of African History, 5(3):351–356. Besley, T., and Persson, T. (2010). State capacity, conflict and development. Econometrica, 78(1):1–34. Bhattacharyya, S. (2009). Root causes of African underdevelopment. Journal of African Economies, 18(5):745–780.

16  Customary or traditional authority is a complex institution in Africa today. While historical studies find evidence of strong traditional authority impeding local development, contemporary surveys show strong local support for traditional authority (Logan 2011). Acemoglu, Reed, and Robinson (2014) argue that this is evidence that local chief have captured local institutions by offering opportunities for civic engagement and the production of local social capital, what others might call ethnicity. 17  Several chapters address this challenge. See the chapters by Myerson (Handbook, this volume), Ncube (Handbook, this volume), Cramer and Chang (Handbook, Volume 2), Addison, Singhal, and Tarp (Handbook, Volume 2).

The Economic Legacies of the African Slave Trades    519 Bottero, M., and Wallace, B. (2013). Is there a long-term effect of Africa’s slave trades. Economic History Working Papers. Banca D’Italia. Brown, E., Phelps, H., and Hopkins, S. (1956). Seven centuries of the prices of consumables, compared with builders’ wage-rates. Economica, 23(92):296–314. Clark, G. (2013). “What Were the British Earnings and Prices Then? (New Series)”, MeasuringWorth. http://www.measuringworth.com/ukearncpi/. Clarkson, T. (1839). The History of the Rise, Progress, and Accomplishment of the Abolition of the Slave Trade by British Parliament. Project Gutenberg e-book #10633s. Curtin, P. (1969). The Atlantic Slave Trade. Madison: University of Wisconsin Press. Curtin, P. (1975). Economic Change in Precolonial Africa: Senegambia in the Era of the Slave Trade. Madison: University of Wisconsin Press. Da Gama, V. (1998). The Diary of His Travels through African Waters, 1497–1499. Singapore: Tien Wah Press. Daaku, K.Y. (1970). Trade and Politics on the Gold Coast, 1600–1720: A Study of the African Reaction to European Trade. London: Clarendon. Dalton, J.T., and Cheuk Leung, T. (2014). Why Is Polygyny More Prevalent in Western Africa: An African Slave Trade Perspective. Economic Development and Cultural Change, 62(4):599-632. Dike, O.K. (1956). Trade and Politics in the Niger Delta. Oxford: Clarendon Press. Diouf, S.A. (2003). Fighting the Slave Trade:  West African Strategies. Athens, OH:  Ohio University Press. Domar, E.D. (1970). The causes of slavery or serfdom: a hypothesis. Economic History Review, 30(1):18-32. Egharavba, J.U. (1968). A Short History of Benin. Ibadan: Ibadan University Press. Eltis, D., and Jennings, L.C. (1988). Trade between West Africa and the Atlantic world in the pre-colonial era. American Historical Review, 9(4):936–959. Equiano, O., and Allison, R.J. (1995). The Interesting Narrative of the Life of Olaudah Equiano. New York: Bedford Books. Evans, E.W., and Richardson, D. (1995). Hunting for Rents:  the economics of slaving in pre-colonial Africa. That Economic History Review, 48(4):665–686. Falconbridge, A. (1788). Account of the Slave Trade on the Coast of Africa. London: J. Phillips, George Yard, Lombard Street. Fenske, J. (2013). African polygamy: past and present. MPRA Working Paper No. 48526. Fenske, J. f. (2013). Does Land Abundance Explains African Institutions? The Economic Journal, 123(573):1363-1390. Fenske, J., and Namratta, K. (2013). Climate, ecosystem resilience and the slave trade MPRA working paper No. 50816. Gemery, H.A., and Hogendorn, J.S. (1979). The economic cost of West African participation in the Atlantic slave trade: a preliminary sampling for the eighteenth century, in H.A. Gemery and J.S. Hogendorn (eds), The Uncommon Market: Essays in Teh Economic History of the Atlantic Slave Trade. New York: Academic Press. Gemery, H., and Hogendorn, J. (1974). The Atlantic slave trade: a tentative economic model. Journal of African History, 15:223–246. Gemery, H., and Hogendorn, J. (1977). Elasticity of slave labor supply and the development of slave economies in the Caribbean: the seventeenth century experience, in V.R.a.T. Tuden (ed.), Comparative Perspectives on Slavery in New World Plantation Societies. New York: New York Academy of Sciences.

