Routledge Handbook of Regionalism and Federalism 2012036633, 9780415566216, 9780203395974

Citation preview

Routledge Handbook of Regionalism and Federalism

Almost all states are either federal or regionalized in some sense. It is difficult to find a state that is entirely unitary and the Routledge Handbook of Regionalism and Federalism necessarily takes in almost the entire world. Both federalism and regionalism have been subjects of a vast academic literature mainly from political science but sometimes also from history, economics and geography. This cutting-edge examination seeks to evaluate the two types of state organization from the perspective of political science, producing a work that is analytical rather than simply descriptive. The Handbook presents some of the latest theoretical reflections on regionalism and federalism and then moves on to discuss cases of both regionalism and federalism in key countries chosen from the world’s macro-regions. Assembling this wide range of case studies allows the book to present a general picture of current trends in territorial governance. The final chapters then examine failed federations such as Czechoslovakia and examples of transnational regionalism – the European Union, North American Free Trade Agreement (NAFTA) and the African Union. Covering evolving forms of federalism and regionalism in all parts of the world and featuring a comprehensive range of case studies by leading international scholars, this work will be an essential reference source for all students and scholars of international politics, comparative politics and international relations. John Loughlin is a Professor and Fellow of St Edmund’s College, Cambridge, and a Senior Fellow and Affiliated Lecturer in the Department of Politics and International Studies in the University of Cambridge. John Kincaid is the Robert B. and Helen S. Meyner Professor of Government and Public Service and Director of the Meyner Center for the Study of State and Local Government at Lafayette College, Easton, Pennsylvania, USA. Wilfried Swenden is a Senior Lecturer in Politics at the School of Social and Political Science, University of Edinburgh, Scotland, UK.

This Page intentionally left blank

Routledge Handbook of Regionalism and Federalism

Edited by John Loughlin, John Kincaid and Wilfried Swenden

First published 2013 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2013 selection and editorial matter John Loughlin, John Kincaid and Wilfried Swenden, contributors, their contributions The right of John Loughlin, John Kincaid and Wilfried Swenden to be identified as editors of this work has been asserted by them in accordance with the Copyright, Designs and Patent Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Routledge handbook of regionalism and federalism / edited by John Loughlin, John Kincaid and Wilfried Swenden. p. cm Includes bibliographical references and index. 1. Regionalism. 2. Federal government. I. Loughlin, John, 1948– JF197.R68 2013 320.4'049–dc23 2012036633 ISBN: 978-0-415-56621-6 (hbk) ISBN: 978-0-203-39597-4 (ebk) Typeset in Bembo by Taylor & Francis Books

Contents

List of illustrations List of contributors Preface and acknowledgements

ix xi xiv

PART I

Theoretical and comparative approaches to federalism and regionalism

1

1

Reconfiguring the nation-state: hybridity vs. uniformity John Loughlin

3

2

Typologies of federalism Ronald L. Watts

19

3

Plurinational federalism and political theory Ferran Requejo

34

4

The penumbra of federalism: a conceptual reappraisal of federalism, federation, confederation and federal political systems Michael Burgess

45

5

Territorial strategies for managing plurinational states Wilfried Swenden

61

6

Federalism, regionalism and the dynamics of party politics Eve Hepburn and Klaus Detterbeck

76

7

Fiscal federalism and the political economy of territorial finance Anwar Shah

93

8

Federalism and public policy: do federalism, regionalism and hybridity make any difference? Evidence from environmental policy Sonja Wälti

116

v

Contents

The ‘new regionalism’ and the politics of the regional question John Agnew

130

10 Economic regionalism in federal and hybrid systems of government Pieter van Houten

140

9

PART II

Case studies by region: North America 11 The United States of America: from dualistic simplicity to centralized complexity John Kincaid

155

157

12 Canada: Federal adaptation and the limits of hybridity James Bickerton and Alain-G. Gagnon

172

13 Mexico: from centralized authoritarianism to disarticulated democracy? Steven T. Wuhs

190

Case studies by region: Europe

209

14 The Belgian Federation: a labyrinth state Kris Deschouwer

211

15 Germany: federalism under unitary pressure Arthur Benz and Jörg Broschek

223

16 Austria: a federal, a decentralized unitary or a ‘hybrid’ state? Relations between the welfare state and the federal state after 1945 Franz Fallend

235

17 Switzerland: Europe’s first federation Paolo Dardanelli

248

18 Russia: Involuted federalism and segmented regionalism Richard Sakwa

259

19 Spain: the autonomic state Francesc Morata

273

20 Italy: between the hybrid state and Europe’s federalizing process Beniamino Caravita

287

vi

Contents

21 Sweden: federalism in the land of centrally guided regionalization Niklas Eklund

302

22 Devolution in the United Kingdom Charlie Jeffery

317

23 Poland: creeping regionalization of the unitary state Paweł Swianiewicz

331

24 France: from the ‘one and indivisible republic’ to the decentralized unitary state John Loughlin

341

Case studies by region: Asia

351

25 India: a hybrid federal-unitary state? Rekha Saxena

353

26 Assessing hybridity in the People’s Republic of China: the impact of post-Mao decentralization Susan J. Henders

371

Case studies by region: Asia-Pacific

387

27 Australia: an ‘integrated’ federation? Cheryl Saunders

389

28 Indonesia: arbitrary polity, unitary state Damien Kingsbury

401

Case studies by region: Africa

413

29 Nigeria: a centralizing federation Rotimi T. Suberu

415

30 Federalism in Ethiopia: hybridity in ambiguity? Sarah Vaughan

428

31 South Africa: the reluctant hybrid federal state Nico Steytler

442

Case studies by region: Latin America

455

32 Brazil: from ‘isolated’ federalism to hybridity Celina Souza

457

vii

Contents

33 Tensions between centralization and decentralization in the Argentine federation Lucas González

471

34 The Bolivarian Republic of Venezuela: an empty constitutional federation? Christi Rangel Guerrero

481

Case studies by region: Middle East

497

35 Lebanon: the hybridity of a confessional state Michael Kerr and Amal Hamdan

499

PART III

Failed federations

511

36 ‘Federalizing the federation’: the failure of the Yugoslav experiment Florian Bieber

513

37 Czechoslovakia: a peaceful disintegration Michal Illner

527

PART IV

Transnationalism regionalism

545

38 Federalism and the European Union Simona Piattoni

547

39 The North American Free Trade Agreement: the limits of integration Douglas M. Brown

559

40 African Union: from practical federalism to fantasy union Thomas Kwasi Tieku

573

Index

viii

584

Illustrations

Figures 6.1 6.2 6.3 11.1 12.1 13.1 13.2 13.3 13.4 13.5 13.6 23.1 23.2

State-wide party positions on ideology and decentralization, 1970s State-wide party positions on ideology and decentralization, 2000s Typology of multi-level party organizations Federal grants-in-aid to state and local governments for social welfare and other public purposes in billions of constant (FY 2005) dollars, 1940–2017 Party systems in Canada, 1945–2011 Political map of Mexico Fiscal decentralization in Mexico, 1995–2010 Expansion and contraction of participaciones in selected states, 1995–2010 Declining budget share toward state and municipal spending, 1995–2010 Proportion of municipal elections won by PRI, PAN and PRD, 1990–2010 Vote share for Mexico’s Chamber of Deputies, 1970–2012 GDP per capita in Polish regions, 2007 Unemployment rate in Polish regions, September 2011

79 80 85 165 176 191 199 199 200 203 205 334 335

Tables 1.1 4.1 5.1 7.1 7.2 7.3 7.4 12.1 13.1 13.2 13.3 14.1 15.1 15.2 16.1

The changing paradigm in regional policy and territorial governance Federal democracies Varieties of territorial management in the context of a plurinational state Trends in multi-order governance structure: 20th century versus 21st century Assignment of resource taxes: ideal vs. practice in Brazil and Canada Resource ownership and resource revenue sharing mechanisms in selected countries Features of fiscal equalization transfers in selected countries Evolution of the constitutional system since 1867 Distribution of state GDP across economic sector, 2010 Effective federalism in Mexico, 2012 Governance in selected Mexican states The distribution of the competences in federal Belgium after the constitutional agreement of 2011 Political, geographic, social and economic indicators Federalism, political parties and party systems in Germany Territorial distribution of tax revenues, 1956–2008

15 56 63 100 102 103 108 174 192 201 202 216 226 232 239 ix

Illustrations

16.2 19.1 19.2 22.1 22.2 22.3 22.4 22.5 22.6

Elections to the National Council (Nationalrat, lower chamber) Population, surface area and GDP per capita Share in general government spending 2006 UK general election results in England, 1997–2010 Scottish Parliament elections, 1999–2011 National Assembly for Wales elections, 1999–2011 Northern Ireland Assembly elections, 1998–2011 What powers for the Scottish Parliament? 2010 Do you think the long-term policy for Northern Ireland should be for it … , 2011 22.7 Overall, do you think that the Northern Ireland Assembly has achieved … , 2011 22.8 Trends in ‘bipolar’ national identity, England, 1997–2011 22.9 Constitutional preferences for England, 2011 22.10 Which institution should have most influence over the way England is run? 2011 22.11 Constitutional preferences by national identity, 2011 22.12 Which should have most influence by national identity, 2011 23.1 Development of the territorial organization in Poland after 1945 23.2 Structure of revenues in different types of local governments, 2009 23.3 Structure of regional budgets, 2009 27.1 Population of Australian states and territories 30.1 The states of the Ethiopian federation and their population figures 32.1 Constitutional reforms 32.2 Financial revenue by level of government, including intergovernmental transfers, 1960/2010 34.1 The relative size of national and sub-national expenditures, 1989–2010 35.1 Registered resident voters and reserved parliamentary seats by religious community, 2011 37.1 Evaluation of the relationship between the Czech and the Slovak Republics, 1996–2009, % of ‘very good’ or ‘rather good’ 39.1 NAFTA’s first decade: trade and investment by the United States with its NAFTA partners, % growth, 1993–2004

x

242 275 278 322 322 322 323 324 325 325 327 327 328 328 329 332 336 336 390 433 460 464 492 502 541 565

Contributors

John Agnew is Professor in the Department of Geography, University of California at Los Angeles, USA. Arthur Benz is Professor of German and Comparative Politics at the Technische Universität of Darmstadt, Germany. James Bickerton is Professor of Political Science at St Francis Xavier University, Canada. Florian Bieber is Professor of South-East European Studies and director of the Centre for South-East European Studies at the University of Graz. Jörg Broschek is Lecturer in Political Science in the Institute of Political Science at the Technische Universität Darmstadt, Germany. Douglas M. Brown is Associate Professor, Department of Political Science, St Francis Xavier University, Antigonish, NS, Canada. Michael Burgess is Professor of Federal Studies and Director of the Centre for Federal Studies at the University of Kent, UK. Beniamino Caravita is Full Professor of Constitutional Law at Sapienza University of Rome, Italy. Paolo Dardanelli is Lecturer in European and Comparative Politics in the School of Politics and International Relations of the University of Kent, UK. Kris Deschouwer is Research Professor and Head of the Department of Political Science, Free University of Brussels (Vrije Universiteit Brussel), Belgium. Klaus Detterbeck is Lecturer in Politics at the University of Magdeburg, Germany. Niklas Eklund is Lecturer in Politics in the Department of Political Science at the University of Umeå, Sweden. Franz Fallend is Senior Scientist in the Department of Political Science, University of Salzburg, Austria. xi

Contributors

Alain-G. Gagnon is Professor of Political Science at the Université de Québec à Montréal, Quebec, Canada. Lucas González is Professor in Political Science at CONICET/Universidad Catolica Argentina and Universidad Nacional de San Martin. Amal Hamdan is graduate student at King’s College, London, UK. Susan J. Henders is Associate Professor in the Department of Political Science and Director of the York Centre for Asian Research, University of York, Canada. Eve Hepburn is Senior Lecturer in Politics and Deputy Director of the Academy of Government, Politics and International Relations, School of Social and Political Science, University of Edinburgh, Scotland. Michal Illner is Senior Researcher, Department of Local and Regional Studies, Institute of Sociology of the Academy of Sciences of the Czech Republic, Prague. Charlie Jeffery is Professor of Politics and Vice-Principal Public Policy at the School of Social and Political Science, University of Edinburgh, Scotland. Michael Kerr is Professor of Conflict Studies at King’s College, London, UK. John Kincaid is the Robert B. and Helen S. Meyner Professor of Government and Public Service and Director of the Meyner Center for the Study of State and Local Government at Lafayette College, Easton, Pennsylvania, USA. Damien Kingsbury holds a Personal Chair in the School of Humanities and Social Sciences and is Director of the Centre for Citizenship, Development and Human Rights in the Faculty of Arts and Education as Deakin University, Melbourne. John Loughlin is Professor and Fellow of St Edmund’s College, Cambridge, and a Senior Fellow and Affiliated Lecturer in the Department of Politics and International Studies in the University of Cambridge, UK. Francesc Morata is Professor in the Political Science and Administration Department at the Autonomous University of Barcelona, Spain. Simona Piattoni is Professor of Political Science at the University of Trento, Italy. Christi Rangel Guerrero is Professor of Political Science at the University of Los Andes, Venezuela. Ferran Requejo is Professor of Political Science in the Department of Political and Social Sciences, Pompeu Fabra University, Barcelona, Spain. Richard Sakwa is Professor of Russian and European Politics and Head of the School of Politics and International Relations of the University of Kent, UK. xii

Contributors

Cheryl Saunders is Laureate Professor in the School of Law, University of Melbourne, Australia. Rekha Saxena is Associate Professor in the Department of Political Science, University of Delhi and Hon. Vice-Chairman, Centre for Multilevel Federalism, New Delhi, India. Anwar Shah is Lead Economist and Program Leader, Governance Program at the World Bank Institute, Washington, DC, USA. Celina Souza is Professor of Political Science at the Institute of Social and Political Studies, State University of Rio de Janeiro, Brazil. Nico Steytler is Professor and Director of the Community Law Centre, University of the Western Cape, South Africa. Rotimi T. Suberu was Professor of Politics at the University of Ibadan, Nigeria, and currently teaches at Bennington College, Vermont, USA. Wilfried Swenden is Senior Lecturer in the School of Social and Political Science at the University of Edinburgh and from 2007–2013 was co-convenor of the ECPR Standing Group of Federalism and Regionalism. Paweł Swianiewicz is Professor in the Department of Local Development and Policy in the Faculty of Geography and Regional Studies, University of Warsaw, Poland. Thomas Kwasi Tieku is Assistant Professor, Lead Researcher at the Centre for International Governance Innovation, Director of African Studies at the University of Toronto, Canada. Pieter van Houten is Fellow of Churchill College and Lecturer in Politics in the Department of Politics and International Studies in the University of Cambridge, UK. Sarah Vaughan is Research Consultant, and Honorary Fellow in Politics in the School of Social and Political Science, University of Edinburgh, Scotland, UK. Sonja Wälti is Assistant Professor in the Department of Public Administration and Policy, School of Public Affairs, American University, Washington, DC, USA. Ronald L. Watts is Principal Emeritus and Professor Emeritus of Political Studies and Fellow of the Institute of Intergovernmental Relations at Queen’s University, Kingston, Ontario, Canada. Steven T. Wuhs is Associate Professor in the Department of Government, University of Redlands, California, USA.

xiii

Preface and acknowledgements

Almost all states are either federal or regionalized in some sense. It is difficult to find a state that is entirely unitary, except perhaps for micro-states such as the Vatican, Monaco, San Marino, or Andorra. This means that a Handbook of Regionalism and Federalism necessarily takes in almost the entire world. It would be possible to compile a handbook that simply lists the characteristics of all the states in the world somewhat like the CIA’s World Factbook. This, however, being little more than a compilation of ‘facts’ and statistics would tell us little about the nature of regionalism and federalism. Both federalism and regionalism have been subjects of a vast academic literature mainly from political science but sometimes also from history, economics and geography. We decided, therefore, in this Handbook to examine the two types of state organization from the perspective of political science and to produce a work that is analytical rather than simply descriptive. We wished also to present some of the latest theoretical reflections on regionalism and federalism, which are treated in Part I of the book. Part II presents a number of case studies with examples of both regionalism and federalism in key countries chosen from the world’s macro-regions. It was not possible to include all the interesting cases, but we have assembled enough case studies to form a general picture of current trends in territorial governance. It is also interesting to explore why some federations fail and we look at a number of cases in Part III. Part IV of the book deals with what we have called ‘transnational regionalism’, which refers to regionalism as generally understood in international relations – that is, regions as comprising groups of countries rather than subnational entities, which is how they are understood in comparative politics. This is quite a vast field, and the book ended up with 40 chapters. In order to provide an overall organizing theme, the editors decided to explore the notion of ‘hybridity’, which was developed in a 2009 article by John Loughlin and seemed to relate to other analyses of territorial governance such as the breakdown of the federal-unitary state distinction and the sophisticated conceptualization of new forms of federalism by Daniel J. Elazar and Ronald L. Watts (see Chapter 2). We did not impose particular research questions on the authors apart from a rudimentary chapter structure and a suggestion that they might respond to the notion of the ‘hybrid state’. Most authors did follow this suggestion, but others felt that it was not really relevant to their analyses. In any case, the large number of theoretical chapters and country case study chapters show that territorial governance remains a rich research seam that can be mined continually. It also is evident that such territorial governance structures are in continual evolution, and older conceptual frameworks, such as the federal-unitary distinction, need to be revised in order to capture these changes. We believe that this volume will be a very useful resource to pursue this research agenda. The book has been too long in the making, and we would like to thank first of all the authors of the chapters who have been very patient and have responded positively to suggestions for improvement made by the editors. We would also like to thank Craig Fowlie, Senior xiv

Preface and acknowledgements

Politics Editor at Routledge, who first commissioned John Loughlin to edit this volume in what seems like an embarrassingly long time ago. Nicola Parkin, Associate Editor at Routledge, has shown great patience and kindness as the book plodded its way through the editorial process. John Loughlin, Cambridge, UK John Kincaid, Easton, Pennsylvania, USA Wilfried Swenden, Edinburgh, UK

xv

This Page intentionally left blank

Part I

Theoretical and comparative approaches to federalism and regionalism

This Page intentionally left blank

1 Reconfiguring the nation-state Hybridity vs. uniformity John Loughlin

Introduction This chapter argues that the nation-state became the primary form of political organization from about the end of the 18th century and that it reached its culmination in the form of the welfare states constructed in Western Europe and, to a limited extent in the United States, after the Second World War. Furthermore, the nation-state model, based on the premise that nations should have states and that states should be co-terminus with nations, while originating in Europe spread across the world through imperialism and colonialism. The ‘Westphalian’ system of international relations exemplified by the United Nations (UN) is really a collection of nationstates rather than simply ‘nations’. The success of the nation-state model can be seen from the growth of the UN from its founding in San Francisco in 1945 by 50 states to its current membership of 193. In fact, so widespread is the nation-state model that we tend to take it for granted as being the ‘natural’ form of political governance. However, the nation-state is the product of an historical development that lasted several centuries and some authors have argued that we are today witnessing its demise (Ohmae, 1996). Both nations and states existed before they became twinned at the time of the French Revolution and they co-existed with other forms of political organization which they came to dominate but which never quite disappeared. The emergence of the nation-state as the dominant political form was largely co-extensive with the arrival of ‘modernity’, however difficult it is to define that concept (Chernilo, 2007). This chapter will begin by examining some of these pre-modern and early modern forms of political organization which, as will become clear, were marked by diversity, asymmetry and hybridity. The modern nation-state swept away much of this complexity and created standardized, uniform and symmetrical political and administrative systems in the name of the modern nation. This process was true of both ‘unitary’ and ‘federal’ states. The archetypical example is France after the Revolution (see the chapter by Loughlin in this volume) but it may also be seen in federations such as the United States (see the chapter by Kincaid in this volume where he describes how ‘coercive’ federalism finally defeats ‘bi-communal’ federalism by the late 1950s). We cannot really speak of the ‘end’ of the nation-state as if this is being replaced either by ‘regions’ as Ohmae thought or by some cosmopolitan and globalized system as is sometimes suggested by 3

John Loughlin

authors such as David Held and his colleagues (Held et al., 1999). However, it does seem clear that, from the 1980s onwards, the central position of the nation-state or at least of national governments has, under the impact of the processes of globalization and neo-liberalism, given way to a more complex system of multi-level governance both at the level of international relations and within nation-states themselves. National governments today operate alongside international organizations such as the UN, the International Monetary Fund (IMF) and the World Bank among many others and, internally, regions and cities have become more prominent in recent years as political actors. Furthermore, the old Westphalian system has given way to a more complex, interpenetrative system in which regions and cities of different kinds have taken on new international roles and activities which is sometimes called ‘paradiplomacy’ (Duchacek, 1986). While it is exaggerated to speak of the ‘demise’ of the nation-state or of ‘new medievalism’ (Bull, 1995), it can be argued that some of the older forms of territorial organization such as regions and cities have re-emerged and the result has been a situation of territorial governance that is more diverse, non-standardized and ‘hybrid’. The term ‘hybrid’ is used here, not in the biological sense of the creation of a new entity from two separate species, but rather in a looser way to designate the co-existence of several competing models of institutional organization and policy approach within the same political system. This will be elaborated further with examples later in the chapter.

Pre-modern and early modern forms of political organization The nation-state was preceded by a variety of pre-modern forms of political organization. Hendrik Spruyt distinguishes six categories of these: older forms such as the papacy, the Holy Roman Empire and feudal arrangements which emerged after the fall of the Roman Empire; and later forms which began to appear from about the 10th century: cities, city leagues and what he calls territorial states, which were, in effect, centralizing monarchies such as France, Sweden and Tudor England (Spruyt, 1994). Spruyt interprets the rise of the modern state as a ‘victory’ of the territorial state over the others although, in practice, some of them, such as the Holy Roman Empire or independent city-states lasted right into the modern period and we can even recognize vestiges of them today.

Pre-modern and early modern Europe The Barbarian invasions (AD 400–800) had brought about the end of the Roman Empire and Roman cities and roads fell into ruin. With the collapse of the old Roman civil service, the Church was almost the only form of organization through its system of monasteries, dioceses and parishes, many of which were based on the old Roman administrative boundaries. The Barbarian tribes later formed kingdoms, of which the most important was that of Clovis (481–511) who founded the Merovingian dynasty. Co-existing with these monarchies was a system of feudalism (from the Latin foedus which is also the root of ‘federalism’), which took several forms across Europe but they shared in common the basic idea of a personal relationship between ‘fiefs’ (lords) and ‘vassals’ (subjects), where the latter provided services of various kinds in return for protection by the former (Brown, 2012). Feudalism was dominant up until about the 12th century but survived in attenuated forms right up to the 19th century. The Merovingian dynasty of Clovis was replaced by the Carolingian dynasty of Charlemagne (AD 742?–814), who was crowned Emperor by the Pope in AD 800, thus founding the Holy Roman Empire which would take many shapes and cover many territories within Europe until it was finally wound up by Napoleon I in 1806 (Bryce, 1864). From about the 11th century there was a revival of city 4

Reconfiguring the nation-state

life as cities became centres of commerce and trade with new burghers and guilds becoming wealthy and politically powerful. Some of these cities evolved into veritable city-states as in Florence, Milan and Venice. Others grouped together into ‘leagues’ of which the most famous is the Hanseatic (ca.13th–17th centuries), which was a group of cities engaged in trade from the North Sea to the Baltic. Underlying this great variety of political forms was a Christian religious culture which was shared by all of the protagonists of mediaeval Europe. All of Western and most of Central Europe were united under the religious system of Catholic Christianity. The East (the Russian lands and most of what we now call the Balkans) was also Christian but following the Byzantine Orthodox traditions and the division between the two was consolidated by the Great Schism of 1054. Despite the common religious heritage of Catholic Christianity, however, political life in the West was dominated by struggles between the different groupings which we can describe here only in a very summary fashion (see Burns, 1988). First, there was the struggle for spiritual and political hegemony between the Holy Roman Emperor and the Pope which dominated much of the late mediaeval and early modern period. Second, there were conflicts between the monarchs who were consolidating their ‘territorial states’ and the feudal barons who were determined to retain their autonomy in the provinces. Third, the cities also struggled against the neighbouring noble or episcopal overlord and often became allies of the monarch or the Emperor. In northern Italy, they were also divided by the famous conflict between Guelphs and Ghibellines which began in the 12th century as a division between city-states that supported either the Emperor (Guelphs) or the Pope (Ghibellines). What is striking about this history is that there was no single, uniform model of political organization but a great variety of forms (Burns, 1988). It was also a system of overlapping orders of law, sovereignty and jurisdiction. Individuals and groups living on the same territory might be subject to ecclesiastic law, feudal arrangements, imperial law, etc. The system was also marked by a great deal of asymmetry in power relationships ranging from the powerful positions of Pope, Emperor, King or Prince-Bishop to small city-states, abbeys or provincial barons. However, even the most powerful entities were constrained by each other and by ecclesiastical or legal bonds, which meant that their sovereignty was never absolute – even during the period of the ‘absolute’ monarchies.

The Reformation and the modern state A key turning point in the evolution and, ultimately, radical reconfiguration of this system leading to the emergence of the modern nation-state was the Protestant Reformation initiated by Luther in 1512 (Burns and Goldie, 1991). It would take us too far outside the scope of this chapter to describe the vast, complex and contradictory movements of change that occurred in 16th-century Europe (Ozment, 1980). What we can say is that by the end of the 16th century, the political and religious landscape of Western and Central Europe had been almost completely transformed. The previously existing religious unity which underpinned the political, social and economic structures was now shattered and replaced by competing versions of Christianity with quite distinctive understandings of church, politics and theology. Wars of Religion between Protestant and Catholic rulers had broken out all over Europe but principally within the Holy Roman Empire. They finally ended there with no clear victory by either side in 1648 with the Treaty of Westphalia. This adopted the principle (already promulgated at the Peace of Augsburg in 1555) of cuius regio, eius religio – the religion of the ruler shall be the religion of the state. This applied only to Catholic and Lutheran states and excluded Calvinists, although the latter did later achieve statehood in Scotland and the Netherlands and in the imperial city of Geneva. 5

John Loughlin

The significance of this is that there appeared, for the first time in Europe, confessional states each with a distinctive way of understanding the relationship between church, state and civil society. This is the origin of the different state traditions in Europe (Dyson, 1980; Loughlin and Peters, 1997), as well as of modern notions of internal and external sovereignty. Among the Protestant states of northern Europe outside the Empire such as Sweden, Scotland or England, it is henceforth the ruler of the state who exercises sovereignty over the territory within the boundaries of that state and not an external power, whether Emperor, Pope or any other state. Within the Empire, the position of the Emperor was greatly diminished vis-à-vis the Protestant states. With regard to external sovereignty, there developed henceforth what became (and is still) known as the ‘Westphalian’ state system of international relations based on the principle that no state can interfere in the internal affairs of another. Originally this meant that Catholic princes could not interfere in the affairs of Protestant states and vice versa but eventually it meant a prohibition on any kind of interference. The form of political organization that could most successfully adapt to these new circumstances was what Spruyt had termed the ‘territorial states’ and these eventually dominated and absorbed the other forms. The other important development that followed the Reformation was the link, explored by historians such as Greenfeld (1992) and Colley (2005), between a particular variety of the Christian religion and the nation. Previously, the term ‘nation’ referred simply to individuals born (natus in Latin) in a particular place and speaking a particular language, but it had little political significance. With the arrival of the Westphalian state, characterized by a particular form of Christianity and newly emerging from the Wars of Religion, nationhood became political in a new way. This was particularly true of the Protestant nations of northern Europe but the Catholic nations also made the link as in Gallican France or Bourbon Spain. Eventually, this religious identification would become secularized as the period of the Reformation gave way to the Enlightenment and anti-religious and anti-clerical movements developed throughout the 18th and 19th centuries.

The distinctive forms of the modern state1 A further development in the configuration of modern territorial governance came in the 17th and 18th centuries. In this period, there were three distinctive historical ‘moments’:  The first was the English ‘Glorious Revolution’ of 1688, which led to the installation of a constitutional monarchy and the hegemony of the Westminster Parliament, followed by the Industrial Revolution which laid the basis of modern industrial capitalism and society.  The second was the American Revolution, when the 13 British colonies in North America overthrew English rule and established first a confederation and then the federation of the United States of America.  The third important historical moment was the French Revolution. Each of these three sets of historical events resulted in distinctive understandings of the state, its administrative structure and its relations with society. From the British constitutional and industrial revolutions emerged the United Kingdom of Great Britain and Ireland, built on a series of Acts of Union between England and the other three nations. This became a multi-national ‘Union’ State. The United States of America was the first modern federal state (see the chapter by Kincaid in this volume). France produced the unitary state par excellence characterized by ‘unity and indivisibility’ (Hayward, 1983; Loughlin, 2007). Each of these state forms – union, federal and unitary – would be imitated by almost all other modern nation-states. As already mentioned above, 6

Reconfiguring the nation-state

it was the French state that invented the idea of the ‘nation-state’, from which developed modern nationalism and, in turn, affected both the British state form and modern federations. It is appropriate, then, to begin with France, even if it came chronologically last (Alter, 1994; Guibernau, 1996; Smith, 2010). The French Revolution, beginning in 1789, was a vast, long-drawn out series of events, the protagonists of which held several contradictory positions regarding the kind of state that should be adopted. Their ideological battles were fought out against the background of a state which was already, in some respects, highly centralized. This was a result of the efforts of the French monarchy which, in previous centuries, and especially during the reign of Louis XIV, had sought to bring under control the nobles who dominated the provinces. The revolutionaries were divided into two main groups with radically differing positions with regard to the territorial organization of the new France: 1 the Jacobins, led by Robespierre, who wished to continue and complete the centralizing tendency of the monarchy; and 2 the Girondins, whose chief spokesman was Jacques Pierre Brissot, who wished to maintain some level of decentralization and diversity (Schmidt, 1990; Ohnet, 1996; Loughlin 2007). In the end, although each group was eliminated in turn by the Terror, including the Jacobins who had initiated it, and who fell with the execution of Robespierre in 1794, it was the Jacobin conception that won the intellectual and political argument and gave rise to the celebrated description of France as the ‘one and indivisible Republic’ (Hayward, 1983). The Republic succumbed to the Napoleonic Empire, which, in turn, gave way to the restored monarchy, thus beginning a chain of regime changes in France which finally settled into the present-day Fifth Republic. Whatever the regime, though, the basic ideas of ‘unity and indivisibility’ and the necessity of expressing this through a centralized state were retained. Furthermore, the Revolution and the Empire created the two basic institutions of the modern French state: the départements set up at the Revolution as a way of abolishing the old system of provinces; and the prefectoral system established by Napoleon as a way of exercising central control over these territorial entities. The extreme Jacobins regarded the Girondins as counter-revolutionaries and described them as ‘federalists’ which henceforth became a disreputable word in the French political lexicon. Later in the 19th century ‘regionalism’ was also regarded in the same way, although the Girondin tradition never totally disappeared (Wright, 2003). The old, pre-Revolution ecclesiastical parishes became the communes, which still, today, number around 36,000 (Loughlin, 2007).

Unitary states and nationalism Nationalism became a powerful force throughout the 19th and 20th centuries and was the driving force behind the unification of politically fragmented territories such as Germany and Italy as well as the break-up of empires such as the Austro-Hungarian, Ottoman, British and French. In the 19th century, the French model of the unitary state was a powerful example and influenced the territorial organization of many of the new states that broke free from imperial rule. In several countries, especially those of Catholic Europe, liberalism was associated with both nationalism and with a strong centralized state capable of wresting control over education and social welfare from the Church. Some countries, such as the Netherlands, Spain and Portugal, had already adopted the French model as a result of the Napoleonic conquests at the beginning of the 19th century. Greece adopted it when it achieved independence from the Ottoman Empire between 1821 and 1829. Despite, or rather because of, its highly fragmented character with thousands of islands, it has been strongly centralist ever since. Belgium broke away from the Netherlands in 1830 to become a monarchy but, despite the presence of a large Flemish-speaking population, opted for 7

John Loughlin

the model of a French unilingual and centralized state and Brussels became transformed by the end of the 19th century from a predominantly Flemish-speaking to a predominantly Frenchspeaking city, situated entirely within Flanders (see the chapter by Deschouwer in this volume). The much diversified states of the Italian peninsula were unified between 1860 and 1870 in a movement known as the Risorgimento, under the leadership of the Piedmontese Camillo Cavour, who became the first Prime Minister of unified Italy. Although there were voices (including that of Cavour) in favour of a decentralized federalist model, in the end the new unified monarchy chose the French model precisely in order to overcome this diversity. Germany, for its part, was no less fragmented than Italy but German nationalists were divided between those who followed Herder in defining nationhood in linguistic and cultural terms, and therefore wished to see a Großdeutschland (Greater Germany), and those liberals who were influenced by the French concept of civic nationalism, who were more in favour of a Kleindeutschland (Smaller Germany). The German-speaking lands, made up of many political entities from kingdoms to bishoprics (the remnants of the pre-modern system described above), were also religiously divided between a Protestant north dominated by Prussia and a Catholic south dominated by Austria. This complexity led to ambiguities about what a German nation-state might look like and whether it should be federalist or unitary. The federalist tradition is probably the older one but during the democratic Weimar Republic and the Nazi Third Reich the model of the unitary nation-state was adopted, which, under the Nazis, evolved into a totalitarian state under the control of the Führer and the Party. This eventually led to the catastrophe of the Second World War. Other states that opted for the French model were Albania, which became independent from the Ottomans in 1912 (Bogdani and Loughlin, 2007), Finland (1918), and many of the states of Eastern and Central Europe (for example, Moldova, Romania and Bulgaria). Turkey, too, became, and remains, a French-style unitary state with Ataturk’s establishment of a secularist state replacing the Ottoman Empire in 1921. The list could continue.

Federal states and nationalism This does not mean that all modern nation-states adopted the unitary and indivisible French model. A minority chose the federal model. As noted above, already in the 18th century, the United States of America passed from being a confederation of colonies to a federal state (see Kincaid in this volume). After the defeat of Germany and Austria in the Second World War, these countries reverted to their federal roots with the encouragement of the victorious allies, especially the United States, for which federalism was synonymous with democracy. Switzerland provides a much older model dating from the ‘Old Confederacy’, which existed between 1291 and 1523 and later confederal models, before it became, in 1848, the Helvetic Confederation, which, despite its name, is a federation rather than a confederation. The United Kingdom was neither a unitary state like France nor a federal state like the United States, but what is sometimes called a ‘Union’ state – that is, a state that has been formed by a series of Acts of Union (Rokkan and Urwin, 1982). Indeed, this was a common way of forming states through dynastic marriages or treaties before the arrival of the modern unitary state according to the French model. Examples are the Union between the Duchy of Brittany and the French crown in pre-revolutionary France and the Austro-Hungarian Empire. What is striking about all these cases of modern federations and union states, nevertheless, is that the nation-state model is retained with the ‘national’ dimension being represented at the federal or union level, where the representative assembly and government are responsible for those affairs that concern the nation as a whole – war, diplomacy, internal security and national 8

Reconfiguring the nation-state

economic development – while the component entities of the state are responsible for those affairs dealt with most appropriately at that level – mainly education, health, social welfare, local government, etc. The important point here is that with regard to the unity of the nation, both unitary and federal states agree that this should not be compromised. The ‘nationalization’ of a federation might occur only over a long period of time, as happened in the United States which had begun, in Kincaid’s terms, as a ‘bi-communal’ federation (i.e. divided between the slaveowning southern states and the northern states opposed to slavery), and ended with ‘coercive’ federalism in which the federal government dominates the states (see the chapter by Kincaid in this volume). Not all unitary, federal or union states have succeeded in maintaining this unity and there are numerous examples of failure or at least of incomplete unification. The ‘first’ United Kingdom of Great Britain and Ireland, which dates from the Act of Union between Great Britain and Ireland in 1801, gave way to the current United Kingdom of Great Britain and Northern Ireland with the partial independence (in the southern counties) of Ireland in 1921. Several federations established by colonial powers after the Second World War also failed: the Malayan Union (1946–48); the Federation of Malaya (1948–63); the Federation of Rhodesia and Nyasaland (1953–63); the West Indies Federation (1958–62); the Mali Federation (1959–60); and the Federal Republic of Cameroon (1961–72). More recently, two former communist federations collapsed: one peacefully in the ‘Velvet Revolution’ – Czechoslovakia; the other with great bloodshed in the Balkan Wars of the 1990s – Yugoslavia. One of the principal reasons for the collapse of these federations, which aimed to unite a number of disparate states and nations, was their failure to construct an overarching and common national identity. Instead, the constituent units adopted individual nationstate building projects, with some of the constituents, for example the Czechs or the Serbs, dominating the federation, which led to a great deal of resentment among the others and undermined the unity of the whole. The Union of Soviet Socialist Republics (USSR) changed its federal structures (which in any case existed largely on paper as it was a system under the strict control of the Communist Party of the Soviet Union) to form a looser Commonwealth of Independent States (CIS). Belgium is an example of a former unitary state, which includes two linguistic communities hostile to each other – the Dutch-speaking Flemish and the French-speaking Walloons – which has become a highly decentralized federation in an attempt to hold the state together. There is uncertainty as to whether this attempt will succeed in the long run mainly because the Flemish seem to have their own nation-building project, while the Walloons are divided and unsure of their political future. Other unitary states have experienced difficulties because of internal nationalisms which challenge the legitimacy of the dominant nation-state, as is the case in Spain where there are powerful Catalan and Basque nationalist movements, and even in France, where there have been challenges especially from Breton and Corsican nationalist movements. The lesson that may be drawn from the latter two cases is that even in countries with a strong unitary and consciously unifying tradition, unification may still be incomplete. Spain, too, despite the Jacobin features of the Francoist state (at least with regard to the notions of unity and indivisibility of the Spanish nation) has been characterized as a ‘multinational’ society because of the continued resistance of Catalonia and the Basque Country to assimilation (Moreno, 2001; Requejo, 2005). On the other hand, the majority of nation-states, whether they be federal, unitary or union, have succeeded in constructing a form of political organization in which the majority of the population do feel an attachment to the ‘nation’, however this is defined, and that this ‘nation’ is identifiable with a ‘state’, whether federal or unitary, with clearly differentiated borders and where the principal source of political legitimacy lies with the core central institutions. 9

John Loughlin

Transformations: from the Trente Glorieuses (1945–75) to globalization The period after the Second World War, called by the French the Trente Glorieuses, was one of economic growth underpinned by Keynesian approaches, and of state expansion through the establishment of welfare states. The latter had already begun before the Second World War, but expanded greatly after the war as part of what Crouch has termed the ‘midcentury consensus’ between the main political families of left and right (Crouch, 1999). From the perspective of this chapter, this period can be seen as the final stage of democratic nation-state building with the introduction of social citizenship. In T.H. Marshall’s formulation, welfare states added the final touch to national citizenship by adding social rights to the already existing political rights of representative democracy, and civil rights (Marshall, 1950). There are various forms of welfare state (Esping-Anderson, 1990), but all have in common a number of basic features: the values of equality and equity for individuals, groups and territories and the duty of the state to intervene in the economy and society in order to achieve these values. The United States adopted increasingly interventionist approaches from the New Deal to the Great Society in the United States (see chapter by Kincaid in this volume). Similar processes of centralization took place outside of the democratic West: in communist states of the USSR and its satellites, and in China, Vietnam, North Korea and Cuba. Finally, the many newly liberated colonies of the developing world adopted the centralized state model, even if some became federations, as with Nigeria. In this section we will concentrate on the experiences of Western states and, in a later section, say something about non-Western states.

Territorial governance in welfare states These developments had implications both for the organization of the central state and administration as well as for territorial governance (Loughlin, 2004). In order better to collect resources from the wealthier sections of society and stronger economic regions and to redistribute them to the weaker sections and to underdeveloped regions, the welfare state found it necessary to centralize. The implication for territorial political organization was that central-local relations took the form of a ‘principal-agent’ relationship: sub-national authorities, whether regions or local governments, increasingly became the ‘agents’ of their ‘principal’, the central state, in the delivery of these services. Furthermore, fiscal policy was controlled by the central government, thus decreasing local fiscal autonomy. More specifically territorial policies, such as regional policy, were conceived mainly in ‘national’ terms – that is, in terms of how policies towards particular weaker regions might help the building up of the overall national economy and society – the nation – rather than in terms specific to those regions themselves. During this period, the European Community was largely ‘residual’: it existed, but more in the background as a support for, and the ‘rescue’ of, the nation-states that were rebuilding themselves after the devastation of the war (Milward, 2000). Other features of state organization and central-local relations during the heyday of the welfare state were territorial symmetry and standardization, and central regulation of sub-national authority activities to minimize variation in service delivery across the territory. The most extreme forms of this approach were in the unitary states of the Nordic countries and in the Napoleonic states of southern Europe, but these general trends could also be found in other states such as the Austrian and German federations and in the United Kingdom. In the United States, similar processes were occurring with the transition from the bi-communal federalism to coercive federalism described by Kincaid (see his chapter in this volume). 10

Reconfiguring the nation-state

The transition from the welfare state to neo-liberalism The period of the Trente Glorieuses came to an end as a result of a series of economic, policy and social crises which began in the late 1960s: the failure of Keynesian economic policies to redress the problems of stagnation and simultaneous inflation; the ‘fiscal crisis of the state’ which found it increasingly difficult to fund the expanding policy programmes associated with the welfare state; the collapse of the Bretton Woods agreement; the oil crises of 1973 and 1979; the social upheavals of the student movements in the 1960s; etc. There were various but related responses to these crises in which the capitalist system successfully ‘reinvented’ itself. First, with the weakening of exchange rate controls brought about by the collapse of Bretton Woods, this was the beginning of a new phase of ‘globalization’ in which truly global financial and commodity markets began to develop (Scholte, 2005; Held et al., 1999). Second, Keynesian macro-economic policy approaches began to be replaced by what subsequently became known as ‘neo-liberalism’ (Harvey, 2004). This was a movement based on the ideas of anti-statist economists, political scientists and philosophers mainly based in the United States, such as Milton Friedman and Friedrich Von Hayek. Their ideas, which had been quite marginal in the 1950s and 1960s, were adopted by Ronald Reagan and Margaret Thatcher in the 1970s and 1980s and became the basis of significant administrative and economic policy reforms. They also were influential in the ‘new globalization’ of this period and formed the background to what became known as the ‘Washington Consensus’, which changed the role and activities of the IMF and the World Bank from Keynesian-type organizations to instruments of neo-liberalism (Stiglitz, 2002). In Europe, the crises and the threat of globalization (meaning the domination of international economic relations by the United States and Japan) sparked off a new round of accelerated European Union integration, beginning with the Single Market project and ending with the Lisbon Treaty. Neo-liberalism, understood in a narrow sense of an approach to economic policy, was part of a wider trend that involved not just the economy but administrative reforms (New Public Management), policy approaches (privatization, deregulation, introduction of market-type processes), which may be summed up as the attempt to reverse the high levels of state intervention and control that had characterized the Trente Glorieuses. It would take us too far outside the scope of this chapter to examine in detail all of these changes (see Harvey 2004 for a survey). What concerns us here is the impact of neo-liberalism on the nation-state and its system of territorial governance.