520    Historical Trajectories and Economic Landscape Gennaioli, N., and Rainer, I. (2007). The modern impact of precolonial centralization in Africa. Journal of Economic Growth, 12:185–234. Grief, A. (2006). Institutions and the Path to the Modern Economy:  Lessons from Medieval Trade. Cambridge: Cambridge University Press. Hopkins, A. (1973). An Economic History of West Africa. New York: Columbia University Press. Iliffe, J. (2007). Africans: The History of a Continent. New York: Cambridge University Press. Johnson, M. (1966). The ounce in 18th century West African trade. Journal of African History, 7(2):197–214. Johnson, M., Lindblad, J.T., and Ross, R. (1990). Anglo-African Trade in the Eighteenth Century: English Statistics on African Trade 1699–1808. Leiden: Centre for the History of European Expansion. Kea, R.A. (1971). Firearms and warfare on the gold and slave coasts from the sixteenth to the nineteenth centuries. Journal of African History, 21(2):185–213. Kuczynski, R.R. (1948). Demographic Survey of the British Colonial Empire, Vol. 1: West Africa. London: Oxford University Press. Kuczynski, R.R. (1949). Demographic Survey of the British Colonial Empire, Vol. 2. London: Oxford University Press. Lange, M. (2009). Lineages of Despotism and Development: British Colonialism and State Power. Chicago: University of Chicago Press. Law, R. (1989). Slave-raiders and middlemen, monopolists and free-riders: the supply of slaves for the Atlantic trade in Dahomey c. 1715–1850. Journal of African History, 30:45–68. Law, R. (1991). The Slave Coast of West Africa, 1550–1750: The Impact of the Atlantic Slave Trade on an African Society. Oxford: Oxford University Press. Law, R. (2004). Ouidah:  The Social History of a West African Slaving ‘Port, 1727–1892. Athens: Ohio University Press. LeVeen, E.P. (1975). The African slave supply response. African Studies Review, 18(1):9–28. Logan, C. (2011). The Roots of resilience: exploring popular support for African traditional authorities. Afrobarometer Working Paper. No. 128. Lovejoy, P.E. (1983). Transformations in Slavery: A History of Slavery in Africa. African Studies Series, 36. Cambridge: Cambridge University Press. Manning, P. (1990). Slavery and African Life: Occidental, Oriental, and African Slave Trades. Cambridge: Cambridge University Press. McEvedy, C., and Jones, R. (1978). Atlas of World Population History. Harmonsworth: Penguin. Michalopoulos, S., and Papaioannou, E. (2010). Divide and rule or the rule of the divided? evidence from Africa. Discussion Paper Series No. 8088. Center for Economic Policy Research. Michalopoulos, S., and Papaioannou, E. (2013). Precolonial ethnic institutions and contemporary African development. Econometrica, 81(1):113–152. Monga, C., and Lin, J.Y. (2015). The Oxford Handbook of Africa and Economics: Policies and Practices, Volume 2. Policies and Practices. Oxford: Oxford University Press. North, D.C., Wallis, J.J., and Weingast, B.R. (2009). Violence and Social Order: A Conceptual Framework for Interpreting Recorded Human History. Cambridge:  Cambridge University Press. Northrup, D. (1978). Trade without Rulers:  Pre-Colonial Economic Development in South-Eastern Nigeria. Oxford: Clarendon Press. Nunn N. (2014). Historical Development, in P. Aghion and S. Durlauf (eds), Handbook of Economic Growth, Vol. 2. North-Holland, pp. 347-402. Nunn, N. (2008). The long term effects of Africa’s slave trades. Quarterly Journal of Economics, 123(1):139–176.

The Economic Legacies of the African Slave Trades    521 Nunn, N., and Puga, D. (2012). Ruggedness: the blessings of bad geography in Africa. Review of Economics and Statistics, 94(1):20–36. Nunn, N., and Wantchekon, L. (2011). The slave trade and the origins of mistrust in Africa. American Economic Review, 111(7):3221–3252. Obikili, N. (2014). The transatlantic slave trade and local political fragmentation in Africa, Economic Research South Africa, working paper 406. Osafo-Kwaako, P., and Robinson, J.A. (2013). Political centralization in pre-colonial Africa, NBER Working Paper. No. 18770. Patterson, O. (1982). Slavery and Social Death:  A  Comparative Study. Cambridge:  Harvard University Press. Richards, W.A. (1980). The import of firearms into West Africa in the eighteenth century. Journal of African History, 21(1):43–59. Ryder, A.F. (1969). Benin and the Europeans, 1485–1897. New York: Humanities Press. Tertilt, M. (2005). Polygyny, fertility and savings. Journal of Political Economy, 113(6):1341–1371. Thomas, H. (1997). The Slave Trade:  The Story of the Atlantic Slave Trade:  1440–1870. New York: Touchstone. Thomas, R.P., and Bean, R.N. (1974). The fishers of men: the profits of the slave trade. Journal of Economic History, 34:885–914. Thornton, J.K. (1983). The Kingdom of Kongo:  Civil War and Transition, 1641–1718. Madison: University of Wisconsin Press. Thornton, J.K. (1999). Warfare in Atlantic Africa, 1500–1800. London: UCL Press. van de Walle, N. (2001). African Economies and the Politics of Permanent Crisis, 1979–1999. Cambridge: Cambridge University Press. Whatley, W. (2014). The transatlantic slave trade and the evolution of political authority in West Africa, in E. Akyeampong, R. Bates, N. Nunn, and J. Robinson (eds), Africa’s Development in Historical Perspective. Whatley, W., and Gillezeau, R. (2011a). The fundamental impact of the slave trade on African economies, in P. Rhode, J. Rosenbloom, and D. Weiman (eds), Economic Evolution and Revolution in Historical Time. Stanford: Stanford University Press. Whatley, W., and Gillezeau, R. (2011b). The impact of the slave trade on ethnic stratification in Africa. American Economic Review, 101(3):571–576. Wilks, I. (1975). Asante in the Nineteenth Century: The Structure and Evolution of a Political Order. London; New York: Cambridge University Press. Wilks, I. (1982). Wangara, Akan and Portuguese in the 15th and 16th Centuries, ii: The Struggle for Trade. Journal of African History, 23(4):463–472.

Chapter 27

The Ec onomi c s of C ol onialism i n A fri c a Gareth Austin

27.1 Introduction A perennial debate casts European rule as either modernizing previously largely static African economies or, in contrast, as retarding their development both at the time and, via institutional path dependence, ever since (Gann and Duignan 1975; Rodney 1972; Acemoglu, Johnson, and Robinson 2001). Both approaches, arguably, understate the continuities in factor endowment before and during colonial rule; the importance of the differences between types of colony; and the significance of African responses to the constraints and opportunities of what proved to be the relatively short period of alien rule (Austin 2008b). This chapter examines colonial interventions in relation to long-term trajectories of economic development in Africa. Specifically, it asks how far colonial interventions, and African responses during the colonial period, altered or accelerated pre-existing patterns or paths of economic change in the continent, paths defined by Africans’ technical and institutional responses to the constraints and opportunities of their resource endowments, in the context of regional and trans-regional markets. The focus is sub-Saharan, though occasional comparisons will be drawn with North Africa.

27.2 Reinterpretations Since c. 1960, which has become the stylized date for African independence—it was the year most French colonies south of the Sahara achieved at least formal independence, along with the Belgian Congo and the largest British colony, Nigeria—the economics and political economy of the colonial period have been approached by a host of scholars from different disciplines but with often interlocking perspectives. In the late 1950s and early 1960s the dominant tone was optimistic, about the future and about changes set in motion under colonial rule. The economic history of the white minority