Trends in territorial governance With regard to territorial governance, these macro-developments led to a number of significant changes which may be summarized as follows (for a fuller account see Loughlin, 2009):

(i) From centralization to decentralization The emphasis during the 1950s and 1960s was on the consolidation of national unity through a process of centralization. The dominant economic paradigm in Western states during this period was Keynesianism which necessitated central government intervention in the economy to manipulate the factors of economic production to produce desirable outcomes such as full employment and the avoidance of market failures. Decentralization did occur during this period (see, for example, Sharpe, 1979). There are, however, different forms of decentralization: political, administrative (sometimes called ‘deconcentration’), industrial or economic. The kind of decentralization that occurred during the Trente Glorieuses tended to be administrative deconcentration 11

John Loughlin

rather than political decentralization in which political decision-making power rather than simply administrative functions were devolved to lower levels. This was basically related to the need to ‘decongest’ highly centralized bureaucratic systems which were becoming increasingly dysfunctional (sometimes called ‘apoplexy at the centre and paralysis at the extremes’). This changed from the mid-1970s when there was a general tendency towards political decentralization as well. A striking example of this tendency is France, which launched a programme of decentralization reforms in 1982 which have significantly reformed the French politico-administrative landscape (Ohnet, 1996; Loughlin, 2009). Today, political decentralization is seen as an element of ‘good governance’ by bodies such as the European Union (EU), the Council of Europe, the UN Human Settlements Programme (UN-HABITAT), the World Bank and the IMF. The imperative for political centralization has left Western Europe and is now seen to apply in all of these organizations to the new democracies of Eastern and Central Europe, to Russia and its former satellites, and to the emergent economies of the developing world.

(ii) From territorial symmetry to asymmetry and the recognition of diversity Part of the ideal of the unified national state was to avoid large disparities across the national territory. In unitary states such as France and Sweden, this meant that systems of territorial governance were basically similar. This ideal of standardized and uniform institutions and policies across the national territory has given way to a greater acceptance of variations in both these regards, sometimes also known as ‘asymmetry’. France, for example, while maintaining the ideal of national unity, accepted some variation in the cases of Corsica and the DOM-TOM2 (Départements d’Outre-Mer/Territoires d’Outre-Mer) and, later, gave to local authorities the right to experiment. Sweden, and then the other Nordic states, experimented with what became known as the ‘selfregulating municipality’, which allowed selected local authorities to free themselves from central regulation and to adopt distinctive approaches in agreed policy areas such as education and health. Italy already had its five ‘special’ regions while Spain in its 1978 Constitution recognized differences between the three ‘nationalities’ (the Basque Country, Catalonia and Galicia) and the ‘ordinary’ Autonomous Communities. Furthermore, the Basque Country was allowed to adopt its own fiscal system, different from the remainder of the country, in which the Basque provinces and not the central state collects taxes and then pays the state for the services it delivers in the region. The question of diversity is even more pertinent (and difficult to achieve) outside of Europe. The former communist states were, in fact, often extremely diverse within and held together by the overarching rule of the Party or the dictator such as Stalin, Mao or Tito. Similarly, the former colonies of Asia and Africa were often artificially constructed states that contained a great diversity of tribes and cultures. With independence, the state held this conglomeration together but often this disintegrated into civil war. The same thing occurred in Yugoslavia when the different republics declared independence. All of these trends have led to a new emphasis on ‘regionalism’.

(iii) From ‘regionalization’ to ‘regionalism’ A useful distinction may be made between ‘regionalization’ and ‘regionalism’, in which the former is understood as a top-down approach to regional issues, controlled by the central state. Regionalization was the typical approach to regional governance and planning from the 1950s until the late 1970s. In contrast to this, ‘regionalism’ is a bottom-up approach in which key political and other actors from within the regions take greater control over their region’s political, social, cultural and economic affairs. This may be done in collaboration with the central state and does not necessarily risk the break-up of the state itself. During the early post-war period, 12

Reconfiguring the nation-state

regionalization was the dominant approach to regional development. From the 1980s, without this being abandoned, there was a much greater element of regionalism. This can be seen in the new approach towards regional policy adopted by the EU with the reform of the Structural Funds in the mid-1980s and the introduction of the principles of subsidiarity and partnership as their operating principles. However, it also occurred in the large nation-states such as France, Spain, the UK and Italy, all of which either introduced or, in the case of Italy, strengthened elected regional governments. The smaller states such as Ireland, Greece, Portugal and the Nordic states such as Denmark and Finland introduced administrative regions, while Sweden set up both administrative regions and elected regional governments. There was a perception in European states that this regional dimension was a prerequisite for accession and, consequently, the new candidate countries of Eastern and Central Europe also began to set up either elected regional governments, as in Poland, or regional administrations as in the other countries. This trend has led to the establishment of political as well as administrative regions. It was also, however, linked to a new regionalist economics approach, with concepts such as the ‘innovative’ and ‘learning’ region (Cooke and Morgan, 1998). The political economy of regionalism meant, therefore, a shift from the top-down ‘regionalization’ approach of the post-war period to a more mobilizing, bottom-up approach carried out by regional actors themselves and primarily concerned with regional economic development. Political regionalism, however, has been much more difficult to achieve outside of Western countries. We have already above adverted to the sometimes volatile mixes of tribal, ethnic, or linguistic groups with the former communist or colonial states. Nevertheless, international organizations such as the Council of Europe and the UN (through its UNHABITAT branch) are now aware of the necessity of finding acceptable solutions to accommodate this diversity, either through federalism or through other ‘consociational’-type arrangements.

(iv) Multi-level governance These developments may be summarized by the concept of ‘multilevel governance’ (MLG) developed by Marks and Hooghe to describe and explain relations among different levels of government in the EU following the passing of the Single European Act in 1986 (Marks and Hooghe, 2001). Of course, there has always been a system of multi-level governance in states if we mean by this simple relations among states, but the term ‘governance’ refers to a new way of conceptualizing such relationships which now involves a range of actors besides governments themselves (Rhodes, 1997). In the formulation of Marks and Hooghe, MLG means that central governments are no longer the exclusive powers in European policy making, as suggested by ‘realist’ and ‘liberal intergovernmentalist’ international relations theories, but now operate alongside a number of other political actors – the EU institutions, regions, local authorities – and even private-sector actors such as business groups. This does not mean that national governments have disappeared or are unimportant. On the contrary, they are undoubtedly still the most powerful and important actors in a system of government. It simply means that they now are much more constrained and have to act in a more collaborative fashion than was the case in the period of the Trente Glorieuses, when European institutions were still largely residual and regional and local authorities were seen as their ‘agents’ acting on their behalf to bring about positive policy outcomes.

(v) From the ‘principal-agent’ model to the ‘choice’ model and nonhierarchical relations among governments Indeed, an important feature of the post-war period of economic expansion and burgeoning welfare state systems was that social and economic policies were decided at the level of central 13

John Loughlin

government (the ‘principal’) while regional and local authorities were often engaged in implementing these policies (as ‘agents’). Since the 1980s, the trend, for example in Sweden and France, has been to have non-hierarchical relations between levels of government. This means a tendency towards a ‘choice’ model in which regional and local authorities may choose distinctive policy approaches and even institutional forms. Of course, all this is within the parameters laid down by central government. In a principal-agent model, it is important that relations among the levels are marked by hierarchy with ‘higher’ and ‘lower’ levels of government. This is obviously the case with regard to relations between the central state and other levels such as the regional and the local. In some cases, there is still a hierarchical relationship between regions (e.g. the Spanish Autonomous Communities or the Belgian régions) and the provinces and local authorities. The tendency today is to adopt a pattern of non-hierarchical relations beneath the national level. Thus, in France, the regions, departments and communes are constitutionally ‘equal’ as is the case in Sweden. In federal systems, of course, there is a division of competences between the federal level and the federated entities, with the latter traditionally being responsible for local governments. However, the complexity of intergovernmental relations in federal systems has led to a more direct relationship between the federation and the other levels, including those ‘below’ the federated entity.

(vi) From fiscal centralization to fiscal decentralization Political decentralization to the regional and local levels implies also fiscal autonomy on the part of these levels (Loughlin and Martin, 2003). However, the post-war period was characterized by fiscal centralization and control by central authorities over spending by regional and local authorities. Since the latter were acting as ‘agents’ of the ‘principal’ (the central government), to carry out public services on its behalf, there was a tendency towards ear-marked grants rather than block grants over which the sub-national authorities could exercise discretion, even in federal states such as Germany. From the 1980s, in line with the above-outlined trends towards greater political decentralization and regionalism, there has, however, been an overall trend in Western democracies towards increased decentralization of spending functions. In effect, political regionalism and local autonomy are not effective unless there is corresponding devolution of financial resources. There has also been a renewed interest in fiscal federalism. In the 1990s there were major reforms of the financing of local authorities in several countries with a trend towards granting greater local fiscal autonomy. There has been a move to give increased spending responsibilities to regional and local authorities, although services tend to be provided through co-operative mechanisms between different tiers of government rather than exclusively through one level of government. That said, it is not always clear that the increased fiscal decentralization, as measured by the proportion of local expenditure as a percentage of overall public expenditure, really reflects increased local decision making. For example, sub-national expenditure figures sometimes include expenditure functions where local government is simply delivering a service effectively controlled by higher levels of government. Examples of where direction is often quite detailed and not simple overall guidance are health and education. Although these functions have been decentralized in many countries, central governments are often still held accountable for them and will tend to use measures (for example specific-purpose grants, or directives) to ensure that local governments meet central goals and targets. Table 1.13 outlines the key features of the shift that has occurred, although it should be kept in mind that these two models are ideal-types and, in practice, in many cases elements of both models may be present. In other words, in reality they are marked by hybridity. 14

Reconfiguring the nation-state

Table 1.1 The changing paradigm in regional policy and territorial governance Classical model Lead organization Central and regional government relationship Policy development Type and nature of decentralization Style of planning Type of regional plan Territorial approach Objectives Mode of operation Special focus Key instruments Government aids Time scale

Contemporary model

Central government Hierarchy

Regional level Equality of levels of government

Top down/centralized Administrative deconcentration but political, fiscal, policy and functional centralization Hierarchical

Collective/negotiated/contractual Political, functional and fiscal decentralization alongside administrative deconcentration National strategic direction, local variation and implementation Strategic with spatial focus

Comprehensive and multi-sector plans Territorial symmetry Promoting growth, employment creation, increased investment Reactive, project-based Problem areas Bureaucratic regulation, public-sector provision Incentive schemes, business aid and hard infrastructure Open ended

Territorial asymmetry Sustainable development and increased competitiveness Pro-active, planned, strategic Balanced and harmonious development of all regions Reduced financial support, mixed public/private/voluntary provision Business environment, and soft infrastructure Multi-annual planning periods

Source: Adapted from Roberts and Lloyd, 1999; Bachtler and Yuill, 2001; and Loughlin, 2009

The concept of hybridity During the period of the Trente Glorieuses, state-driven Keynesian and social welfare policy was hegemonic. Other models from the traditional Marxist (the old-style Stalinist communist parties), or the neo-Marxist New Left (e.g. the Frankfurt School or the parties of the extreme Left), or the New Right (Von Hayek, Friedman, etc.) were quite marginal in public discourse. As mentioned above, the mainstream parties of the Right and Left – Social Democrats, Christian Democrats, UK Conservatives and Labour – entered into what Colin Crouch (1999) terms the ‘mid-century consensus’ to support and develop the welfare state/Keynesian model. As mentioned above, this led to a standardized model of territorial governance for the whole of the national territory which emphasized symmetry and uniformity. Today, these states are characterized by ‘hybridity’. What we mean by this concept is that there may exist within the same national state a variety of institutional and policy models, which are sometimes in competition with each other but where none of them is hegemonic. To some extent this was previously true of federal states which may have been constructed precisely to accommodate such diversity. In Germany, for example, the different Länder retained quite different political and cultural traditions, from Social Democracy in the northern parts to Christian Democracy mainly in the south. In the United States, as we have seen, the political cultures of the south differed from those of the north and north-east. In the United Kingdom, Scotland, Wales and Northern Ireland had distinctive administrative traditions. However, as we saw above, the tendency in both unitary and federal states has been to reduce this diversity and impose a common approach and institutional set-up. At least, there developed a hegemonic approach 15

John Loughlin

which dominated other models to such an extent that the latter became marginal, as we saw with the ideas of the New Right mentioned above. This hegemony has now given way in both unitary and federal states to the re-emergence of these other models alongside the previously dominant ones. A good example is the United Kingdom, where the south-east of England is much more wedded to Thatcherite and New Labour ‘neo-liberal’ approaches, while Scotland, Wales and Northern Ireland have retained a more Old Labour or social-democratic approach to public policy. There has also been a ‘loosening up’ of what had previously been rigid institutional relationships between the different parts of the political system. Again, the UK may be taken as an exemplar. Following devolution, it now has a UK parliament and government, a Scottish Parliament (SP), a National Assembly for Wales (NAW), and a Northern Ireland Assembly (NIA). Relations between the SP and the UK government resemble those of a federation while those between the UK government and the NAW and NIA resemble much more those of a regionalized unitary state. Similarly both Spain and Italy have ‘special’ regions along their ‘ordinary’ regions. Even France, the paragon of the unitary and uniform state, now recognizes several different territorial arrangements and even has three different systems of public administration.4 In fact, France in 2003 modified its Constitution to give expression to this diversity (Loughlin, 2007). The shift that we have described above has led to either the creation of such diversity or its intensification where it already existed. As a result the old clear dichotomy between federal and unitary states has given way to more complex patterns of both types of state. Both Daniel Elazar and Ronald Watts have developed new typologies of federal states, while others such as Bullmann (1996) and Loughlin (1996) have attempted to refine in a more exact way the complex patterns of unitary states.

Conclusions This shift should be related to the broader changes such as globalization, Europeanization, and the political, economic and social transformations that have accompanied them. The nationstate has not disappeared but now exists in new internal and external configurations. It is tempting to return to the early part of this chapter and to see contemporary developments as the re-emergence of those older patterns characteristic of European societies before the emergence of the nation-state as the hegemonic model of territorial governance. There are indeed some similarities. The centrality and absolute sovereignty of the post-Westphalian state has been largely relativized even if national governments do remain the most important political actors in both international and domestic politics. Although a Europe of the Regions has not materialized in the sense of a federal Europe in which regions replace nation-states as the units of the federation, nevertheless regions of various kinds have emerged as key political actors alongside national governments and, indeed, a variety of arrangements of sub-regional or local authorities, captured by the term ‘multi-level governance’. Is this a purely Western or even just European set of developments? I would argue that while many of the changes have originated in the West (more in the United States than in Europe in many instances), what we have described above is relevant to all parts of the world today. Furthermore, the European experience in many respects provides a template for changes in other parts of the world which are dealt with in this Handbook. First, Europe is the birthplace of both modernity and the nation-state and it was the experiences of imperialism and colonization that exported this model to the rest of the world. Most independent states since the 19th century, but especially following the Second World War and, later, the collapse of the communist systems have adopted the nation-state model. Second, it is in Europe that political and economic 16

Reconfiguring the nation-state

regionalism have been most strongly developed in all their varieties. This is true both of Europe’s unitary regionalized states and its federal states. Even the example of the EU has been a model for similar regional organizations such as the Mercado Común del Sur (Mercosur) in South America and, to some extent, the North American Free Trade Association (NAFTA). It is also true, however, that there exist models of political organization that owe relatively little to Europe. The most striking example of this is China, which has developed a unique political and economic system that combines both one-party communist rule with capitalism and engagement in the globalized world economic system. However, even in China there are issues of territorial governance, centralization vs. decentralization, and regional diversity. Already, there are distinct regional regimes in Hong Kong, Macao and Tibet. One might expect to see further pressures towards diversification and hybridity even in countries such as this.

Notes 1 The following draws largely on Loughlin, 2011. 2 The French overseas territories, which were former colonies and dependencies. 3 The table was compiled by Dr Mohamed Nada, with whom the author is working at UN-HABITAT to devise a new system of regional governance in Egypt. 4 The national administration, the territorial administration serving the regions and departments, and the hospital administration.

References Alter, Peter (1994) Nationalism, 2nd edn (London: Edward Arnold). Bachtler, John and Douglas Yuill (2001) Policies and Strategies for Regional Development: A Shift in Paradigm? Regional and Industrial Policy Research Paper No. 46, University of Strathclyde. Badie, Bertrand (1995) La Fin des territoires (Paris: Fayard). Bogdani, Mirela and John Loughlin (2007) Albania and the European Union: The Tumultuous Journey Towards Integration and Accession (London: I.B. Tauris). Brown, Elizabeth A.R. (2012) ‘Feudalism’, Encyclopædia Britannica Online (Encyclopædia Britannica Inc.), www.britannica.com/EBchecked/topic/205583/feudalism (accessed 9 July 2012). Bryce, James (1864) The Holy Roman Empire (Oxford: T. & G. Shrimpton). Bull, Hedley (1995) The Anarchical Society: A Study of Order in World Politics, 2nd edn (Basingstoke: Macmillan). Bullmann, Udo (1996) ‘The Politics of the Third Level’, Regional and Federal Studies Vol. 6, Issue 2 (June): 3–19. Burns, James Henderson (1988) The Cambridge History of Medieval Political Thought c. 350–c. 1450 (Cambridge: Cambridge University Press). Burns, James Henderson, with the assistance of Mark Goldie (eds) (1991) The Cambridge History of Political Thought, 1450–1700 (Cambridge: Cambridge University Press). Chernilo, Daniel (2007) A Social Theory of the Nation State: The Political Forms of Modernity Beyond Methodological Nationalism (London: Routledge). Colley, Linda (2005) Britons: Forging the Nation 1707–1837, 2nd edn (New Haven, Conn. and London: Yale Nota Bene). Cooke, Philip and Kevin Morgan (1998) The Associational Economy: Firms, Regions and Innovation (Oxford: Oxford University Press). Crouch, Colin (1999) Social Change in Western Europe (Oxford: Oxford University Press). Duchacek, Ivo (1986) The Territorial Dimension of Politics: Within, Among, and Across Nations (Boulder, CO and London: Westview Press). Dyson, Kenneth (1980) The State Tradition in Western Europe: A Study of an Idea and Institution (Oxford: Martin Robertson). Esping-Anderson, Gøsta (1990) The Three Worlds of Welfare Capitalism (Cambridge: Polity Press; Princeton: Princeton University Press). Greenfeld, Liah (1992) Nationalism: Five Roads to Modernity (Cambridge, MA: Harvard University Press). Guibernau, Montserrat (1996) Nationalisms: The Nation-state and Nationalism in the Twentieth Century (Cambridge: Polity Press). 17

John Loughlin

Harvey, David (2004) A Brief History of Neoliberalism, 2nd edn (Oxford: Oxford University Press). Hayward, J.E.S. (1983) Governing France: The One and Indivisible Republic, 2nd edn (London: Weidenfeld and Nicolson). Held, David, David Goldblatt and Jonathan Perraton (1999) Global Transformations: Politics, Economics and Culture (Cambridge: Polity Press). Loughlin, John (1996) ‘Representing Regions in Europe: The Committee of the Regions’, Regional and Federal Studies Vol. 6, Issue 2 (June): 147–65. ——(2004) ‘The ‘Transformation’ of Governance: New Directions in Policy and Politics’, Australian Journal of Politics and History Vol. 50, No. 1 (March): 8–22. ——(2007) Subnational Government: The French Experience (Basingstoke: Palgrave Macmillan). ——(2009) ‘The ‘Hybrid’ State: Reconfiguring Territorial Governance in Western Europe’, Perspectives on European Politics and Society Vol. 10, No. 1: 49–66. ——(2011) ‘Federal and Local Government Institutions’, in Daniele Caramani (ed.) Comparative Politics, 2nd edn (Oxford: Oxford University Press). Loughlin, John and Steve Martin (2003) International Lessons on Balance of Funding Issues (Whitehall: Office of the Deputy Prime Minister), Nov., www.local.odpm.gov.uk/finance/balance/interles.pdf. Loughlin, John and Guy B. Peters (1997) ‘State Traditions, Administrative Reform and Regionalization’, in M. Keating and J. Loughlin (eds) The Political Economy of Regionalism (London: Routledge), 41–62. Marks, Gary and Liesbet Hooghe (2001) Multi-Level Governance and European Integration (Lanham, MD and Oxford: Rowman & Littlefield). Marshall, T.H. (1950) Citizenship and Social Class, and Other Essays (Cambridge: Cambridge University Press). Milward, Alan S. (2000) The European Rescue of the Nation-state, 2nd edn (London: Routledge). Moreno, Luis (2001) The Federalization of Spain (London: Routledge). Ohmae, Kenichi (1996) The End of the Nation State: The Rise of Regional Economies (New York: Free Press Paperbacks). Ohnet, Jean-Marc (1996) Histoire de la décentralization française (Paris: Livre de poche). Ozment, Steven (1980) The Age of Reform: An Intellectual and Religious History of Late Medieval and Reformation Europe (New Haven and London: Yale University Press). Requejo, Ferran (2005) Multinational Federalism and Value Pluralism: The Spanish Case (London: Routledge). Rhodes, R.A.W. (1997) Understanding Governance: Policy Networks, Governance, Reflexivity, and Accountability (Oxford: Oxford University Press). Roberts, P. and G. Lloyd (1999) ‘Institutional Aspects of Regional Planning, Management and Development: Models and Lessons Learned from the English Experience’, Environment and Planning B 26: 517–31. Rokkan, Stein and Derek Urwin (eds) (1982) ‘Introduction: Centres and Peripheries in Western Europe’, in The Politics of Territorial Identity: Studies in European Regionalism (London: Sage), 1–17. Schmidt, Vivien (1990) Democratizing France: The Political and Administrative History of Decentralization (Cambridge: Cambridge University Press). Scholte, Jan Aart (2005) Globalization: A Critical Introduction, 2nd edn (Basingstoke: Palgrave Macmillan). Sharpe, L.J. (ed.) (1979) Decentralist Trends in Western Democracies (London: Sage Publications). Skinner, Quentin (1978) The Foundations of Modern Political Thought (Cambridge: Cambridge University Press). Smith, Anthony D. (2010) Nationalism: Theory, Ideology, History, 2nd edn (Cambridge: Polity Press). Spruyt, Hendrik (1994) The Sovereign State and its Competitors: An Analysis of Systems Change (Princeton: Princeton University Press). Stiglitz, Joseph (2002) Globalization and its Discontents (London: Penguin). Wright, Julian (2003) The Regionalist Movement in France, 1890–1914: Jean Charles-Brun and French Political Thought (Oxford: Oxford University Press).

18

2 Typologies of federalism Ronald L. Watts

Introduction The major hypothesis of this book is that the traditional distinction between federal and unitary states has to some degree given way to a complex typology of ‘hybrid’ states. Although in many cases states may have a predominant form that is primarily federal or unitary, increasingly there are examples of political entities combining some elements of both forms in widely varying combinations. These hybrids have occurred because statesmen, faced with a variety of factors relating to economic and strategic issues and the need to recognize and accommodate internal territorial diversity, have often been more interested in pragmatic political solutions than in theoretical purity. Today, few federations meet the criteria of pure federalism. Many predominantly federal systems, especially more recent ones, have incorporated constitutional provisions or developed practices involving some unitary elements in their intergovernmental relations or financial arrangements. Even the United States, usually regarded as the classic model of federalism, has evolved into what has been described by John Kincaid as ‘coercive federalism’ (Kincaid, 1990). At the same time, many undoubtedly unitary systems have incorporated some territorial decentralization including federal elements. France, Japan and Sweden provide examples. Then there are those political systems such as Spain, Italy and the United Kingdom that have evolved such a mixture of federal and unitary elements that their classification as unitary or federal has become a matter of debate (re: Spain, Watts, 2010; re: Italy, Caravita, 2002; re: United Kingdom, Watts, 2007). As John Loughlin has emphasized, the pressures of changing economic conditions, financial interdependence, conflicting demands of the welfare state, and neo-liberalism have led to a preponderance of hybrid types of state in which federal and unitary internal relations have come to exist in widely varying combinations influenced by their particular economic and political conditions and orientations (Loughlin, 2009: 17). Notable features of these developments are that no single new model has emerged and that there has developed an enormous variety of forms and hybrids. Indeed, looking first just at those political systems that are federal or claim to be federal, the number of these has expanded enormously over the past half century, but there have been many variations (Watts, 2008: 18–19). 19

Ronald L. Watts

The latter half of the 20th century saw a proliferation of federations as well as other federal forms to unite or accommodate multi-ethnic communities in former colonial areas in Asia, Africa, South America and Europe. Between 1945 and the early 1960s, new federations, quasifederations or hybrids were founded in Asia, for example in Indo-china (1948), Burma (1948), Indonesia (1949), India (1950), Pakistan (1956), Malaya (1948 and 1957) and Malaysia (1963); in the Middle East, for example in the United Arab Emirates (UAE, 1971); in Africa, for example in Libya (1951), Ethiopia (1952), Rhodesia and Nyasaland (1953), Nigeria (1954), Mali (1959), the Congo (1960), and Cameroon (1961); and in the Caribbean, for example the West Indies Federation (1958). Among federations founded or restored in Europe were also those of Austria (1945), Yugoslavia (1946), Germany (1949), and Czechoslovakia (1969). Between 1960 and the late 1980s, a number of these federations were temporarily suspended or abandoned as it became clear that federal systems were not a panacea and that there were limits to the appropriateness of federal solutions in particular circumstances. Nevertheless, the two decades since the early 1990s have seen a further burst of federal solutions of varying types. Belgium transformed its constitution into a fully fledged federal one in 1993. South Africa in 1996 confirmed a constitutional hybrid of federal and unitary features. Spain, as a result of the operation of its 1978 constitution, has become, in practice, increasingly a federation in all but name. In South America, following earlier failures, Brazil (1988), Argentina (1994), and Venezuela (1999) adopted new federal constitutions, and Mexico from 2000 onward has been bringing greater reality to its federalism. After the break-up of the Union of Soviet Socialist Republics (USSR) in 1993, Russia adopted a new federal constitution. Ethiopia in 1995 and Nigeria in 1999, after the restoration of civilian rule, established new federal constitutions. Also, in the effort to resolve severe previous conflicts, federal experiments, not always successful, have been attempted in Bosnia-Herzegovina (1995), Iraq (2005), Sudan (2005), and the Democratic Republic of the Congo (2006). Federal systems have also recently been under consideration in Nepal and the Philippines. In Italy, too, there has been an evolution towards a federal system. The United Kingdom, while not a federal system, has adopted a significant measure of devolution for Scotland, Wales and Northern Ireland. Thus, as Daniel J. Elazar noted, in the early years of the 21st century, the world appears to be in the midst of a paradigm shift from a world of sovereign unitary nation-states to a world of states with interstate and internal linkages of a constitutionally federal character. Many of these, though in varying forms and degrees, involve hybrid characteristics with both federal and unitary elements.

The federal-unitary distinction The fundamental distinction between federal and unitary systems lies in the location of ultimate political authority or sovereignty. In unitary political systems, the ultimate authority, constitutionally or in practice, lies with the central government. There may be administrative, legislative, or financial decentralization to constituent units, but that occurs at the discretion or will of the central government, which may, if it so determines, overrule constituent units on any matter. Thus, in unitary political systems, there is a single central source of ultimate political and legal authority. Federal political systems are, by contrast, polities in which there are two (or more) orders of government combining elements of shared rule (collaborative partnership) through common institutions with territorial self-rule (constituent-unit autonomy) for the governments of the constituent units in an intergovernmental constitutional relationship that is not determined by the central government alone. The key here is not the degree of decentralization, but the degree of constitutionally guaranteed autonomy that the constituent units may exercise (Kincaid and Tarr, 20

Typologies of federalism

2005). The broad genus of federal systems exhibiting shared rule and regional shared rule encompasses a whole spectrum of more specific non-unitary species ranging from constitutional unions, quasi-federations and federations, to confederations, federacies and beyond. As in a spectrum, the categories are not exclusively delineated, but shade into each other at the margins.

Types of federal political systems Elazar identified as specific categories within the range of federal systems: unions, constitutionally decentralized unions, federations, confederations, federacies, associated statehoods, condominiums, leagues and joint functional authorities (Elazar, 1987: 38–64). To these we may add hybrids that combine the characteristics of different kinds of systems. Those that are predominantly federations in their constitution and operation, but have some overriding federal government powers more typical of a unitary system, may, following K.C. Wheare’s example, be described as ‘quasi-federations’ (Wheare, 1963: 19). In any attempt at classification of types of federal systems, there are two important points to note. First, we must take account of the difference between constitutional form and operational reality. In many political systems, political practice has transformed the way the constitution operates. In Canada and India, for example, the initial constitution was clearly quasi-federal, containing some central overriding powers more typical of unitary systems. However, in both cases, the use of these powers has been moderated (virtually completely in the Canadian case) such that the operational reality today comes closer to that of a full-fledged federation. Other particularly notable examples of the impact of operational practice in a more unitary direction have occurred in Malaysia, Pakistan, Russia, South Africa and Venezuela. To characterize federal systems, therefore, it is necessary to study both their constitutional law and their politics and how these have interacted. Second, while knowledge about the structural character of a federal political system is important to gain an understanding of its character, equally important is the nature of its political processes. Significant characteristics of federal processes include the degree of predisposition to democracy, non-centralization as a principle expressed through multiple centres of decision making, open political bargaining as a major feature of the way political decisions are arrived at, the operation of checks and balances to avoid the concentration of political power, and a respect for constitutionalism, since each order of government derives its authority from an agreed constitution. Within the genus of federal political systems, we may identify the following distinct species:

Unions These are polities compounded in such a way that the constituent units preserve their respective integrities primarily or exclusively through the common organs of the general government rather than through dual government structures. New Zealand and Lebanon have been such examples. So also was Belgium, where central legislators served also with a dual mandate as regional or community councillors, prior to Belgium becoming a federation in 1993. Another example was the United Kingdom prior to the substantial devolution in 1999.

Constitutionally decentralized unions These are basically unitary in form, in the sense that ultimate authority rests with the central government, but incorporate constitutionally protected sub-national units of government or constitutionally entrenched rights for certain minorities. There are numerous examples, including 21

Ronald L. Watts

(constitution dates in parentheses) Antigua and Barbuda (1981), Republic of Cameroon (1972, amended 1996), Republic of Colombia (1991), Republic of Fiji Islands (1997), French Republic (1958), Republic of Ghana (1992), Georgia (1995), Republic of Indonesia (1959), Italian Republic (1947), Japan (1947), Republic of Namibia (1990), Kingdom of the Netherlands (1815, amended 2002), Papua New Guinea (1975), Portuguese Republic (1976), Solomon Islands (1978), Democratic Socialist Republic of Sri Lanka (1978), Kingdom of Sweden (1975), Republic of Tanzania (1977, amended 1984), United Kingdom of Great Britain and Northern Ireland (1999), Ukraine (1996), and Republic of Vanuatu (1980) (Watts, 2008: 16, table 6).

Federations These are compound polities combining strong constituent units and a strong general government, each possessing powers delegated to it by the people through a supreme constitution, each empowered to deal directly with the citizens in the exercise of its legislative, administrative and taxing powers, and each with its major institutions directly elected by the citizens. Federations represent a particular species of federal system in which neither the federal nor the constituent units are constitutionally subordinate to the other (i.e., each has sovereign powers derived from the constitution that is not unilaterally amendable rather than power derived from another level of government). Currently, some two dozen countries meet or claim to meet the basic criteria of an established functioning federation. These (including quasi-federations, and with dates of original foundation and current constitution in parentheses) include the Argentine Republic (1853, 1994), Commonwealth of Australia (1901, 1901), Federal Republic of Austria (1920, 1945), Republic of Belau (1981, 1981), Kingdom of Belgium (1993, 1993), Federative Republic of Brazil (1891, 1988), Canada (1867, 1867 and 1982), Union of the Comoros (1978, 2001), Federal Democratic Republic of Ethiopia (1995, 1995), Federal Republic of Germany (1949, 1949), Republic of India (1950, 1950), Malaysia (1963, 1963), United Mexican States (1824, 1917), Federated States of Micronesia (1979, 1986), Federal Republic of Nigeria (1954, 1999), Islamic Republic of Pakistan (1973, 1973), Russian Federation (1993, 1993), Federation of St Kitts and Nevis (1983, 1983), Republic of South Africa (1996, 1996), Kingdom of Spain (1978, 1978), Swiss Confederation (1848, 2000), UAE (1971, 1996), the United States of America (1789, 1789), and Bolivarian Republic of Venezuela (1811, 1999). In the cases of South Africa and Spain, their constitutions have avoided the label ‘federation’, but their structures are those of a quasi-federation (Watts, 2008: 41–42, 49–50). Although less clear, some Italians claim that Italy is already a federation. On the other hand, while the UAE explicitly describes itself as a federation, its structure and process are more akin to those of a confederation (Watts, 2008: 55–56). In addition to the examples of established federations listed above, post-conflict federal experiments have been attempted in the Republic of Bosnia and Herzegovina (1995), Iraq (2005), and the Democratic Republic of the Congo (2006). A similar post-conflict experiment in Sudan (2005) failed with the country being split in 2011.

Confederations These occur where pre-existing polities join together to form a common government for certain limited purposes (usually for foreign affairs, defence or economic purposes), but the common government is dependent upon the will of the constituent governments, being composed of delegates from the constituent governments, and therefore has only an indirect electoral and fiscal base. Historical examples include Switzerland (1291–1798 and 1815–47) and the United States of America (1781–89). In the contemporary world, the European Union (EU) is primarily a 22

Typologies of federalism

confederation, although it has increasingly incorporated some features of a federation (see below). Other examples have been Benelux (1944), the Caribbean Community (CARICOM, 1973), and the Commonwealth of Independent States (CIS, 1991). As mentioned above, while the constitution of the UAE refers to it as a ‘federal state’, in structure it is fundamentally confederal.

Federacies This term, coined by Elazar (1987: 7, 54–57) refers to political arrangements where a smaller unit or units, usually islands, are linked to a larger polity, but the smaller unit or units retain substantial autonomy, have a minimum role in the government of the larger one, and the relationship can be dissolved only by mutual agreement. Examples are the relationships of the Åland Islands to Finland, the Azores Islands to Portugal, the Faroe Islands to Denmark, Greenland to Denmark, Guernsey to the United Kingdom, the Isle of Man to the United Kingdom, Jammu and Kashmir to India, the Madeira Islands to Portugal, the Northern Mariana Islands to the United States, and Puerto Rico to the United States.

Associated states These relationships are similar to federacies, but differ in that they can be dissolved by either of the units acting alone on pre-arranged terms established in the constituting document or treaty. Examples include the relationships of Bhutan to India, the Cook Islands to New Zealand, Lichtenstein to Switzerland, Monaco to France, the Netherlands Antilles to the Netherlands, Niue to New Zealand, and San Marino to Italy.

Condominiums These occur where political units function under the joint rule of two or more external states in such a way that the inhabitants have substantial internal self-rule. Examples have been Andorra, which functioned under the joint rule of France and Spain (1278–1993), Vanuatu, which operated under a British-French condominium (1906–80), and Nauru, which was under a joint Australia-New Zealand-United Kingdom condominium (1947–68).

Leagues These are linkages of politically independent polities for specific purposes that function through a common secretariat, rather than a government, and from which members may withdraw unilaterally. Examples have been the Arab League, the Association of Southeast Asian Nations (ASEAN), the Baltic Assembly, the Commonwealth of Nations, the North Atlantic Treaty Organization (NATO), the Nordic Council, and the South Asian Association for Regional Cooperation (SAARC).

Joint functional authorities These are agencies established by two or more polities for joint implementation of a particular task or tasks. Three examples from many such agencies are the North Atlantic Fisheries Organization (NAFO), the International Atomic Energy Agency (IAEA), and the International Labour Organization (ILO). Such joint functional authorities may also take the form of transinternational border organizations established by adjoining sub-national units, as in the case of 23

Ronald L. Watts

the interstate grouping for economic development involving four regions in Italy, four Austrian Länder, two then-Yugoslav republics and one West German Land established in 1978, and the case of the interstate Regio-Baseliensis involving Swiss, German and French cooperation in the Basel region.

Federal-unitary and federal-confederal hybrids Some political entities combine characteristics of different political systems. Hybrids occur because statesmen need pragmatic solutions to particular political and economic circumstances. Broadly speaking, two types of hybrids are notable. One type consists of federal-unitary hybrids. Many political systems, which in their constitutional form and political operation predominantly have the character of federations, have incorporated some provisions or practices more typical of a unitary system, enabling the general government to override the autonomy of the constituent units in certain circumstances. Wheare (1963: 19) labelled such systems ‘quasifederal’. While such hybrids have proliferated in recent decades, they are by no means new. The original British North America Act founding the Canadian federation in 1867 included a number of provisions (Canada Constitution Act, 1867, ss.56 and 57, as extended by s.90 and s.92(10)(c)), enabling the federal government to override provincial legislation by disallowing provincial legislation or legislating in certain fields that would normally fall under provincial jurisdiction. In the early decades of the Canadian federation, these federal overriding powers were used frequently. These powers remain in the Canadian Constitution, although, by convention, they have not been employed for half a century. Consequently, in practice, Canada has evolved from a quasi-federation into a more completely federal one. A number of federations established during the 20th century fall into the category of ‘quasi-federations’. Notable examples are India, Pakistan and Malaysia, the constitutions of which were predominantly federal but have included significant overriding central emergency powers that have been exercised quite frequently, although in the case of India there has now been some attenuation in their use. Among the countries listed above in the category of federations, others that may be classified as predominantly federations but with some significant unitary elements in their constitutions or practice, and therefore described as quasi-federations are Russia, Argentina, Venezuela and Comoros (Watts, 2008: 14). Both South Africa and Spain have most of the features of a full-fledged federation, but avoid the label and incorporate some unitary features. One area in which constituent-unit autonomy in nominal federations has often in practice been undermined is financial dependence upon federal transfers. In most federations, constituent units have become heavily dependent on federal transfers. Indeed, in South Africa, Nigeria, Mexico, Spain and Belgium, transfers from the federal government constitute between 68% and 96% of the revenue of the constituent governments – a very high degree of dependence (Watts, 2008: 105). By contrast, in some decentralized unitary systems, such as Sweden and Japan, the proportion of own-source revenue in the total of constituent revenues is high; hence, their financial dependence on the central government is significantly lower (Watts, 2005). A second notable type of hybrid involves the combination of features typical of a federation and a confederation. Germany, while predominantly a federation, has a confederal element in the Bundesrat, its federal second chamber, which is composed of instructed delegates of the Land governments who have a veto over a significant portion of federal legislation. A unique hybrid combining much more fully the characteristics of a confederation and a federation is the EU following the Maastricht Treaty of 1993. The EU is basically an economic confederation but with some of the features of a federation. Derived from its confederal roots 24

Typologies of federalism

are the prominence of the intergovernmental Council of Ministers, the Commission composed of one commissioner from each member state, the fiscal minimalism and reliance largely on financial transfers from the member states, and the retention by member states of most traditional powers over domestic and foreign policy. On the other hand, there are a number of features not normally found in a confederation and more akin to those found in federations. These include the roles of the European Commission, which in many respects is similar to that of an executive body, of the directly elected European Parliament with its co-decision powers, and of the Court of Justice of the EU as a judicial body enforcing the supremacy of EU law. In addition, voting in the Council of Ministers on a wide range of issues has been increasingly based on the qualified majority principle rather than unanimity. In terms of functions, there has been a distribution of responsibilities, with trade and commerce heavily in the hands of the EU and social policy in the hands of the member states. Yet another federal-confederal hybrid is the UAE. The 1971 provisional constitution, which was made permanent in 1996, proclaimed the UAE to be a ‘federal state’, but in form it is largely confederal (Watts, 2008: 55–56). The Supreme Council of Rulers consisting of the seven non-elected traditional emirate rulers is the highest federal authority, having both legislative and executive authority. Because the emirs derive their status from their position within the emirates, this gives the Supreme Council a confederal character. There is a unicameral federal legislature, but it is a consultative body only. It has been oil wealth that has largely contributed to the UAE’s stability.