The Economics of Colonialism in Africa    523 regimes of southern Africa was interpreted in terms of Lewis’s model of “economic development with unlimited supplies of labour”, contrary to Lewis’s own insistence that sub-Saharan Africa was labor-scarce (Barber 1961; Lewis 1953). The rapid expansion of export agriculture in parts of early colonial West Africa helped inspire Myint’s “vent-for-surplus” model of growth achieved by the mobilization of previously “surplus” reserves of land and labor (Myint 1958). Whereas Myint envisaged peasants simply reacting rationally to market opportunities, Hill’s fieldwork (1963) represented the pioneers of Ghanaian cocoa farming as risk-taking entrepreneurs. Economic historians began to integrate resource endowments, markets, and political economy into fuller and more nuanced syntheses (Hopkins 1973). During the 1970s, disappointments with the immediate fruits of independence stimulated much more critical reappraisals of the colonial record, often formulated in terms of dependency theory (Amin 1972; Rodney 1972). Historical evidence as well as perspectives from both market economics and dependency theory led Arrighi and others to refute the application of the Lewis model to the settler economies. They showed that the “subsistence-level” wages of the mid-twentieth century were the result not of a static traditional agriculture but of state interventions to replace surplus-producing, price-responsive peasants with a coercively constructed system of migrant labor (Arrighi 1970; Palmer and Parsons 1977). More recently, rational-choice political economists have interpreted the political influence of settler lobbies in public choice terms and argued that colonial governments failed to establish individual property rights in land, thereby discouraging investment (Bates 1981; Firmin-Sellers 1996). Acemoglu, Johnson, and Robinson—albeit partly conflating colonial rule with the external slave trades of the pre-colonial era—saw colonial regimes as essentially extractive (Acemoglu, Johnson, and Robinson 2001; Acemoglu and Johnson 2010). A basic constraint in researching colonial economies is our ignorance of per capita gross domestic product (GDP). On the numerator, we have relatively good data on exports and the public sector, but face conceptual as well as data gaps in estimating the output of the most labor-consuming activities, food production, and internal trade. On the denominator, the colonial governments introduced censuses, but the early results were often based on estimation rather than counting. Though their coverage gradually improved, it was restricted by modest administrative capacity and the incentive to people to evade enumeration in the hope of avoiding taxation (Manning 2010). In this context, as we will see, studies of welfare outcomes have focused on real wages and, increasingly, anthropometrics. Meanwhile the progressive opening of the archives of governments and firms has revealed more about colonial perceptions and motives, and court records have illuminated how property rights worked in practice (e.g. Austin 2005; Fenske 2012).

27.3  Structure and Change in Pre-colonial Economies Until well into the colonial period, and often beyond, most of sub-Saharan Africa, most of the time, was characterized by an abundance of cultivable and graze-able land in relation to the labor available to exploit it. This did not mean “resource abundance”: much of Africa’s mineral endowment was either unknown or inaccessible with pre-industrial technology, or

524    Historical Trajectories and Economic Landscape was not yet valuable even in overseas markets. Worse, thin soils made it costly or difficult to pursue intensive cultivation, especially where animal manure was absent. Sleeping sickness prevented the use of large animals, whether for plowing or transport, in the forest zones and much of the savannas. Over wide areas the extreme seasonality of the annual distribution of rainfall rendered the core of the dry season effectively unavailable for farm work. The consequent low opportunity cost of dry-season labor reduced the incentive to raise labor productivity in craft production. Conversely, the characteristic choices of farming techniques were land-extensive and labor-saving; but the thinness of the soils constrained the returns on labor. All this helps explain why the productivity of African labor was apparently higher outside Africa, over several centuries—the underlying economic logic of the external slave trades, which in turn aggravated the scarcity of labor within sub-Saharan Africa itself. Meanwhile, the major source of innovation to improve productivity and food security was the selective adoption of exotic cultigens (including plantains, maize and cassava) imported from Asia or the Americas to supplement what, except in Ethiopia, was a relatively meager range of endemic cultivable plants (Austin 2008a). The structure of incentives encouraged a high degree of self-sufficiency. By the mid-twentieth century social scientists tended to assume that pre-colonial economies had necessarily been overwhelmingly subsistence oriented and, further, that “traditional” African culture and governance rejected the logic of optimizing under scarcity. The latter view took its most sophisticated form in Polanyi’s Substantivism (Polanyi 1966). The last 50–60 years of research has progressively changed these assessments. Polanyi’s proposition that prices in pre-colonial economies were set by custom or command rather than by the interaction of supply and demand, has been falsified even for his chosen case, the kingdom of Dahomey (Law 1992). Again, research has uncovered strong tendencies towards extra-subsistence production, most notably in West Africa. For example, the currency materials (cowries, etc.) imported via Saharan caravans and European ships were not used to lubricate external trade; rather, they were used as currencies only in intra-African trade (Inikori 2007). The external slave trades bid resources into the generation and export of captives and directly damaged peaceful economic activities. But with the effective beginning of the abolition of the largest of these trades, the Atlantic, in 1807, West African production of agricultural and forestry products expanded, for internal as well as overseas markets (Hopkins 1973; Inikori 2009). Because of the relative scarcity of labor, and in the absence (generally) of significant economies of scale in production, it was rare for the reservation wage (the minimum wage rate sufficient to persuade someone to sell their labor rather than work for themselves) to be low enough for a would-be employer to afford to pay it. Hence pre-colonial labor markets (except for casual work) mainly took the form of slave trading, not least in West Africa (Hopkins 1973; Austin 2005). In sub-Saharan Africa relative abundance of land made political centralization difficult to achieve and sustain (Herbst 2000). Political fragmentation created a free-rider problem which facilitated the external slave trades, in that larger states would have had stronger incentives and capacities for rejecting participation (Inikori 2003). This fragmentation also later facilitated the European conquest. Ethiopia was the exception that proved the rule: its fertile central provinces and large agricultural surplus supported a long-established and modernizing state with an economic base sufficient to defeat the Italian invasion during the late-nineteenth-century European partition of Africa. Emperor Menelik II of Ethiopia can be likened to Mehmet Ali Pasha in post-Napoleonic Egypt. The latter had used an even