Variations in federations Focusing more on the specific category of federations (including quasi-federations), one can identify certain structural features and political processes common to them. These are:  The existence of at least two orders of government – one for the whole federation and the other for the constituent regional units, each acting directly on its citizens;  A formal constitutional distribution of legislative and executive authority and allocation of revenue sources between the two (or more) orders of government ensuring some areas of genuine autonomy for each government;  Provision for the designated representation of distinct regional views within the federal policy-making institutions, usually provided by the particular form of the federal second chamber;  A supreme written constitution not unilaterally amendable and requiring for amendments the consent of a significant portion of the legislatures, governments, or voters of the constituent units;  An umpire (in the form of courts, provision for referendums, or an upper house with special powers for resolving intergovernmental conflicts); and  Processes and institutions to facilitate intergovernmental collaboration for those areas where governmental responsibilities are shared or inevitably overlap. While these features generally characterize federations, federations and quasi-federations have exhibited many variations in the application of the federal idea. There is no simple ‘ideal’ or ‘pure’ form of federation. Ultimately, federalism represents a pragmatic prudential technique, the effectiveness of which depends on the relation of the particular form in which it is adopted or adapted to its particular political and economic circumstances. No single typology can account adequately for the rich range of variations found in federations. Consequently, in comparative terms, we may identify 10 types of significant difference among federations: 25

Ronald L. Watts

1 Maturity of federations Broadly speaking, we may classify federations into four groups depending on their degree of maturity. In one category are the ‘mature’ federations that have operated effectively for at least half a century or more. In this category are the United States (1789), Switzerland (1848), Canada (1867), Australia (1901), Austria (1945), Germany (1949), and India (1950). Each of these exhibits virtually all of the characteristics of a federation described above and has displayed a prolonged period of federal stability. Unlike some of the more recent, ‘emergent’ federations referred to below, these mature federations have, in their evolution, developed both federal and state governments which have not only formally autonomous powers, but which have exercised them fully in practice. Interestingly, the two examples that began with clearly quasi-federal characteristics have, in their evolution, virtually eliminated (Canada) or reduced (India) the operation of these quasi-federal elements. A second category may be described as ‘emergent’ federations. These are federations created during the past 50 years and are still in the process of establishing their equilibrium. European examples are Spain and Belgium, but there are many others in Latin America, Asia and Africa. As a group, they tend to be more centralized than the mature federations, examples being Mexico, Nigeria and Ethiopia. Seven others are highly centralized quasi-federations, as exemplified by South Africa, Russia, Argentina, Malaysia, Venezuela, Comoros and Pakistan. Many of them have been prone to instability as illustrated by the Latin American examples of Argentina, Brazil, Mexico and Venezuela, or have even experienced frequent periods of military rule as in Nigeria and Pakistan. A third category consists of four recent efforts in post-conflict situations to establish federal experiments in order to resolve those conflicts. One in Sudan has already failed with the splitting of Sudan in 2011. The other three, Bosnia and Herzegovina, Iraq and the Democratic Republic of the Congo have yet to operate functionally. The fourth category is that of ‘failed federations’. These are important because much can be learned from the pathology of federations (Watts, 2008: 179–88; Franck, 1966; Hicks, 1978; Young, 1995: chs 10, 11). In this category are two groups of federations. One is the group of communist federations, for instance the USSR, Czechoslovakia and Yugoslavia, which all eventually disintegrated. The other is a significant number of overly ambitious post-colonial federations such as Indochina (1945), Burma (1948), Indonesia (1949), Libya (1951), Ethiopia (1952), Rhodesia and Nyasaland (1953), Mali (1959), the Congo (1960), Cameroon (1961), and the West Indies (1958), which did not last long.

2 Bases of internal diversity Since William Livingston (1956), scholars have been conscious of the fundamental importance of the social forces underlying the pressures for diversity within federations. The causal interaction of these forces with political institutions and processes has affected both the creation and subsequent operation of federations (Watts, 2008: 21; Moreno and Colino, 2010). Since the basis of internal diversity has varied among federations, analysis of these differences provides important insight into understanding their creation and operation. In general terms, one may distinguish between those federations where regional diversity is largely territorial or historical and those where regional diversity is deeply rooted in internal cultural, linguistic, ethnic, religious and even national differences (Burgess, 2007). Among the former would be Argentina, Australia, Austria, Brazil, Germany, Mexico and the United States, where there may be subtle but not profound cultural, regional differences, but geography and history provide the fundamental basis for institutionalizing their diversity. 26

Typologies of federalism

By contrast, in such federations as Belgium, Bosnia and Herzegovina, Canada, Ethiopia, India, Malaysia, Nigeria, Pakistan, Spain and Switzerland, internal linguistic, religious, ethnic and even national, regional differences have provided insistent pressures both for maintaining regional distinctiveness and autonomy and for inclusiveness in the operation of their federative institutions.

3 Creation by aggregation or devolution Federations have been created in three different ways, and this has affected their design and subsequent operation. One way is the aggregation of formerly independent separate states. The United States, Switzerland and Australia provide classic examples. In these federations, the process led, in the early stages, to an emphasis on retaining a substantial element of autonomy for the federating units. A second pattern has been through devolution from a previous unitary regime. Examples of this pattern are Austria, Belgium, Germany (after the Third Reich), Nigeria, Ethiopia, South Africa and Spain. This pattern has, at least initially, usually resulted in an emphasis on retention of substantial central powers, although Belgium provides a notable exception. A third pattern has been the combination of these two processes – Canada, India and Malaysia providing examples. The creation of the Canadian federation involved a devolution to two new provinces (Ontario and Quebec) from what had previously been a single unitary Province of Canada, and also the addition of two previously separate colonies (New Brunswick and Nova Scotia) as provinces of the new federation. The independent federation of India established by the 1950 constitution constituted both a devolution to states that had previously been provinces and an incorporation of previously separate princely states. The Malaysian Federation of 1963 combined the previous Federation of Malaya established in 1957 (a centralized federation that had grown by devolution in 1948 from the unitary Malayan Union of 1945), with three additional states, Sabah and Sarawak in Borneo and Singapore (the last of which was expelled from the federation two years later). As a result of this process, some special concessions for the two Borneo states were incorporated in the asymmetrical Malaysian federation.

4 Size of federation and number and relative size of constituent units There is an enormous variation in the size of federations and in the number and relative size of their constituent units. Those variations have affected the functions assigned to the different orders of government and the internal political dynamics of their operation. Some federations are massive in terms of continental territorial size or population, such as Australia, Brazil, Canada, India, Russia and the United States. In these, size itself has been one of the inducements to federalization. At the other extreme are four micro-federations. Largely the product of decolonization, these ‘Lilliput federations’ as Dag Anckar (2003) has referred to them, are federations of small islands. Comoros (1978), Micronesia (1979), Belau (1981), and St Kitts and Nevis (1983), with total federal populations ranging from 17,000 in Belau (with 16 states), to 630,000 in Comoros (with three constituent units) represent a very different scale of operations that affects the capacities of their governments and the dynamics of inter-island relations. The number of constituent units also plays an important role in shaping the dynamics of political relationships within federations. In nine federations, there are 20 or more basic constituent units, the largest number being 83 (currently) in Russia and 50 states in the United States. The others are Argentina with 23, Brazil with 26, India with 28, Mexico with 31, Nigeria with 36, Switzerland with 26 and Venezuela with 23. Where there are such a large number of constituent units, usually none has been in a position to dominate politics within the federation or to counterbalance the federal government. At the other extreme have been federations with only two to four constituent 27

Ronald L. Watts

units. Examples include three of the micro-federations (Comoros with three, Micronesia with four, and St Kitts and Nevis with two), Pakistan since 1973 with four, Bosnia and Herzegovina with two entities, and Nigeria in its early years as a federation until 1963 with three regions. These examples taken together with the experience of Pakistan 1956–71, Czechoslovakia prior to separation in 1992, and Serbia and Montenegro 1992–2006 provide strong evidence that where the number of constituent units is so few, often with one dominant region, it is possible for individual units to challenge or dominate the federal government, typically producing markedly unstable political relationships. The remainder of the federations fall between these two extremes with 6–17 constituent units. In these cases, the individual constituent units have been able to exert more political influence than in federations with a larger number of constituent units, but have not experienced the degree of instability displayed by federations with only two to four constituent units. The relative size and wealth of constituent units in relation to each other within a federation has also been significant. The ratio of the population in the largest constituent unit to that in the smallest ranges from the extreme of 307.2 in India to 1.2 in Belau (Watts, 2008: 73). The variations in population and wealth among constituent units within a federation may cause wide discrepancies in their ability to perform their functions. Furthermore, where a single unit or pair of units contains a majority of the federal population, their predominance in federal politics has tended to be a source of resentment in the smaller constituent units. Examples are the Flemish Region in Belgium, Punjab in Pakistan, Ontario and Quebec in Canada, New South Wales and Victoria in Australia, Oromia and Anhara in Ethiopia, and Abu Dhabi and Dubai in the UAE. Interestingly, despite the large absolute size of Uttar Pradesh in India and California in the United States, they represent only 16% and 12% of their respective total federal populations, thus moderating their influence in federal politics.

5 Symmetrical and asymmetrical federations In most federations, the formal constitutional distribution of legislative and executive jurisdiction and of own-source financial resources applies symmetrically to all the full-fledged member constituent units. This constitutional symmetry exists in Argentina, Australia, Austria, Brazil, Ethiopia, Germany, Mexico, Micronesia, Nigeria, Pakistan, South Africa, Switzerland, the UAE, the United States and Venezuela. However, there is a group of federations where the constitution explicitly provides for asymmetry in the jurisdiction constitutionally assigned to full-fledged member states. Where this has occurred the reason has been to recognize significant variations among the full-fledged constituent units relating to geographic size and population, their particular social and cultural composition and distinctiveness, or their particular economic situation. While the forms and degree of asymmetry vary, federations with asymmetric, full-fledged constituent units include Belgium, Bosnia and Herzegovina, Canada, Comoros, India, Malaysia, Russia, St Kitts and Nevis, and Spain (Watts, 2008: 128). Internal asymmetry among the regional units within a federation clearly adds to complexity. Moreover, the tension between pressures for distinctiveness of particular constituent units and for equality of states, as exhibited in Canada and Spain, can be a source of continuing dissension. Nevertheless, some federations have found that the only way to accommodate strongly varied pressures for regional autonomy has been to incorporate some degree of asymmetry in the constitutional distribution of powers to constituent units.

6 Variations in the form of the distribution of legislative and executive authority While a basic common feature of all federations (as noted above) is a constitutional distribution of legislative and executive authority among the orders of government, the form taken by that 28

Typologies of federalism

distribution has varied (Majeed et al., 2006). In many federations, particularly in the United States and former British colonies, each order of government has been generally assigned executive responsibilities in the same field for which it has legislative powers.1 Arguments favouring such an arrangement are that it reinforces the autonomy of each legislative body, assures each government of the effective administration of its own legislation. Furthermore, in cases where there is a parliamentary executive (as in Canada, Australia and most of the former British colonies), the legislature can exercise control over the body executing its laws only if legislative and executive jurisdiction coincides. In the European federations with a civil-law tradition, such as Switzerland, Austria and Germany, however, jurisdiction of legislation and of administration has often been assigned to different orders of government. This has enabled federal legislatures to lay down considerable uniform legislation while leaving the application of that legislation to constituent unit governments taking account of varying regional circumstances. Such federations tend to be more centralized in legislative terms and more decentralized in administrative terms. Such an arrangement does, in practice, require more extensive collaboration and coordination between orders of government, however. In its extreme form, exhibited by Germany and Austria, it has created a virtually interlocking relationship of governments at different levels. The contrast between the two approaches is somewhat moderated by the fact that, in the former group of federations, federal governments have, in practice, delegated considerable responsibilities for federal programmes to constituent governments, often by providing financial assistance through grant-in-aid programmes, but the contrast in the linkage between legislative and administrative jurisdiction remains significant.

7 Major sources of regional revenues and reliance on transfers The allocation of finances to each order of government in a federation is important for two main reasons: first, these resources enable or constrain governments in the exercise of their legislative and executive responsibilities; second, taxing powers and expenditure are themselves important instruments for affecting and regulating the economy. Balancing considerations of administrative efficiency, avoiding tax competition, achieving equity, enabling accountability, and supporting regional autonomy has led in most federations to arrangements whereby most of the major tax sources have been assigned to the federal government with substantial transfers to the constituent unit governments being provided to match their expenditure needs (Shah, 2007). To mitigate the financial dependency of constituent unit governments, these transfers often take the form of constitutionally specified, unconditional grants or shares of specific federal tax revenues and, in a number of cases, with allocations assigned by an independent finance commission. Among federations, however, there are wide variations in the actual ‘own source’ tax revenues of the constituent units, the proportion of constituent unit revenues consisting of federal transfers, the conditionality of these transfers, and the proportion and nature of equalization transfers to mitigate disparities of wealth among constituent units. Generally speaking, the predominance of federal taxing powers (federal government revenues as a percentage of total federal-state-local revenues) has been higher in the emergent federations (70%–98%) than in the mature federations (40%–65%), with Australia being an exception (Watts, 2008: 102). Generally, as a consequence, intergovernmental transfers as a percentage of provincial and state revenues have been significantly higher in the emergent federations, indeed 68%–96% in Belgium, Spain, Mexico, Nigeria and South Africa compared to the mature federations, where, for instance, the comparable figures in Canada, Switzerland and the United States range between 13% and 26% (Watts, 2008: 105). It is these contrasts that have led many emergent federations to follow the Australian example of 29

Ronald L. Watts

constitutionally requiring that a high proportion of federal transfers be unconditional and dispensed by finance commissions (Watts, 2008: 106–16).

8 Federal institutions based on the separation of powers or parliamentary principles Generally, the legislative and executive institutions within federations have fallen into two basic categories: those involving the separation of legislative and executive powers and those involving a fusion of legislative and executive powers in a parliamentary executive responsible to the popularly elected house of the legislature (LeRoy and Saunders, 2006). The former is based on extending the federal principle of dispersion of power between orders of government to dispersion of power within each order of government. One example is the presidential-congressional institutions of the United States, a model copied by all four Latin American federations and by Nigeria in its 1999 Constitution. Another is the collegial, fixed-term executive in Switzerland. By contrast, parliamentary institutions are based instead on the principle that while authority in the federation is divided between orders of government, within each government executive and legislative power is concentrated so that the executive can be democratically controlled by the legislature. There are two sub-types of parliamentary executives: those modelled closely on the majoritarian British institutions at Westminster, as often found in federations that were previously British colonies (e.g. Canada, Australia, India and Malaysia), and those following European traditions of responsible cabinet governments, usually with coalitions, as found in Austria, Germany, Belgium and Spain. A third category might be called the hybrid presidential-parliamentary form of government, with Russia and Pakistan providing different variants of this form. These forms of executive and legislative institutions have affected the operation of federations in several fundamental ways (Watts, 2008: 137–44). First, they have had an impact on the representativeness and effectiveness of their federal governments. For instance, parliamentary executives provide opportunities for coalition arrangements, but this is counter-balanced by the weakened veto powers of second chambers because they are not confidence houses. Second, the different forms of executive have also affected the capacity of the federal executive to generate federal consensus (Watts, 2008: 142). Third, the form of executive within both federal and regional orders of government has also had a significant impact on the character and processes of intergovernmental relations. Where there has been a separation of legislative and executive powers, intergovernmental relations have tended to involve multiple channels of federal-state relations involving executives, officials, legislators and agencies interacting not only with their opposite numbers but also in a web of criss-crossing relationships that one American scholar characterized as ‘marblecake federalism’ (Grodzins, 1967: 257). In parliamentary federations where cabinets supported by majorities in their legislatures have tended to dominate, a common prevailing characteristic has been what has come to be called ‘executive federalism’, the dominance of executives in intergovernmental relations. ‘Executive federalism’ has been most marked in Germany, Australia and Canada, but has also been a distinctive characteristic of intergovernmental relations in India, Malaysia, Austria, Belgium and Spain.

9 Common law and civil law federations The presence of a common law tradition (as in the United States, Australia, India, Malaysia and Nigeria), a civil law tradition (as in the European and Latin American federations such as Switzerland, Germany, Austria, Belgium, Spain, Brazil and Mexico), or a mixed common law and civil law legal system (as in Canada and South Africa) has had a strong bearing on how the 30

Typologies of federalism

constitutional law of a federation is applied and interpreted (Poirier and Saunders, 2013). In federations where the civil law tradition has prevailed, the result has usually been a much more explicit delineation of jurisdiction and a more limited scope for judicial review (Majeed et al., 2006: 325; LeRoy and Saunders, 2006: 348). Two types of courts for ultimate constitutional interpretation may be found among federations. Broadly speaking, those federations either in the common law tradition or those closely following the US model have established a supreme court as the final adjudicator in relation to all laws, including the constitution. Examples are the United States, Canada, Australia, India, the four Latin American federations, Malaysia, Nigeria, Pakistan, Comoros, Micronesia, Belau, and the High Court of St Kitts and Nevis. The alternative, a constitutional court specializing just in constitutional interpretation, has been the pattern followed in Germany, Austria, Russia, Bosnia and Herzegovina, the UAE, Belgium and Spain. A unique third approach is that found in Switzerland, where the Federal Tribunal may rule on the validity of cantonal laws but not of federal laws. The validity of the latter is determined instead by the federal electorate through legislative referendums.

10 Degrees of decentralization and non-centralization In assessing comparatively the degree of decentralization in different federations and unitary states, distinction of two aspects is crucial. One is scope of jurisdiction exercised by each order of government. The other is the degree of autonomy or freedom from control by the other orders of government with which a particular government performs the tasks assigned to it. ‘Non-centralization’ refers to the latter aspect in which a central government is not able to control or revoke the jurisdiction of the constituent units and is what distinguishes decentralization in federal systems from that in unitary systems. Thus, it can be said that some federations may allocate fewer responsibilities to their constituent states or provinces than some decentralized unitary systems, but leave those states or provinces with greater freedom or autonomy over the exercise of those responsibilities (Watts, 2005, 2008: 111–12). A major problem in any comparative assessment of the degree of autonomous decentralization in different federations is that no single quantifiable index can adequately measure the scope of effective jurisdictional decentralization and the degree of autonomy of decentralized decision making within a political system. Among the multiple indices, although not of equal weight, that must be considered in any such assessment are legislative jurisdiction, administrative jurisdiction, own-source revenues, expenditures, impacts of unfunded mandates, decentralization to non-governmental agencies and constitutional limitations applying to all governments. While taken alone, it is a crude measure; nevertheless, a comparison of federations in terms of the proportions of federal and constituent unit expenditures after intergovernmental transfers indicates an enormous range in the percentage of combined total expenditures carried out by federal governments and giving different federations a distinctively different character. Malaysia (84%), Brazil (60%), Nigeria (60%), Australia (60%), and Mexico (59%) are among the more centralized, while Switzerland (32%), Canada (37%), Germany (37%), and Belgium (38%) are the most decentralized in terms of scope of jurisdiction. Austria (55%), Spain (51%), South Africa (50%), Russia (46%), the United States (48%), and India (45%) represent a middle group (Watts, 2008: 103). However, this does not take into account the degree of autonomy with which this jurisdiction is exercised. Taking account of this qualitatively, suggests that among the most highly centralized federations and quasi-federations are Venezuela, Pakistan, Malaysia, Nigeria, Argentina, Mexico, Ethiopia, Russia and the four micro-federations. Distinctly contrasted in their measure of autonomous non-centralization are Switzerland, Canada and Belgium. Between these two 31

Ronald L. Watts

extremes would come the relatively centralized federations of Brazil, Austria and Australia, and the more moderately centralized federations of Germany, Spain, the United States and India.

Conclusion This review of the broad category of federal political systems and within that of the more specific category of federations makes it clear that there is a considerable variety in the patterns of social conditions accommodated and an enormous range in the institutional arrangements and political processes they have encompassed. All these systems have attempted, many with considerable success, to combine elements of autonomous self-rule for the constituent units in certain matters and an over-arching shared rule in other matters in order to reconcile the simultaneous desires for distinctive diversity and united action. The variations among them make it clear that there is no single, pure, ideal form of federal system applicable everywhere. Even within the single category of federations, these have varied greatly in their institutional design and in their operation to meet their particular conditions and context. In the effort to meet these particular circumstances, statesmen have often resorted to hybrids combining elements of federal and unitary, and federal and confederal combinations. There is, consequently, no single, simple typology, but rather a range of typologies required for understanding the differences within the general category of federal political systems and the more specific category of federations.

Note 1 The United States, Canada and Australia are classic examples, but note that in the Canadian constitution an exception is made in the field of criminal law, where legislative jurisdiction is assigned to the federal government (s.91(27)) but administration to the provinces (s.92(14)).

References Anckar, Dag (2003) ‘Lilliput Federalism: Profiles and Varieties’, Regional and Federal Studies Vol. 3, No. 2: 107–24. Burgess, Michael (ed.) (2007) Multinational Federations (London: Routledge). Canada (1867) Constitution Act, 1867 (originally titled British North America Act, 1867). Caravita, Beniamino (2002) ‘Italy: Toward a Federal State?’ Federations, Special Issue: Themes of the International Conference on Federalism 2002: 25–26. Elazar, Daniel J. (1987) Exploring Federalism (Tuscaloosa, AL: University of Alabama Press). Franck, Thomas (1966) Why Federations Fail: An Inquiry into the Requisites for a Successful Federation (New York: New York University Press). Grodzins, Morton (1967) ‘The Federal System’, in Aaron Wildavsky (ed.) American Federalism in Comparative Perspective (Boston: Little Brown). Hicks, Ursula K. (1978) Federalism: Failure and Success: A Comparative Study (London: Macmillan). Kincaid, John (1990) ‘From Cooperative to Coercive Federalism’, Annals of the American Academy of Political and Social Science Vol. 509: 139–52. Kincaid, John and G. Alan Tarr (eds) (2005) A Global Dialogue on Federalism, Vol.1: Constitutional Origins, Structure and Change in Federal Countries (Montreal and Kingston: McGill-Queen’s University Press). LeRoy, Katy and Cheryl Saunders (eds) (2006) A Global Dialogue on Federalism, vol. 3: Legislative, Executive and Judicial Governance in Federal Countries (Montreal and Kingston: McGill-Queen’s University Press). Loughlin, John (2009) ‘The “Hybrid” State: Reconfiguring Territorial Governance in Western Europe’, Perspectives on European Politics and Society Vol. 10, No. 1 (April): 49–66. Majeed, Akhtar, Ronald L. Watts and Douglas Brown (eds) (2006) A Global Dialogue on Federalism, vol. 2: Distribution of Powers and Responsibilities in Federal Countries (Montreal and Kingston: McGill-Queen’s University Press). Moreno, Luis and César Colino (eds) (2010) A Global Dialogue on Federalism, Vol. 7; Diversity and Unity in Federal Countries (Montreal and Kingston: McGill-Queen’s University Press).

32

Typologies of federalism

Poirier, Johanne and Cheryl Saunders (eds) (forthcoming 2013) A Global Dialogue on Federalism Vol. 8: Intergovernmental Relations in Federal Systems (Montreal and Kingston: McGill-Queen’s University Press). Shah, Anwar (ed.) (2007) A Global Dialogue on Federalism, Vol. 4: The Practice of Fiscal Federalism: Comparative Perspectives (Montreal and Kingston: McGill-Queen’s University Press). Watts, R.L. (2005) ‘Autonomy or Dependence: Intergovernmental Financial Relationship in Eleven Countries’, Working Paper of Institute of Intergovernmental Relations, No. 5, 1–103, www.iigr.ca/pdf/ publications/362_Autonomy_or_Dependenc_I.pdf. ——(2007) ‘The United Kingdom as a Federalized or Regional Union’, in A. Trench (ed.) Devolution and Power in the United Kingdom (Manchester: Manchester University Press): 239–68. ——(2008) Comparing Federal Systems, 3rd edn (Montreal and Kingston: McGill-Queen’s University Press). ——(2010) ‘España: Una federación multinaciónal encubierta?’ in José Tudela Aranda y Felix Kneupling (eds) España y modelos de federalismo (Madrid: Centro de Estudios Politicos y Constitucionales): 55–81. Wheare, K.C. (1963) Federal Government, 4th edn (London: Oxford University Press). Young, Robert A. (1995) The Secession of Quebec and the Future of Canada (Montreal and Kingston: McGill-Queen’s University Press).

33

3 Plurinational federalism and political theory Ferran Requejo

Plurinational democracies and federalism in the 21st century In the last two decades there have been a number of political, social and cultural changes that have clearly influenced the sphere of political legitimacy in liberal democracies and federalism. Among these changes are a rapid process of technological and economic globalization, and what we might call a cultural and national turn associated with the claims of different groups so that they can be constitutionally recognized and politically accommodated within the rules of current liberal democracies. The most significant empirical cases of this turn are those related to minority nations, national minorities, indigenous peoples and transnational migrations.1 In this chapter I will focus on the case of minority nations in plurinational polities, such as Quebec in Canada, Scotland in the United Kingdom, and Catalonia and the Basque Country in Spain, dealing with the possibility that these collectives may achieve a stable recognition and accommodation through federal rules. The question of national pluralism has usually been absent in the theoretical and practical approaches of classical federalism. Minority nations are territorially concentrated collectives with a basic national identity that does not coincide, at least for a significant number of their members, with the national identity of the majority group of the polity. These collectives display distinguishing features, such as a different history from the rest of the state, a specific language, a different religious culture, etc. Some of them may even have been independent powers sometime in the past. They also express a will to be recognized as a different collective and a clear desire for self-government. Polities that include one or more collectives of this kind are known as plurinational polities. In the field of political theory, the importance that cultural and national collectives have for the self-understanding and self-esteem of individuals has been highlighted, first by the debate between the liberal and communitarian perspectives in the 1980s, and subsequently, from the beginning of the 1990s onwards, between two general approaches within political liberalism (so-called Liberalism 1 and Liberalism 2). Broadly speaking, Liberalism 2 criticizes the notions of individualism, universalism, ‘stateism’ and nationalism associated with traditional political liberalism (Liberalism 1) for being normatively biased in favour of majority national groups. In many cases, national groups play an important moral role in the interpretation of the basic values of liberty, pluralism and political equality in plurinational contexts. In other words, the basic criticism is that 34

Plurinational federalism and political theory

traditional political liberalism displays a ‘stateist’ bias that results in unfair practical consequences for minorities. This debate between both approaches of liberalism has revealed the inability of mere individualistic liberal, democratic and social constitutional rights to guarantee an egalitarian and equitable treatment of individuals belonging to national minorities. Moreover, this debate has revealed a number of misleading features of institutional federalism, as well as the inability of democratic states to play a culturally and nationally neutral role similar to that which they can adopt – at least in principle – in relation to other phenomena, such as religion. This theoretical and empirical turn opens a new normative and institutional perspective in the revision of federalism in plurinational democracies (Gagnon and Tully, 2001; McGarry, 2002; Nagel, 2004; Máiz and Requejo, 2005; Burgess and Gagnon, 2010). In general terms, the normative and institutional challenges posed by some minority nations create the need to improve present-day federal democracies. The traditional conceptions of democracy, federalism and constitutionalism have traditionally been unable to implement this improvement, which is fundamental in order to progress towards polities of greater ‘ethical’ quality. The following is a list of analytical elements (factual, conceptual and normative) which have been highlighted in recent years in the fields of political theory, political science and constitutionalism that, I think, must be taken into account in the analyses of plurinational federal democracies.

Factual analytical elements  In practical terms, most human beings are culturally and nationally rooted. Individuals are born and socialized in specific national, cultural, historical and linguistic contexts.  Classical liberal and federal political theories as well as traditional constitutionalism were created in much more simple and homogeneous social contexts than present-day societies.  The abstract and ‘universalist’ language that underlies the liberal values of liberty, equality and pluralism has contrasted, in practical terms, with the exclusion of a number of voices with regard to the regulation of specific liberties, equalities and pluralisms of contemporary societies (those who did not own property, women, indigenous people, ethnic, linguistic and national minorities).  Each of these movements gives rise to specific questions regarding the recognition and political accommodation (group rights, self-government, defence of particular cultural values, presence in the international arena, etc.). Insufficiency of liberal, democratic and social rights for protecting and developing the cultural and national features of minorities.  The existence of nation-building processes in all states, including the liberal-democratic ones: all states act as nationalist agencies. At the same time, the existence of at least partially competitive values, interests and identities in plurinational democracies. Existence of different narration and reconstruction of history and collective memories.  States usually treat the internal national differences of liberal democracies as ‘particularist deviations’. A practical response has been to promote the cultural and national assimilation of minorities in order to achieve their ‘political integration’. The practical consequence in plurinational democracies is the marginalization of the internal national minorities in the name of notions of ‘citizenship’ and ‘popular sovereignty’ (of the state), which are usually culturally and nationally biased notions interpreted in favour of majorities.  Nation-building and state-building processes have conditioned the conception (theories) and evolution (practices) of democratic federalism. 35

Ferran Requejo

Conceptual/normative analytical elements  Cultural and national liberties are very important components for the quality of a democracy and for the self-understanding and self-esteem of individuals (as recognized by the United Nations in its Human Development Report 2004).  Historical events usually play an important normative role. The political contexts in which individuals are socialized are often the result of historical processes that usually include violent elements – wars of annexation, exterminations, mass deportations, etc. – which are sometimes at the root of modern-day struggles for the recognition and self-government of minority nations.  Existence of two general theoretical attitudes in political theory and in policy making when dealing with national pluralism: 1 pragmatic (to avoid conflicts in the least costly way possible); and 2 moral (to approach it as a question of ‘justice’ – fair relations between permanent majorities and minorities). While the conflict between national majorities and minorities is based on the latter attitude, potential solutions may be achieved through the former.  Within the moral theoretical attitude, we find two general paradigms in relation to questions of socio-economic or cultural/national justice in pluralist societies: a paradigm of equality (distribution) versus a paradigm of difference (recognition and political accommodation). Existence of flaws in traditional liberalism and constitutionalism based on their individualist, universalist and ‘statist’ elements when they applied to plurinational societies.  In a plurinational polity different national groups (demoi ) which believe they constitute legitimate self-ruling collectives always coexist. At the same time ‘the people’ of the state are usually constitutionally presented as the only self-ruling demos of the polity. However, there is a surprising absence of a theory of the demos or demoi in traditional theories of democracy (whether they be of a more liberal or a more republican nature). There is also a lack of a theory of legitimate political borders in liberal approaches.  Liberal-democratic states as ‘culturally neutral’ entities is a myth of traditional liberalism.  Individual and collective identities are not fixed, but we make choices from them. The belief that we are ‘autonomous individuals’ who choose our national, ethnic, linguistic, religious, etc., identities is, to a great extent, another myth of traditional liberalism.

‘Classical’ institutional responses The first condition for solving a problem is to try to define or describe it correctly, and defining and describing a problem correctly involves establishing at least three aspects. First is knowing how to identify what the basic issue is, identifying the decisive question that needs to be considered. Second, defining a problem also involves knowing how to describe it with the maximum precision possible. This implies both a careful conceptual treatment and the inclusion of elements of an historical nature and the most important empirical data related to the problem. Third, defining a problem is knowing where one has to look to find possible solutions, both in the sphere of political theory and in that of comparative politics. When we have a question and do not know where to go to find the answers, this normally means that from an epistemological perspective we are not on the right track. As has been mentioned already, one of the most important questions with regard to the case of plurinational democracies is the recognition and political accommodation of the national pluralism of these democracies. Obviously, in addition to this question there is probably a whole series of aspects that are interrelated with it: economic development; inequalities of income; the multicultural character of society; integration in supranational organizations such as the European 36

Plurinational federalism and political theory

Union. However, it is methodologically improper to mix all these elements from the outset. In this case, the key point is not to establish how the demos becomes kratos (self-ruling power) – this would be the traditional vision of democracy – but how the different national demoi that coexist within the same democracy are politically and constitutionally recognized and accommodated in terms of equality (between the national majorities and minorities) in the kratos of the polity. This involves dealing with and introducing aspects of both a ‘democratic’ nature – participation between majorities and minorities in the ‘shared government’ – and, above all, of a ‘liberal’ nature – the protection and development of minority nations confronting the ‘tyranny of the majority’, both in the internal sphere of this democracy and in the international sphere. It is, therefore, a matter of establishing the ‘checks and balances’ also in a collective dimension which have received little or no attention from traditional political conceptions, but which constitute specific dimensions of core questions of liberal political theory, such as the ‘negative liberties’ and the ‘tyranny of the majority’. Whatever the most suitable liberal-democratic solution or solutions may be will obviously depend, among other things, on the context of each specific case (its history, international situation, types of actors, political culture, etc.). However, it seems to be clear that in contexts of national pluralism it is necessary to establish a much more refined interpretation than that offered by the basic values of traditional liberal-democratic constitutionalism: liberty, equality, individual dignity and pluralism. This complexity demands theories that are more sensitive and modulated to the variations of empirical reality when one attempts clearly to identify its basic legitimizing values. Moreover, it demands, above all, practical, institutional and procedural solutions that are much more suitable for the type of pluralism that one wishes to accommodate. Both qualifications of the liberal-democratic agenda have yet to be satisfactorily addressed and resolved. The three ‘classic’ institutional responses for societies with a strong component of national diversity (Norman, 2006; Amoretti and Bermeo, 2004; Watts, 1999; Lijphart, 1999) have been:  Federalism: (in a wide sense, including all kind of federations, associated states, federacies, confederations and regional states).  Consociationalism: institutions and processes of a ‘consociational’ nature (based on consensus between the majorities and permanent national minorities). One can find examples of these institutions and processes in the democracies of Switzerland, Belgium or Northern Ireland, together with clear federal solutions in the first two cases.  Secession. Let us now look at some elements offered by political theory with regard to federalism. The generic question is whether federalism offers a suitable framework for establishing the recognition and accommodation of plurinational democracies. If this is so, what are the most suitable concepts and values on which to base this framework?

Federalism and political theory Theoretical approaches to federalism in plurinational contexts The fundamental challenge facing plurinational federations nowadays can be summarized as a liberal, democratic and national challenge of polity building. The main question, in essence, is whether it is possible to combine, in the same federation, the political perspective of the construction of a federal union that normally predominates among the majority national group of the state and the perspective of a confederal union that usually predominates in the minority national 37

Ferran Requejo

entities of the federation. Both kinds of realities are usually based on different legitimizing concepts and values, although the same terms are often used to refer to them. This difference is expressed in the diverse conceptions regarding which rights, duties, institutions and political collective decision-making processes are legitimate in liberal-democratic terms when they are applied to realities characterized by strong components of national pluralism. The question regarding the probability of combining these two perspectives or, in other words, the probability of establishing a political and constitutional accommodation of de facto national pluralism cannot be answered in abstract terms, but must involve the institutional practices of comparative politics and case analysis. Broadly speaking, federalism is a notion that has been neither historically nor normatively related to national pluralism until quite recently. In fact, it is evident that both classic institutional analyses and those of a normative nature regarding federalism have been heavily influenced by the historical example of the United States, the first contemporary federation (Requejo and Caminal, 2011, 2012; Requejo and Nagel, 2011; Karmis and Norman, 2005; Hueglin, 2003). This is an empirical case that is not related to national pluralism. If we remain within the realm of US federalism (J. Madison, Federalist Papers, 10, 51), the response to the question concerning the possibilities of the political accommodation of plurinational societies by means of federal formulas is basically negative. The fundamental reasons for this are both historical and organizational. This is essentially a uninational model that avoids, but implicitly responds to, a fundamental question that, paradoxically, democratic theories have failed to answer: what is ‘the people’, the demos, and who decides to what ‘the people’ refers? If we take empirical data into account, it would appear to be practically impossible to politically empower the different demoi of a plurinational society within the uninational rules of the game of the federal model of the United States. Similarly, the fact that the first modern federation was the influential case of the United States – which was built using strong uninational and symmetrical components, and a strong Supreme Court that acted as a polity maker during its practical development – has not been unrelated to the evolution of federations and federal thought that was dominant until recently. Here the centre of gravity is located in the governance of a modern nation-state and the subsequent supremacy of the central power over the federated powers. One of the explicit objectives is to avoid the instability that confederations have repeatedly displayed at an empirical level. In contrast with the school of thought represented by Althusius and Montesquieu, the establishment of the federation should not involve existing social and territorial divisions but should attempt to construct a new polity that subsumes the old divisions by establishing new processes of state-building and nation-building. Here the union is more important than the units. This is an evolution that is very different from the more ‘confederal’ logic that characterizes the political systems and the political thought of the classic form of federalism before the American federation (Althusius, Politica Methodice Digesta VIII) – a conception that survived into the modern era in Switzerland and, albeit not for long, in the Netherlands. Depending on the federal conception within which we locate ourselves, we will obtain different conclusions in all the areas of territorial accommodation. The interpretation of the values of liberty, equality and pluralism will be different depending on whether one is dealing with uninational or plurinational liberal democracies, above all with regard to collective or group liberties and rights, the subjects of equality or the type of pluralism which is to be protected or guaranteed. Let us look at the present, for example. The classical questions ‘equality, of what?’ or ‘who are the equals?’ will receive different ‘federal’ responses depending on where we situate ourselves in the Madisonian or Althusian tradition of federalism, and depending on where we situate ourselves in a conception that is linked to Liberalism 1 or Liberalism 2 of the analytical and 38

Plurinational federalism and political theory

normative debate on liberal democracies. Therefore, for example, if we approach the subject from a Madisonian and Liberalism 1 approach, we will be more prone to base the answer to these questions on the concept of ‘equality of citizenship and individual rights’ – regardless of the cultural or national characteristics of the empirical citizens. In contrast, if we take the Althusian and Liberalism 2 approach, the answers will be more likely to consider the ‘equality of the demoi’ and a combination of individual and collective rights in the constitutional rules of the polity. In contemporary federations and regionalized states, the tension between liberal, democratic, national and functional logics has been resolved in ‘national’ terms, usually in the federal governments’ favour. Moreover, the history of federalism or, to be precise, the history of federations, has mainly been characterized by the development of models that are basically symmetrical. Comparative politics shows in what way symmetrical models have not been a particularly propitious option when they are coexistent or juxtaposed with nation-building processes within the same political system. These models are encouraged under the perspective of the welfare state policies and equal social rights across the entire territory of the state. However, they also lead to uniformity in the ‘entrance requirements’ of the political system and that makes achieving real political accommodation difficult. In fact, the pluralism of national minorities constitutes a form of de facto asymmetry which requires that the recognition of plurinationality be established using the same ‘entrance requirements’ of the constitutional system. This normative and institutional tension seems unavoidable in present-day plurinational federal democracies. To sum up, the symbolic and institutional challenges, as well as those relating to the rules of decision making, which plurinational societies pose for federalism are usually more complex than those posed by uninational societies. It seems clear, too, that Madisonian and Liberalism 1 approaches display difficulties and flaws in plurinational democracies. Both fail to ensure the political recognition and accommodation of the internal national pluralism of these kind of polities. In addition to the search for ‘common ground’ in the federation, the issue which most concerns minority nations is the establishment of institutions and protection mechanisms of a ‘liberal’ nature in the constitutional sphere that protects them from the decisions taken by the majorities. This political accommodation involves the establishment of broad forms of self-government and participation in the shared government of the federation that is based on one’s own national characteristics. A final theoretical feature to be considered is the individualistic Kantian philosophical framework usually established by Madisonian and Liberalism 1 approaches. Let us turn to this more abstract aspect of political legitimacy in plurinational federal democracies.

Political recognition and moral collectivism: a Hegelian turn Somewhat paradoxically, it could be said that some central aspects of Kantian philosophy are ‘too straight’ for the ‘crooked timber’ which, to quote Kant himself, characterizes humanity.2 Kant maintains the perspective of moral individualism, which I summarize here by means of two assertions: 1 the autonomy of the self as a subject – conceived as ‘prior to its ends’ – is the liberal value par excellence; and 2 the individual is the last source of any legitimate moral claim. Despite the fertility of Kantian philosophy in the field of political legitimacy in traditional liberal theories, I believe that Kant fails in his attempt to link the notions of moral individualism, state nationalism and cosmopolitanism.3 This is a particularly important failure in the case of plurinational liberal democracies. Despite Isaiah Berlin’s double warning about Hegelian philosophy,4 let us see if some elements of Hegel’s critique of Kant are better able to frame pluralism in plurinational democracies at the beginning of the 21st century. In general terms, Hegel’s more social vision opens the door to two important elements for analysis: the politics of recognition and the role of moral collectivism. Both connect with the Hegelian 39

Ferran Requejo

concept of ethicity (in contrast to morality). This perspective makes it possible to approach individualism and universalism in a different way. Let us look at this more closely.

Ethicity and the politics of recognition As is well known, Hegel sets out a number of criticisms of Kantian philosophy, introducing a more social perspective (social interaction) into his philosophy. Hegel does not deny that the natural roots of conflict are to be found in the passions and desires of individuals, just as Kant establishes when, following on from Hobbes, he deals with the ‘unsocial’ component of human beings. Nevertheless, unlike Kant and the earlier liberals, at the collective level, Hegel dispenses with the issue of state formation (contract) in his analytical agenda. Through his philosophy, Hegel does not intend to say how things should be, but how they really are. He considers that the analysis should spring from the reality of the existence of states, as they are the political contexts in which individuals are politically born. Hegel identified antagonism towards civil society as a typically modern phenomenon, and it is in civil society where the particularities that create conflicts reside, as well as the most important socializing and paradoxically disintegrative features. Along this path we pass from the sphere of Kantian morality to the sphere of Hegelian ethicity. Moreover, Hegelian philosophy accentuates its well-known struggle against moralism, displaying a sceptical attitude towards the Kantian ideals of the ‘cosmopolitan society’ and ‘perpetual peace’. Both are no more than a moral sermon, as beautiful as they are ineffective. Deontology (morality based just on principles and rights) appears to be an incomplete approach to understand individual dignity. In fact, it is never possible to know a priori the consequences of actions based on the exercise of rights. It is also essential to add a consequentialist approach, as individual dignity is a notion that always refers to particular social contexts; that is, to societies with specific historical, linguistic, cultural and national characteristics. To abstract these characteristics from normative analyses by means of deontological concepts based on an abstract form of individualism impoverishes these analyses. Hegel stresses that the antagonism of civil society is the source of conflict, but also a factor in socialization (Philosophy of Right, sec. 142, 182). Moral imperatives do not have enough force to end conflicts. From this perspective, we can deduce that the main political task is to establish a set of political institutions that help to prevent and resolve conflicts (today: federalism? consociationalism? secession?). The real constitution of a state lies in the interactive framework of its institutions. In this way, Hegel introduces a new analytical approach to the study of modern political societies. The basic idea is that the underlying, strictly individual, perspective of classical liberalism leaves too many normatively relevant elements out of focus. In addition to the dignity and identity of the individual considered in isolation, it is important also to consider the relationships between individuals in order to understand their dignity and identity. Recognition is the aim of this interaction. Our identities are formed through our relationships; individual freedom is neither solipsistic nor fragmented, it is not an atomized notion comprehensible by self-introspection. The recognition of other individuals is part of the self. Identities are partly shaped by the social relations that make up our ethical outlook. In contrast with the principles of some versions of liberalism, the individual does not come before his/her aims. We seek a kind of recognition that satisfies the desire to be admitted in a specific way into the polity. Here there is a human need: that others recognize our status as independent entities with our own characteristics. This implies a social relationship that is not necessarily peaceful, but based on the confrontation between different ‘subjectivities’. In Hegel’s Phenomenology of the Mind, we are faced with a process that is a new stage in the progress of the consciousness of freedom, that is, of the development of the mind.5 Self-consciousness 40

Plurinational federalism and political theory

does not exist on its own, but is transformed through practical contrast with other self-consciousnesses. The mind is the collective subject, the knowledge that ‘gradually appears’ and its phenomena are intersubjectivity – an ‘I’ that is a ‘we’ – although consciousness does not realize this at times. Demand for recognition is thus always mutual and reciprocal. Therefore, the initial relationship of mutual recognition is conflictive. Violent human relations are not anecdotal; rather they characterize the immediacy of social relations through the demand for recognition. This is the Hobbesian element of Hegel’s conception. The first movement produces the master-slave relationship through action. An action that is based on the desire for recognition by others and which ends in non-egalitarian situations. Consciousnesses oppose each other in a fight to the death which ends when one of the adversaries prefers liberty to death and recognizes the other without being recognized by him. The slave will later free himself through work, which only he carries out in contact with nature, not the master. In fact, the action is unique, indivisible and belongs to the two self-consciousnesses, but for the process to be satisfactory for both parties, the action and the universal language that accompanies it must be also mutual. However, the process does not end here, in a state of negativity, split and war, but continues through struggle and must reach positivity, reconciliation and peace. Reconciliation represents a new unity of the self-consciousnesses. These are the three moments of Hegelian ‘dialectics’: simplicity or abstraction; split or negativity; and reconciliation or specific accommodation of differences. This is a fundamental element of the progress of consciousness and freedom in history. Thus, it is possible to understand some of the central concepts of liberal-democratic legitimacy and federalism in different ways, depending on whether we adopt a Kantian or a Hegelian perspective. For example, the former establishes the notion of ‘citizenship’ as an abstraction, faced with which subjects are subsequently divided by their ethnic, linguistic, or national differences. The Kantian perspective would tend to maintain this contrast, asserting the greater legitimacy of the notion of ‘citizenship’ over the private differences displayed by individual identities, because that notion preaches a notion of equality ‘above’ these differences, even in symmetrical federal models. In an extreme case, a Jacobin model of rule of law and democracy will be defended. From a Hegelian perspective, in contrast, what will be asserted when faced with this contrast is a third moment, a revised notion of citizenship that is able to accommodate those differences so that the practical freedom achieved situates the ‘we’ at a higher level of liberal-democratic legitimacy. Reciprocal understanding is guaranteed only by instituted recognition, for example, through plurinational asymmetrical models. Thus, recognition is one dimension of the value of political liberty. Individual autonomy outlines in part our subjectivity, but the struggle for recognition is what frames our political relations. Recognition is also an aspect of political equality, and this fact involves the cultural and national spheres of individuals. The search for recognition occurs both among individuals and between collectives, as individual autonomy only occurs within a specific community (characterized by its history, language, etc.). Individuals are simultaneously independent of, but also dependent on the collectives to which they belong, irrespective of their voluntary (profession) or involuntary (language, history) nature. Thus, recognition, which presides over the transition from morality to ethicity, makes it necessary to go beyond the Kantian morality and the individualistic perspective of classical liberalism and federalism. This approach requires recognition by collectives of each other.

From the politics of recognition to moral collectivism Nobody has established more clearly than Hegel the human need for recognition. Thus, from the politics of recognition inherent in Hegelian ethicity comes the need to introduce the perspective of moral collectivism besides that of moral individualism. 41

Ferran Requejo

From the perspective of moral collectivism, 1 national groups are seen as legitimate sources of rights and moral claims – that is, they become legitimate actors through the normative links of their members to certain values, institutions and collective projects; and 2 moral autonomy of individuals is not necessarily the supreme or only liberal value – other values can take its place in specific contexts, such as collective freedom and tolerance, along with individual autonomy. These would be two requirements to establish a successful constitutional and political accommodation of national pluralism in a liberal state (in addition to the individualistic perspective). The ‘liberal’ key of the recognition between majorities and minorities will be one that is reciprocal and established on an equal footing (Seymour, 2008; Taylor, 1998). This makes it possible, from a perspective situated beyond moral individualism, to tackle the relations between different national groups within the state, and this is possible despite the stateist emphasis inherent in Hegelian thought.6 This is, in a manner of speaking, the establishment of an a posteriori social contract, whose legitimacy is no longer purely and simply ‘moral’, but includes a modus vivendi-type component based on the mutual recognition of partially disjointed ethicities, but which the latter recognize as a normatively superior agreement to those mere political agreements of a moral nature. In a plurinational liberal democracy, the perspective of moral collectivism is pluralistic by definition. This is a point which takes us away from Hegel’s monist view of the state. Moral collectivism in plurinational polities refers to a set of values, interests and identities of an agonistic character (conflict understood as something unavoidable in political relations), which encourages agreements of a pragmatic nature (modus vivendi agreements). Berlin (value pluralism) and Taylor (political recognition) meet within a more diverse and complex ethics than that stipulated by Hegel (Berlin, 1998; Taylor, 1992), but both are needed, the more individual perspective of Berlin’s liberalism and the more collective perspective of Taylor’s recognition. To follow the path of the ‘atomized’ individualism and the monist moral perspective that accompanies traditional state liberalism means legitimizing de facto relations of domination that exist between national groups within plurinational democracies. In other words, to stay exclusively within the perspective of moral individualism implies to legitimize the status quo of factual relations of domination present in the institutions, rules and decision-making processes of traditional liberal democracies. Clearly, to highlight the ethical importance of national groups for individuals does not involve accepting that these groups are of a static, eternal, or non-plural character. As with almost everything that is human, they are internally dynamic, historical and pluralistic entities. Over time, they change their values, their priorities and their internal composition, but they will probably be replaced by other forms of collective ethicity that will also be a legitimate source of rights, moral claims, constitutional recognition and political accommodation. Hegel provides a theoretical perspective that, despite and beyond his stateism, is a shift towards a more interactive (dialectal) approach which is normatively and institutionally relevant for the relationship between majorities and minorities in nationally diverse democracies. It provides a normative and institutional democratic refinement required to break the monopoly of state nationalism and a notion of citizenship based purely on moral individualism which are still very present in most approaches of political liberalism, federalism and constitutionalism. In the language of the liberal tradition, this requires establishing collective rights for national minorities alongside individual rights, and alternative institutional models. Potential conflicts between individual rights and collective rights should be regulated in a similar way to conflicts between individual rights (courts, modus vivendi agreements, etc.), but to do so from the premises of pluralist and egalitarian recognition, the composition and procedures of the high courts (supreme courts or constitutional courts) and intergovernmental relations in plurinational polities should take 42

Plurinational federalism and political theory

into account national pluralism. However, the analysis of comparative federalism shows that the two general objectives of plurinational democracies – constitutional recognition and political accommodation of national pluralism – are often done in a very incomplete and biased way through uninational and symmetrical traditional federal formulae. Actually, all federal plurinational states show problems of internal legitimacy (Requejo, 2010, 2005; Tierney, 2004; Baldi, 2003). In this way, the Hegelian legacy of political recognition and moral collectivism, as an enlargement of the Kantian perspective, facilitates a better implementation of national pluralism through institutions and procedural rules based on plurinational federalism, partnership and consociational models. Both the ethical refinement of liberal-democratic theory in relation to the relationship between national majorities and minorities, and the institutional practice that permits a fair recognition and political accommodation of national pluralism, continue to be two challenges to democratic constitutionalism in the 21st century.

Notes 1 The notion of ‘minority nations’ is used here as the equivalent to that of ‘stateless nations’ commonly used in the analytical literature on nationalism. However, in this chapter I do not include the case of ‘national minorities’, which are collectives that live in a different state from that in which the majority of people of the same national group reside (e.g., the case of the Hungarian minority in Romania, the Russian minority in Lithuania, etc.). Minority nations and national minorities differ both from a descriptive analytical perspective and from a normative perspective. 2 ‘Alus so krummen Holze, als woraus des Mensch gemacht ist, kann nichts ganz Gerades gezimmert werden’ [Out of the crooked timber of humanity, no straight thing was ever made], Idea for a Universal History from a Cosmopolitan Point of View, 1784. Berlin cites this classic quotation of Kant’s in order to establish his critique of the Platonic and positivist background of Western thought and of the utopian positions sometimes associated with it (see Berlin 1998). 3 However, the same Kantian work offers elements with which to rethink the articulation of these concepts when we move away from the individualistic approach of human unsocial sociability. An analysis of this Kantian concept and its continuity in Hegel’s philosophy, in Requejo and Valls 2007. 4 ‘ … the Hegelian system had the greatest influence on contemporary thought. It is a vast mythology which, like many other mythologies, has great powers of obscuring whatever it touches. It has poured forth both light and darkness – more darkness perhaps than light, but about that there will be no agreement’, ‘In Hegel we do see history through the eyes of the victors, certainly not through the eyes of the victims’ (Berlin 2002: 74, 90). 5 I follow P. Singer’s suggestion to use the word ‘mind’ rather than ‘spirit’ as a translation of the German term ‘Geist’ (see Singer 2001). In fact, this is the concept that acts as the ‘principle’ of a kind of philosophy that would like to be ‘scientific’, in the sense of a rigorous and well-founded form of knowledge that is not mere ‘opinion’. However, it is a principle that one deduces from the most immediate consciousness or the sensitive consciousness (ch. 1 of Phenomenology). When the deductive process ends, the work reaches the ‘absolute knowledge’ or epistemological knowledge (ch. 8), which goes beyond subjective opinions and allows the consciousness to be, at last, fused once again within the logic of the whole epistemological process. That is, finally Heraclitus is inside Parmenides. The end of the process means that we understand all the logic that has been present since the beginning. The soloist (consciousness) joins the choral finale (mind). See the last paragraph of Hegel’s ‘Introduction’. 6 In his ‘technical’ language, Hegel defines the state as ‘the actual reality [Wirklichkeit] of the ethical idea’. See Philosophy of Right, section 257.

References Amoretti, U. and N. Bermeo (eds) (2004) Federalism and Territorial Cleavages (Baltimore: Johns Hopkins University Press). Baldi, B. (2003) Stato e territorio (Roma-Bari: Laterza). Berlin, I. (1998) ‘My Intellectual Path’, New York Review of Books (14 May). 43

Ferran Requejo

——(2002) Freedom and its Betrayal: Six Enemies of Human Liberty (Princeton and Oxford: Princeton University Press). Burgess, M. and A. Gagnon (eds) (2010) Federal Democracies (London: Routledge). Gagnon, A. and J. Tully (eds) (2001) Multinational Democracies (Cambridge: Cambridge University Press). Hueglin, T. (2003) ‘Federalism at the Crossroads: Old Meanings, New Significance’, Canadian Journal of Political Science 36, 2. Karmis, D. and W. Norman (eds) (2005) Theories of Federalism: A Reader (New York and Houndsmills: Palgrave Macmillan). Lijphart, A. (1999) Patterns of Democracy (New Haven: Yale University Press). Máiz, R. and F. Requejo (eds) (2005) Democracy, Nationalism and Multiculturalism (London: Routledge). McGarry, J. (2002) ‘Federal Political Systems and the Accommodation of National Minorities’, in K. Neremberg and A. Griffiths (eds) Handbook of Federal Countries (Montreal and Kingston: McGill-Queen’s University Press), 416–47. Nagel, K.J. (2004) ‘Transcending the National/Asserting the National: How Stateless Nations like Scotland, Wales and Catalonia React to European Integration’, Australian Journal of Politics and History 50 1: 58–75. Norman, W. (2006) Negotiating Nationalism (Oxford: Oxford University Press). Requejo, F. (2005) Multinational Federalism and Value Pluralism (London: Routledge). ——(2010) ‘Federalism and Democracy. The Case of Minority Nations: a Federalist Deficit’, in M. Burgess and A. Gagnon (eds) Federal Democracies (London: Routledge), 275–98. Requejo, F. and M. Caminal (eds) (2011) Political Liberalism and Plurinational Democracies (London: Routledge). ——(eds) (2012) Federalism, Plurinationality and Democratic Constitutionalism (London: Routledge). Requejo, F. and K.J. Nagel (eds) (2011) Federalism Beyond Federations (Farnham: Ashgate). Requejo, F. and R. Valls (2007) ‘Somos conflictivos, pero … Actualidad de la tesis de Kant sobre la insociable sociabilidad de los humanos y su prolongación por parte de Hegel’, Isegoría No. 37: 127–63. Seymour, M. (2008) De la Tolérance a la Reconnaissance (Montréal: Les Éditions du Boréal). Singer, P. (2001) Hegel (New York: Oxford University Press). Taylor, Ch. (1992) ‘The Politics of Recognition’, Multiculturalism and the ‘Politics of Recognition’ (Princeton, NJ: Princeton University Press). ——(1998) Hegel et la societé moderne (Paris: Cerf). Tierney, S. (2004) Constitutional Law and National Pluralism (Oxford: Oxford University Press). Watts, R. (1999) Comparing Federal Systems (Montreal and Kingston: McGill-Queen’s University Press).

44

4 The penumbra of federalism A conceptual reappraisal of federalism, federation, confederation and federal political system Michael Burgess

Introduction: old concepts and new developments During the last 30 years there has been a relative stability in the intellectual process of concept formation and reformation in federal studies. Since the publication of Preston King’s Federalism and Federation in 1982 most established scholars of federalism have adopted this conceptual distinction or are at least fully aware of it (King, 1982). Subsequently some useful small refinements and clarifications have appeared in respect of terminology and definitions but in general the broad contours of the subject would seem to be quite settled. In this chapter I want to look again at the conventional concepts of federalism, federation, confederation and federal political system with a view to their re-appraisal. The principal reason for this revisionist objective springs from the realities of contemporary international change. The collapse of the Berlin Wall in November 1989 signalled the end of the Cold War in Europe and sent shockwaves around the world. The implications of such a critical juncture in world politics have yet to be fully understood. Over 20 years later, the direct and indirect consequences remain largely unknown; the whole story has yet to be told but we are still too close to these tumultuous events to acquire the clear historical perspective that comes only with distance from the object. However, one consequence of this historical watershed in the post-Cold War era has been the sudden appearance in the 1990s of a stream of new federal models. The cluster of federal models has not been restricted only to formal federations but has also included formally non-federal models that are either federal in practice or have evolved in that direction by incorporating conspicuous federal elements in their constitutions or political systems. The scope of the list is worldwide and includes the following countries: Belgium (1993); the Russian Federation (1993); the European Union (EU) following the ratification of the Treaty on European Union, known as the Maastricht Treaty (1993); Argentina (1994); Ethiopia (1995); Bosnia-Herzegovina (1995); South Africa (1996); Nigeria (1999); and Venezuela (1999). Moreover, the trend toward territorial decentralization in the United Kingdom (UK), Spain and Italy by the turn of the century had gathered so much pace that it seemed as if the federal idea had acquired the status of a zeitgeist; it was the spirit of the age. Since then, the early years 45

Michael Burgess

of the 21st century have witnessed Iraq (2005) as another new federal model and Nepal (2007) as moving closer toward a federal republic. Scholars of comparative federalism could therefore be forgiven for seeing in these contemporary developments a process of federalization. There appeared to be good reasons to believe that at last the 20th century had delivered an ‘Age of Federalism’, as predicted in the 1960s by William Riker in his now classic Federalism: Origin, Operation, Significance (Riker, 1964). We can now appreciate why the post-Cold War era has witnessed what some scholars have claimed was ‘the revival of federalism’ (Karmis and Norman, 2005: 1–30). The appearance of these new federal models in the 1990s and early 2000s confirmed the ‘new awakening’ of identity politics and ushered in a new age of post-conflict management and resolution in comparative federalism. Intellectually it exposed and emphasized the different approaches to the study of federalism between comparative politics and conflict studies. The former approach saw in the federal idea a structural response and reaction to the dawn of a new era in world politics while the latter school of thought regarded it largely as a useful instrument of power sharing for post-conflict management in deeply divided societies. However, both approaches accepted that the federal idea was worth adopting and adapting to the new conditions of state building and (multi)national integration presented by these fresh challenges. Both approaches also recognized the formidable complexity of these challenges, the common characteristic of which lay in the hugely heterogeneous nature of their societies. It is important to underline this point: the federal idea has been utilized in the post-Cold War era in countries that are recognized as multinational, multiethnic, multilingual and increasingly multi-religious. This is what we mean when we refer to such countries as having prominent cultural-ideological social cleavages with political salience. The chapter will begin by taking stock of the conventional definitions and terminology used to understand and explain federalism. This will also involve a modest revisionist critique of the conceptual landscape in the light of contemporary international change, leading to a new classification of federal models and some concluding remarks about its implications for federal theory. The concluding remarks will briefly discuss the meaning implicit in the title ‘the penumbra of federalism’. We will begin with our conceptual stock-taking exercise and highlight the important distinctions between the different terms.

Federalism, federation, confederation and federal political system Federalism and federation Since the early 1980s more and more studies of federalism have been based upon the conceptual distinction between ‘federalism’ and ‘federation’. The distinction was first made by King in 1973 but it was only formally adopted in 1982 and most scholars of federalism since then have recognized that these two terms are no longer synonymous (King, 1973). In political science analysis, they are discrete concepts that mean different things. Broadly speaking, federalism is widely acknowledged to be a normative principle that recommends a particular set of values and principles which, when translated in practice, create a level of institutionalization that produces federation as one among other federal forms. A federation is a state but it is a particular kind of state. King’s original definition of federation was put this way: An institutional arrangement, taking the form of a sovereign state, and distinguished from other such states solely by the fact that its central government incorporates regional units in its decision procedure on some constitutionally entrenched basis. (King, 1982: 77) 46

The penumbra of federalism

The impact of this novel distinction, while not accepted by every scholar of the subject, had an immediate practical consequence. It was formally acknowledged by the International Political Science Association (IPSA) in 1985 when it confirmed the creation of the Comparative Federalism and Federation Research Committee (RC28) and my own edited book Federalism and Federation in Western Europe appeared in 1986 (Burgess, 1986). Since then the conceptual distinction has been modestly revised so that a federal state can be defined in the following way: A distinctive organizational form or institutional fact the main purpose of which is to accommodate the constituent units of a state or a union of states in the decision making procedure of the central government by means of constitutional entrenchment. (Burgess, 2006: 2) This current modification does not at first glance seem to constitute much of a revision but as we will observe later in the chapter it has an important contemporary significance that reflects the sort of empirical realities identified above. At this juncture all we need to note is that the term ‘federal’ qualifies the word ‘state’ so that it incorporates certain values, beliefs and interests which give it distinctive institutional features and an innate moral character directly related to political community building. After all, as King observed, ‘for all of its institutional character, a federation is still governed by purpose, and thus reflects values and commitments’ (King 1982: 146). Its character is also ‘moral’ in the sense that it is specifically designed for the organization of human relations in ways that are designed to preserve, protect and promote difference and diversity having political salience. Since there are different differences and diverse diversities, there is a basic presumption of the value and worth inherent in the variety of human characteristics and identities that symbolize human dignity. However, it was in the conceptual relationship between federalism and federation that new insights, understandings and interpretations of federalism first emerged. The approach had important comparative empirical and theoretical implications for the subject because it opened up new possibilities for engaging contemporary change. This can be best illustrated by King’s observation that ‘although there may be federalism without federation, there can be no federation without some matching variety of federalism’ (King 1982: 76). His claim proved to be empirically valid but it was also interesting, innovative and useful in a heuristic sense for the political scientist who could now contemplate comparing federalisms as well as federations. The latter subject had long been established and was commonplace in political science but the former comparative analysis of federalisms had been relatively neglected. Here there was a fascinating opportunity to analyze precisely what different kinds of federalisms animated different kinds of federations. In order to do this, however, federalism had to be placed firmly in its context because meaning derives from context. Consequently it logically followed that the concept had to be located in its own distinct setting: historical, cultural, intellectual, philosophical, social, economic and ideological. This consideration in turn revealed its enormous multidimensional complexities. However, as we have already noted above, federalism and federation are both a response to and a formal recognition of the complexity of human relations. King’s conceptual distinction, then, made explicit what had actually been implicit in the mainstream literature on federalism for many years but it actively encouraged scholars to investigate and monitor the federalism in federation. He summarized the relationship succinctly and in so doing drew attention to the subtle and complex interactions between them when he observed: ‘Federation might best be understood in terms of the problems to which it has constituted a set of historically varying answers. If we understand the problems, the understanding of structure more clearly follows’ (King, 1973: 151–76). Clearly this statement prompted scholars and students 47

Michael Burgess

to direct their attention toward the nature of the problems to which federation had appeared as a solution. This itself was a revealing approach if we reconsider, compare and contrast the reasons for the creation of the federal model of the United States in the late eighteenth century – and widely referred to today as a territorial, mono-national federation – with those that created the new multinational federal models in the late twentieth and early twenty-first century. However, it is not enough just to contrast the different problems that confronted these federations over two centuries apart; it is also necessary to emphasize that they were responses to different kinds of diversities. In other words, the sorts of problems to which federation has come to be a structural response today have changed markedly since the American federal experience of the 1780s. We will therefore demonstrate in this chapter that both the nature of the problems and the kinds of federal responses have radically altered over time. Logically this also implies that the nature of the diversities and therefore the kinds of federations have changed in concomitant fashion. In such a way outlined above has the genius of federalism and federation been appropriately underlined. The inherently flexible nature of federation suggests that while it is certainly not a panacea for resolving every problem of difference and diversity in the polity, it can be adopted and adapted to suit many different circumstances. This is why as a concept it is simultaneously both simple and complex, and this is also why there can never be a basic template for federation or, as Ronald Watts has put it, ‘no single pure model of federation that is applicable everywhere’ (Watts, 1996: 1). Nonetheless, scholarly interest in the subject has been reignited precisely because of the appearance of the new federal models each one the response to a different set of circumstances but having in common the federal aspiration in at least one or more of the following three ways: a formal written federal constitution; the operation of a federal political system; and the values, beliefs or norms characteristic of the federal idea. This interest is also buttressed by the fact that these two concepts are firmly a part of political belief, practice and behaviour. This is patently obvious when we ask what the advocates of federation seek to achieve or what they understand federation to be. As King astutely observed in his classic text, ‘the observer’s criteria [cannot] be totally at odds with those professed by actors or proponents within federations since this would imply that federal institutions involve no element of conscious self-direction’ (King, 1982: 14). This brief outline of the conceptual relationship between federalism and federation suggests that the subject has in many ways become stronger as a distinct field of scholarly enquiry, much more clearly defined in its conceptual analysis and much more sophisticated in terms of its comparative empirical and theoretical implications than at any time hitherto. Bearing these observations in mind, let us turn to look at the third federal concept identified in this section of the chapter – namely, confederation.

Confederation Historically it has been common to juxtapose federation with confederation, often portrayed in the German language and legal thinking as Bundesstaat and Staatenbund. The former refers to a composite state of a single people or demos while the latter relates to a union of states rather than to a single state. By far the best detailed study of confederation today remains Murray Forsyth’s Unions of States published in 1981 (Forsyth, 1981). The enduring clarity of his definitions makes it appropriate for us to utilize them in this short chapter. The sort of world that Forsyth conveyed in his book was what he called ‘interstate’ and ‘intrastate’ relations, the former being relations between states and the latter being relations within the state. This was the first step toward his conception of confederation, which he saw as ‘occupying the intermediary ground between the 48

The penumbra of federalism

interstate and the state worlds, of going beyond the one but of not unequivocally reaching the other’. It was the conceptual area that he dubbed ‘the area of the half-way house’ (Forsyth, 1981: 6). He concluded that confederation was ‘not a state’ because it was ‘not a union of individuals in a body politic, but a union of states in a body politic’. It was based on a treaty between states rather like normal interstate relations but it had the distinction of going beyond this: as a confederation it manifested itself ‘as a constituted unity capable of making laws for its members’, but it was not ‘the constituted unity of one people or nation, but a unity constituted by states’ (Forsyth, 1981: 7, 15). Classic confederation, then, was a union of states bound together by a permanent treaty which while not creating a new state, nonetheless involved ‘a profound locking together of states themselves as regards the exercise of fundamental powers’ (Forsyth, 1981: 15). One consequence of this kind of union was that it established a familiar feature of confederation: that the powers and competences of the central authority thus created would impact directly upon the constituent states of the union rather than directly upon the peoples of the union. The states would therefore occupy the important intermediary position between the new central authority and the peoples who together constituted the confederation. In other words, it was not empowered to act directly upon each individual person but only upon the constituent states themselves. It was the states that acted upon their own peoples. There are two important points to make about this conventional wisdom. First, the assumption that confederation can act only upon the constituent states of these unions and not the individual person remains unclear and is possibly misleading. We only have to turn to look at the Federalist Papers in which in 1788 James Madison underlined the real capacity of the then confederation (called the ‘confederacy’) to act upon individual persons in the areas of piracy and matters relating to the postal services: In cases of capture; of piracy; of the post office; of coins, weights and measures; of trade with the Indians; of claims under grants of lands by different states; and, above all, in the case of trials by courts-martial in the army and navy, by which death may be inflicted without the intervention of a jury, or even of a civil magistrate – in all these cases the powers of the Confederation operate immediately on the persons and interests of individual citizens. (Rossiter, 1961: 250) This important statement confirmed that the conceptual distinction between federation and confederation in this particular respect was actually more a question of degree than principle. Indeed, in his quest to legitimize the proposed new ‘national’ government in the United States, Madison confessed that its ‘great principles’ were not novel at all. They should be considered ‘less as absolutely new than as the expansion of principles which are found in the Articles of Confederation’ (Rossiter, 1961: 251). On this reckoning, then, confederations did pass a limited range of laws that were binding on the individual citizens of constituent states in those spheres where they already had the agreed competence to do so. The second point derives from a series of lectures delivered by Henry Sidgwick at the University of Cambridge during the closing years of the nineteenth century and subsequently published as The Development of European Polity (Sidgwick, 1903). In these lectures Sidgwick noted that ‘much learning and subtlety’ had been applied to distinguishing the conception of a federal state (Bundesstaat) from that of a ‘confederation of states’ (Staatenbund). He also remarked that perhaps ‘undue importance’ had been attached to the ‘aim of getting a clear and sharp distinction’ between these two discrete concepts. ‘The two notions – confederation of states, federal state – represent’, he argued, ‘two stages in the development of federality’, which embraced a wide divergence of views about federal principles, akin almost to a spectrum of federality 49

Michael Burgess

(Sidgwick, 1903: 433). His phrase ‘federal polity’, then, can be conveniently reduced to two main types – namely, a federal state (or federation) and a confederation (or league of independent states), but as he observed, ‘in neither case is the distinction simple and sharp since the balanced combination of ‘unity of the whole aggregate’ with ‘separateness of parts’, which constitutes federality, may be realised in very various modes and degrees’ (Sidgwick, 1919: 532). The new federal models identified above confirm that both federal and confederal elements sit alongside each other to produce huge variations in their relationships in the federal polity. These two observations emphasize that our understanding of confederation must not be based on false premises. Indeed, it must be qualified in terms of both its direct impact upon the individual person and in its fundamental distinction from federation. Madison himself had acknowledged that no language was ‘so copious as to supply words and phrases for every complex idea, or so correct as not to include many unequivocally denoting different ideas’ (Rossiter, 1961: 229). Confederation, then, was an inherently ambiguous concept. Although there were certain fixed characteristics that constituted its hallmark, as Forsyth has demonstrated, its meaning in practice could vary according to its context. The position adopted by Sidgwick came very close to understanding confederation as merely part of a process – tantamount to a transitional phase – in the history of state and nation building. As he put it, a confederation, if it held together, had ‘a tendency to pass into a federal state’ (Sidgwick, 1919: 519). King discussed confederation in terms of sovereignty and the conventional unitary/confederal/federal typology but quickly came to an unsurprising conclusion: ‘the basic principle underscored in confederations is that the centre is not sovereign – it holds its authority at pleasure from the constituent (peripheral) units of the confederation.’ Moreover, the difference between federal and confederal authority was that in the former case the decision-making procedures were ultimately majoritarian while in the latter they operated on the unanimity principle. And in respect of the typology itself, he observed that it had ‘never told us a great deal’ (King, 1982: 133–34, 142). The most recent contributions to the mainstream literature on confederation in the 1990s have come from two authors – namely, Daniel Elazar and Fredrick Lister – but it is to the former that we must look for a novel construction of the concept (Lister, 1996; Elazar, 1998). Contemporary world developments introduced the new term ‘governance’ to sit alongside, rather than replace, the traditional term ‘government’ in recognition of a fundamental structural change in the nature of the state and its relationship to global financial capitalism. Political authority was gradually dispersed both within and without the state so that new forms and levels of decision making evolved, making the conventional nation-state just one unitary actor among a range of non-governmental organizations with the capacity both to compete and cooperate with it. Today, therefore, we speak of government and governance. State sovereignty, while not completely redundant in domestic affairs, has been outstripped in an age of growing interdependence in external relations. The conceptual and empirical implications of these momentous developments for federalism have been far-reaching. In particular they have led to what both Elazar and Lister have dubbed a ‘revival of confederation’ in the new era of globalization. Elazar saw in these contemporary global circumstances huge international change in the traditional state system but he construed it in a special way. He called it ‘the post-modern epoch’ of an emerging world order that had its roots in the period since the end of the Second World War. This involved what he called ‘a paradigm shift’ from statism to federalism which suggested to him a new world of ‘diminished state sovereignty and increased interstate linkages of a constitutionalized federal character’ (Elazar, 1995: 1–25; Elazar, 1998: 17). These global perspectives, in turn, led Elazar to construe the new state configurations as new ‘postmodern confederations’ which he distinguished from the earlier ‘premodern’, ‘medieval’ and ‘modern’ confederations. New post-war confederal experiments began to emerge predominantly, though not exclusively, 50

The penumbra of federalism

in the form of economic and trade relationships, enhanced security, environmental protection and more efficient service delivery, and it was early post-war Europe that donned the mantle of pacesetter. Apart from the early post-war organizations like the Council of Europe in the Western European Union (WEU), the North Atlantic Treaty Organization (NATO) and the Conference on Security and Co-operation in Europe (CSCE), Elazar witnessed confederal arrangements and ‘confederal-type’ organizations in the post-Maastricht EU, the post-Soviet Commonwealth of Independent States (CIS), and outside Europe in CARICOM (the Caribbean Area) and the Association of Southeast Asian Nations (ASEAN). He saw in these developments and levels of institutionalization ‘a new federal reality’, a form of union that had the enormous advantages of ‘partnership and sharing through constitutionalized multicentered forms of polity’ that brought ‘the advantages of federalism without the threats of tighter union’ (Elazar, 1998: 41). This new ‘federal reality’ rooted in essentially ‘functional’ activities, he believed, seemed to have resolved ‘the problems of constitutional design in confederations’, but it remained to be seen whether or not it would also ‘solve the problem of governance’ (Elazar, 1998: 137). In his thesis of the post-modern epoch of confederal governance, then, Elazar characteristically cast his empirical net so wide as to encompass the whole world and while not certain that these new confederations and confederal arrangements would necessarily succeed, he insisted that the ‘development of worldwide, multilateral, overlapping, linking institutions embracing formerly sovereign states’ would at the very least contain within them ‘federal limitations of one kind or another, somewhere between the constituent states of federations and the sovereign states defined and recognized by modern international law’ (Elazar, 1998: 147).

Federal political system Our fourth and final federal concept in this section of the chapter brings into focus the phrase ‘federal political system’ and in federal studies it is, as we will see, a deceptively complex term. It is a concept that has been and continues to be used by some scholars in a somewhat lazy descriptive fashion that lacks conceptual precision. Before we examine it in detail, however, it is first important to locate the concept primarily in its federal usage. Today it is Ronald Watts with whom we largely associate the notion of federal political system and it is to the evolution of his thinking that we must first turn our attention. The long evolution of its usage began with Watts’s classic New Federations: Experiments in the Commonwealth, which was published in 1966 and has been perpetuated in the most recent edition of his Comparing Federal Systems that appeared in 2008 (Watts, 1966, 2008). In his early work, Watts was mercifully immune to some of the more extreme objectives of the intellectual revolution in the social sciences – known as the ‘behavioural revolution’ – that occurred during the late 1950s and 1960s in the United States and did not allow it to intimidate the clarity of his thinking in the first detailed comparative study of six of the new federations in the British Commonwealth. While emerging unscathed from the more extreme rigours of a value-free science of politics, however, Watts nevertheless paid very little attention to the conceptual implications of the term federal political system. He appeared to use it as merely a descriptive label, slipping easily between the terms ‘federation’, ‘federal political system’ and ‘federal system’ without pausing to differentiate their meaning in conventional political science language (Watts, 1966: 7–9). In retrospect, it is probably correct to argue that this hardly seems to be of any great conceptual or theoretical significance for the comparative study of federalism, but it is in the perpetuation of the use of these terms that a problem in conceptual analysis has recently arisen. It is important to note that the conceptual problem to which we will shortly refer was originally detected in the late 1960s by Michael Stein. In a well-known detailed scholarly review 51

Michael Burgess

of Watts’s New Federations, Stein brought into sharp focus the conceptual implications of his use of the concept of federal political system (Stein, 1968: 721–47). Stein claimed that Watts was ‘the first student of federalism to use this concept in conjunction with “federal society”’, and that he had ‘borrowed the concept from the writings of David Easton and others’. Very quickly he noted in this survey that Watts did not really ‘apply the concept of political system in a systematic manner’. Instead he claimed that Watts focused ‘almost exclusively on the formal structures of government’ that expressed federal features. He did not, for example, include ‘those political or power relationships outside the formal structures of authority’ that ultimately produced ‘authoritative decisions for the members of a society’. In particular, Stein argued that Watts had not dealt ‘systematically with those aspects of political parties, interest groups, elite groups and political attitudes’ that were ‘shaped by, and in turn shape, the operations of federal political systems’ (Stein 1968: 729–31). In a nutshell, Stein criticized Watts’s use of federal political system for being undeveloped and understated. He summarized it in the following way: Watts also does not distinguish between the more inclusive aspects of a society’s political behaviour encompassed by the concept of political system and those aspects that are peculiarly influenced by and influence the federal structure. The concept of federal political system is narrower in meaning and scope than is the general concept of political system. (Stein, 1968: 731) Stein provided his own definition of federal political system which he felt was clearly and unequivocally expressed: A federal political system, then … is that form of political system (of a nation-state) in which the institutions, values, attitudes, and patterns of political action operate to give autonomous expression both to the national political system and political culture and to regional political subsystems and subcultures (defined primarily by ethnic-linguistic factors). The autonomy of each of these systems and subsystems is counterbalanced by a mutual interdependence. This balance maintains the overall union. (Stein, 1968: 731) Having nailed his own colours to the mast, Stein remarked that political scientists had come to ‘include the informal political structures in their analyses of political systems’ and that concepts had been developed in comparative politics which gave some precision to discussions of such systems and patterns of action as political parties, pressure groups, political movements, cliques and factions; competing elites; and political cultures and subcultures. It was the informal institutions, informal patterns of political action, and patterns of political attitudes that together comprised the infrastructure of any political system and was likely to show some striking similarities, particularly as they related to the federal political structure (Stein, 1968: 737). In summary, Stein underlined the poverty of Watts’s use of the concept federal political system. Stein had made an important conceptual distinction in two particular ways. The first was between Watts’s broad descriptive reference to the concept of political system as something that included states but was not restricted to them. This was conceptually imprecise and presumably never intended to be anything more substantive than a loose umbrella term under which he could gather together examples of the federal principle at work in the world. Second, he drew a sharp conceptual boundary, as we have seen, between the political system used in a general sense in political science terminology and the federal political system which had a more specific meaning. It would appear that this is where the debate about the concept of federal political 52

The penumbra of federalism

system faded away. It does not appear to have been the source of any further intellectual concern until much later. It is difficult to be certain about when the question was reignited, but it would appear to be correlated with the impact of King’s Federalism and Federation (1982), after which the debate about federal theory received a marked stimulus in the 1980s and 1990s. It also had something to do with the evolution of Watts’s own thinking about the subject in the light of Daniel Elazar’s conceptual framework of federalism, which had its origins in the 1970s but which he developed more clearly in these two subsequent decades (Elazar, 1979: 13–57). Watts seems to have engaged the debate and assisted further toward a conceptual groundclearing exercise in 1994 when his ‘Contemporary Views on Federalism’ sought to provide a timely update on how the concepts and terminology had evolved up to that point (Watts, 1994: 1–29). This is where he first suggested, ‘for the sake of clarity’, three terms that should be distinguished – namely, ‘federalism, federal political systems, and federations’. He emphasized that the concept federal political system was the descriptive term that referred to the genus of political organization itself marked by Elazar’s combination of shared rule and self-rule. However, he himself persisted with his broad descriptive reference to ‘federal systems’ used interchangeably with ‘federal political systems’ in the first edition of his Comparing Federal Systems in the 1990s published in 1996. A qualitative change toward a much more substantive appreciation of the concept of federal political system was evident in his important contribution to the conceptual debate in 1998, when his article ‘Federalism, Federal Political Systems, and Federations’ was published in The Annual Review of Political Science (Watts, 1998). In the article he returned to his 1994 formulation in seeking to differentiate these three federal concepts in order to draw a firm line under their discrete meanings and usages. However, he appeared to want to bell the cat. Here we can clearly identify the intellectual impact that Elazar’s conceptual approach had had on him. In reiterating his earlier formulation, however, he added a significant qualification that is worth a moment’s reflection: It should be emphasized that the empirical study of federal political systems is not simply a matter of categorizing various species and subspecies in terms of their institutional structures; it also involves an examination of the various relationships found within the species, including the processes and dynamics of their operation and the interrelation of their political structures with the social, cultural, and economic environment within which they operate. (Watts, 1998: 121) Stein’s 1968 chickens had finally come home to roost. Watts had at last added the important infrastructural elements to his systemic federal jigsaw and in consequence fleshed out the concept to provide a much more inclusive and sophisticated definition. Where, then, does our short survey of federal political system leave us in terms of conceptual analysis? What are we to make of Watts’s understanding of it? It is important to note that in stipulating federal political system as a genus itself rather than a species of another, larger genus, he had adopted a distinctive position. His reference to it as a genus of political organization related to the combination of shared rule and self-rule in a sense begged the question as to precisely what he understood ‘shared rule and self-rule’ to be. It is logical for us to assume that he construed it in the same way as Elazar and this meant that it was part and parcel of his extremely broad interpretation of ‘federalism’ itself. Both Elazar and Watts, then, subordinated federation (the federal state) to a mere species within the genus of federal political system, along with unions, confederations, federacies, associated states, condominiums, leagues, decentralized unions and joint functional bodies (Watts, 1998: 120–21). 53

Michael Burgess

There is no definitive way to resolve this conceptual question of federal political system as either genus or species. It would appear to rest solely upon the choice that each scholar wishes to make – the approach that he chooses to adopt – in conceptual analysis. We are unavoidably forced back to our basic definitions of the terminology used. It ultimately depends upon how useful our concepts are in explaining and understanding the world as it is. This means that the issue of genus or species must be rooted in empirical reality. In such a complex way are the conceptual and empirical worlds interrelated. There is much merit in the conceptual construction(s) of Watts and Elazar as they relate to federal political system, but there is as always a price to be paid for it. Their approach clearly transcends the state without being confined by it. The concept of federation therefore is relegated to the status of a mere subspecies of federal political system. However, while it certainly has the theoretical quality of flexibility and adaptability necessary to find a place for just about every form of federal or federal-type polity or arrangement in the world, it also has the drawback of casting the net so wide as to muddy the conceptual waters. We are reminded of King’s observation that if our categories are too broad we can always find what we want ‘if one looks hard enough’ (King, 1982: 35). As Murray Forsyth famously put it, ‘with sufficient effort’ federalism ‘can be detected almost everywhere’ (Forsyth, 1981: 7). There is, then, a trade off between concepts that are highly inclusive and embrace empirical reality with great gusto and those that are more exclusive and have quite narrow empirical implications. This is, of course, an extremely old conundrum in political science, and especially in comparative politics, and King’s conceptual distinction between federalism and federation is a good example of it. The problem with using the concept of federation is that it is rigid; it recognizes only those federal models that are formally federal states according to their written constitutions so that the scope for including new federal models that defy this standard categorization is quite limited. The conventional way of understanding the concept of federal political system in this context is, contrary to that of Watts and Elazar, to see it as conceptually subordinate to the federal state. Logically every state has a political system – and in Eastonian terms a social and an economic subsystem – so that in this scheme we would expect to find a federal political system in a federal state and not the other way around. Here the federal political system is construed in Stein’s sense as an outgrowth or consequence of federation which, if it works well, serves to buttress the state. This relationship in turn is closely related to the existence or emergence of a federal political culture. The gist of this analytical discussion leads us to reflect deeply upon these two very different approaches to understanding federalism. Generally North American scholars do not have as strong a conception of the state as do those steeped in the European school of thinking about federalism and federation. They often point to the danger of exaggerating the importance of the state in federal studies and prefer to reconfigure it as part of a much broader conceptual apparatus which, so to speak, brings the state down from its lofty conceptual position to a much more accommodative theoretical level. Watts and Elazar fit perfectly into this intellectual position as notable critics of statism, albeit for different reasons. There are, then, strong conceptual and empirical grounds for the intellectual predisposition to look beyond the federal state at non-state forms of federalism, especially in the world of international relations theory that includes interdependence, transnational and regime theories. There is another way of looking at this conceptual problem, though, which enables us to combine theoretical flexibility with empirical reality without marginalizing the state, and it encourages us to suggest a novel classification of federal states and federal political systems. We will follow this up in the next section of the chapter, but for the moment let us summarize where this conceptual reappraisal has taken us. 54

The penumbra of federalism

In this section of the chapter we have explored the concepts of federalism, federation, confederation and federal political system. We have shown that the intellectual debate about these concepts since the early 1980s has produced a series of contributions to federal theory which while unresolved has nevertheless increased our understanding of the chameleon-like qualities of the federal idea. Not only does the issue of different structural forms and expressions of the idea continue to be contested, but the conceptual relationships inherent in the idea are also revealed as both simple and complex. The conceptual reasoning and argumentation endures largely because the world keeps changing and in seeking to adapt and adjust to contemporary change, federalism reveals a conceptual versatility that is necessary, inherent and remarkable. This is precisely why the intellectual debate will always be an endless one.