The Economics of Colonialism in Africa    525 larger, more labor-abundant, agricultural base to promote modernization, in his case including manufacturing, in the face of political and economic pressure from France and Britain and from his nominal Ottoman overlords. By no coincidence, most of the sub-continent was colonized at a time when the industrialization of Europe was creating or expanding markets for various commodities that could profitably be produced in Africa. The land–labor ratio, the environmental constraints on intensive agriculture, and also the specific qualities of particular kinds of land in various parts of the continent, gave Africa at least a potential comparative advantage in land-extensive export agriculture (Austin 2013). By the time of colonization, especially in Western Africa, indigenous populations were increasingly taking advantage of the combination of these supply-side features and of access to expanding overseas markets. From Senegal to Cameroon, thousands of tonnes of groundnuts and palm oil, and from the 1880s rubber, were being produced for sale to European merchants (Law 1995).

27.4 Colonial states Historians distinguish three main categories of colony in Africa: “settler” (more precisely, settler-elite colonies) in which most of the cultivable land was appropriated for European use; “peasant” colonies in which the land remained overwhelmingly in the hands of Africans, partly producing crops for export; and “concession” colonies in which much of the land was reserved for Europeans, but mainly for mining or plantation companies rather than individual settler-farmers. We will see that these distinctions had major implications for markets, indigenous entrepreneurship, manufacturing and income distribution. Despite early starts by the Portuguese and Dutch on the fringes of southern Africa, as well as by the French in Algeria, most of the continent was conquered only late in the history of European empires, in the Scramble for Africa, 1879 to c. 1905. Facilitated by the adoption of quinine against malaria, and prompted partly by merchant and mining interests, it was intended to cost European taxpayers little. Like pre-colonial governments, colonial administrations found their revenues constrained by the often modest size of marketed output, high costs of collecting taxes from often scattered populations, and the risk of revolt. Responding to the last two constraints, they preferred customs duties to direct taxes. But, while this option was seized upon in the wealthier of the “peasant” colonies, such as the Gold Coast (Ghana) and Nigeria, its applicability was restricted in colonies that generated fewer exports per capita, such as French West Africa and Tanganyika (mainland Tanzania). In Kenya, direct taxation of Africans was preferred also because of the influence of white settlers. The most systematic investigation so far is Frankema’s study of eight British African colonies, including both “peasant” and “settler” colonies (Table 27.1). Thus, measured by the time required to pay it, the tax burden was relatively light; which does not mean that it was not painful for the poor, especially where its distribution was regressive, as in settler economies. Frankema found, across his sample, that as per capita government revenue rose, so did the share of government expenditure on health and education, whereas spending on the police and army remained roughly constant (Frankema 2011).