New federal models It is time for us to consider the foregoing conceptual survey in terms of the two concepts of federation and federal political system. This is the product of contemporary change that periodically compels scholars to rethink, re-examine and ultimately reappraise their conventional conceptual categories in the light of new developments. This was always what drove Elazar in his admirable efforts constantly to engage change and to take from it novel conceptual adjustments and fresh insights into the study of federalism. In what follows I shall adopt the phrase federal democracies as my conceptual point of departure and argue that today we should be investigating every federal model from the standpoint of the relationship between federalism and democracy, a relationship that has recently been examined in detail elsewhere (Burgess and Gagnon, 2010). In presenting this novel classification it is important to note that the framework of analysis proposed derives directly from our perception of the world as it is and as it appears to be evolving. It is not intended to make any theoretical claims for this new classification which has been constructed in the attempt to engage contemporary events and developments. We are trying to capture the emerging complexity evident in the recent formation of federal states and federal political systems. Consequently, it is perfectly possible to have a formally federal state with a formally federal constitution, such as Ethiopia or Malaysia, but a political system that is not always federal in practice, while it is also possible to refer to a state that is formally non-federal with a written constitution that is also formally non-federal, such as Spain or Italy, but with a political system that is increasingly federal in practice. In short, I have presented the novel classification to reflect the new federal models that have appeared since the end of the Cold War (Burgess, 2012). What we are witnessing today, it would seem, is an increasing number of hybrid federations and federal political systems that do not always conform to our existing conceptual apparatus. They defy conventional definition and conceptual conformity, making them almost impossible to classify. We need, therefore, to acknowledge the hybrid nature of these new models, often with flawed, defective or incomplete democracies, and try to find some new way to classify them. Let us very briefly explore what the distinctions between the proposed models mean so that we can present a set of differentiated forms of federal democratic cases in the following way.

Model 1: mature federal democracies This category of case studies includes those that are widely considered to be a combination of both the classic federal models – the United States, Switzerland, Canada and Australia – together with those that, though not as old, nevertheless reflect strong liberal democratic constitutional cultures and can be construed as established federal models. Belgium is included here because 55

Michael Burgess

Table 4.1 Federal democracies Mature federal democracies

Flawed federal democracies

Incomplete federal democracies

Model 1 Formal federal constitution Federal state Federal political system in practice Federal government in practice

Model 2 Formal federal constitution Federal state Not a federal political system in practice Not a federal government in practice

Model 3 Formally non-federal constitution Not formally a federal state Federal political system in practice Not a federal government but behaves as one in practice

Case studies

Case studies

Case studies

United States Switzerland Germany Austria Canada Belgium

Russia Ethiopia Nigeria Malaysia Pakistan Mexico Venezuela

Spain Italy India South Africa EU

Emergent federal democracies Model 4

Transitional federal democracies Model 5

Aspiring federal democracies Model 6

Formal federal constitution Federal state Federal political system in practice but managed by the international community

Formally non-federal constitution Formally not a federal state Federal political system in practice but managed by the international community

Federal government in practice

Federal government in practice

Formally non-federal constitution Formally not a federal state Not a federal political system in practice but some federal elements managed by the international community Not yet a federal government in practice but an aspiration

Case studies

Case studies

Case studies

Iraq

Bosnia-Herzegovina Nepal Cyprus

Democratic Republic of the Congo Somalia Sudan Afghanistan

although it was not formally a federation until 1993, it has been evolving via a process of incremental federalism into a federal state at least since the late 1960s.

Model 2: flawed federal democracies The justification for this category of federal model is that it includes states which although formally federal with a written federal constitution, are highly defective and deficient in terms of federal political practices – the operation of the federal political system – and executive abuses of power in terms of the rule of law, human rights, corruption and the intimidation of legitimate political opposition. There is evidence of some efforts to improve different aspects of its liberal democratic credentials and behaviour but it is patchy and largely ineffective. 56

The penumbra of federalism

Model 3: incomplete federal democracies This category of case studies revolves around the idea of a pronounced gap between constitutional theory and practice. These states and unions of states (the EU) are not formally federal but in practice they operate in ways that are federal in all but name. The practices and processes of an evolving liberal democratic constitutionalism are palpably evident, although there may be some formal structural-institutional deficiencies involving inadequate forms of political participation and representation, and even some lingering historical legacies, as in India, of strong executive power inimical to the development of local autonomy. There is, then, evidence of an evolving federal political culture.

Model 4: emergent federal democracies This is a very recent category which together with Model 5 suggests that it might be an indication of the future trend of federal state formation in terms of post-conflict state and nation building where there is little or no liberal democratic constitutional culture or even a tradition of political party competition. The existence of a federal constitution and a ‘managed’ federal political system must be set against the absence of a federal political culture and suggests that this has to be created over the long term via the process of political socialization by the international community (IC) as an external coercive force. Nancy Bermeo labelled this ‘forced together federalism’, but it could equally be dubbed ‘forced together federalism’ as a new model (Bermeo, 2002: 110). The coercive basis of the federal model means that the motives for elite cooperation in the new federation are based upon a reluctant recognition that no alternative exists to rebuild the state and forge a new political community ultimately with shared values and common goals.

Model 5: transitional federal democracies This category of case studies is similar to Model 4 in the extent to which it relies upon the IC to sustain what is also an imposed federation. However, it is differentiated from Model 4 in that it does not have a formal written federal constitution and is not formerly a federal state although, like Model 4, it does operate a federal political system in practice which is overseen by the IC. Bosnia-Herzegovina is the main case study in this new federal model but it could also include both Nepal and Cyprus, the latter of which has a strong IC in the form of the United Nations (UN) and the EU as a combined external though not as coercive a force as exists in Bosnia-Herzegovina and Iraq. These case studies can be construed as transitional federal democracies expressing federalism as a ‘federalization process’, which portrays them as ‘federations in the making’.

Model 6: aspiring federal democracies The final model is of a residual character. It is proposed as a category where the constitutionallegal language and terminology produces a federal discourse, often of an extremely limited nature and scope, but one that aspires ultimately to a federal destination. These case studies, like the Sudan and the Democratic Republic of the Congo, are usually examples of post-conflict states from the developing world that adopt formal constitutional declarations of intent to trumpet a liberal democratic federal future, but other cases, such as Somalia and Afghanistan, remain mired in civil war or other kinds of violent conflict where once again the conditions for federal democracy are, to say the least, unpromising and clearly indicate the complete absence of a federal political culture. 57

Michael Burgess

The conceptual framework that structures these six federal models is only tentatively proposed to clarify and put into perspective the revival of federalism that we identified earlier in the chapter. Without wishing to speculate too much about the future trends of federalism, it is nonetheless important to comment upon the somewhat dramatic appearance of the new federal models and in some cases the reformation of old models. The decade of the 1990s was clearly unusual, and from a longer historical perspective today is increasingly looking unique, in the area of federal state formation and subsequent maintenance. Consequently the organization of contemporary federal democracies above is intended to be broadly indicative of current trends and developments during the last quarter of a century that we have dubbed the ‘post-Cold War era’. The scheme is a modest one and very few claims are made here for what is merely a descriptive but relatively novel way of organizing the contemporary world of federal states and federal political systems. Its principal purpose is heuristic: it is meant to be one small step toward explaining the formation and evolution of so many federations and federal political systems, especially during the decade of the 1990s. Political scientists have to try to explain how and why this happened when it did. In some respects it is still probably too early for us to be able to do this with any certainty but the content of this chapter, if it tells us anything about contemporary federal state building, points to the broad context of globalization, including in some cases European integration, and the implications of international change occasioned in particular by the end of the Cold War. Global institutional innovations require global explanations embedded in local contexts and circumstances. Against this broad background context we must situate the individual case studies identified above and try to integrate their unique contexts and circumstances into this larger picture. There is plenty of intellectual scope for comparative federalism, federation, confederation and federal political system in this scheme and the new classification proposed here is both descriptive and dynamic. It is descriptive in the extent to which it distinguishes between different types of federal formations and evolutions, and it is dynamic in so far as it accommodates change by facilitating movement from one model to another according to practical, concrete developments that are empirically verifiable. These might be tendencies toward more or less democratization or a discernible shift in the direction of more or less genuine autonomy for the constituent or sub-national units in federations and federal political systems. Currently what the framework suggests in terms of federal democracies is that in order to make sense of what is happening we must continue to concentrate our analytical attentions upon the complex relationship between federalism and democracy – or federalization as part of the larger process of democratization in some empirical contexts – which are so intimately intertwined.

Conclusion: the penumbra of federalism It is time to bring the chapter to an end with some concluding remarks about the relationship between political concepts and empirical verifiability. This short conceptual reappraisal serves as a timely reminder that it remains important constantly to check if the various technical terms used in the conceptual analysis of federalism, broadly construed, correspond with the reality of the world as it is, or, to put it another way, whether or not the concept adequately captures or explains the thing that is to be examined. The chapter confirms above all the complex and dynamic relationship between federalism, federation, confederation and federal political system as primary federal concepts. It demonstrates that the clear and precise distinctions between them as concepts surveyed above often become blurred and are only dimly perceived when the conceptual world engages with empirical reality. The interaction between the conceptual and empirical worlds is ultimately symbiotic: each is dependent on the other. Concepts, however basic, are useful only to the extent that they help us to explain and understand the world and this means that empirical reality can 58

The penumbra of federalism

always impose itself on conventional conceptual analysis, rendering it redundant and compelling us to adjust and adapt it in order to try to capture the complexity of what is happening. If we focus upon comparative federalism it is obvious that the new federal models have presented classic federal theory with a huge challenge in terms of their formation and maintenance. The existing mainstream theories and approaches to the study of federalism – dating from the second half of the 20th century – simply do not explain the sort of contemporary change identified in this chapter. It is time for a new theoretical response that can provide an appropriate conceptual framework to accommodate federal developments in the post-Cold War era. This is why the framework of federal democracies was proposed in the chapter. The reality of contemporary change has stretched and always will stretch existing conceptual analyses, definitions and frameworks to their limits. It is necessary that this should be so and it is in the nature of the subject that it will remain so. The federal concepts surveyed here have strong historical and philosophical dimensions so that it should come as no surprise to discover that there are no fixed, unalterable conceptual boundaries. If meaning derives from context and the context keeps changing we should not be surprised if terminological meanings are also subject to periodical change. The construction and reconstruction of federal concepts reflect this penumbra of meanings. They exist in the equivalent of a twilight zone between light and shade that constantly changes, making them difficult to discern and hard to distinguish one from another. A good example of this conundrum is the idea of an emergent confederation or confederal relations developing inside an existing federation. It is not as far-fetched as it initially might appear. We can already discern strong elements of just such a practical arrangement in Belgium with its parallel policy spheres for Flanders and Wallonia that indicate a quality of separation within union more akin to confederal than federal relations. There is also the example of BosniaHerzegovina, which is formally a diarchy that incorporated the Federation of Bosnia and Herzegovina with 10 cantons as one of two units when it was founded in 1995. These are two cases that confound the standard conceptual categories outlined in the chapter and point to the future flexible possibilities of the federal idea. Finally the EU has always defied classic theoretical constructions in combining federal and confederal elements in its institutional and decision-making procedures and practices, as well as accommodating three formal federations in its membership: Austria, Germany and Belgium. The penumbra of federalism, then, conveys the sense in which the frontiers and boundaries between these concepts are not always sharp and precise but often blurred and indistinct, so that one concept shades off into another. Indeed, as we have demonstrated, they overlap and intermingle, making it possible to anticipate more federal-confederal hybrids, as Watts has already noted in his work, and forcing us to tread very carefully when the conceptual world meets its empirical counterpart (Watts, 2008: 55–61).

References Bermeo, Nancy (2002) ‘The Importance of Institutions: A New Look at Federalism’, Journal of Democracy 13(2): 96–110. Burgess, Michael (ed.) (1986) Federalism and Federation in Western Europe (London: Croom Helm). ——(2006) Comparative Federalism: Theory and Practice (London: Routledge). ——(2012) In Search of the Federal Spirit: New Empirical and Theoretical Perspectives in Comparative Federalism (Oxford: Oxford University Press). Burgess, Michael and Alain-G. Gagnon (eds) (2010) Federal Democracies (London: Routledge). Elazar, Daniel J. (1979) ‘The Role of Federalism in Political Integration’, in Federalism and Political Integration (Ramat Gan, Israel: Turtledove Publishing, for the Jerusalem Institute of Federal Studies), 13–57. ——(1995) ‘From Statism to Federalism: A Paradigm Shift’, Publius: The Journal of Federalism 25(2): 1–25.

59

Michael Burgess

——(1998) Constitutionalizing Globalization: The Postmodern Revival of Confederal Arrangements (Lanham, MD: Rowman and Littlefield Inc. Publishers). Forsyth, Murray (1981) Unions of States: The Theory and Practice of Confederation (Leicester: Leicester University Press). Karmis, Dimitrios and Wayne Norman (eds) (2005) Theories of Federalism: A Reader (New York: Palgrave Macmillan). King, Preston (1973) ‘Against Federalism’, in R. Benewick, R.N. Berki and B. Parekh (eds) Knowledge and Belief in Politics (London: Allen and Unwin), 151–76. ——(1982) Federalism and Federation (London: Croom Helm). Lister, Frederick (1996) The European Union, the United Nations and the Revival of Confederal Governance (Westport, CT: Greenwood Press). Riker, William (1964) Federalism: Origin, Operation, Significance (Boston: Little, Brown & Company). Rossiter, Clinton (ed.) (1961) The Federalist Papers (New York: New York American Library). Sidgwick, Henry (1903) The Development of the European Polity (London: Macmillan & Co.). ——(1919) Elements of Politics, 4th edn (London: Macmillan & Co.). Stein, Michael (1968) ‘Federal Political Systems and Federal Societies’, World Politics 20: 721–47. Watts, Ronald L. (1966) New Federations: Experiments in the Commonwealth (Oxford: Clarendon, Oxford University Press). ——(1994) ‘Contemporary Views on Federalism’, in B. DeVilliers (ed.) Evaluating Federal Systems (Cape Town, South Africa: Jutta & Co., Dordrecht, Martinus Nijhoff), 1–29. ——(1996) Comparing Federal Systems in the 1990s, 1st edn (Montreal: McGill-Queen’s University Press). ——(1998) ‘Federalism, Federal Political Systems, and Federations’, The Annual Review of Political Science 1: 117–37. ——(2008) Comparing Federal Systems, 3rd edn (Montreal: McGill-Queen’s University Press).

60

5 Territorial strategies for managing plurinational states1 Wilfried Swenden

Introduction This chapter critically assesses territorial strategies for managing plurinational states. After defining the plurinational state, the chapter identifies a number of ‘varieties’ of territorial strategies. Subsequently the chapter reviews the debate among scholars of ‘governing divided societies’ with respect to the cost and benefits of territorial management. According to some, territorial solutions set plurinational states on a path of disintegration with the emergence of new – often unitary-sovereign – states as their most likely outcome. For others, territorial solutions provide the best chance for holding together plurinational states. In this chapter, I argue that territorial strategies can indeed be successful for managing plurinationalism, but only if they arise under favourable conditions (within a consolidated democratic context) and adopt a suitable territorial design. Furthermore, territorial strategies are often part of a wider toolkit of institutional devices for managing divided societies, and as such they are rarely sufficient in their own right for governing plurinational democracies.

Defining the plurinational state Plurinational or multinational states are states that are marked by the presence of at least two territorially distinct communities. Their territorial distinctiveness can be linked to the presence of a particular language, religion, tribe, a shared history, but above all a shared understanding of being part of a separate political community with a distinctive identity separate from or in addition to that of the state as a whole (Keating, 2001; Tully, 1995; Ghai, 2000; but Stepan et al., 2011). Plurinational states are a type of plural state, yet many plural states are not plurinational as the ethnic groups living within lack a clear territorial basis. For instance, Lebanon is plural, but not plurinational because its Sunni, Shia, Maronite Christian, Druze and Orthodox communities live territorially dispersed. The grievances of such non-territorially concentrated groups are often more appropriately accommodated through power sharing or consociational devices than through territorial strategies. Plurinational states are sometimes federal, but not all federal states are plurinational. Some federal states have relatively weak sub-state identities. This is the case, for instance, for Austria, 61

Wilfried Swenden

Australia, Germany (with the possible exception of Bavaria) and the United States. Citizens in these states see themselves first and foremost as Austrian, Australian, German or American. Here, territorial strategies may have been used to make the polity more market-efficient and democratic (by building in additional arenas for political competition) but not to manage territorially concentrated diversity (Kymlicka, 2001). In fact, in the United States the boundaries of the states were purposefully drawn in such a way as to prevent indigenous or non-WASP (white Anglo-Saxon Protestant) Americans from constituting a majority in any of them. Finally, plurinational states should also be distinguished from states in which the boundaries of the territorial units do not respect identifiable markers of ethnicity. For instance, Switzerland is plurilingual and multi-religious but the boundaries of its 26 cantons cross-cut this linguistic and religious diversity. In this sense, while Switzerland may be a plural society or federation, it is not a plurinational state or federation (Dardanelli, 2011). Earlier, we defined states as plurinational because they contain distinct political communities within the state. Yet the boundaries of these political communities can change over time, as is the nature or relevance of the markers on the basis of which political entrepreneurs build these communities. Furthermore, the nation or political community to which individuals belong is highly subjective. Let me briefly discuss these three points in turn. First, the (de facto) boundaries of a political community can change over time. For instance, as a result of immigration, suburbanization, mixed marriages, etc., there is now a considerable number of French-speakers who live in the municipalities surrounding the predominantly Francophone (but officially bilingual) Brussels Capital Region. Although located in the Flemish Region, a large share of these Francophone citizens identifies with the French Community and not the Flemish Region in which they reside, and hence they seek the inclusion of these municipalities into the territory of the Brussels Capital Region or the French Community. Second, nationalism and sub-state nationalism are constructed or imagined (Anderson, 1991) and therefore the toolkits of nation building may change. To stick with the Belgian example, in the 19th and early 20th centuries Flemish nationalism was very much linked with the historic struggle for linguistic emancipation of the Dutch-speaking north against the demographically smaller, but politically superior French-speaking south (Witte et al., 2000). Although contemporary Flemish nationalists keep that history alive, the demarcation of clear language boundaries coupled with the imposition of the principle of linguistic territoriality have largely settled the Flemish linguistic grievances (save the language status of the boundary areas near Brussels). Therefore, contemporary Flemish nationalism is linked as much with differences in the socio-economic and political profile of Flanders compared with that of the other Belgian regions. Flemish nationalists assert that a relatively poor Wallonia prevents the Flemish nation from developing its full economic potential. Third, feelings of national belonging or answers to the question of what constitutes the political community are highly subjective. For instance, some citizens who reside in Scotland feel exclusively Scottish, others feel exclusively British, whereas another group may hold multiple or dual identities (they may feel as Scottish as British, more British than Scottish or vice versa). It follows that the management of plurinational states is an ongoing task since political entrepreneurs identifying with the state or the sub-state nation or both are continuously engaged in building the nation that best reflects their political priorities and loyalties.

Varieties of territorial strategies for governing plurinational states What is the possible set of institutional repertoires that states can choose from when seeking to manage plurinationalism within their borders? We can distinguish between territorial strategies 62

Strategies for managing plurinational states

on the basis of: 1 the extent to which they constitutionally or legally entrench sub-state levels of autonomy (self-rule); 2 the extent to which such autonomy arrangements apply across the entire territory of the state; 3 the degree to which differences in autonomy apply among the substate entities of the state; 4 the extent to which autonomy arrangements involve international monitoring; and 5 the degree of sub-state participation in the governance of the centre (shared rule). Table 5.1 presents a set of ideal types of territorial arrangements based on the five criteria identified above. They are discussed (with examples) in the remainder of this section.2 Usually, when thinking of strategies for managing plurinationalism, varieties of territorial autonomy or self-rule come to mind. Yet, sub-state nationalities can be accommodated, to some degree, with limited territorial self-rule. If we leave aside the Home Rule provisions that applied to Northern Ireland between 1921 and 1972, the UK was governed as a union and not a unitary state for most of the 20th century. Within it, Scotland and Wales obtained recognition as ‘composite nations’ of the UK through their representation in the union cabinet (territorial secretaries) and their over-representation relative to their share of the population in the UK Parliament (shared rule). The UK secretaries of state presided over territorial field administrations which in Scotland, Wales and (since 1972) Northern Ireland implemented UK policy (in certain fields – most notably education and health) in a territorially distinctive way (Mitchell, 2008; Bulpitt, 1983). In this sense, plurinationalism was accommodated by projecting sub-state nationalist interests at the centre. However, a ‘union’ strategy is always contingent upon a centre that is willing to endorse territorially distinctive policies. Furthermore, the centre has to be perceived as sufficiently legitimate to speak on behalf of the sub-state nation(s). In the UK, the Conservative government under Prime Minister Margaret Thatcher (1979–90) was felt to have sidestepped Scottish, Welsh and Northern Irish preferences. When the party lost nearly all of its electoral representatives in these nations – in part because of its (perceived) territorially ignorant policies – it could no longer claim legitimately to represent them. Therefore, in time, the discredited union state (by necessity) gave way to an arrangement in which the distinctiveness of the sub-state nations was more politically secure. Unlike the ‘union state’, the other strategies for managing plurinationalism (federalism, devolution, federacies) incorporate an important element of self-rule. Federalism is the best known of these. In federal states (internal) sovereignty is shared or ‘divided’ between the centre and the Table 5.1 Varieties of territorial management in the context of a plurinational state Union state Constitutional Not applicable entrenchment of sub-state autonomy Self-rule affects Not applicable entire territory Variations in the Not applicable degree of self-rule? International monitoring? Shared rule

No Territorial ministries and ministers

Federalism

Devolution

Federacy

Yes

No

Yes

Yes

Possibly

No

Possibly (asymmetric federalism) No

Possibly (e.g. UK devolution) No (except Northern Ireland) Weak

Yes (only applicable to small share population) Possibly

Yes (second chamber, intergovernmental relations)

Possible representation in national legislature, but not in executive

63

Wilfried Swenden

states. The centre and the sub-state entities (cantons, Länder, states, provinces, autonomous communities … ) each have certain powers on which they can make final decisions (Riker, 1964). The competencies of the federal centre, the sub-state entities, or both are set out in a supreme constitution which cannot be amended unilaterally by the centre or sub-state entities alone. An umpire (usually a Supreme Court, or as in Switzerland, the people by referendum) have the final say in adjudicating ‘competence struggles’ between levels of government or substate entities. Federal states apply territorial self-rule and shared rule provisions, i.e. the governance of the centre must recognize the federal nature of the state by providing mechanisms to incorporate the sub-states in central decision making, not only on issues of territorial constitutional reform but also in matters that require centre-sub-state cooperation – for instance, when sub-state entities implement federal legislation. In federal states, provisions of self- and shared rule are normally applicable across the board, i.e. there is not a part of the state’s territory without some form of sub-state autonomy in place (Elazar, 1987; Hooghe et al., 2010). However, the degree of sub-state autonomy can be variable: India (through special provisions for Kashmir and the north-east), Canada (due to the exemption of Quebec from certain federal policies), Spain (as a result of more extensive autonomy arrangements that apply to the Basque Country, but also Catalonia, Navarre and Galicia), and Belgium (following the merger of Community and Region in Flanders, and the constitutionally inferior position of the Brussels Capital Region and the German Community) all exhibit forms of territorial asymmetry. In fact, asymmetry is the corollary of plurinationalism, since the governments of minority nations may demand asymmetric powers as recognition for their status as nations (hence, Quebec’s demand to be recognized as a ‘distinct society’ different from the English-speaking provinces), while majority nations or as O’Leary put it, the ‘Staatsvolk’ (such as ‘the Castilians’ in Spain) may be more interested in projecting their interests at the centre, which they can control more easily (O’Leary, 2001). In contrast with federalism, devolution results in a less secure constitutional entrenchment of sub-state autonomy. In theory, the centre could unilaterally revoke the autonomy of the sub-state governments. The lack of strong shared rule mechanisms at the centre (compared with some, but not all, federal states) also puts sub-state governments in a comparatively weaker position to stop such retrenchment from happening. Finally, whereas federal states can emerge from a centripetal (or ‘coming together’) or a centrifugal (or ‘holding together’) process (Stepan, 1999), devolved states always emerge from a unitary or union setting and thus are always part of a centrifugal dynamic. The UK is a devolved, not a federal state, because the UK Parliament has retained all (constitutional) sovereignty. As UK constitutional commentators have put it, ‘power devolved is power retained’ (Bogdanor, 2001).3 For similar reasons, constitutional lawyers often question the federal nature of Spain and India. In Spain, the mediation of territorial interests at the centre is weak (the central second chamber does not adequately represent territorial interests), and is often contingent upon the brokerage power of small regionalist parties whenever the party in central office lacks a legislative majority on its own. Furthermore, the revision of the constitution, including sections covering the ‘state of the autonomies’ does not require the explicit consent of the sub-state entities (yet the autonomous communities must endorse proposed revisions to their autonomy statutes). In constitutional terms, India too is considered a ‘union’ and not a ‘federation’. Kenneth Wheare described it as a quasi-federal state at best in view of the extensive provision (and use) of Emergency powers (including the centre’s option – subject to certain conditions – to substitute or suspend state governments) and the right of the centre to legislate in areas of state competence without first requiring state approval. This said, there can be an important disjuncture between the legal or constitutional nature of a state and the actual dynamics of multi-level government within that state. Constitutional lawyers tend to emphasize the former (but there are exceptions, for instance Tierney, 2004), while political 64

Strategies for managing plurinational states

scientists pay more attention to the latter. For instance, many political scientists would argue that contemporary Spain and India function as de facto federal states (see for instance Colino, 2009; Mitra and Pehl, 2010) even though their constitution may not be ‘federal’. In India, the fragmentation of the party system since 1989 has led to the formation of broad-based federal coalition governments in which small, state-based parties often play the role of kingmakers (Sridharan, 2010). A more fragmented centre in tandem with a more activist Supreme Court (policing the political abuses of the constitutional Emergency provisions) have generated a polity that is much more ‘federal’ than ‘union’ like in its actual practice (Sáez, 2002). Similarly, by convention, the UK government will not amend the parliamentary acts that stipulate the selfrule provisions for Scotland, Wales and Northern Ireland without securing the prior consent of the devolved legislatures. Furthermore, Scottish and Welsh self-rule were put in place following a referendum in 1997 which would render their unilateral retrenchment illegitimate in the eyes of the Scottish and Welsh public.4 The Good Friday agreement, envisaging self-rule for Northern Ireland was endorsed by Northern Irish and Irish voters in 1998. Territorial strategies for managing plurinational states are often introduced to contain the demands of sub-state (or minority) nations, but sometimes these nations are so tiny in relation to the overall population of the state, that the proposed solution is not devolution or federalism, but an arrangement that recognizes their separate identity in the context of a unitary state with far-reaching territorial self-rule as the result. Arguably, UK devolution could be seen as such an ‘autonomy arrangement’ (Wolff, 2011), since the combined populations of Scotland, Wales and Northern Ireland only represent about 15% of the UK total. Yet, that share is still considerably above the comparable shares of Greenland, the Åland Islands, Aceh, Bougainville, South Tyrol, Gaugauzia or Hong Kong, relative to Finland, Indonesia, Papua New Guinea, Italy, Moldova and China, respectively (Henders, 2011; Stepan et al., 2011). In these cases, the autonomy arrangement that is put in place is referred to as a ‘federacy’. Stepan et al. (2011: 204) define a federacy as a ‘political-administrative unit in an independent unitary state with exclusive powers in certain areas, including some legislative power, constitutionally or quasi-constitutionally embedded, that cannot be changed unilaterally and whose inhabitants have full citizenship rights in the otherwise unitary state’. Compared with devolved arrangements, federacy arrangements are more secure; that security may not only derive from its entrenchment in the constitution of the ‘parent state’, but also from the role that outside actors play in monitoring the adequate observance of the arrangement. Federacies are often linked to cross-border arrangements, because they seek to accommodate minority nations that may have a strong cultural, historic or economic link with another state: as the (Catholic) Northern Irish with Ireland, or the German-speakers in South Tyrol with Austria.5 Hence, an international treaty between Sweden and Finland, or between Austria and Italy secures the special status of the Åland Islands or South Tyrol in Finland and Italy, respectively. McGarry and O’Leary (2008: 8) have argued that the Good Friday agreement also turned Northern Ireland into a ‘near’ federacy insofar as its status, unlike that of Scotland or Wales is entrenched in an international treaty. They argue that the unilateral suspension of devolution in Northern Ireland between October 2002 and May 2007 was consistent with the framework of devolution, but breached the terms of the international agreement.

The perils and virtues of territorial strategies for plurinational management The context in which territorial strategies emerge Theorists of conflict management have often criticized territorial autonomy arrangements for being inherently unstable (for useful review essays see Amoretti and Bermeo, 2004; Erk and Anderson, 65

Wilfried Swenden

2009; Wolff, 2011). The most prominent criticism is that they empower minority nationalist elites by giving them political, administrative, economic and fiscal resources which they can use to launch demands for further autonomy, putting the state on a slippery slope towards secession or disintegration (see in particular Nordlinger, 1972, for a powerful and early criticism, but also Horowitz, 1985). In recent times, the dangers of federalism or autonomy strategies have been emphasized in particular by scholars working on Russia or Central and Eastern Europe. They frequently point at the failure of the Soviet Union, Yugoslavia or Czechoslovakia to hold together and contrast this with the territorial integrity of the Central and Eastern European states that lacked such self-rule arrangements prior to their democratization. For these scholars the combination of ethno-federal structures with democratization provided a fatal (and sometimes violent) cocktail in which secessionism had become ‘the only game in town’ (see Bunce, 1999; Hale, 2004; Roeder, 2007, 2009; Treisman, 2007). In this chapter, I adopt a more contextual approach to this issue and hence provide a more qualified view on the relationship between territorial autonomy arrangement and plurinational management. More in particular, I argue that one should distinguish the effects of territorial arrangements on state integrity, depending on whether they occur in a non-democratic, a democratizing or a consolidated democratic context (Linz and Stepan, 1996 for a distinction between the three terms). Territorial autonomy is always partially fictitious in a non-democratic regime, because it is subject to the approval of the ruling political and military elites who could rescind it whenever it (is seen to) undermine(s) their authority. Czechoslovakia, Yugoslavia and the Soviet Union were not democratic but they were ‘federal’ in a sense that they had political-administrative units below the level of the central state, each of which had constitutionally assigned responsibilities and resources. They were also ‘plurinational’ since the territorial units overlapped with national republics or minority nations (though not always perfectly so, especially not in the Soviet Union). Ultimately, however, the goal of national federalism was to spread the socialist project and: to deepen central control over the periphery through administrative policies that secured the economic and political hegemony of the [Communist] Party and through social and economic policies that over time would even out the socioeconomic disparities among republics and among nations that constituted those republics. (Bunce, 2004: 427) Like Western modernists, communists assumed that economic development, urbanization and education would erode sub-state identities in the long term. The failure of the communist project to deliver on its economic promises, combined with the very limited career prospects for regional party elites (say to ‘hop’ from the level of the Soviet Union Republic to that of the Soviet Union state) solidified and strengthened sub-state identities. For Bunce, not the ethnic or national diversity of Central and Eastern Europe, but its combination with national federalism (and the institutional resources for minority nations which this implied) explained where territorial disintegration occurred (i.e. only in those states that recognized sub-state nations; see also Brubaker, 1996). In contrast, territorial self-rule has more integrative effects where it is implemented in a plurinational and fully consolidated democratic context. As McGarry and O’Leary (2009: 18) assert, ‘there is not yet an example of an established democratic plurinational federation failing’. Democratic federations can be said to be more successful in accommodating sub-state nationalism because, in comparison with non-democratic or democratizing states, they allow more space for a discussion on civil and political rights across multiple levels, and they combine this with a greater sense of openness for accommodating territorial group rights by a combination of 66

Strategies for managing plurinational states

self- and shared rule provisions. The rule of law and the frequency of competitive elections at the state and sub-state levels serve to sustain the legitimacy of the state and of democracy (Bunce, 2004: 432). In non-democratic federations, like the Soviet Union, that openness was much more constrained and the relative weakening of one level (the centre) at the time of democratization could be more easily exploited by sub-national elites through their monopoly on sub-state power. The disintegration of the Soviet Union shows that democratization, where it occurs in the context of a state that already recognized sub-state nationalities, can trigger a process of territorial disintegration (Filippov et al., 2004). Hence, while plurinational autonomy arrangements and democracy may reinforce the plurinational state, plurinational autonomy arrangements and democratization may undermine it, especially for states in which minority nations had access to important political, institutional or economic resources prior to their democratization. This finding leads Bunce (2004: 435) to the observation that in Spain democratization and plurinational accommodation may have been relatively successful because and not in spite of Franco’s highly unitary regime which preceded it. The complete disregard of territorial self-rule arrangements in the Francoist state meant that sub-state national elites had to negotiate with a powerful central state on obtaining territorial self-rule. This, so Bunce suggests, would have helped to channel (or perhaps force) the sub-state national elites into accepting a solution for territorial accommodation within the context of the Spanish state. Similarly, notwithstanding the calamity of Partition, in 1947 the Indian princely states and the former provinces of British India faced a ‘union’ government that had the political, legal and if needed military firepower to address further threats to the territorial integrity of the Indian state (Adeney, 2007). Without a strong centre (admittedly, helped by the towering figure of Nehru and the ideological binding force of Indian nationalism), India might not have been able to address these territorial challenges successfully. Applying the same reasoning to two different contemporary cases, one could argue that autonomy arrangements stand a better chance of holding together the Iraqi state (which was governed as a unitary state under Saddam Hussein) than Pakistan (where a process of democratization could strengthen peripheral provincial elites who could use their resource strength to weaken the grip of the Punjabi-Muhajiri combine on the Pakistani administrative and military apparatus).

The institutional design of territorial autonomy arrangements: the number and relative size of sub-state units Other than the context (non-democratic, democratizing or fully democratic) in which autonomy arrangements emerge, scholars of governing divided societies have often pointed at the institutional design of these arrangements as factors determining successful territorial management. The design question can be broken up into several parts. First of all, how many sub-units should the federal state have? Even if there is only one key fault line on the basis of which a state could be divided into two territorially concentrated sub-groups, is a two-unit federation the best possible institutional design choice from the viewpoint of federal stability (Bednar, 2008)? Despite its very compact size and population, Switzerland is made up of 26 cantons, but this high number helps to neutralize any divisions that otherwise might have developed in a three- or four-unit federation structured along linguistically homogeneous cantons. Similarly the 10 province-strong Canadian federation helps to fragment the political preferences of the Englishspeaking majority (dispersed across nine provinces) in relation to French-speaking Quebec. The (relative) stability of these multi-member federations contrasts with the territorial disintegration of two-unit Czechoslovakia, Malaysia-Singapore, and pre-1971 Pakistan (which as of 1955 was divided between ‘one unit’ West Pakistan and East Pakistan, the latter seceded as Bangladesh in 67

Wilfried Swenden

1971; Talbot, 2009: 125–48).6 For the same reason, Cyprus or Sri Lanka might not hold together if they were to be governed as bi-communal federations (McGarry and O’Leary, 2009: 20). In contrast, the prospects for Belgium look slightly better because it is a ‘hyphenated’ state (Swenden and Brans, 2006), with Brussels as a third region of mutual interest to Flanders and Wallonia. A second ‘unit’ design issue concerns the relative size of the units. For Hale (2004), the Soviet Union not only disintegrated because democratization imbued its sub-state nationalities with the power to contest the legitimacy of the state, but also because Russia, its core unit, contained more than half of the population. Where the core unit is too strong in demographic and economic terms it is likely to pay much less heed to the demands of the minority nations. In Spain, Canada or India, the ‘majority nation’ lives fragmented across multiple autonomous communities, provinces or states, and as such English-speaking Canadians, Castilians or Hindi-speaking Indians are weakened in their capacity to dominate the state. Therefore, a weakness of UK devolution lies in the overpowering strength of England in the union. The English example also reminds us that there is a limit as to how far a state can go in breaking up its majority nation into territorial fragments. In England, citizens of the north-east (where feelings of sub-English identity were thought to be the strongest) voted against a proposition in 2004 to create a regional assembly, putting on hold plans to roll out devolution to other English regions too. Especially in federacytype arrangements, the desire for self-rule is often limited to only a very small part of a state’s territory, making its extension to other parts of the state unlikely.

The institutional design of territorial autonomy arrangements: self-rule, shared rule and symbolic recognition Next to the number and relative size of territorial units in the state, decisions must be made with regard to what should be their powers, their degree of involvement in co-governing the centre and their recognition as nations in the state. Indeed, as Keating (2001, 2009) reminds us, there are at least three important components to territorial strategies for managing plurinational states: self-rule, shared rule and symbolic recognition (see also McEwen and Lecours, 2008; Swenden, 2010; Moreno and Colino, 2010).

Self-rule Self-rule can take some of the grievances of minority nations away since it gives them the opportunity to make ‘final decisions’ (Riker, 1964) on issues that are dear to their concerns. However, while self-rule may (temporarily) weaken the ‘will’ of minority nations to secede, it also increases their ‘capacity’ to launch claims for further autonomy (or secession; Erk and Anderson, 2009). In democratic regimes, territorial self-rule assumes the presence of directly elected sub-state legislatures. In time voters may shift their political allegiance to the sub-state nation. Through the control of culture (possibly including broadcasting) and education, sub-state governments can ‘rewrite’ national history and mobilize support for more regional autonomy. Most importantly, self-rule is likely to increase the visibility and relative importance of regionalist or secessionist parties (Hepburn, 2009). Brancati (2009) considers the relative strength of such parties as the most important variable for predicting whether self-rule will intensify claims for secession. Especially in plurinational states where sub-state nationalism is confined to a relatively small minority of the overall population (like in the UK), such parties are often too small to weigh substantially on national politics.7 In a context of territorial self-rule, they stand a much better prospect of capturing a sizeable share of seats in the sub-state parliament. This increased leverage can 68

Strategies for managing plurinational states

be used to bring their ultimate goal of independence a few steps closer, or at the very least to intensify further grievances which the centre may feel compelled to accommodate. The extent to which minority nationalist parties will be successful in projecting such a strategy not only depends on whether they are strong enough to govern the minority nation (or need to rely on the support of state-wide parties), but paradoxically also on whether they need the centre to fulfil some of their aspirations. Secession is more likely where the cost of disentanglement is too small, in other words where the centre has been left with few competencies of significance. There is of course no ‘one-size-fits-all blueprint’ of competencies that should stay with the centre and competencies that should be allocated to the sub-state entities in plurinational states. Since citizens of minority nations often have a strong sense of national identity, they are likely to claim (and obtain) self-rule in culture, education or the provision of health or social care. Conversely, most modern multi-level states, all of which are engaged in redistributive policies, tend to keep the financing thereof (though not necessarily their provision) largely central (Watts, 1999; Obinger et al., 2005). Redistributive objectives also imply that the centre often retains a key role in extracting revenues from mineral resources where these are geographically concentrated in one specific area. Yet, in Canada, Alberta can keep most of its oil-extracting revenues to itself, and this is also the case – albeit it to a lesser degree – for the Niger Delta Region and Iraqi Kurdistan. Citizens in federal states with a strong welfare state tradition often expect that the level and criteria for obtaining social benefits should be universally applicable irrespective of one’s place of living. There are efficiency grounds to organize social security at the highest possible level, simply because social risk can be spread across a larger pool of citizens. However, social policies also provide a potent tool of identity building (Beland and Lecours, 2005), which is why sub-state nations have successfully contested the centre’s legitimacy in dominating the provision of welfare (McEwen, 2005). In some states, they have stepped in where the centre retrenched from its previous welfare commitments (as in contemporary Britain where Scotland, unlike the UK, will not charge tuition fees to Scottish undergraduate students), they have contested the centre’s monopoly in providing key social entitlements (as in the case of the Flemish government which provides a Flemish care insurance scheme, over and above such a federal scheme) or they have engaged in a ‘race to the top’ (and not the bottom) in providing the most attractive mix of welfare schemes, especially where they have the financial resources to do so.

Shared rule The stability of a plurinational state not only depends on finding an appropriate balance between centre and sub-state competencies, but also on identifying ways for sub-state nations to influence key policy and constitutional decisions at the centre. Shared rule gives minority nations a sense of co-ownership over central decisions that affect their interests. In a federal setting, the centre usually retains an important role in macro-economic management, social security, foreign policy and defence, and quite often it also has a strong input in determining important tax sources such as income tax, VAT or corporate tax. Often these policies have wide-ranging implications for what the sub-state entities can and cannot do with their own set of competencies. Strong shared rule mechanisms can weaken the need for separate (bilateral) intergovernmental channels between the centre and the minority nations since their involvement in central decisions is structurally guaranteed. However, shared rule is always difficult to organize in a plurinational context, because most shared rule mechanisms (second chamber, intergovernmental forums) do not entrust sub-state nations with a veto-power vis-à-vis the centre (or other regions/nations in the state). Therefore, minority nations tend to prefer bilateral contacts with the centre that serve to underline their ‘special character’ in the state. In federacies or highly asymmetric federal states, 69

Wilfried Swenden

such bilateral arrangements may be the only feasible option for communicating with the centre. For instance, in Spain and Canada, the failure to transform the second chamber into a more effective chamber of territorial representation is not only due to the central government’s fear that such a reform would weaken its capacity to control decision making at the centre (Roller, 2002; Franks, 1999). It is also due to the lukewarm support of the minority nations who fear they may have to trade their informal but often effective voice in relation to the centre for a formalized but also collective sub-national voice in which they may be sidelined by other regions. Hence, Quebec, Catalonia or the Basque Country fear that Senate reform could actually weaken and not strengthen their influence in central decision making. The relationship between self- and shared rule in a plurinational context requires some further thought. For states that are highly asymmetric and that have developed out of a union, it is sometimes said that more territorial self-rule has come at a cost of less shared rule. This observation applies to the UK, in which following devolution English MPs, in view of the lack of self-rule for England, have pushed to reduce the influence of the devolved nations in Whitehall and Westminster (Keating, 2009). Some commentators (Keating, 2009; but also McGarry, 2007) see this as ‘the Achilles Heel’ of UK devolution, since it appears to signal a lack of commitment from the English-dominated centre to invest in the future of the union. In contrast, Belgium and Spain (during the Francoist period) were unitary states and not unions. Therefore, the centre did not recognize the sub-state nations until the first federalization reforms took place. Furthermore, these reforms affected the entire territory (albeit not necessarily in a symmetric way). Consequently, the empowerment of the regions and sub-state nations in both countries has led to the development of a stronger and not weaker set of mechanisms for channelling sub-state grievances at the centre. Comparative scholars may still consider these mechanisms as comparatively weak, especially in the Spanish case, but the key point is that they have become stronger and not weaker (though not necessarily in a linear way) as the process of federalization progressed (Bolleyer, 2009; Colino, 2009). While the UK case may serve to illustrate the potential danger of insufficient shared rule, the Belgian case may demonstrate the difficulty of having too much shared rule. The consociational Belgian federal government entrusts the French-speaking minority (whose interests largely overlap with that of the Walloon Region and French-speaking Community) with a permanent veto-right. With growing political divergence between the two language communities on many issues from how to address the economic crisis to immigration policy, many Flemish MPs have become keener on hollowing the powers that are still left with the centre. They argue that the French speakers have used their federal veto-right irresponsibly and have blocked the modernization of the state. However, the consent of the two language communities is not just required to sanction steps for enlarging the competencies of the sub-state entities, but also for forming a federal government tout court (which must be balanced equally between Dutch- and Frenchspeaking ministers). Decision making at the federal level has been paralyzed as the result of different views on the territorial organization of the state and on the ideological direction of federal policy. Belgium was without a fully furnished federal government for 541 days after the June 2010 federal elections because parties from the two language communities could not bridge their opinions. However, McGarry and O’Leary (2009) would dispute that the excess of shared rule is the root of Belgium’s current woes. After all, so they would argue, the Flemish majority is too small (60–40) to warrant anything less than a full-blown power-sharing arrangement at the centre. Furthermore, in the Brussels Capital Region comparable consociational mechanisms benefit the much smaller (15–85) Flemish minority. The alternative for immobilism could be civil unrest or a control strategy of the Flemish in relation to the French-speaking minority. Furthermore, Belgian ‘consociationalism’ operated relatively well until 2007. The recent 70

Strategies for managing plurinational states

‘problematization’ of shared role is connected to the split party system and the growing ideological polarization between a centre-right Flemish and a centre-left Francophone party system. It would be a step too far, though, to blame federalism for this. The process of federalization (and the uncoupling of federal and regional elections since 2004) may have created more room for the political expression of inter-community differences, but the structure of the Belgian federation has been shaped by and has not shaped the linguistically split party system. The parties were split (or in the process of splitting) when the country became federal. In sum, it is difficult to point at federal design as the main cause of the current Belgian crisis of governability: nationalist opportunism and rent-seeking in the context of a split party system are equally to blame (Swenden, 2013).

Symbolic recognition Symbolic recognition is the final feature of importance when crafting successful territorial strategies for a plurinational state. Symbolic recognition assumes that national political elites are willing to identify the state as plurinational and recognize that it encompasses multiple, but possibly complementary identities (Punjabi and Indian, Sindhi and Pakistani, Tamil and Sri Lankan). Symbolic recognition can be reflected in the choice of national anthem or flag, the wording of the constitution (constitutional preamble), the acceptance that the state has more than one national or official language, religion or tribe. Symbolic recognition may also imply that the constitution or federal practice tolerate ‘asymmetric’ arrangements that underline the ‘distinctiveness’ of the various (minority) nations comprising the state (Kymlicka, 2001). However, symbolic recognition is not only about accommodating sub-state nationalities, it is also about projecting an image that transcends these nationalities. To put it in Stepan et al.’s (2011: 1–39) terms, it is about recognizing that the state is plurinational, in a sense which makes it more than simply ‘robustly multinational’ (they define a ‘robustly multinational state’ as a state in which sub-state nationalities dominate and where there is little room left for the expression of other national identities). In their terms, it is about projecting the state as a ‘state-nation’ (in which a majority of citizens hold multiple loyalties). Political elites need to think through this plurinational logic when designing the federation and crafting policies. For instance, where sub-state nations are closely linked with sub-state languages, the idea of the state-nation assumes that the right to express oneself in the language of one’s nation (linguistic accommodation) should not entirely free oneself from the moral obligation to learn and practice the language of other nations within the state; or – where there are simply too many, as in India – to master a ‘connecting language’ that is understood by most citizens across the state (linguistic integration). There may be strong normative grounds to support linguistic territoriality (van Parijs, 2007), but where linguistic accommodation equals linguistic segmentation, the common ground on which to build a state-wide integrative discourse is likely to be much narrower; possibly too narrow. It is in this sense that symbolic recognition cannot be easily detached from the other two dimensions (self-rule and shared rule); self-rule may provide sub-state nations with a degree of recognition which they seek from the state; shared rule may commit (some) politicians from sub-state nations to a ‘state-nation’ instead of a ‘robustly multinational’ discourse, while a sufficiently powerful federal centre can use its capacity to help cement multiple but complementary identities.

Conclusion: contextualizing territorial design There is no one-size-fits-all solution for managing plurinational states. Choices on the number and relative size of units in a plurinational state can only be steered to a limited extent since 71

Wilfried Swenden

there are few federal(izing) or devolved plurinational states that did not already have a ‘plurinational’ character before they applied territorial strategies. In this chapter I have tried to argue that the record of territorial management strategies is not as dismal as assumed. The failure of non-democratic or democratizing multinational federalism in Central and Eastern Europe cannot be extrapolated to the wider universe of plurinational states. Furthermore, as Ronald Watts once aptly remarked, multinational federations are often difficult to govern not because they are federal or have applied territorial management devices, but rather because they have turned federal or have applied such devices because they seek to manage multinational societies that otherwise would be very difficult (or impossible) to govern (Watts, 1999). Territorial management always involves a combination of self-rule, shared rule and symbolic recognition. Each of these seeks to ‘accommodate’ territorial pluralism within the state. Yet, I have tried to argue that there is a parallel need to balance these accommodative devices with ‘integrationist’ incentives (McGarry et al., 2008); that is, if the purpose of territorial management strategies is to ‘sustain’ and not simply to dissolve or disintegrate the plurinational state (Bednar, 2008). The higher the degree of self-rule, the lower the cost of secession. Similarly, self-rule without shared rule reduces the loyalties that sub-state entities will hold towards the centre. However, the opposite is also true: too much shared rule at the centre on issues that are of little concern to sub-state interests can paralyze central decision making and undermine the legitimacy of the state. Finally, successful symbolic recognition implies that the nation is built on the compound image of sub-state or minority nationalist identities, but also expresses the separate – and added – value of the plurinational state an sich. The right mix of these three elements does not of course guarantee that territorial differences will be managed successfully. There are limits to what institutional design can do. Furthermore, territorial management is only part of the equation; how territorial design interacts with other devices for conflict management (the choice of electoral system, the party system, the presence of a parliamentary, presidential or semi-presidential system, consociationalism) is equally important for assessing the extent to which territorial management contributes to appeasing (or aggravating) plurinational divisions (on this point, see Reynolds, 2010; Wolff, 2011).

Notes 1 A first draft of this paper was presented at the Territorial Politics Research Group Seminar at Edinburgh University, 27 October 2011. I am grateful for the comments from Dani Cetrà, Michael Keating, Nicola McEwen, Cera Murtagh, David Marti Tomás, Dejan Stepjanovic and David Stewart. I am solely responsible for any remaining errors. 2 Some authors (McGarry and O’Leary, 2008; Stepan et al., 2011; Wolff, 2011) also include confederacies in this overview, but I have purposefully left them out of Table 4.1 and the discussion for managing plurinational states. A confederal arrangement applies to more than one state, as state sovereignty remains intact and the confederal centre cannot act without the consent of the states that partake in the confederation. Confederations have either transformed into successful federations (the United States, Switzerland), unitary states (the Netherlands) or disintegrated (Senegambia, or more recently Sudan, which prior to the secession of South Sudan in 2011 was governed under a confederal interim constitutional arrangement) (Reynolds, 2010). The European Union has confederal and federal traits. The European monetary crisis in 2010–11 illustrates the difficulty of sustaining political systems of a primarily confederal nature. 3 Power can be devolved all round, or, as is the case in the UK, to just a few minority nations. UK devolution is peculiar due to the dual asymmetry of the arrangement: not only is England without selfrule, but the powers that have been devolved to Scotland, Wales and Northern Ireland are variable, with Scotland obtaining the largest and Wales the lowest level of self-rule. 4 The suspension of devolution in Northern Ireland for much of the period between 2000 and 2007 was due to the gridlock among coalition partners in the power-sharing Northern Irish executive. Without 72

Strategies for managing plurinational states

workable power sharing in place, the UK government felt it necessary (with the implicit consent of the Northern Irish coalition partners) to suspend self-rule for as long as required. 5 Not all minorities of this type benefit from comparable autonomy arrangements; for instance, the Hungarian-speaking minority in Serbia, Slovakia or Romania does not benefit from such arrangements. 6 Serbia-Montenegro also disintegrated, but it was never governed as a federation. 7 Though they could affect the campaign strategies of polity-wide parties in those regions, see Meguid, 2008; Toubeau, 2011, or they may gain more significant representation in the federal second chamber where it is composed on the basis of equal representation for the territorial units.

References Adeney, K. (2007) Federalism and Ethnic Conflict Regulation in India and Pakistan (Basingstoke: Palgrave-Macmillan). Amoretti, U. and N. Bermeo (eds) (2004) Federalism and Territorial Cleavages (Baltimore: Johns Hopkins University Press). Anderson, B. (1991) Imagined Communities: Reflections on the Origins and the Spread of Nationalism (London: Verso). Bednar, J. (2008) The Robust Federation: Principles of Design (Cambridge: Cambridge University Press). Beland, A. and A. Lecours (2005) Nationalism and Social Policy: The Politics of Territorial Solidarity (Oxford: Oxford University Press). Bogdanor, V. (2001) Devolution in the United Kingdom (Oxford: Oxford University Press). Bolleyer, N. (2009) Intergovernmental Cooperation: Rational Choices in Federal Systems (Oxford: Oxford University Press). Brancati, D. (2009) Peace by Design: Managing Intra-State Conflict through Decentralization (Oxford: Oxford University Press). Brubaker, R. (1996) Nationalism Reframed: Nationhood and the National Question in the New Europe (Cambridge: Cambridge University Press). Bulpitt, J. (1983) Territory and Power in the United Kingdom (Manchester: Manchester University Press). Bunce, V. (1999) Subversive Institutions: The Design and Destruction of Socialism and the State (Cambridge: Cambridge University Press). ——(2004) ‘Federalism, Nationalism and Secession: The Communist and Postcommunist Experience’, in U. Amoretti and N. Bermeo (eds) Federalism and Territorial Cleavages (Baltimore: Johns Hopkins University Press), 417–40. Colino, C. (2009) ‘Constitutional Change without Constitutional Reform: Spanish Federalism and the Revision of Catalonia’s Statute of Autonomy’, Publius: The Journal of Federalism 39 (2): 262–88. Dardanelli, P. (2011) ‘Multi-lingual but Mono-national – Exploring and Explaining Switzerland’s Exceptionalism’, in F. Requejo and M. Caminal (eds) Federalism, Plurinationality, and Democratic Constitutionalism: Theory and Cases (Abingdon: Routledge). Elazar, D.J. (1987) Exploring Federalism (Tuscaloosa: University of Alabama Press). Erk, J. and L. Anderson (2009) ‘The Paradox of Federalism: Does Self-Rule Accommodate or Exacerbate Ethnic Divisions’, Regional & Federal Studies 19 (2): 191–202. Filippov, M., P. Ordeshook and O. Shvetsova (2004) Designing Federalism: A Theory of Self-sustainable Federal Institutions (Cambridge: Cambridge University Press). Franks, C.E.S. (1999) ‘The Canadian Senate. Not Dead Yet but Should it be Resurrected?’ in S.C. Patterson and A. Mughan (eds) Senates: Bicameralism in the Contemporary World (Columbus: Ohio State University Press), 120–62. Ghai, Y.P. (2000) Autonomy and Ethnicity: Negotiating Competing Claims in Multi-ethnic States (Cambridge: Cambridge University Press). Hale, H. (2004) Why Not Parties in Russia? Democracy, Federalism and the State (Cambridge: Cambridge University Press). Henders, S. (2011) Territoriality, Asymmetry and Autonomy: Catalonia, Corsica, Hong Kong and Tibet (Basingstoke: Palgrave-Macmillan). Hepburn, E. (2009) ‘New Challenges to Stateless Nationalist and Regionalist Parties’, Special Issue, Regional & Federal Studies (19): 4–5. Hooghe, L., G. Marks and A. Schakel (2010) The Rise of Regional Authority: A Comparative Study of 42 Democracies (London: Routledge). Horowitz, D.L. (1985) Ethnic Groups in Conflict (Los Angeles: University of California Press).

73

Wilfried Swenden

Keating, M. (2001) Plurinational Democracy: Stateless Nations in a Post-Sovereignty Area (Oxford: Oxford University Press). ——(2009) The Independence of Scotland (Oxford: Oxford University Press). Kymlicka, W. (2001) ‘Minority Nationalism and Multination Federalism’, in W. Kymlicka, Politics in the Vernacular: Nationalism, Multiculturalism and Citizenship (Oxford: Oxford University Press), 91–120. Linz, J.J. and A. Stepan (1996) Problems of Democratic Transition and Consolidation: Southern Europe, South America and Post-Communist Europe (Baltimore: Johns Hopkins University Press). McEwen, N. (2005) Nationalism and the State: Welfare and Identity in Scotland and Quebec (Brussels: PIE-Peter Lang, Federalism and Regionalism Series). McEwen, N. and A. Lecours (2008) ‘Voice or Recognition? Comparing Strategies for Accommodating Multinationalism in Multinational States’, Commonwealth & Comparative Politics 46: 220–43. McGarry, J. (2007) ‘Asymmetry in Federations, Federacies and Unitary States’, Ethnopolitics 6 (1): 105–16. McGarry, J. and B. O’Leary (2008) ‘Territorial Pluralism: Its Forms, Flaws and Virtues’, EDG Paper presented at the Assessing Territorial Pluralism Conference, Kingston, 16–17 May. ——(2009) ‘Must Pluri-national Federations Fail?’ Ethnopolitics 8 (1): 5–25. McGarry, J., B. O’Leary and R. Simeon (2008) ‘Integration or Accommodation? The Enduring Debate in Conflict Regulation’, in S. Choudhry (ed.) Constitutional Design for Divided Societies (Oxford: Oxford University Press), 41–87. Meguid, B. (2008) Party Competition between Unequals: Strategies and Electoral Fortunes in Western Europe (Cambridge: Cambridge University Press). Mitchell, J. (2008) Devolution in the UK (Manchester: Manchester University Press). Mitra, S. and M. Pehl (2010) ‘Federalism’, in N. Gopal Jayal and P.B. Metha (eds) The Oxford Companion to Politics in India (Delhi: Oxford University Press), 43–60. Moreno, L. and C. Colino (2010) ‘Introduction’, in L. Moreno and C. Colino (eds) A Global Dialogue on Federalism – Volume 7. Diversity and Unity in Federal Countries (Montreal and Kingston: McGill-Queen’s University Press), 3–15. Nordlinger, E.A. (1972) Conflict Regulation in Divided Societies (Center for International Affairs, Harvard University). Obinger H., S. Leibfried and F.G. Castles (2005) Federalism and the Welfare State: New World and European Experiences (Cambridge: Cambridge University Press). O’Leary, B. (2001) ‘An Iron Law of Nationalism and Federation? A (Neo-Diceyian) Theory of the Necessity of a Federal Staatsvolk, and of Consociational Rescue’, Nations and Nationalism 7(3): 273–96. Reynolds, A. (2010) Designing Democracy in a Dangerous World (Oxford: Oxford University Press). Riker, W. (1964) Federalism: Origin, Operation and Significance (Boston: Little Brown). Roeder, P. (2007) Where Nation-states Come From: Institutional Change in the Age of Nationalism (Princeton: Princeton University Press). ——(2009) ‘Ethnofederalism and the Mismanagement of Conflicting Nationalisms’, Regional & Federal Studies 19 (2): 203–20. Roller, E. (2002) ‘Reforming the Spanish Senate: Mission Impossible?’ West European Politics 25 (4): 69–93. Sáez, L. (2002) Federalism without a Centre: The Impact of Political and Economic Reform on India’s Federal System (Delhi: Sage). Sridharan, E. (2010) ‘The Party System’, in N. Gopal Jayal and P.B. Metha (eds) The Oxford Companion to Politics in India (Delhi: Oxford University Press), 117–35. Stepan, A. (1999) ‘Federalism and Democracy: Beyond the U.S. Model’, Journal of Democracy Vol. 10 (October): 19–34. Stepan, A., J.J. Linz and Y. Yadav (2011) Crafting State-Nations: India and Other Multinational Democracies (Baltimore: Johns Hopkins University Press). Swenden, W. (2010) ‘Beyond UK Exceptionalism? Comparing Strategies for Territorial Management’, in Klaus Stolz (ed.) Ten Years of Devolution: Snapshots at a Moving Target (Augsburg: Wißner Verlag), 13–35. ——(2013) ‘Rebooting Territorial Pluralism: Belgium and the Crisis of Governability 2007–11’, in J. McGarry and R. Simeon (eds) Assessing Territorial Pluralism (Vancouver: University of British Columbia Press). Swenden, W. and M. Brans (2006) ‘The Hyphenated State, Multi-level Governance and the Communities in Belgium: The Case of Brussels’, in M. Burgess and H. Vollaard (eds) State Territoriality and European Integration (London: Routledge), 120–44. Talbot, I. (2009) Pakistan: A Modern History, 3rd edn (London: Hurst). Tierney, Stephen (2004) Constitutional Law and National Pluralism (Oxford: Oxford University Press). 74

Strategies for managing plurinational states

Toubeau, S. (2011) ‘Regional Nationalist Parties and Constitutional Change in Parliamentary Democracies: A Framework for Analysis’, Regional & Federal Studies 21(4/5): 427–46. Treisman, D. (2007) The Architecture of Government: Rethinking Political Decentralization (Cambridge: Cambridge University Press). Tully, C. (1995) Strange Multiplicity: Constitutionalism in an Age of Diversity (Cambridge: Cambridge University Press). van Parijs, P. (2007) ‘Europe’s Linguistic Challenge’, in D. Castiglione and C. Longman (eds) The Language Question in Europe and Diverse Societies (Oxford: Hart), 217–54. Watts, R.L. (1999) Comparing Federal Systems (Montreal and Kingston: Queen’s-McGill University Press). Witte, E., J. Craeybeckx and A. Meynen (2000) Political History of Belgium (Brussels: VUB Press). Wolff, S. (2011) ‘Consociationalism: Power-sharing and Self-governance’, in S. Wolff and C. Yakinthou (eds) Conflict Resolution: Theories and Practice (London: Routledge), 23–56.

75

6 Federalism, regionalism and the dynamics of party politics Eve Hepburn and Klaus Detterbeck

Introduction Unlike many other scholarly fields covered in this Handbook, political party scholarship has been relatively slow in responding to changed understandings of state structures resulting from federalism and regionalism. Whilst territorial politics has been on the agenda of political science for at least the last 30 years (albeit in the background – see Loughlin, 2009; Keating, 2009), comparative studies of stateless nationalist and regionalist parties began to emerge in the 1990s (de Winter and Türsan, 1998), whilst conceptual analyses of state-wide party adaptation to federal, regionalized and multi-level states have only just surfaced in the last decade. Why is this? Why did party scholars come so late to the territorial politics table? One explanation is practical: some of the most radical changes to state structures, resulting in the wholesale transferral of powers to sub-state levels, have only occurred in the last 20 or so years (despite the persistence of territory as a constant factor shaping politics since the birth of the ‘nation-state’ in the late eighteenth century; see Keating, 1998). This is especially true for formerly centralized states such as Spain, the UK and Italy, which have witnessed unprecedented levels of decentralization from the 1980s onwards. Of course, ‘coming together’ federations such as the United States, Switzerland and Australia, or more ‘recently’ formed federal democracies like Germany and India, empowered the sub-state level decades ago. The lack of scholarly attention to territorial politics in these cases may result from a second explanation, which is normative in nature: party scholars, like most political scientists, tend to focus on the nation-state as the only meaningful unit of political science (Jeffery and Wincott, 2010). Indeed, scholarly understandings of parties are overwhelmingly state-centric, whereby parties are largely seen as seeking to control the machinery of the state and to represent the interests of the state citizenry. These understandings fall through, however, when pitched against the reality of most democracies today. Party politics no longer (if it ever did) exclusively take place at the state level; it does so at multiple – intertwining – levels: neighbourhood, local, regional, provincial, state and in some cases (such as the European Union (EU) member states), supranational. It is the multi-level nature of the state that has caused so much head-scratching amongst scholars on how to conceptualize the role, structure and functions of parties. Political parties are generally considered to be the main instruments of national integration across states, through 76

Federalism, regionalism and party politics

their representative functions and coordinated policy making (Caramani, 2004). However, the challenges of spatial rescaling along federal and regional lines, or ‘territorialization’ to put it another way (see below), have forced state-wide parties to adapt both programmatically and organizationally to the new political realities of multi-level political structures and have challenged their integrative role and capacity. Spatial rescaling requires that state-wide parties adapt to the creation, or strengthening, of several institutional loci of decision making at different territorial levels, which may contain diverse electoral systems with distinct structures of party competition (Swenden and Maddens, 2009). However, recent institutional change is not the only territorial factor causing state-wide parties to alter their focus and strategies. Some issues have long caused state-wide parties to differentiate their organization in different parts of the country. The first is the mounting evidence of the fact that sub-state regions can ‘generate and sustain political cultures, serving as ‘small worlds’ for citizens’ (Henderson, 2010). Regions have become containers of political attitudes and behaviour, as well as social and political identities, which diverge from other parts of the state even in the absence of regional political institutions (Elkins and Simeon, 1980; Moreno, 1999). The second factor is the pressure state-wide parties have faced from the inexorable rise of stateless nationalist and regionalist parties (SNRPs) (Hepburn, 2009). Massetti (2009) maintains that the number of SNRPs operating in Western Europe has more than tripled in the last 30 years, from 29 to 93. Furthermore, these parties are gaining in strength and power, regularly entering government at the sub-state, and sometimes the nation-state, level (Elias and Tronconi, 2011). As a result of these three factors – institutional reforms, societal conditions and party competition – state-wide parties must refocus their strategies for different regional contexts and tackle regional issues head-on. This requires a deviation from the ‘state-wide’ logic of party politics as state-wide parties must increasingly cope with an era of ‘de-nationalization’ (Hough and Jeffery, 2006: 7). Although it has taken some time and effort to convince scholars that territory is not an irrelevance or a ‘residual’ factor in shaping party politics (see Keating, 2009; Jeffery and Wincott, 2010), it appears that issues of federalism and regionalism are increasingly on the radar of mainstream party scholarship, evident in a recent special issue of the journal Party Politics dedicated to decentralization and state-wide parties (Hopkin and van Houten, 2009). Indeed, there has recently been a small surge of academic inquiry into the ‘resurgence of territory’ in party politics. First steps have been made towards analysing the effects of state decentralization on formal party structures and programmes (Carty, 2004; van Biezen and Hopkin, 2006; Hopkin and Bradbury, 2006; Fabre, 2008; Swenden and Maddens, 2009; Detterbeck and Hepburn, 2010), the impact of devolution on electoral politics (Hough and Jeffery, 2006), the mechanics of party systems at different levels (Deschouwer, 2003, 2006; Thorlakson, 2006, 2009; Detterbeck, 2012), and the degree of symmetry (or congruence) between coalition governments at different territorial levels (Stefuriuc, 2009). However, we still lack some of the conceptual tools for understanding how parties – both state-wide parties and SNRPs – have responded to the dynamics of multi-level systems. The aim of this chapter is to make a modest contribution to addressing this gap, by exploring what we call the ‘territorialization’ of parties and party competition in federal and regionalized systems.

Federalism, regionalism and parties Over the last few decades, states throughout the world have witnessed significant spatial restructuring of powers down to lower levels (Marks et al., 2008). The result of these territorial reforms of states has been the empowerment of the sub-state level, and in some cases the 77

Eve Hepburn and Klaus Detterbeck

creation of new political regions. This trend is particularly marked in Europe, where all of the large Western European states, such as Germany, France, the UK, Spain and Italy, and some of its smaller ones, like Austria, Belgium and Switzerland, have created new, or strengthened existing, regional tiers of authority.1 Marks, Hooghe and Schakel (2008) created a ‘regional authority index’ that illustrates the scope and depth of the institutionalization of regions during the period 1950–2006. Of the 42 mainly Organization for Economic Co-operation and Development (OECD) countries examined, half of them had created regional tiers of authority and only two became marginally more centralized. This astonishing finding confirms that ‘the current trend is away from centralization and the nation-state’ (Jeffery and Wincott, 2010: 168). States no longer constitute the ‘main locus of power and decision’ (Loughlin, 2009), if, indeed, they ever did. Sub-state legislatures have carved out their own spaces as sources of power and decision making, becoming focal points for territorial interest representation (Hough and Jeffery, 2006). As such, we can confidently say that sub-state regions have become a ‘core feature of west European politics’ (Hough and Jeffery, 2006: 7). This radical transformation of the state poses important questions for the role, aims, focus and structure of political parties. Studies of political parties and party systems, like the rest of social science, have been dominated by a state-centric bias and theories of ‘nationalization’. Scholars like Caramani (2004) and Chhibber and Kollman (2004) have argued that party systems and competition have become more nationalized in much of the industrialized West, with a homogenization of political structures and processes across the state. These studies assume that political parties are single, homogenous organizations with authority and decision making centralized in the hands of the national executive. However, there is good reason to review and question the classical assumption on the integrative function of political parties in multi-level political systems. Political parties might once have been ‘geared to obtaining control of the apparatus of the central state’ (Loughlin, 2009), but that is no longer the case in federal or regionalized states, whereby many of the functions and powers of states have been devolved downwards. The decentralization of states has put the integrative functions of parties under severe pressure, forcing parties to re-evaluate how best to respond to, and represent, the diverse needs and interests of the electorate. This challenge has become all the more pressing in light of other significant challenges such as the decline of statewide party membership, the rise of smaller, single-issue parties (and in particular, radical-right parties), supranational integration and globalization, amongst other factors. Parties now operate in an increasingly complex and volatile environment with increased competition for votes. Political parties have, of course, responded to these challenges in different ways – and unpacking these different strategies is a further aim of this chapter. In particular, ‘context varies and so do the internal preconditions for organizing multi-layered parties’, thereby resulting in diverse party solutions to these problems (Detterbeck, 2012). We can, however, make some general preliminary comments. The first is that parties have adopted diverse, changeable and often contradictory stances on territorial issues. Certainly, some of these parties were at the helm of the decentralization reforms, arguing that regional autonomy was a solution to administrative overload at the centre, a democratic gesture recognizing the identities of stateless nations and regions, or a panacea to problems emanating from the challenges of nationalist parties (such as the Belgian Christian Democrat, Liberal and Socialist parties that presided over several rounds of federal reforms). Others criticized regionalization reforms, arguing that they would lead to the eventual break-up of the state or, at the very least, a significant loss of sovereignty (such as the UK Conservatives and Partido Popular in Spain). However, many parties changed their position in an opportunistic and instrumental fashion, moving repeatedly back and forth on the ‘territorial question’ (Meguid, 2008; Hepburn, 2010; Toubeau, 2011). We have sought to account for this repositioning of parties on the issue of 78

Federalism, regionalism and party politics

decentralization in Figure 6.1 and Figure 6.2. The most dramatic change over time was evident in the European Left, which moved from a highly centralist position during the post-war period through to the 1960s and 1970s, to becoming staunch advocates of decentralizing policies from the 1990s as part of a ‘good governance’ strategy. The most notable parties here were the Spanish Socialist Party (PSOE), which moved from a Jacobin-oriented perspective of the state to supporting the creation of Spanish autonomous communities (Moreno, 2001); the Left in Italy (represented initially by the Italian Communist and Socialist Parties, then the Democrats of the Left, and currently the Democratic Party), which moved from supporting a centralized, unitarist model of the state to advocating a true ‘regionalization’ in the 1980s and moves towards federalization in the 2000s (Hepburn, 2010); the French Communist Party and various elements of the French Socialist Party (e.g. Chevènement, Mouvement des Citoyens), which moved from anti-decentralization in the 1960s and 1970s to pro-déconcentration reforms in France in the 1980s (Loughlin, 2007); and finally the British Labour Party, which was firmly opposed to decentralization in the 1960s, divided on the first devolution referendum in the 1970s and the primary instigator of devolution following a second referendum in the 1990s (Bradbury, 2006). The Right in Europe, meanwhile, repositioned itself less radically than the Left, from endorsing degrees of administrative, though not political decentralization (called in French déconcentration) – often from a neo-liberal standpoint (see Loughlin, 2009), to outright contempt for some political devolutionary measures (especially in the UK, Spain). This position was most evident amongst the British Conservatives, which did support administrative decentralization in the 1960s (unlike Labour); however, it was entirely hostile to political devolution in the 1970s and 1990s (Hepburn, 2010). In contrast, the position of the Right in Spain and Italy has been generally hostile to the political decentralization of power or recognition of sub-state identities,

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

1- -

Spanish PSOE, French PCF, British Labour, ' , ' -' - - - - - - - - - 'r Italian PCI & PSI , , German SPD, - - - - - - - - - - - - - - , Austrian SPO L...

________

__

-

-

-

-

--1

Belgian BSP-PSB

,

-------~

I

, UK Liberals , : , German FDP, , Belgian

Ideological position

L

_

~V_V~~~ __ ,

r---------l

Italian DC, Belgian CVP/PSC 1 - - - - - - - - - - ...

:

, German CDU, _______ ~ __Austrian OVP

' ,

~- - - - - - -

_I

British

,- - - - - - - - -I Conservatives , , Spanish AP

,

---------

,L _________ ,

I

Support for Decentralization

J

High

Figure 6.1 State-wide party positions on ideology and decentralization, 1970s 79

Eve Hepburn and Klaus Detterbeck

Belgian : PS

French PS

Belgian SP.a German SPD, Austrian SPO r--------I ~ - - - - -- -- -- -- -- -- ~ - : UK Liberals :

Ideological position

,

~

,

French UMP

~:~~uhr

t---------

:' - - - - - - - -" Belgian - - - - - - -': Italian UDC, :: VLD , German FDP , ~ - - - - - -

---------r------=----------

,

: German CDU, ' ,, Austrian OVP :

L _______ I

, British ,, Conservatives

_________

, ,,

Spanish PP, , Italian PdL

r

Support for Decentralization

,

Belgian CD&V

'

L __________ I

.J

I

High

Figure 6.2 State-wide party positions on ideology and decentralization, 2000s

and thus the positions of these parties have not changed a great deal between the 1970s and the 2000s, as shown in Figures 6.1 and 6.2. Of course, some party families have a more principled approach to decentralization, evident in their more-or-less stable support for bringing power ‘closer to the people’ – and the Christian Democrat defence of the principle of subsidiarity is notable here, especially in Italy, Austria and Germany (Swenden and Maddens, 2009). Here, we see little ‘movement’ in Figures 6.1 and 6.2; the positions of Christian Democrats remain more or less the same. An exception to this is provided by CD&V in Flanders which, given the new institutional and competitive context in Belgium, has radicalized its demands for autonomy since the 1970s. Liberal (Democratic) parties, in their support for individual rights, tend to be in favour of decentralized federal structures (though backtracking on this aim if the political climate is not amenable – i.e. the UK Liberal Democrats since joining a Conservative-led coalition government in 2010). Finally, we have witnessed only moderate party repositioning on decentralization in long-standing federal countries such as Germany and Austria, where all parties – be they Left or Right – generally endorse the current structures of the state (Detterbeck, 2012). The responses of stateless nationalist and regionalist parties (SNRPs) have also varied, with some delighted at the creation of new regional polities and engaging in (coalition) governments at that level (Convergència i Unió, CiU, in Catalonia; Plaid Cymru in Wales; Scottish National Party, SNP), and with others radicalizing their demands to include independence, wishing to have nothing to do with the ‘pretend’ regional assemblies (Independence Republic of Sardinia, iRS) (see Gómez-Reino et al., 2006; Elias and Tronconi, 2011). Following the implementation of reforms, however, very few (mainstream) parties have sought to overturn decentralization measures. This does not mean, however, that parties fully comprehend the scope of what has happened, or indeed, how to respond effectively to the challenges of multi-level party systems. Ideology has provided some broad parameters for the strategic and organizational responses of 80

Federalism, regionalism and party politics

parties. We have found, for instance, that in contemporary European party systems, it is the parties on the Left that tend to be more in favour of decentralization and also more likely to regionalize their own party organizations, while parties on the Right generally seek to maintain tighter centralized control of the party (see Figure 6.2; also see Detterbeck and Hepburn, 2010). However, it is important to remember that this used to be the opposite case, with the Left in favour of the unitary Jacobin state, while the Right promoted degrees of (administrative) decentralization (and there were – and still are – also factions within each party that disagree with the party leadership on the territorial question). As such, there are, as we shall see below, no hard and fast rules as to how parties respond to the creation of multi-level party systems or the flourishing of territorial movements seeking the radical restructuring, or indeed complete disintegration, of the state. With the rise in prominence of a number of SNRPs from the 1970s onwards, the capacity of state-wide parties for national integration and representative linkages has been (further) compromised. Given the significance of SNRPs’ territorializing influence on politics, it is worth dwelling on these political actors in more detail.

The rise of SNRPs Stateless nationalist and regionalist parties have been the subject of a great deal of academic debate in the last two decades, and especially in the last few years. In countries throughout the world, but particularly in Europe, these parties have been responsible for pushing the agenda for radical constitutional change, resulting in decentralization and federalism, and some of them have been able to enter government at the regional, and even state, level (Hepburn and Zaslove, 2009; Elias and Tronconi, 2011). Their core business is sub-state territorial empowerment, whereby empowerment involves seeking to represent and advance the particular interests of the sub-state territory (Hepburn, 2009). According to Gómez-Reino, de Winter and Lynch (2006: 258), the decentralization of political structures ‘represents a major policy victory of autonomist parties, as transfer of power from the national to regional authorities constitutes the core demand of regionalist parties’. This is especially true given that many of these parties have waited so long to see some of their goals realized. While a handful of SNRPs emerged at the turn of the twentieth century, these parties made their first big splash in the 1970s (de Winter and Türsan, 1998; Keating, 1998; de Winter et al., 2006). During the heyday of the decolonization struggles, the birth of the civil rights movement and peace movement, worldwide student protests and economic turmoil following the oil crisis of 1973, many SNRPs won the sympathy of regional populations and were elected to parliament.2 Stein Rokkan and his colleagues (Lipset and Rokkan, 1967; Rokkan and Urwin, 1982) developed the most authoritative framework for explaining the emergence of these ‘peripheral movements’. They argued that during the first wave of centralizing state formation, whereby territories were expanded and governed by a single authority, state agencies faced the existence of relatively autonomous peripheral communities. A centre-periphery cleavage was drawn between the dominant national culture and the ethno-linguistic minorities, which mobilized along territorial lines (using whatever territorial, economic or cultural resources were available to them) in resisting centralizing and standardizing policies from the centre (Rokkan and Urwin, 1982). As such, parties and party systems are formed around such cleavages and have the ability to ‘freeze’ them even when the relevance of the cleavage has declined (Lipset and Rokkan 1967: 50). However, at the time, Rokkan and Urwin (1983: 165) were unable to find evidence of the ‘unfreezing’ of politics resulting from peripheral movements, which they concluded had ‘not been very successful’. They would probably draw different conclusions if they applied their framework to today’s politics. 81

Eve Hepburn and Klaus Detterbeck

Once considered to be ‘throwbacks to the past’ (Hobsbawm, 1990), SNRPs have established themselves as both reputable and influential political players in most Western European democracies. Whilst Rokkan and Urwin (1983) identified 29 ‘peripheral parties’, Lane et al. (1997) put this number at 44 ‘ethnic parties’, and Massetti (2009) has more recently estimated that 93 such parties exist. Clearly, SNRPs are increasing in numbers, thereby contradicting modernization theory that posits that territorial cleavages will be eventually ironed-out during the process of nationalization (see Keating, 2009). Not only that, but SNRPs have also moved from ‘niche’ actors in party systems to mainstream political players, or ‘protest to power’ as Elias and her colleagues argue (Elias, 2009; Elias and Tronconi, 2011). No longer the ‘outsider’ in party politics, these parties have successfully entered government at the regional and state levels and many have been responsible for pushing the agenda for radical constitutional change. SNRPs have also sought to adapt to multi-level politics, by differentiating their party machines at different territorial levels – in particular, the supranational level. Of note, SNRPs have prolifically taken advantage of opportunities at the European level for building alliances and networking, with many endorsing the popular (but ultimately disappointing and short-lived) goal of a ‘Europe of the Regions’ (Lynch, 1996; Elias, 2008, 2009; Hepburn, 2008, 2010). SNRPs have also had success, to varying degrees, in entering government at different territorial levels (Gómez-Reino et al., 2006; Hepburn and Zaslove, 2009; Stefuriuc, 2009; Elias and Tronconi, 2011). With the creation of new sub-state electoral arenas, many SNRPs have sought to transform themselves into governing parties at that level. This, however, presents multiple challenges for these parties; in crossing the threshold of government (especially if they consider themselves ‘anti-system’ parties), in entering coalition government and possibly having to moderate/abandon some of their goals as part of a compromise deal, and the effects of government incumbency, whereby a poor performance may be severely punished by their supporters (see Elias and Tronconi, 2011). As such, operating in a multi-level political environment often forces these parties to change their strategies, behaviour, and in some cases to compromise some of their principles. Another major challenge for these parties, which results from the development of a multidimensional policy space within multiple political arenas, is the creation of a new political rival, in the form of regional branches of state-wide parties that have become more autonomous and territorially focused (Hepburn, 2009; Elias, 2011). The decentralization of state-wide party organizations means that SNRPs are no longer the only ones seeking to represent territorial interests or presenting alternative constitutional goals. A difficult challenge for SNRPs is when the territorial goals of the state-wide party prove more popular than their own aims (for instance, when regional electorates choose autonomy over independence in a referendum – a scenario likely to happen in Scotland in 2014–15); however, the worst-case scenario is when state-wide parties effectively ‘steal’ the goals of the SNRPs, rendering them obsolete. This situation occurred in Belgium in the 1960s and 1970s, when SNRP success encouraged the Christian Democrats to split into two autonomous, unilingual parties in 1968 to pursue their separate visions of the future of the Belgian state (followed by the splitting of the Liberal and Socialist parties along similar Flanders–Walloon lines in the 1970s). Furthermore, the new Belgian ‘regional’ parties essentially adopted the SNRPs’ positions in their efforts to reform the state along federal lines. On the plus side, the aims of the SNRPs had been met; on the down side, this act removed their very reason for existing, resulting in a steep electoral decline (Deschouwer, 2009). However, regionalist parties continue to have a place in federal Belgium – often as coalition brokers. As Gómez-Reino et al. (2006: 258) argue, ‘no autonomist party is entirely satisfied with the current depth, width and speed of these [decentralization] transfers’. Few parties will see that such reforms constitute a ‘mission accomplished’; instead they will keep fighting for more powers. 82

Federalism, regionalism and party politics

The next question is: how do state-wide parties respond to this competition from SNRPs? The obvious answer is that they must present themselves, and their programmes, as serious regional rivals to the nationalist/regionalist party vision. However, the way in which they do this varies enormously from place to place and party to party.