526    Historical Trajectories and Economic Landscape Table 27.1  Number of working days required of an unskilled urban African to equal average annual per capita revenue in British African colonies, 1910–1938 Year

1910/13

Kenya 6.9 (5) Nigeria 3.9 (1) Unweighted average of 8 colonies* 7.1

1925

1938

16.1 (9) 3.7 (1) 10.8

23.3 (13) 4.7 (1) 14.6

Source: (Frankema 2011, 139–42) Figures in parentheses exclude customs duties. * Gambia, Sierra Leone, Gold Coast, Nigeria, Nyasaland (Malawi), Kenya, Uganda, Mauritius

Colonial administrations entered the post-1945 era with a new public commitment to actively promote the development of the economies over which they presided. “Developmental” language was partly redeemed by greater spending. In principle this came partly from the metropolitan taxpayer. However, in the French case, Manning (1988: 123–125) has calculated that the government continued to receive more in tax from Africa than it spent there. In British West Africa, the new statutory export marketing boards accrued substantial surpluses by keeping a large margin between the price paid to producers and the price that the boards received for the crop on the world market. The surpluses were kept in London, in British government bonds: forced savings from African farmers, which assisted the British metropolitan economy to recover from its postwar dollar shortage (Rimmer 1992: 41–42). Reflecting and perpetuating their low revenues, colonial administrations could afford relatively few European personnel, including in the “native reserves” of settler economies. In the 1930s, the ratio of white administrative officials to the African population was 1:19 000 in Kenya, 1:27 000 in French West Africa and 1:54 000 in Nigeria. In c. 1939 the supposedly 43 114 000 (actually many more) inhabitants of British tropical Africa were presided over by a total of 938 white police and army personnel, 1223 administrators, and 178 judges: an overall ratio of 1:18 432 (Kirk-Greene 1980: 35, 38, 39). Indeed, the ratios were actually lower, given that the censuses under-counted. Given the paucity of their financial and human resources, colonial regimes relied on African intermediaries, such as chiefs, as the front line of government: to save money, and in the hope that chiefs possessed greater legitimacy with the population. It was a compromise: chiefs were considered legitimate by the population only to the extent that occasionally they were allowed to influence colonial actions, and in colonies such as the Gold Coast individual chiefs were not infrequently deposed by pressure from their subjects. Expect perhaps for the Belgian Congo, neither chiefs nor European governors were as powerful in everyday rule as has often been depicted (Young 1994; Mamdani 1996; compare Berman and Lonsdale 1979). This capacity constraint was one reason for the caution with which the colonial authorities generally approached social engineering. Britain and France entered the partition of Africa having banned slavery in their existing colonies. Yet in many of the new colonies in Africa, slave-holding, though usually not slave trading, was tolerated for years or decades, partly because a rapid emancipation would undermine the economic and social position of chiefs, and exacerbate the labor shortages faced by enterprises beyond the scale of the family (Miers and Klein 1999). Whereas in the mid-nineteenth century, when they annexed

The Economics of Colonialism in Africa    527 Lagos, the British were enthused by the notion of individual property as the universal key to economic advance (Hopkins 1980, 1995). After the Scramble, however, they and their counterparts preferred to maintain family and communal land rights under the supervision of the rural chiefs to avoid the risk that the poorer farmers would sell up and become proletarians or, worse, lumpen-proletarians in the towns. In the peasant—or, in part, rural capitalist—colonies of British West Africa, another reason for maintaining the existing land tenure systems was that they proved consistent with massive investment in the expansion of tree-crop cultivation. This was spectacularly true in the Gold Coast, which rapidly became the world’s largest cocoa producer: benefiting the African farmers, but also government customs revenue and the profits of European merchants. The Akan land tenure system, upheld in this respect in the colonial courts, protected the right of someone who planted a tree to ownership of it and its fruits, at least during his lifetime (Austin 2005, 2008b).