The territorialization of state-wide parties The reconfiguration of political authority across different territorial levels, combined with the persistence of regional identities and the rise of SNRPs, has necessitated an adjustment to how state-wide parties organize and compete. The creation of multi-layered political arenas means that parties can no longer pursue one strategy for office in a single state-wide political arena. Instead their priorities are split: they must target different arenas for decision making at several territorial levels. In particular, at the sub-state level, state-wide parties must refocus their strategies for different regional contexts (and in European countries, parties must also address and compete on European policies and issues). The resulting reorganization and repositioning of parties to compete on these multiple territorial levels may be described as the ‘territorialization’ of political parties. We use the term territorialization, as opposed to the ‘regionalization’ (Swenden and Maddens, 2009), ‘federalization’ (Koole, 1996) or ‘decentralization’ (Fabre, 2008; Hopkin and van Houten, 2009) of parties, for a number of reasons. All of these terms seem to imply that a degree of power is given to substate party branches, to act autonomously from the central party. However, this ignores two different aspects of power relations. First, there is a need to account for the direction of power redistribution. Political parties, in responding to multi-level politics, may choose to concentrate power at the state or supranational levels, as opposed to devolving authority downwards. This is evident, for instance, in the case of Australia whereby local party branches have sought to work together to strengthen their national party strategy (Koop and Sharman, 2008), or in countries in Europe, where parties have developed a strong European arm, which may even interact with regional branches independently of the state level (Moon and Bratberg, 2010). In these cases, power is redistributed across multiple levels, which could mean upwards to the state or Europe, as well as downwards to the regions. Second, there is a need to account for power-sharing and interdependence within parties. The notion of devolving authority downwards in parties is unable to account for the fact that this autonomy may be constrained by what Carty (2004: 9) calls ‘a fundamental interdependence of parts’, whereby authority is shared across a party, rather than concentrated in a single place. It is therefore necessary to take into account the interlocking as well as autonomous aspects of political parties as they become more focused on territorial issues (alongside other old and new cleavages) as well as organizationally attuned to multi-level political structures The ‘territorial rescaling’ of state-wide parties has a number of dimensions. First, regional branches of state-wide parties have adopted stronger territorial party identities and rhetoric. This involves altering party logos, letterheads, posters and other party literature to reflect the importance of the locality to the party. It also involved a qualitative change in party discourse, emphasizing the importance of the territory (be it culture, identity, social values), territorial representation within the state, and often pledges to defend the interests of the territory. In some cases this has led to sub-state branches declaring themselves to constitute the party representing the nation/ region in opposition to SNRPs. Second, different sectors of parties have adopted different policy goals. In particular, sub-state branches of state-wide parties may develop constitutional alternatives to independence to defuse support for SNRPs. This is facilitated by the exploitation and repackaging of party traditions with regard to their positions on regional autonomy. All of the main party families have both 83

Eve Hepburn and Klaus Detterbeck

centralizing and decentralizing traditions, as argued earlier. For instance, whilst Liberal (Democratic) parties have at times supported the creation of a federal state, in which the identities and traditions of a territory are recognized, at other times they have opposed ‘particularism’ in any form. Left-wing parties have shifted back and forth between centralism and regionalism, the latter especially when they entered alliances with SNRPs. Christian Democrat parties have advocated bringing powers to local communities in line with the principle of subsidiarity. In every case, however, when competing with SNRPs, regional sub-state branches of state-wide parties must emphasize the commitment of the party to a stance on decentralization that is seen to outweigh the benefits of secession. Third, the territorial rescaling of parties results in the organizational differentiation of regional parts from the centre. The fact that sub-state branches often need to differentiate their policies from the centre to ‘fit’ the local setting has resulted in a desire for greater sub-state organizational independence, with control over candidate selection, finance and electoral strategies. This transition has not always met with the blessing of the central party. In many cases, the central party’s reluctance to grant concessions to the sub-state branch has led to intra-party tensions, with regional units threatening to secede. Fourth, the territorial rescaling of parties has resulted in new forms of power-sharing amongst constituent units. In rescaling party organizations, power and authority no longer rest in one single place, but rather different organizational units within parties possess different powers and autonomous functions. This raises questions for how the different parts interact with, and influence, one another. In this sense, parties are developing new stratarchical organizational structures, replacing the hierarchical structures of old (Carty, 2004). In the remainder of this chapter, for reasons of space, we will focus in particular on the last two dimensions of territorial rescaling, concerning state-wide party organizational responses.

State-wide party responses to multi-level competition Following our argument, we suggest looking at the interaction of two key dimensions when exploring the relationships between the central party level and its sub-state branches: the strength of joint decision-making structures within party organizations, and the degree of organizational and programmatic autonomy enjoyed by sub-state branches (see Dyck, 1996; Deschouwer, 2006). Indicators of joint decision-making are formal and informal linkages between state and sub-state units that determine the extent of ‘shared rule’ within parties, including mechanisms for regional input into state-wide decision making, joint party structures and the inclusion of regional officials in the state executive. Indicators of sub-state autonomy, or ‘self-rule’, are regional control over candidate and leadership selection, policy programmes, campaign strategies, coalition building and finance. The combination of different degrees of joint decision making and autonomy creates a set of ideal types of state-wide party organizations, which are captured in Figure 6.3. Most of these party types are integrated across territorial levels. However, the typology also covers bifurcated parties which compete at multiple levels but have cut organizational linkages between territorial party layers (Smiley, 1980). We will illustrate each party type by giving some examples. These examples also point to the fact that individual parties may move between types over time.

Consensualist parties Consensualist parties have a strong tendency towards uniformity in political appeal and to find consensus through joint decision making. Sub-state branches have only limited autonomy but possess privileged access to central party decision making. There is a high degree of multi-level 84

Federalism, regionalism and party politics

I I

---------------,

Consensualist parties

I

L ______________ _

Federalist parties

Joint Decision making

L

__________

-J-----------, Confederalist parties

I

,------------I

,I

- - - - - - - - - -

--,

Centralist parties

I

Decentralist parties

1------------1

Bifurcated parties

I

Regional autonomy

I

I

High

Figure 6.3 Typology of multi-level party organizations

cooperation inside the party organizations resulting in common policies and strategies suiting all territorial interests. In terms of membership structures, party bureaucracy, parliamentary coordination and career patterns, the links between party levels are extensive. Party elite coordination in state executives is one of the main features here. In consensualist parties, everything comes together at the centre, which is the ‘heart’ of the party. In the joint federal systems of Austria and Germany parties have come to mirror the strong interdependence between territorial levels in their internal structures. However, there are differences between the more hierarchical approach of the Social Democrats and the more decentralized traditions of the bourgeois parties in both countries. Moreover, in recent years, Austrian and German sub-state party branches have made more extensive use of their formal rights in going their own ways (Detterbeck and Jeffery, 2009). Thus, there has been a gradual move from the consensualist to the federalist model of territorial party organization.

Federalist parties Federalist parties combine strong shared rule with strong self-rule. Sub-state branches have considerable freedom to decide over internal procedures while at the same time being strongly involved in central decision-making processes. There will be a delicate, sometimes contested, balance between the institutionalized will to find common ground and the need to allow for diversity among the different sub-state branches. Federalist parties are most likely to thrive in joint federal systems. The Austrian and German Christian Democrats, which have been established in the Länder first before building the state-wide organization, are probably the best examples here. Yet, as noted above, the other parties in these countries have come to resemble the federalist type more closely over time. 85

Eve Hepburn and Klaus Detterbeck

Confederalist parties Like federalist parties, the confederalist type is characterized by elaborate joint decision-making mechanisms and a strong element of sub-state autonomy. Here, however, the locus of control rests with the constituent parts. The central party coordinates rather than leads the party, and the sub-state branches delegate organizational competences according to their own preferences. Confederalist parties are most likely to thrive in dual federations. Despite differences in the actual degree of standardization in policy making, party organizations are centred on the constituent units (cantons, provinces or states) of the federation in countries like Australia, Switzerland and the United States. Regional party branches dispose much of the organizational resources and have considerable leeway in devising their strategies and policies. Nevertheless, the central party level has taken on an increasingly important role in coordinating territorial interests and levelling out differences. Central bodies have thus become stronger in leading the parties and in managing diversity without imposing uniformity (Rydon, 1988; Ladner, 2007). The internal flow of power has thus been reconfigured into a ‘two-way street’ (Wekkin and Howard, 2012). The Latin American federations of Argentina and Brazil show similar developments. In both countries, the base of political support is traditionally concentrated at the regional level. Provincial and state governors and party bosses are key figures at both central and regional party levels. Again, though, national party bodies have grown stronger in recent decades in enforcing party loyalty and discipline (Gordin, 2004; Santos and Pegurier, 2011).

Centralist parties Centralist parties are integrated by virtue of hierarchical control. The central party interferes heavily in sub-state party matters, whereas the regional branches have limited impact on central affairs. Both shared rule and self-rule are weak. As a result of centralist coordination, which extends to all facets of party activities, party cohesion is rather strong. Depending on the strategies of the central leadership, these parties will be capable of having uniform party policies and electoral tactics across the territory. An important route to party centralism seems to be one-party hegemony. In countries like India, Malaysia, Mexico and South Africa a dominant party has been (or still is) at the centre of federal politics. These ‘catch-all parties’ have brought together a variety of social forces in their quest for power. The centralism in political organizations, often based on clientelistic networks, has strengthened the position of state-wide leaders in such parties. Looking at developments over time, however, federalism has proved to be a powerful weapon for opposition parties to challenge one-party hegemony. In a similar vein, sub-state party elites have grown more self-confident in asking for more autonomy or a stronger voice at central level. In the Mexican Partido Revolucionario Institucional (PRI), for example, there has been a shift in authority towards the regional party leadership after the party lost power at the state level in 2000 (Prud’homme, 2012). More generally speaking, party centralism may be conceived of as a means of ‘holding together’ in plurinational settings (Stepan, 2001). Top-down hierarchy prevents sub-state branches from adopting autonomist and separatist positions. In this sense, central control can serve both the aims of leadership ambitions and national integration. ‘Holding-together’ has been an important motto not just of parties like the Malaysian Barisan Nasional (BN) or the African National Congress (ANC) in South Africa, but also of state-wide parties in Spain and Italy. In these countries, we may again note differences between party families. Conservative parties tend to be more hierarchical than their left-wing competitors which show more open leadership structures (Hopkin, 2009; Fabre and Méndez-Lago, 2009). 86

Federalism, regionalism and party politics

Decentralist parties Decentralist parties put a strong emphasis on self-rule. In a stratarchical way, there are multiple territorial centres of power which operate with a considerable degree of independence from one another. While still being integrated parties, most vital activities have been separated between territorial levels and vertical links are rather weak. Decentralist parties can have quite distinct policy profiles and electoral strategies in different parts of the country. Traditional examples of this party type are the Swiss parties or the Australian Liberal Party. Today, there are several cases of asymmetrical federalism and regionalism where the relations between the central party and individual sub-state branches show decentralist traits. In Great Britain, Labour and the Conservatives have devolved substantial powers to their Scottish and Welsh branches while establishing only weak instruments of joint decision making (Bradbury 2006). In Spain, the Catalan Socialists and the Conservatives in Navarre (until 2008) enjoyed a special status allowing them strong autonomy rights at the sub-state level (Fabre and Méndez-Lago, 2009). In Italy, the centre-left Democratic Party has inherited a formally federal party network in which, in particular, the sub-state branches in the party strongholds yield significant autonomous resources and powers. This, however, rarely translates into influence at the national level (Conti et al., 2009).

Bifurcated parties Finally, bifurcated parties compete in multiple territorial arenas under the same party label but have cut all organizational linkages between state-wide and sub-state levels. The lack of common membership and leadership structures allows each party unit to develop its own position autonomously. At the same time, bifurcated parties are unable to engage in internal multi-level coordination activities with respect to electoral campaigns and public policy making. In Canada, the Liberal Party has formally separated into independent federal and provincial organizations in four of the nine provinces, including Quebec and British Columbia (Thorlakson, 2006). In the other provinces, vertical linkages are comparatively weak. The Conservative Party at state-wide level has no organizational connection to the provincial parties which bear the same name (Cross, 2004). In Belgium, parties have split along linguistic lines. All major parties compete at federal and sub-state levels but restrict themselves to either Dutch-speaking or French-speaking electorates. With the exception of the Greens, organizational linkages to the ideological ‘sister party’ in the other language community have become dormant (de Winter, 2006). Thus, Belgian parties are multi-level organizations which operate on a non-state-wide basis.

Explaining variance in party responses Clearly, state-wide parties have developed a range of different responses to the increasingly asymmetrical nature of party competition in federal or decentralized states. Responses have varied across countries, as well as across parties within one system. In explaining variation, we take clues from some of the major schools of party research (Ware, 1996): sociological approaches, which focus on social cleavages; institutional approaches, which focus on either structural aspects of the polity or the organizational structures of parties; and rational choice approaches, which focus on party strategies in competitive environments (see Deschouwer, 2006; Swenden and Maddens, 2009). The first explanation concerns the salience of territorial interests and identities within society. The existence of territorial cleavages – including distinct sub-state identities or regional economic disparities – can act as an independent factor influencing the organizational decisions of 87

Eve Hepburn and Klaus Detterbeck

state-wide parties (even in the absence of an SNRP that seeks effectively to mobilize these interests). In the Spanish or Italian parties, for example, sub-state branch autonomy tends to be higher in regions where there is stronger support for sub-state nationalism. Second, party responses are affected by the institutional structures of the multi-level state. Within this category, two aspects define the environment in which parties operate: the territorial structures of the state and the structures of the electoral process. The territorial structures of the state determine the degree of interdependence between political levels. In federal and regionalized states, these levels can be heavily intertwined or strongly separated. To illustrate: the strong linkages between party levels in Austria and Germany mirror the extensive cooperation within their federal systems. Furthermore, structures of the electoral processes – primarily electoral systems and electoral timing, but also parliamentary rules of incompatibility of mandates and state regulations on public funding of parties – must be taken into account. As with territorial state structures, electoral structures influence the degree of interdependence between political levels. In the UK, for example, differences in electoral formula and the separation of devolved elections from Westminster elections have added to the distinctiveness of the Scottish and Welsh party systems. Our third set of factors relates to ‘sticky’ party traditions, which shape the ways in which individual parties distribute power internally and link territorial levels. Party traditions can be understood with respect to party ideologies and constitutional aims. In many cases parties adopt the organizational structures that they wish to see implemented within the state. Christian Democrats, along with Liberals and Greens, may find intra-party decentralization more suited to their programmatic profiles and tradition than Social Democrats and the more radical Left. Internal party dynamics, while being based on historical trajectories and party ideologies, will also be affected by the specific position a party has in a competitive environment. This may lead to a stronger propensity to adapt strategically to changing structures of competition. Among the most relevant aspects of this fourth explanatory dimension are the access to government at state-wide and sub-state levels, variations in electoral strength between different arenas and the impact of SNRPs on competitive dynamics. All of these factors will have an impact on internal power balances between the central party and regional branches. Quite often, parties in government at the state-wide level have seen a diminishing role of the regional party ‘barons’ inside their party organizations.

Conclusion To conclude, this chapter has argued that political parties, which were once instruments of national integration, are now faced with the challenge of regionalism, federalism and ‘denationalization’. It seems that the territorial cleavage, which Rokkan and his colleagues posited had only a negligible effect in shaping state-wide party competition in 1982, has since borne fruit with the institutionalization of regions over the last 30 years. In addition to the persistence of regions as ‘small worlds’ as containers of political attitudes, behaviour and identities, and the consolidation of stateless nationalist and regionalist parties as mainstream players in party systems, it is clear that state-wide parties operating in multi-level systems must adapt to the territorial cleavages cemented by the institutionalization of political regions. As we have shown, there are numerous ways in which political parties have responded to state structural change and the strengthening of the territorial dimension of politics. Whilst some have sought to reflect new divisions of powers within the state in their own organizational strategies, others have held on to the belief that a united party will underpin a unified state. We have identified six modes of adaptation amongst the parties: centralist (with centralized decision making and weak joint 88

Federalism, regionalism and party politics

structures), consensualist (implying consensus through joint decision making), federalist (where parties mix strong intra-party coordination with high degrees of flexibility), confederalist (supremacy of the sub-state party level within integrated structures), decentralist (where parties devolve significant control to regional branches), and bifurcated parties (showing a lack of vertical linkages between levels). The repositioning and reorganization of parties at the regional level heralds a new type of political representation in multi-level states. Whilst parties have traditionally claimed to represent citizens across the state by appealing to a common political vision, in multi-level states it has become difficult to commit the party as a whole to a single policy programme. Parties must now adapt to several arenas of political authority, compete with SNRPs on the territorial dimension, and accommodate the territorial interests of the regional electorate. This has had a significant impact on party systems, causing greater divergence between regional and state-wide parties and systems, and leading to new forms of coordination and interdependence within parties. Though it is too soon to herald the decline of state-wide political representation, statewide parties must urgently re-think how to maintain their integrative functions in increasingly disintegrating party systems, otherwise they run the risk of disintegrating themselves.

Notes 1 This was evident in Italy from 1970 when powers were transferred to 15 ordinary regions (Palermo, 2005); in Spain during the 1980s with the implementation of a regional system of 17 autonomous communities (Moreno, 2001; Colino, 2008); in France with the creation of 22 newly elected regional councils in 1982 (Cole, 2006; Loughlin, 2007); Belgium in 1995 when it became a federation (Deschouwer, 2005); the UK from 1998 when new regional legislatures were created in Scotland, Wales and Northern Ireland (Mitchell, 2009); and Germany from the late 1990s with the strengthening of Länder powers (Detterbeck and Jeffery, 2009). Cases outside Europe include Argentina, Brazil, Mexico, Bangladesh, Malaysia, the Philippines, Kenya, Nigeria and South Africa. 2 For example, Partido Andalucista, Volksunie in Flanders, SNP, Unione Démocratique Bretonne, Liga Veneta, Partido Nacionalista Vasco, and CiU.

References Bradbury, J. (2006) ‘British Political Parties and Devolution: Adapting to Multi-level Politics in Scotland and Wales’, in D. Hough and C. Jeffery (eds) Devolution and Electoral Politics (Manchester: Manchester University Press). Caramani, D. (2004) The Nationalization of Politics: The Formation of National Electorates and Party Systems in Western Europe (Cambridge: Cambridge University Press). Carty, K. (2004) ‘Parties as Franchise Systems: The Stratarchical Organizational Imperative’, Party Politics Vol. 10, No.1: 4–24. Chhibber, P. and K. Kollman (2004) The Formation of National Party Systems: Federalism and Party Competition in Canada, Great Britain, India and the United States (Princeton: Princeton University Press). Cole, A. (2006) ‘Decentralization in France: Central Steering, Capacity Building and Identity Construction’, French Politics Vol. 4: 31–57. Colino, C. (2008) ‘The Spanish Model of Devolution and Regional Governance: Evolution, Motivations and Effects on Policy-making’, Policy & Politics Vol. 36, No. 4: 573–86. Conti, N., C. Roux and F. Tronconi (2009) ‘Introduction: Parties and Voters in a Multi-level Electoral Setting’, Modern Italy Vol. 14, No. 2: 127–33. Cross, W. (2004) Political Parties (Vancouver: University of British Columbia Press). Deschouwer, K. (2003) ‘Political Parties in Multi-layered Systems’, European Urban and Regional Studies Vol. 10/3: 213–26. ——(2005) ‘The Unintended Consequences of Consociational Federalism: The Case of Belgium’, in I. O’Flynn and D. Russell (eds) Power Sharing: New Challenges for Divided Societies (London and Ann Arbor: Pluto Press), 92–106. 89

Eve Hepburn and Klaus Detterbeck

——(2006) ‘Political Parties as Multi-level Organizations’, in R. Katz and W. Crotty (eds) Handbook of Party Politics (London: Sage). ——(2009) ‘The Rise and Fall of the Belgian Regionalist Parties’, Regional & Federal Studies Vol. 19, Nos. 4–5: 559–677. Detterbeck, K. (2012) Multi-level Party Politics in Western Europe (Basingstoke: Palgrave Macmillan). Detterbeck, K. and E. Hepburn (2010) ‘Party Politics in Multi-level Systems: Party Responses to New Challenges in European Democracies’, in J. Erk and W. Swenden (eds) New Directions in Federalism Research (London: Routledge). Detterbeck, K. and C. Jeffery (2009) ‘Rediscovering the Region: Territorial Politics and Party Organizations in Germany’, in W. Swenden and B. Maddens (eds) Territorial Party Politics in Western Europe (Houndmills: Palgrave Macmillan). de Winter, L. (2006) ‘Multi-level Party Competition and Coordination in Belgium’, in D. Hough and C. Jeffery (eds) Devolution and Electoral Politics (Manchester: Manchester University Press). de Winter, L., M. Gómez-Reino and P. Lynch (2006) ‘Introduction: Autonomist Parties in European Politics’, in L. de Winter, M. Gómez-Reino and P. Lynch (eds) Autonomist Parties in Europe: Identity Politics and the Revival of the Territorial Cleavage (Barcelona: Institut de Ciènces Polítiques i Socials (ICPS)). de Winter, L. and H. Türsan (1998) Regionalist Parties in Western Europe (London: Routledge). Dyck, R. (1996) ‘Relations Between Federal and Provincial Parties’, in B. Tanguay and A. Gagnon (ed.) Canadian Parties in Transition (Toronto: Nelson). Elias, A. (2008) ‘Whatever Happened to the Europe of the Regions? Revisiting the Regional Dimension of European Politics’, Regional & Federal Studies Vol. 18, No. 5: 482–92. ——(2009) Minority Nationalist Parties and European Integration (London: Routledge). ——(2011) ‘Party Competition in Regional Elections. A Framework for Analysis’, ICPS Working Papers, No.295. Elias, A. and F. Tronconi (2011) ‘From Protest to Power: Autonomist Parties in Government’, Party Politics Vol. 17, No. 4: 505–24. Elkins, D. and R. Simeon (1980) Small Worlds: Provinces and Parties in Canadian Political Life (Toronto: Methuen). Fabre, E. (2008) ‘Party Organization in a Multi-level System. Party Organizational Change in Spain and the UK’, Regional & Federal Studies Vol. 18, No. 4: 309–29. Fabre, E. and M. Méndez-Lago (2009) ‘Decentralization and Party Organizational Change: The British and Spanish State-wide Parties Compared’, in W. Swenden and B. Maddens (eds) Territorial Party Politics in Western Europe (Houndmills: Palgrave Macmillan). Gómez-Reino, M., L. de Winter and P. Lynch (2006) ‘Conclusion: The Future Study of Autonomist and Regionalist Parties’, in L. de Winter, M. Gómez-Reino and P. Lynch (eds) Autonomist Parties in Europe: Identity Politics and the Revival of the Territorial Cleavage (Barcelona: Institut de Ciènces Polítiques i Socials (ICPS)). Gordin, J. (2004) ‘Testing Riker’s Party-based Theory of Federalism: The Argentine Case’, Publius: The Journal of Federalism Vol. 34, No. 1: 1–14. Henderson, A. (2010) ‘Why Regions Matter: Sub-state Polities in Comparative Perspective’, Regional & Federal Studies Vol. 20, No. 4–5: 439–45. Hepburn, E. (2008) ‘The Rise and Fall of a “Europe of the Regions”’, Regional & Federal Studies Vol. 18, No. 5: 537–55. ——(ed.) (2009) ‘New Challenges for Stateless Nationalist and Regionalist Parties’, Special Issue of Regional & Federal Studies Vol. 19, Nos. 4–5: 477–650. ——(2010) Using Europe: Territorial Party Strategies in a Multi-level System (Manchester: Manchester University Press). Hepburn, E. and A. Zaslove (2009) ‘From Alpine Protest to National Influence: The Territorial Strategies of the Italian Lega Nord and the Bavarian CSU’, paper presented at the ECPR General Conference, Potsdam, 9–12 September. Hobsbawm, E. (1990) Nations and Nationalism since 1780 (Cambridge: Cambridge University Press). Hopkin, J. (2003) ‘Political Decentralization, Electoral Change and Party Organizational Adaptation: A Framework for Analysis’, European Urban and Regional Studies Vol. 10, No. 3: 227–327. ——(2009) ‘Decentralization and Party Organizational Change: The Case of Italy’, in W. Swenden and B. Maddens (eds) Territorial Party Politics in Western Europe (Houndmills: Palgrave Macmillan). Hopkin, J. and J. Bradbury (2006) ‘British State-wide Parties and Multi-level Politics’, Publius: The Journal of Federalism Vol. 36, No. 1: 135–52. 90

Federalism, regionalism and party politics

Hopkin, J. and P. van Houten (2009) ‘Decentralization and State-wide Parties: Introduction’, Party Politics Vol. 15, No. 2: 131–36. Hough, D. and C. Jeffery (2006) ‘An Introduction to Multi-level Electoral Competition’, in D. Hough and C. Jeffery (eds) Devolution and Electoral Politics (Manchester: Manchester University Press). Jeffery, C. and D. Wincott (2010) ‘The Challenge of Territorial Politics: Beyond Methodological Nationalism’, in C. Hay (ed.) New Directions in Political Science (Basingstoke: Palgrave Macmillan). Keating, M. (1998) The New Regionalism in Western Europe (Cheltenham: Edward Elgar). ——(2009) ‘Thirty Years of Territorial Politics’, West European Politics Vol. 31, No. 1: 60–81. Koole, R. (1996) ‘Cadre, Catch-all or Cartel?’ Party Politics Vol. 2: 507–23. Koop, R. and C. Sharman (2008) ‘The Elusive Nature of National Party Organization in Canada and Australia’, paper presented at Canadian Political Science Association annual conference, University of British Columbia, Vancouver, 3–6 June. Ladner, A. (2007) ‘Political Parties’, in U. Klöti et al. (eds) Handbook of Swiss Politics (Zürich: Neue Zürcher Zeitung Publishing). Lane, J., D. McKay and K. Newton (1997) Political Data Handbook: OECD Countries, 2nd edn (Oxford: Oxford University Press). Lipset, S. and S. Rokkan (1967) ‘Cleavage Structures, Party Systems, and Voter Alignments: An Introduction’, in S. Lipset and S. Rokkan (eds) Party Systems and Voter Alignments: Cross-national Perspectives (New York: Free Press). Loughlin, J. (2007) Subnational Government: The French Experience (Palgrave Macmillan). ——(2009) ‘The ‘Hybrid’ State: Reconfiguring Territorial Governance in Western Europe’, Perspectives on European Politics and Society Vol. 10, No. 1 (April): 49–66. Lynch, P. (1996) Minority Nationalism and European Integration (Cardiff: University of Wales Press). Marks, G., L. Hooghe and A. Schakel (2008) ‘Measuring Regional Authority’, Regional & Federal Studies Vol. 18, No. 2: 111–21. Massetti, E. (2009) Political Strategy and Ideological Adaptation in Regionalist Parties in Western Europe, unpublished DPhil thesis, University of Sussex. Meguid, B. (2008) Party Competition between Unequals: Strategies and Electoral Fortunes in Western Europe (Cambridge: Cambridge University Press). Mitchell, J. (2009) Devolution in the United Kingdom (Manchester: Manchester University Press). Moon, D. and Ø. Bratberg (2010) ‘Conceptualizing the Multi-Level Party: Two Complementary Approaches’, Politics Vol. 20, No. 1: 52–60. Moreno, L. (1999) ‘Local and Global: Mesogovernments and Territorial Identities’, Nationalism and Ethnic Politics Vol. 5, Nos. 3–4: 61–75. ——(2001) The Federalization of Spain (London: Frank Cass). Palermo, F. (2005) ‘Italy’s Long Devolutionary Path Towards Federalism’, in S. Ortino, M. Zagar and V. Mastny (eds) The Changing Faces of Federalism: Institutional Reconfiguration in Europe from East to West (Manchester: Manchester University Press), 182–201. Prud’homme, J. (2012) ‘Federalism, Political Parties and Civil Society in Mexico’, in W. Renzsch and K. Detterbeck (eds) Parties and Civil Society in Federal Systems (Montreal: McGill-Queen’s University Press). Rokkan, S. and D. Urwin (1982) ‘Introduction: Centres and Peripheries in Western Europe’, in S. Rokkan and D. Urwin (eds) The Politics of Territorial Identity (London: Sage). ——(1983) Economy, Territory, Identity: Politics of West European Peripheries (London: Sage). Rydon, J. (1988) ‘The Federal Structure of Australian Political Parties’, Publius: The Journal of Federalism Vol. 18: 159–71. Santos, F. and F. Pegurier (2011) ‘Political Careers in Brazil: Long-term Trends and Cross-sectional Variation’, Regional and Federal Studies Vol. 21, No. 2: 165–84. Sharman, C. (ed.) (1994) Parties and Federalism in Australia and Canada (Canberra: Federalism Research Centre, Australian National University). Smiley, D. (1980) Canada in Question: Federalism in the Eighties (Toronto: McGraw-Hill-Ryerson). Stefuriuc, I. (2009) ‘Government Formation in Multi-level Settings: Spanish Regional Coalitions and the Quest for Vertical Congruence’, Party Politics Vol. 15, No. 1: 93–115. Stepan, A. (2001) ‘Toward a New Comparative Politics of Federalism, (Multi)nationalism, and Democracy: Beyond Rikerian Federalism’, in A. Stepan (ed.) Arguing Comparative Politics (Oxford: Oxford University Press). Swenden, W. and B. Maddens (2009) ‘Introduction: Territorial Party Politics in Western Europe’, in W. Swenden and B. Maddens (eds) Territorial Party Politics in Western Europe (Basingstoke: Palgrave Macmillan). 91

Eve Hepburn and Klaus Detterbeck

Thorlakson, L. (2006) ‘Party Systems in Multi-level Contexts’, in D. Hough and C. Jeffery (eds) Devolution and Electoral Politics (Manchester: Manchester University Press). ——(2009) ‘Patterns of Party Integration, Influence and Autonomy in Seven Federations’, Party Politics Vol. 15, No. 2: 157–77. Toubeau, S. (2011) ‘Regional Nationalist Parties and Constitutional Change in Parliamentary Democracies: A Framework for Analysis’, Regional & Federal Studies 21(4/5): 427–46. van Biezen, I. and J. Hopkin (2006) ‘Party Organization in Multi-level Contexts’, in D. Hough and C. Jeffery (eds) Devolution and Electoral Politics (Manchester: Manchester University Press). Ware, A. (1996) Political Parties and Party Systems (Oxford: Oxford University Press). Wekkin, G. and J. Howard (2012) ‘Political Parties, Civil Society and American Federalism’, in W. Renzsch and K. Detterbeck (eds) Parties and Civil Society in Federal Systems (Montreal: McGill-Queen’s University Press).

92

7 Fiscal federalism and the political economy of territorial finance Anwar Shah

Introduction To date, there have been 28 federal or quasi-federal and 20 decentralized unitary countries with a combined total of about two-thirds of the world’s population. These governance arrangements have led to a resurgence of interest in fiscal federalism principles and practices as federal systems are seen to provide safeguards against the threat of centralized exploitation as well as decentralized opportunistic behaviour while bringing decision making closer to the people. Fiscal federalism principles are concerned with the division of economic powers and the structure and design of intergovernmental finance to secure a common economic union. Each country adapts these principles to suit its political, institutional and cultural characteristics; as a result, each country has a unique structure of economic governance. More often than not, though, politics trumps fiscal federalism principles. This chapter is concerned with four areas of concern: division of taxing powers for sales and resource taxes, fiscal equalization transfers, tax competition, and shared responsibility for fiscal discipline.

The choice of fiscal constitution under multi-order governance: political economy considerations A stylized grouping of fiscal constitutions on a centralization/decentralization spectrum would show ‘one republic’ having a unitary fiscal constitution as the most centralized and Montesquieu’s confederate republic as the most decentralized multi-order governance structure. Under a unitary constitution, a country has a single or multi-tiered government wherein effective control of all government functions rests with the central government. In a confederate republic of city states or ‘civic republics’ (Kincaid, 1967), the central government acts as the agent of the member units and may not have any independent taxing and spending powers. James Madison’s vision of a ‘compound’ republic would occupy an intermediate space on this spectrum. The political and economic tradeoffs and choices in territorial finance offered by alternate fiscal constitutions are briefly discussed below to serve as a background for our discussion of taxing and conflicting choices in territorial finance. 93

Anwar Shah

One republic A unitary constitution facilitates centralized decision making. It places a greater premium on national security, law and order, uniformity and equal access to public services than it does on democratic rights, voice and exit, and diversity of preferences. Facing a threat of foreign invasion and internal conflict, this was a preferred system of governance under autocratic or dictatorial rule or single-party regimes. A large majority of countries retain unitary constitutions to ensure uniform and equal access even though concerns about foreign invasion and internal uprising have largely dissipated and autocratic rule has given way to entrenchment of democratic rights and freedom. Over time, these countries have also recognized the diverse preferences of localities and the virtues of decentralized public decision making. Hence, some unitary countries (e.g. Japan, South Korea, Denmark, Norway, Sweden, Finland, China and Poland) are more decentralized than most federal countries. A decentralized unitary constitution has an added virtue of constraining duplicate layers of legislative and executive organs and associated costs, especially of intermediate tiers of government. Such tiers are either non-existent as in Japan, South Korea, Denmark, Norway, Sweden and Finland or have relatively minor significance as in China and Indonesia. A unitary government typically has a greater degree of centralization of taxing powers than expenditure responsibilities. Exceptions to this rule are the Scandinavian countries where lower government orders are primarily self-financed.

Confederate republic A confederate republic comprising autonomous self-governing city-states or civic republics or Tiebout communities offering diverse choices of local tax-and-service menus and enjoying veto power over national government decision making represents the most decentralized form of multi-order governance. Such a constitution places great premium on democratic choice, exercise of voice and exit through voting by ballots or with one’s feet, diversity of preferences, and virtues of home rule and community governance. Civic republics are seen as catalysts for collective action by harnessing good Samaritan values to improve economic and social outcomes. Small and open jurisdictions, which are self-financed, ensure that decision making is closer to the people and that median voter preferences matter. Because such small jurisdictions are vulnerable to hostile action, the confederate government, which is financed by contributions from member republics, assumes national security and defence obligations. It also discharges other economic responsibilities authorized by member states such as having a common currency and a joint monetary authority. Switzerland and the United States originated as confederate republics. Confederate republics, however, are vulnerable to political instability arising from external aggression or internal conflict. The confederate government is hamstrung to take timely and appropriate military and economic actions because of the internal political divisiveness. Protection of the poor and disadvantaged minorities is also not assured due to the autonomy of the civic republics. Democratic choice is also not guaranteed, as was witnessed in Singapore, a citystate. In the absence of a choice of small, open economies with a wide array of public services and tax burdens and costless mobility, efficiency is also not assured in a Tiebout economy. Equity, equality and redistribution goals are compromised for the sake of economic efficiency. Since most societies place some premium on altruistic goals, Montesquieu’s confederate republic gained little popularity in practice. Various nations looked for political and fiscal constitutions that would marry efficiency and equity objectives while preserving democratic choice. An important alternative vision, the ‘compound republic’ comprising a compact between states and an independent central government, was conceptualized by the framers of the United States Constitution 94

Fiscal federalism and territorial finance

and subsequently redefined and replicated in a large number of countries (see Inman, 2006; Inman and Rubinfeld, 1997).