27.5  Economic specialization and dynamics The European partition of Africa occurred during a major expansion, pre-1914, of overseas markets for the actual or potential products of tropical agriculture. The single most successful response, pioneered in Nigeria and the Gold Coast by Africans, was the adoption of a South American crop, cocoa beans. This can be viewed as a further step on the long-established African path of raising incomes by increasing returns to labor through the selective adoption of exotic cultigens. Again, though the adoption of a permanent crop entailed a new production function, the most economically successful methods of cultivating it were land-extensive. In the Gold Coast, the devotion of European planters to a more capital and labor-intensive approach explains their commercial failure in competition with African growers (Austin 1996). Crucially, this was in the setting of a “peasant” colony in which the European producers did not enjoy the advantage of a supply of directly or indirectly coerced labor, as in the settler colonies. Where the physical environment did not suit the more lucrative cash crop, and food security remained farmers’ overwhelming priority, there were no breakthroughs comparable to the adoption of cocoa in the West African forests. In the Niger Valley, and later in Tanganyika, the French and British launched grand agricultural projects: with results, respectively, modest and derisory (Roberts 1996; Van Beusekom 2002; Hogendorn and Scott 1981). These projects confirmed the inefficiency of agricultural intensification in the circumstances. It was only when colonial rule was half a century old in most of Africa that scientific agriculture, state or private, began to contribute significantly to raising productivity in African export-crop farming, still less food-cropping (Richards 1985). Meanwhile handloom weavers survived, at least in West Africa, often using imported, machine-made yarn, and selling their produce to the more prosperous cash-crop farmers (Austin 2013). We will return to manufacturing below. Where European technology made a great difference was in introducing deep-mining methods, to spectacular effect in the diamond and gold industries of South Africa, plus mechanized transport—the latter all the more important where sleeping sickness was endemic. In principle, Africa’s longstanding shortage of investment could have been remedied through colonization by countries which were not only themselves industrialized, but major

528    Historical Trajectories and Economic Landscape exporters of capital. A 1938 study by Frankel remains the only attempt at a comprehensive count of foreign investment in colonial sub-Saharan Africa. The per capita figures are surely exaggerated because, again, of the census underestimations of population (Table 27.2). The fact that nearly 45 percent of the total was public investment (grants and loans from the imperial metropoles) underlines the paucity of foreign private investment outside the mining industries of South Africa, what is now Zimbabwe, and the copperbelt of Central Africa (Zambia and the Belgian Congo). While colonial administrations were generally reluctant to register individual land titles in agriculture, exceptions were made for foreign investors in many colonies (but, notably, not in Nigeria), and expropriation risk was negligible. Thus the lack of foreign investment in agricultural Africa cannot be sufficiently explained in institutional terms. Rather, some key environmental obstacles to the profitable embodiment of capital, such as the precariousness of soil fertility in a setting in which supplies of water and fertilizer were unreliable or costly, remained severe to the end of colonial rule and beyond (Austin 2008a). The colonial occupation also confronted foreign rulers and investors with another structural problem: the scarcity of labor in relation to the availability of agricultural land. The new governments responded in three ways. One, already mentioned, was their gradual approach to the elimination of slavery. While they usually suppressed slave raiding and trading pretty rapidly, in many African colonies they initially tolerated the continued use of slave labor. In West Africa, the growth of export agriculture enabled former slaves to become free peasants in some cases, and migrant wage laborers in others (Austin 2009). The second colonial approach was use of coercion by governments to recruit labor, for themselves or for private European employers, or to direct peasants to grow specific crops, usually cotton, in areas where this entailed planting less food, because of a short planting season (Fall 1993; Likaka 1997; Tosh 1980). Forced labor was generally fairly ineffective, partly for lack of government capacity. But it was only slowly and unevenly phased out, persisting in the French empire until abolished by law in 1945 (Cooper 1996), and lasting longer still in the Portuguese empire. The third government strategy characterized the settler economies, in contrast to the peasant colonies. This was the attempt to force Africans to quit the produce market and sell their labor instead, to European agriculturalists or mine owners. The archetypal expression of this policy (though it was not entirely enforced) was the Natives Land Act of 1913 in

Table 27.2  Foreign investment in sub-Saharan Africa, 1870–1936 Aggregate Union of South Africa Southern & Northern Rhodesia (Zimbabwe & Zambia) Angola and Mozambique (Portuguese) Belgian Africa (Congo and Rwanda-Burundi) French Africa south of the Sahara British Eastern Africa (Kenya, Uganda, Tanganyika, Nyasaland) British West Africa (Nigeria, Gold Coast, Gambia, Sierra Leone) All colonial sub-Saharan Africa (including Sudan, Zanzibar, but excluding Portuguese Guinea)

Per head of population

554 681 102 403 66 732 143 337 70 310 110 189

55.8 38.4 9.8 13.0 3.3 8.1

116 730

4.8

1 221 686