Compound republic – federation of states and provinces In principle, a federal form of government has a multi-order governance structure, with all orders of government having some independent as well as shared decision-making responsibilities. This opens up the possibility of having a compound republic that comprises a federation of states (provinces) as discussed below. Shared rule and territorial finance remain contentious issues in most federations. Federalism has been seen as either a ‘coming together’ or a ‘holding together’ of constituent geographic units (Iaryckzower et al., 2000). ‘Coming together’ has been the guiding framework for mature federations such as the United States, Canada and more recently the European Union (EU). The ‘holding together’ view, also called the ‘new federalism’, represents an attempt to decentralize responsibilities from a central government to states or provinces with a view to overcoming regional discontent with central policies and forestall secession. This view is the driving force behind the current interest in principles of fiscal federalism in unitary countries and relatively newer federations such as Brazil, India, and Pakistan and emerging federations such as Iraq, Nepal, Spain, Sri Lanka and South Africa. In Pakistan, this was the primary motivation for the unanimous consent to the 2010 passage of the 18th amendment to Pakistan’s constitution to empower provinces. Federal countries broadly conform to one of two models: dual federalism or cooperative federalism. A third model, the so-called ‘competitive federalism’, wherein all governments have overlapping responsibilities and compete vertically and horizontally to establish their clientele of services, is mostly a theoretical construct and not practised anywhere (Kenyon and Kincaid, 1991; Breton, 1995, 2006; Salmon, 2006). Under dual federalism, the federalism compact is between the federal and provincial governments, and they have separate and distinct responsibilities, while local governments are typically creatures of the provinces as in Canada, India, Pakistan and the United States. Under cooperative federalism, central and provincial roles can assume one of three forms: interdependent spheres as in Germany, ‘marble cake’ with overlapping responsibilities as in Belgium, or independent spheres as in Brazil. In all these models, with the sole exception of the independent spheres model, provinces have a strong constitutional role, and local governments remain creatures of provinces and states (Kincaid, 1990). The dual federalism model empowers provinces and states, and moves decision making somewhat closer to people. It also has the advantage of dealing with ethnic and linguistic conflicts if provinces are numerous and geographically small enough to represent populations with relatively homogeneous characteristics and similar tastes and preferences for certain taxes and public services (e.g. cantons in Switzerland). If provinces are delineated as economic regions, they can also enhance efficiency of the internal common market by exploiting economies of scale and scope. They have the potential, as well, to deal with inter-local spill-overs and intra-regional inequities. Provincial governments can also be responsive to citizen preferences if provincial government is not captured by feudal, industrial, or military elites. The absence of a well-developed communication and transportation system and a lack of urbanization also makes provinces almost a necessity for geographically large countries. The dual federalism model, nevertheless, has significant shortcomings:  Tragedy of commons associated with common pool resources: Under dual federalism, both the centre and the provinces compete to claim a larger share of the fixed national pie. This accentuates 95

Anwar Shah

universalism and pork-barrel politics, leading to a tragedy of commons where all federating units out-compete each other in profligate spending and giveaways in taxes and subsidies.  A Leviathan model of governance: Empowering provinces leads to a potential for greater duplication of government structures and processes at the central and provincial levels, leading to increased costs for the exchequer and higher transactions costs for citizens. This may also lead to overgrazing by politicians and bureaucrats. As a result, the growth in the size of government becomes unrelated to the quality and quantity of service delivery. Opportunism and pork-barrel politics lead governments to act as employment-creation agencies detracting them from their primary role in financing public services and providing a supportive enabling environment for private-sector-led growth.  Agency problems with incomplete contracts: In most large countries, empowering states and provinces does not necessarily move decision making closer to people. States do not correspond to Tiebout communities or Kincaid’s ‘civic republics’. Provinces and states are often larger in geographic size and population than smaller countries. New York, California, Ontario, Sao Paulo, Sindh and Bihar, for example, have governments larger than those of many countries. Having decision making far removed from people implies that provincial governments have incomplete contracts with their citizens and cannot be held to account by people at large (Seabright, 1996). In countries where politics is dominated by feudal, military and industrial elites, such as Pakistan, this leads to complete alienation of governments from their people. This lack of accountability in governance is further accentuated by a constraining of voice and exit options under provincial empowerment due to large areas and interstate barriers to mobility.  Weaker and fragmented local governance: The fiscal federalism approach is focused on internalizing benefits and costs of service provision to the same jurisdiction and treats local government as a subordinate tier under multi-order governance and outlines principles for defining the roles and responsibilities of various orders of government (see Boadway and Shah, 2009). Hence, in most federations, as in Canada and the United States, local governments are creatures of state (intermediate order) governments (dual federalism). In a few isolated instances, as in Brazil, they are equal partners with higher-level governments (cooperative federalism), and in Switzerland they are the main source of sovereignty and have greater constitutional significance than the federal government. Thus, depending on the constitutional and legal status of local governments, intermediate order governments in federal countries assume varying degrees of oversight of the provision of local public services. Generally, this constrains the role of local governments because expansion of their role comes at the expense of the intermediate order of government. As globalization and the information revolution diminish the economic relevance of the intermediate order of governments, these conflicts are accentuated and intermediate order governments have a tendency to play a more intrusive role in local government to stay politically relevant. Both in Canada and Pakistan, provincial governments have recently moved to constrain local government powers. These fiscal federalism perspectives serve as a response to market failures and heterogeneous preferences with little recognition of government failures or the role of entities beyond government. Empowered states and provinces under dual federalism create incentives for weaker and numerous local governments. The exigencies of provincial politics dictate that local governments are given straitjacket mandates with few resources and are kept under tight provincial reigns as in India and Pakistan. Empowered provincial governments typically encourage local fragmentation so as to allow more intrusive controls. In India, there are 254,119 local governments responsible for a pittance (5%) of national expenditures. Most of these expenditures finance the salaries and allowances of civil servants and 3 million elected officials with little left for public services. In contrast, in China where 96

Fiscal federalism and territorial finance











the provincial role is restrained and local governments are empowered, there are only 43,965 local governments, accounting for 51.4% of national expenditures (see Qiao and Shah, 2006). Stifling local innovations: Provinces and states often impose one-size-fits-all mandates that constrain local flexibility and stifle innovation. In Pakistan, provincial ordinances in 2000 required all local governments, small or large, to have 16 departments and a fixed number of positions. In the United States, outdated state laws that are rooted in unjustified distrust of local decision making have stifled successful cities from developing and implementing coherent visions of their future and better services for their residents (see Frug and Barron, 2008). Constraining good governance and strangling metropolitan growth: Empowered provinces typically block rationalization of local government functions, especially when local empowerment implies chipping away at their own powers. A classic example is the powers assigned to metropolitan areas under dual federalism. The fiscal federalism literature suggests that large metropolitan areas should have autonomous two-tier regional governments with powers equivalent to those of a province and with a direct interface with the centre. For this reason, the Shanghai, Beijing, Bangkok and Seoul local governments are treated by China, Thailand and South Korea, respectively, as provinces. In India and Pakistan, where states and provinces are relatively more powerful, metropolitan areas such as Mumbai, New Delhi, Karachi and Lahore are treated at par with smaller local governments with limited autonomy. Fragmentation of internal common market: Empowered provinces also have the potential to create internal barriers to trade and factor mobility through domicile (residence) requirements and regulatory and trade barriers. Mature federations like the United States have circumvented these problems through a constitutional interstate commerce clause. These barriers, however, are formalized in the political and bureaucratic system of India and Pakistan, resulting in a fragmented economic union. Increased threat of succession: Empowering provinces poses a threat to political union, especially in countries with ethnic, linguistic and religious divides and with a smaller number of provincial jurisdictions with one or more dominant provinces, such as Pakistan. As a rule of thumb, all dual federalism models with fewer than 10 provincial jurisdictions tend to face internal conflicts and potential political instability. Diminished economic relevance of intermediate-order governments (provinces and states) under ‘glocalized’ governance: Globalization and the information revolution are making the economic role of provinces largely redundant. Globalization empowers supranational regimes and local governments at the expense of national and provincial governments. Globalization also implies that a country’s international competitiveness is decoupled from its resource base but directly linked with its knowledge base. This suggests a greater role of national governments in financing education and training. National governments also assume greater importance in social risk management due to the vagaries of the global system and social dumping by corporations trying to stay internationally competitive. National governments also assume a greater role in securing an economic union. The provincial economic role, however, is on the wane as the information revolution makes national coordination and oversight over local governments and horizontal coordination at the local level through inter-local partnerships feasible, as done in Finland. In view of the above pressures, states and provincial governments are being pressed to reposition their roles in order to retain economic relevance. The political role of states and provinces, however, remains strong in all nations, and is even on the rise in some nations, such as Germany, Canada, Pakistan and India. Germany’s Länder have assumed a central role in implementing EU directives and in policy making for regional planning and development. In Canada, Alberta and Ontario recently re-centralized secondary and postsecondary education. In India, states have effectively blocked implementation of the 73rd 97

Anwar Shah

and 74th amendments to the Constitution empowering local governments. In Pakistan, provinces recently moved to scale back the fiscal and administrative autonomy of local governments. Thus, while economic considerations warrant a leaner role for provinces and fatter role for local governments, few countries are likely to make such role reversals. Given the weaknesses of Montesquieu’s ‘confederate republic’ and Madison’s ‘compound republic’, a hybrid of the two – a ‘compound republic’ comprising a federal compact between Montesquieu-Kincaid civic republics and an independent central government – is worth exploring.

Compound republic as a federation of civic republics A compound republic as a federation of civic republics but having an overarching central government directly accountable to citizens at large for its autonomous functions and to civic republics for contracted functions offers the advantages of Tiebout communities in democratic choice and accountability while assuring efficiency and equity in a federal economy. Member states in a compound republic are self-financed and accountable to local residents. A compound republic can also be justified if accountability in governance, while keeping citizens’ transactions costs low, is a prime concern in multi-order governance, as is the case in neo-institutional economics (NIE). Shah and Shah (2006; and Shah and Shah 2007) apply NIE principles to develop a framework for analysing fiscal systems and local empowerment and for comparing mechanisms for local governance. This framework is helpful in designing multiple orders of government and in clarifying local government responsibilities in a broader framework of local governance. According to the NIE framework, various orders of governments (as agents) are created to serve the interests of the citizens as principals. The jurisdictional design should ensure that these agents serve the public interest while minimizing transactioncosts for the principals. The existing institutional framework does not permit such optimization because the principals have bounded rationality; that is, they make the best choices on the basis of the information at hand but are ill informed about government operations. Enlarging the sphere of their knowledge entails high transactioncosts, which citizens are unwilling to incur. Those costs include participation and monitoring, legislative costs, executive decision-making costs, agency costs or costs incurred to induce compliance by agents with the compact, and uncertainty costs associated with unstable political regimes (Horn, 1997; Shah, 2005). Agents (officials of various orders of government) are better informed about government operations than are principals, but they have an incentive to withhold information and to indulge in opportunistic behaviours or ‘selfinterest seeking with guile’ (Williamson, 1985: 7). Thus, the principals have only incomplete contracts with their agents. Such an environment fosters commitment problems because the agents may not follow the compact. The situation is further complicated by weak or extant countervailing institutions, path dependency and the interdependency of various actions. Countervailing institutions such as the judiciary, police, parliament and citizen activist groups are usually weak and unable to restrain rent-seeking by politicians and bureaucrats. Historical and cultural factors and mental models by which people see few benefits to, and high costs of, activism prevent corrective action. Further, empowering local councils to take action on behalf of citizens often leads to loss of agency between voters and councils because council members may interfere in executive decision making or get co-opted in such operations while shirking their legislative duties. The NIE framework stresses the need to use various elements of transaction costs in designing jurisdictions for various services and in evaluating choices between competing governance mechanisms. 98

Fiscal federalism and territorial finance

Given the high transaction costs and perceived infeasibility of market and hierarchical mechanisms of governance for partnerships of multiple organizations, a network mechanism of governance is a possible mode of governance for such partnerships – the kind to be managed by local governments. The network form of governance relies on trust, loyalty and reciprocity between partners with no formal institutional safeguards. Networks formed on the basis of shared interests (interest-based networks) can provide a stable form of governance if membership is limited to partners that can make significant resource contributions and if there is a balance of powers among members. Members of such networks interact frequently and see cooperation in one area as contingent on cooperation in other areas. Repeated interaction among members builds trust. Hope-based networks are built on the shared sentiments and emotions of members. Members have shared beliefs in the worth and philosophy of the network goals and have the passion and commitment to achieve those goals. The stability of such networks depends greatly on the commitment and style of their leaders (Dollery and Wallis, 2001) and the catalytic and mediating role played by local governments (Shah, 2010). This framework has important implications for reforming the structure of government and territorial finances (see Table 7.1). Top-down mandates on local governments need to be replaced by bottom-up compacts. The role of local government must be expanded to serve as a catalyst for the formulation, development and operation of a network of both government providers and entities beyond government. Local government’s traditionally acknowledged technical capacity becomes less relevant in this framework. More important are its institutional strengths as a purchaser of services and as a facilitator of alliances, partnerships, associations, clubs and networks for developing social capital and improving social outcomes. Two distinct options of multi-order governance are possible in this regard: 1 a compound republic with local government as the primary agent, subcontracting to local providers, provincial/regional and federal or central government authorities and engaging networks and entities beyond government; and 2 a compound republic with local, provincial/regional and federal/national governments as independent agents.  Option A: compound republic with local governments as primary agents of citizens: In this option, a local government serves as 1 purchaser of local services, 2 facilitator of networks of government providers and entities beyond government, and 3 gatekeeper and overseer of province/state and national governments for the shared rule or responsibilities delegated to them. This role represents a fundamental shift in the division of powers from higher to local governments. It has important constitutional implications. Residual functions reside with local governments. Provincial legislatures would not be elected directly; they would be constituted from local government representatives to perform inter-municipal services. The provincial chief executive (governor) could either be directly elected or nominated by the centre subject to confirmation by the provincial council. The provincial council would make policies on inter-local issues and oversee the provincial executive headed by the governor. The governor could be removed by a three-quarters majority of the provincial council. In Finland – a country with no intermediate-order governments – inter-municipal functions are performed through voluntary partnerships among local governments. The national government is assigned redistributive, security, foreign relations and interstate (inter-regional) functions such as harmonization and consensus on a common framework. The Swiss and the Chinese systems bear some affinity to this model.  Option B: compound republic with various orders of government as independent agents: An alternative framework for establishing the supremacy of the principals is to clarify the responsibilities and functions of various orders as independent agents. This framework limits shared rule. 99

Anwar Shah

Finance follows function strictly, and fiscal arrangements are reviewed periodically for finetuning. Local governments enjoy home rule, with complete tax and expenditure autonomy. Brazil’s fiscal constitution incorporates some features of this model, albeit with significant deviations.  Feasibility of options: Option A is well grounded in the history of modern governments and is most suited for countries with a continuing history of internal or external conflict. It is already practised to some degree in Switzerland, Finland, Denmark, Sweden and China. Although a majority of governments have tinkered with their fiscal systems, the radical change recommended here is not in the cards anywhere. The unlikelihood of overcoming path dependency – a tall order for existing institutions and vested interests – makes such reform infeasible. Option B, therefore, might be more workable, but the clarity of responsibilities might not be politically feasible. There is unlikely to be political will to undertake bold reforms, but piecemeal adaptation of this model might be forced on most countries by globalization and citizen empowerment facilitated by the information revolution (Shah, 2010). Table 7.1 Trends in multi-order governance structure: 20th century versus 21st century 20th century

21st century

Unitary or federal Centralized or provincialized Centre that manages or coerces Citizens as agents, subjects, clients and consumers Focus on government

Decentralized unitary or compound federal republic Globalized and localized Centre that leads Citizens as governors and principals

Focus on division of the fiscal pie Competitive edge for resource-based economies Federalism as a tool for coming together or holding together Residuality principle, Ultra vires, ‘Dillon’s rule’ Limited but expanding role of global regimes with democracy deficits Emerging federal prominence in shared rule

Strong state (province) role Diminishing role of local government

Tax and expenditure centralization with revenue sharing and input-based conditional grants to finance sub-national expenditures Source: Adapted from Boadway and Shah, 2009

100

Focus on governance with interactive direct democracy Focus on responsive, responsible, fair and accountable governance Competitive edge for human capital-based economies Global collaborative federalism with a focus on network governance and reaching out Community governance principle, subsidiarity principle, home or self-rule and shared rule Wider role of global regimes and networks with improved governance and accountability Leaner but caring federal government with an enhanced role in education, training and social protection Ever-diminishing economic relevance of states (provinces) and tugs-of-war to retain relevance Pivotal role of local government as the engine of economic growth, primary agent of citizens, gatekeeper of shared rule, facilitator of network governance; wider role of ‘beyond government’ entities Tax and expenditure decentralization with fiscal capacity equalization and output-based national minimum standards grants for merit services

Fiscal federalism and territorial finance

In practice, in the United States, a compound republic of independent civic republics was not feasible due to path dependency. Such a compact was only feasible among states given the institutional context of the United States in the aftermath of its civil war. Switzerland is the only federal country conforming to this view of federalism. Swiss cantons come closer to the original concept of civic republics. The Swiss cantons initially struck a compact to create a confederate republic but moved over time to strengthen the overarching responsibilities of a federal government directly accountable to Swiss citizens. They also eliminated the veto power of individual cantons and replaced it with a supermajority of cantons and majority of citizens through public referenda. What is unique about Switzerland among federal countries is its higher degree of commitment to the principles of citizen sovereignty, local home rule and community governance.

Taxing choices in raising and sharing revenues from natural resources One of the most vexing issues in federal assignment of powers is resource ownership and sharing of natural resources. Ideally to strengthen common economic and political union, natural resources should belong to the country as a whole, and all net resource revenues should be deposited in a national heritage fund (Norwegian style) and owned by all citizens regardless of their place of residence. The assets of this fund are held in perpetuity and cannot be drawn down, but capital income is available for current use – both as dividends distributions to citizens and revenues for all orders of government. Another justification for national ownership of natural resources is the exhaustible nature of some resources such as minerals, oil and gas. Once a deposit is depleted, the resource rents become unavailable. Hartwick (1978) has argued that economic efficiency and intergenerational equity would then point to the need to capture and save all rents from the resources. Assuming that the ownership of all resources is vested in the federal government, it would capture all ‘rents’ – that is, the economic surplus in excess of market returns to the capital and labour involved in exploration and extraction. However, for ‘hidden’ resources or those with uncertain quality, a share of rents must be awarded to discoverers and developers of resource deposits so as to compensate them for the risks of exploration and extraction, thereby promoting the socially optimal levels of discovery and development. Additionally, with exhaustible resources, resource extraction will ultimately cease to provide a base for regional economic activity. To this end, resource revenues should be reinvested in capital that helps regions transition to a sustainable industrial base as resource deposits are depleted. State and local ownership of natural resources creates difficulties in sharing revenues among members of the federation. Of course, resources are immobile and can be easily taxed by subnational jurisdictions (see Table 7.2 for a representative assignment and the practice in Brazil and Canada). However, because resources tend to be distributed unequally across a nation, subnational taxation perpetuates regional inequalities and may encourage fiscally induced migration to resource-rich regions, resulting in inefficient allocation of resources. Also, because of their instability and unpredictability, natural resource revenues are not an appropriate source of revenue for sub-national governments. Some resource taxes are, however, designed to cover costs of local service provision, such as property taxes, royalties and fees. Severance taxes on production and output could be assigned to local governments. In addition, sub-national governments could impose taxes to discourage local environmental degradation. This explains why in Australia, Canada and the United States, intermediate-order governments (and some local governments in the United States) impose such taxes on natural resources. Despite these economic efficiency and equity considerations, national ownership of natural resources has not been possible in most federal nations, except Russia and South Africa (unitary in constitution but quasi-federal in practice) (see Table 7.3). States and local governments often 101

Anwar Shah

Table 7.2 Assignment of resource taxes: ideal vs. practice in Brazil and Canada Tax instrument

Representative assignment

Brazil

Canada

Resource rent taxes (income and profits) Royalties, fees, severance, production, output, property taxes Conservation charges Carbon taxes Energy taxes Motor fuel taxes

Federal

Federal with tax sharing with state and local Federal with tax sharing with state and local

Federal-State with tax base sharing States except property tax as a local tax

Federal – Federal Local

State and local – State State

State and local

State and local Federal Federal/state/local Federal/state/local

contend that they should have the right to ownership and revenues from natural resources located within their boundaries. Many sub-national governments claim a right to convert resource wealth (their ‘heritage’) into financial capital – turning ‘oil in the ground into money in the bank’ (McLure, 1994). They also demand preferential access to resource revenues due to the public infrastructure needed for resource exploitation. The natural resource sector also requires significant investment in exploration and capital-intensive development, especially in regard to hydrocarbon and mineral resources. As well, profitable exploitation of the resource requires complementary infrastructure, particularly transportation. Additionally, resource development creates environmental and social externalities that afflict the resource-producing region most deeply. These considerations provide a case for giving sub-national governments adequate tax bases to compel resource producers to internalize costs and fund sector-specific infrastructure. Therefore, resource royalties and environmental and developmental charges represent appropriate avenues of sub-national taxation. However, to the degree that revenues from such bases are insufficient for the required public services, some preferential revenue sharing from federal revenues to the resource-rich region may be appropriate. Indeed, the federation’s cohesion may require preferential sharing of resource revenues on a derivation basis because resource-rich regions may demand a proportion of resource revenues as a condition of staying in the federation. McKenzie (2006) argues that the federation may be viewed as generating an additional surplus that can be shared between its federated entities depending on the political bargain that can be struck between its resource-rich and resourcepoor members. For resource-rich members, secession would sacrifice any additional ‘federation’ surplus, but the region would retain all of its resource revenues. Assuming that a sub-national government has this credible alternative or some other bargaining power, the outcome is biased in favour of the resource-rich region capturing a greater share of the federal surplus. Allocation of some preferential share of resource revenues may be viewed, then, as a bargaining solution where the share is the ‘bribe’ required for the region to stay in the federation. This is done by Malaysia for Sabah and Sarawak, by Canada for Newfoundland, and by Indonesia for Papua and Aceh. Despite the practical need for some bargained revenue-sharing arrangement, special autonomy or region-specific agreements lead to an unstable revenue-sharing system with inevitable jockeying by each region for the most-favoured arrangement. Where federal governments concede to preferential sharing, it is important that the arrangements follow consistent principles across the federation as was done recently in Canada. In the presence of state and local resource ownership, a second-best alternative is to have centralization of resource-rent taxes only and then redistribute them through a federal fiscalequalization programme. If politics does not permit such an arrangement and resource rent taxes 102

United Arab Emirates

Federal

Federal

United States

Emirate

State

Provincial

Federal/ provincial Federal/State

Central Central Central Federal Central

Ownership of natural resources

Source: Bishop, Qiao and Shah, 2010; Bahl and Tumennasan, 2002

Decentralized

Joint base sharing

Federal

Federal

Russia

Canada

Federal

Pakistan

Centralized revenue sharing Joint revenue sharing

Unitary Unitary Unitary Federal Unitary

Azerbaijan Norway Indonesia Nigeria China

Centralized

Country type

Country

Arrangement type

Federal and provincial separately Federal and state separately Emirate

Federal/ provincial Federal/State

Central Central Central Federal Central

Authority on revenue bases and rates

Federal and provincial separately Federal and state separately Emirate

Central Central Federal Federal Central/ provincial/ local Federal/ provincial Federal/state

Federal/Emirate

Federal and provincial separately Federal and state separately

Federal/state

Federal

Central Central Federal Federal Central/provincial

Authority on Authority on revenue collection sharing arrangement

Tax assignment and revenue sharing

Table 7.3 Resource ownership and resource revenue sharing mechanisms in selected countries

100%, Emirate collected

100%, state collected

100%, state collected 100%, state collected

98%

No No 9% >=13% 100%, subnational collected

Arrangement related to origin of revenue

46.7

1.6

3.5

30.0

0.5

24.7 12.9 10.1 35.7 –

Share of mining and quarrying in GDP

NA

46.3

57.4

29.2

24.1 32.4 12.2 28.9 69.9

Subnational share of total government expenditure

Anwar Shah

are also decentralized, then inter-state (net) fiscal equalization is in order. Such equalization is practised in Germany but this accommodation has been difficult to work out in other nations. Another contentious issue is the sales tax assignment, especially since the advent of multistage value added taxes (VATs). Assigning such taxes to the centre leads to centralized tax powers and democratic accountability deficits for state and local governments. Assignment to state governments leads to complexity in tax administration and fiscal wars among states.

Sales tax assignment in a federal economy Sales taxes are good candidates for decentralization to the states, especially if significant revenueraising responsibility is required. Nonetheless, sales taxes are centralized in some federations. In Australia, Switzerland and Pakistan, only the federal/confederal government can levy a VAT or special consumption taxes (Schmitt, 2005). In South Africa, provinces and municipalities are prohibited from levying a VAT or a general sales tax (Steytler, 2005). Typically, general sales taxes are levied on consumption goods defined with varying degrees of inclusiveness, and on a destination basis. As such, they are essentially general taxes on the residents of the taxing jurisdiction. Given the relatively low degree of mobility of households, they are likely to be much less distorting than, say, taxes on mobile bases like capital (so long as investment goods are not included in the base). As well, because general sales taxes are not significant instruments for redistribution, little is lost from an equity point of view from decentralizing them to the states. However, some distortions and administrative problems arise from state sales taxes. The main source of inefficiency stems from cross-border shopping. Residents of high-tax states shop in states with low tax rates. Given the absence of border controls in a federation, this is difficult to avoid. Hence, tax competition will likely result in rate levels and structures that do not vary greatly across jurisdictions. On the surface, this might pose a problem for poorer jurisdictions that might otherwise need higher tax rates to finance basic services. However, this disadvantage can be mitigated by equalizing transfers from the federal government, as discussed below, so that states can provide comparable public services at comparable tax rates. There are some serious administrative issues with state sales taxation, especially if there is no federal sales tax with the potential for a single tax-collection authority. If the state sales taxes take the form of sub-national credit-method VATs, the taxation of inter-state transactions creates potentially major administrative difficulties. Such transactions can either be taxed on a destination basis (taxing final consumption) such that imports are taxed and exports are zero rated, or on an origin basis (taxing production) such that imports are exempted from taxation while exports are fully taxed in the state of origin (i.e. not zero-rated). The origin basis has the advantage that cross-border sales need not be monitored as closely, but since origin-based sales taxation is analogous to taxing production, inter-state differences in tax rates – which one must expect if states are exercising independent revenue-raising authority – translate into inefficiencies in the allocation of production across states. The use of the destination basis for a VAT requires applying taxes appropriately at state borders within a federation, including crediting for taxes paid on intermediate business inputs in selling jurisdictions. In the absence of border controls, reliance must be placed on self-assessment by firms being taxed. Not only will this add to compliance costs, but also enforcement may be difficult unless auditing information flows freely across borders. Having a single federal administration will make things simpler, but the complexity involved in firms keeping track of their input credits on a state-by-state basis is significant. It could impede the free flow of goods and services within the nation. There are some possible alternatives to strict application of the destination principle but these invite further complexity in administration (see Boadway and Shah, 2009). The most common 104

Fiscal federalism and territorial finance

practice is to avoid these problems by employing single-stage sales taxes despite their inefficiencies. These include especially the fact that under a single-stage sales tax, it is difficult to avoid a cascading of the tax through purchases of taxed business inputs, especially if the tax is levied at the retail stage. Systems in which taxes paid on purchases from registered dealers are credited to or exempted from later levies reduce the incidence of this problem at some administrative cost, but do not eliminate it. A related issue is the difficulty in ensuring that sales to buyers outside state boundaries have been purged of taxes on intermediate inputs. These same problems arise at the federal level and are typically addressed by adopting a multi-stage VAT. Under a VAT, taxes on business inputs are eliminated by the system of crediting for input purchases, while under the destination principle, exports are given full credit for taxes paid and imports are fully taxed. As noted above, adopting the same remedy at the state level is difficult because of the absence of border controls. Furthermore, because states are much more open than entire countries, the administrative complexities of operating a system of taxing and crediting on all cross-border transactions would be high and would likely distort inter-state trade. For these reasons, single-stage state sales taxes may well be preferred in a developing country context, at least until viable alternatives, such as the deferred-payment method, have been tried and tested. Apart from the difficulties of dealing with cross-border transactions by both producers and consumers, separate state sales taxes entail another layer of administrative machinery for government, and additional compliance costs for businesses required to collect the tax. These costs are especially high where there are both federal and state sales taxes. In the United States, companies can face widely divergent tax regimes because sales taxes are levied on a destination basis. For example, Colorado allows local governments to establish both the base and the rates for their own sales tax. In Virginia, the state government determines both. The potentially high compliance costs led the US Supreme Court to rule that a company must have a physical presence in a state before it can be required to collect the sales tax for that state (Fox, 2007). The practical difficulties associated with sub-national administration of a multi-stage sales tax are well illustrated in Brazil where the federal government levies a manufacturer-level sales tax (IPI); states are assigned a broad-based credit-method VAT (ICMS at a 17% rate); and municipalities administer a services tax (ISS). Under the ICMS, inter-state sales are taxed on the origin principle (at a 12% rate for North–South and a 7% rate for South–North transactions), and international trade is taxed on a destination basis. Thus, relatively less developed northern states get preferential treatment in domestic trade. In international trade, as most of the imports are destined for the southern states and a disproportionate amount of exports go through the northeastern states, most of the revenues are collected by the richer states and export rebates are given by poorer states. Another emerging area of major potential inter-state conflict is the use of the ICMS as a tool for state industrial development. Some north-eastern states offer 15-year ICMS tax deferrals to industry. In a highly inflationary environment such as that of Brazil in the past, unless such tax liabilities are indexed, they can wipe out much of the ICMS tax liability. A separate source of conflict is classification of emergent IT industries such as communications and software. If they sell goods, they are subject to the state ICMS; if they are service providers, municipalities can levy the ISS. Inter-jurisdictional disputes between states and municipalities are of continuing concern (Rezende, 2007). Recognizing these difficulties, China introduced a centrally administered VAT with proceeds shared with the provinces. Effective 1 January 1994, the provinces are given 25% of VAT revenues by origin; the central government gets the remainder. India faces major difficulties in reforming its sales tax system. Sales taxes are assigned to the states. The federal government administers excise taxes, and the proceeds are shared with the states, while the octroi is a local tax on intermunicipal trade. Although these tax assignments are exclusive, in practice there are overlaps that 105

Anwar Shah

contribute to complexity. The central manufacturing excises have evolved into a central VAT (CENVAT) at the manufacturing stage. Thus, production is taxed by the centre, while the sale of the goods is taxed by the states, essentially sharing the same base (Rao, 2007). Sales taxes are administered on narrow bases: the number of rates varies by state from six in Orissa to 17 in Bihar and Gujarat. Some states consider the sales tax an important element of redistributive policy. To reform the sales tax, a broad-based national VAT has been proposed, but this is strongly opposed by the states. The states are also dissatisfied with the centrally administered excise tax because it limits their tax powers. The federal government prefers to raise additional revenues from administered prices rather than from excises because the proceeds from excises have to be shared with state governments. The octroi tax on inter-municipal trade is a source of significant revenues for local governments and remains popular in spite of its anti-trade bias. Other inefficiencies of state sales taxation stem from administrative problems often present in developing countries. One is that the broader the base, the more difficult it is to enforce compliance. To get a fully general consumption base, including both goods and services, it is practically necessary to collect the tax at the level of final sale to the consumer, the retail stage, under either a single-stage or a VAT system. This increases the compliance costs considerably because the number of taxpaying firms can be extremely large. In practice, exempting small firms mitigates this problem. However, in developing countries, a reasonable exemption could encompass a large number of firms, narrowing the tax base considerably. Furthermore, enforcement of the tax becomes very difficult if it is applied at the retail level where evasion is likely to be high. Some of these difficulties can be avoided by levying the tax at an earlier stage, though again at the cost of making the base much narrower.

State and local tax autonomy and fiscal wars State or local governments may inefficiently compete down taxes for fear of loss of tax base. Tax rates on mobile factors may be set inefficiently low due to strategic tax competition. If this practice becomes pervasive, then firms will reap the benefits of such incentives regardless of their location decisions. Thus, for local governments, these incentives lead to self-defeating outcomes (i.e. reduced revenues without attracting new capital). In Brazil, use of ICMS (origin-based) to attract capital from other regions has generated interstate conflict. Despite the fact that the National Council on Fiscal Policy (CONFAZ with state finance ministers as members) sought to harmonize ICMS bases and rates, some of the tax concessions refused by the council are practised by many states anyway. States can also resort to tax-base reductions or grant un-indexed payment deferrals (Longo, 1994). In Pakistan (until 1998) and in India, state and local governments imposed taxes on interjurisdictional trade in order to protect local industry and limit internal trade. In India, manufacturing states impose higher taxation on goods intended for internal trade so as to pass the tax burden onto non-residents. Some states allow local urban governments to tax goods entering their jurisdiction. The result is a series of tariff zones within India (Rao, 2007). In Brazil, industrial states offered tax deferrals for extended periods on the state VAT. State governments often indulge in ‘guerra fiscal’ (fiscal war) by which they attempt to shift local tax burdens to non-residents (see Salomao, 2000). Location-specific tax holidays are offered by sub-national governments in a large number of countries. These incentives attract fly-by-night industries. Sub-national governments in India, Pakistan, Brazil, Malaysia, Mexico, the Philippines and others use taxes actively for industrial policy. In the United States, state and local governments offer concessions on a range of taxes (e.g. property, sales, corporate income and personal income) so as to attract industry. These incentives can be targeted to an individual firm or offered to all firms with particular 106

Fiscal federalism and territorial finance

characteristics (Fox, 2007). Firm-based incentives become a source of perpetual negotiations for footloose service industries. For example, the Marriott corporation has repeatedly won greater concessions from Maryland with threats to move its headquarters across the Potomac River to Virginia. Even in China, with a tax system harmonized by unitary tax laws, local jurisdictions offer incentives to attract investment. Richer jurisdictions, easily capable of meeting their assigned ‘tax revenue task’, can waive local taxes or refund the local portion of shared taxes to enterprises (Dong, 2007). There may be opportunities to export taxes levied on products and services used by nonresidents. This is especially true for taxes on business incomes and natural resources. This practice is common in most federations, but there are limits to such a strategy due to mobility of factors in the long run and by demand responses.

Fiscal equalization – ensuring a fair suck of the sauce bottle Economic rationales for equalization payments in a federal economy are subject to debate. Proponents argue that equalization is needed to ensure efficiency and regional equity. The efficiency case is based on differential fiscal capacity and associated tax burdens leading to misallocation of resources as capital and labour may move because of the fiscal surplus in resourcerich regions rather than in response to economic returns on capital or labour. The equity case is based on ensuring that citizens with the same income are treated alike regardless of their residence. Opponents argue that if fiscal differentials are fully capitalized in property values, then there is neither an efficiency nor an equity case for equalization as residents in rich jurisdictions pay less for public services but more for private services, especially for land and property, and vice versa for poorer jurisdiction. This explanation is among those advanced for having no federal equalization in the United States (see Oates, 1972; Kenyon and Kincaid 1996). Proponents have won the day politically; hence, most federal countries have fiscal equalization. Even in the United States, state transfers to local governments have strong equalization features. All equalization programmes are concerned with inter-jurisdictional equity or horizontal fiscal equity, not interpersonal (vertical) equity. Which order of government finances and administers an equalization programme is determined either by the constitution (as in Canada, Germany and Switzerland), or by the legislature (as in Australia). Paternal programmes, in which higher-level governments finance equalization for lower levels, are common (e.g. Australia and Canada). Fraternal or Robin Hood-type programmes, in which governments at the same level establish a common pool, to which richer jurisdictions contribute and the poorer jurisdictions draw, are rare (exceptions include Germany at the Länder level and Finland, Sweden and Denmark at the local level). Robin Hood programmes are preferred, as they represent an open political compromise balancing the interests of the union and the contributing jurisdictions, as done by the Solidarity Pact II in Germany. Such programmes foster national unity, as poorer jurisdictions clearly see the contributions made for their well-being by residents of other jurisdictions. Paternal programmes lack the discipline of fraternal programmes because unless enshrined in the constitution (as in Canada), they are guided largely by national politics and the budget of the federal and state/provincial (for local equalization) governments. Some countries combine fraternal and paternal components (see Table 7.4). In Switzerland, effective 2007, the federal government finances two-thirds of the programme; the remaining third is financed by the rich cantons. The programme has a fiscal capacity equalization component based on factor income, with 59% of the financing from the federal government and 41% from rich cantons. The cost-equalization component is financed solely by the federal government. 107

Anwar Shah

Germany’s equalization programme has a small supplement financed by the federal government. In Denmark, equalization at the local level uses the Robin Hood approach for both fiscal capacity and fiscal need equalization for counties (using 85% national average standard) and large cities (90% of the national average standard for fiscal capacity and 60% of the national average standard for fiscal need); for smaller municipalities, it uses the paternal approach for fiscal capacity equalization (using 50% of national average as the standard of equalization) and the Robin Hood approach for fiscal need equalization (using 35% of the national average as the equalization standard). Fiscal equalization programmes differ in terms of how the total pool of resources devoted to such programmes is determined. In Canada and Germany, both the total pool and its allocation to provinces/states are formula driven. Under the Australian and Swiss programmes, the total pool is arbitrarily determined by the federal government through an act of parliament – total Table 7.4 Features of fiscal equalization transfers in selected countries Feature

Australia

Canada

Germany

Objective

Build capacity to provide services at same standard with same revenue effort and same operational efficiency Federal law Federal parliament

Achieve reasonably comparable levels of public services at reasonably comparable levels of taxation across provinces Constitution Federal parliament

Paternal

Paternal

Equalize differences Provide minimum in financial capacity acceptable levels of of states certain public services without much heavier tax burdens in some cantons than others Constitution Constitution Federal parliament, Federal parliament initiated by the upper house (Bundesrat) Fraternal Mixed

Ad hoc

Formula

Formula

Ad hoc

No

Yes

Yes

No

Formula Yes, representative tax system Yes

Formula Yes, representative tax system No

Formula Yes, major macro tax bases Some

High

Low

Formula Yes, actual revenues No (only population size and density) Low

Medium

No (?) Independent agency No Supreme Court

Yes (?) Intergovernmental committees Yes, five years Supreme Court

Yes (?) Solidarity Pact II

Yes Federal government

Legal status Legislation

Paternal or fraternal Total pool determination Equalization standard determines pool and allocation Allocation Fiscal capacity equalization Fiscal need equalization Programme complexity Political consensus Who recommends Sunset clause Dispute resolution

Source: Boadway and Shah, 2009

108

Switzerland

No No Constitutional Court Supreme Court

Fiscal federalism and territorial finance

proceeds of the general sales tax in Australia and an arbitrarily determined level of funding from the federal government and rich cantons in Switzerland. The method of equalization also differs across programmes. Australia, Canada and Germany equalize per capita fiscal capacity using the representative tax system. Switzerland uses macro tax bases. It devotes 19% of equalization financing to cost equalization using eight factors: population size, area, population density, population older than 80, number of large cities, number of foreign adults resident for more than 10 years, unemployment, and number of people requesting social assistance from the canton. In Germany, actual rather than potential revenues are used in these calculations because both actual and potential revenues are the same due to the uniformity of state tax bases and tax rates through federal legislation. Germany makes simple expenditureneed adjustments based on population size, density and whether a city has a harbour. China uses potential revenues, although they equal actual revenues when there is uniformity of tax bases and tax rates as mandated by central government legislation. Canada’s programme does not include fiscal need compensation. Australia uses a comprehensive equalization programme, equalizing fiscal capacity as well as need for all state expenditures. Unlike Canada, Australia implicitly equalizes the above-average states down (by adjusting their transfers) as well as the below-average states up. In that sense, it is equivalent to a ‘net’ system. The introduction of expenditure needs compensation introduces complexity and controversy and dilutes political consensus. As a result, Australia’s programme is the most complex and controversial of all programmes and has garnered the least political consensus. Most equalization programmes are introduced as permanent; an exception is Canada where there is a sunset clause for quinquennial review and renewal by the national parliament. Such a clause is helpful in providing regular evaluation and fine-tuning. Almost all programmes in mature federations specify formal mechanisms for resolving disputes over the working of these transfer programmes. Overall, the experience of mature federations suggests that in the interests of simplicity, transparency and accountability, it would be better for such programmes to focus only on fiscal capacity equalization to an explicit standard that determines the total pool as well as the allocation among recipient units. Fiscal need compensation is best dealt with through specific-purpose transfers for merit goods, as is done in most industrial countries. Most transition economies have equalization components in their grant programmes for sub-national governments. China, Latvia, Lithuania, Poland, Romania, the Russian Federation and Ukraine have adopted transfer formulae that explicitly incorporate either fiscal capacity and/or expenditure need equalization concerns. For local fiscal equalization, these countries nevertheless use one-size-fits-all approaches to diverse forms of local government, creating equity concerns. Except for Indonesia, developing countries have not implemented programmes using explicit equalization standards, although equalization objectives are implicitly attempted in the general revenue-sharing mechanisms used in Argentina, Brazil, Colombia, India, Nigeria, Mexico, Pakistan and South Africa. These mechanisms typically combine diverse and conflicting objectives in the same formula and fall significantly short on individual objectives. Because the formulae lack explicit equalization standards, they fail to address regional equity objectives satisfactorily. Even in the Indonesian programme, the equalization standard contains complex statistical criteria having to do with regional variation in fiscal gap allocations, and the total pool is not determined by an explicit equalization standard; instead, the equalization standard is implicitly determined by the ad hoc determination of total funds available for equalization. In all federal countries, equalization payments are subject to much controversy and debate. In Australia, the richer states (i.e. New South Wales and Victoria) want greater fiscal autonomy and 109

Anwar Shah

lower equalization payments, whereas the majority of states prefer centralization of tax authority and receiving revenues through revenue sharing and equalization. As a result, a high degree of tax centralization is accompanied by substantial equalization transfers. Although Canada avoided tax centralization, fiscal equalization still invites strong political support from relatively poorer provinces, and it was an important element of the political compact to keep Quebec from seceding from the federation. Academic scholars have referred to it as the ‘glue’ that holds the federation together (Boadway, 1992). Thus, most federal jurisdictions ensure continuation of their programme against complaints by richer provinces.

Institutional arrangements for fiscal equalization payments Who should be responsible for designing such federal-state fiscal relations? There are various possibilities. The most obvious is to make the federal government solely responsible because it is responsible for the national objectives to be delivered through the fiscal arrangements. In many countries, this is the norm. A problem with it might be the natural tendency for the federal government to be too involved with state decision making and not allow the full benefits of decentralization to occur. To some extent, this can be overcome by imposing constitutional restrictions on the ability of the federal government to override state decisions. Alternatively, one can have a separate body involved in the design and ongoing reform and enforcement of fiscal arrangements. It can be an impartial body or a body comprising both federal and state representatives. It can have true decision-making authority or be advisory. In any case, it at least needs to be able to coordinate decision making at the two orders of government. For determining the system of grants, one finds four types used in practice. The most commonly used practice is for the federal government to decide on it. This has the disadvantage of biasing the system toward a centralized outcome, whereas the grants are intended to facilitate decentralized decision making. In India, the federal government is solely responsible for the Planning Commission transfers and the centrally sponsored schemes. These transfers have strong input conditionality with potential to undermine state and local autonomy. Brazil’s 1988 constitution provided strong safeguards against federal intrusion by enshrining the transfers’ formulae factors in the constitution. These safeguards represent an extreme step as they undermine the flexibility of fiscal arrangements to respond to changing economic circumstances. The second approach is to set up a quasi-independent body, such as a grants commission, the purpose of which is to design and reform the system. These commissions can have a permanent presence as in South Africa and Australia, or they can be convened periodically to make recommendations for the next five years as done in India. These commissions have proven to be ineffective in some countries largely because many of the recommendations have been ignored by the government as in South Africa. In other cases, while the government may have accepted and implemented the recommendations, they have been ineffective in reforming the system due to the constraints they have imposed on themselves as in India. In some cases, these commissions become too academic, thereby creating an overly complex system of intergovernmental transfers as with the Commonwealth Grants Commission in Australia (Shah, 2007). The third approach is to use executive federalism or federal-provincial committees to negotiate the system’s terms as in Canada and Germany. Germany’s system is enhanced by having state governments represented in the Bundesrat. This allows for explicit political input from the jurisdictions involved and attempts to develop a consensus. The fourth approach is a variation on the third approach. It uses an intergovernmental-cumlegislative-cum-civil society committee with equal representation from all constituent units but chaired by the federal government to negotiate changes in the arrangements. The so-called Finance 110

Fiscal federalism and territorial finance

Commission in Pakistan represents this model. This approach has the advantage that all stakeholders – donors, recipients, civil society and experts – are represented on the commission. This keeps the system simple and transparent. An important disadvantage of this approach is that due to the unanimity rule, such bodies may be deadlocked for long periods as happened in Pakistan during the 1990s.

Territorial fiscal autonomy and implications for fiscal management A compound federal republic is commonly perceived to be prone to fiscal populism leading to unsustainable fiscal policies. This is particularly true when there is a separation of executive and legislative functions as under the presidential form of government in the United States. Electoral pressures result in short time horizons for politicians and encourage profligate spending that threatens macro stability. Political institutions typically express systematic bias toward large and visible projects, although the social marginal costs of such projects may be much higher than their social marginal benefits. Public debt is often used to bind future governments to current policy choices, thus leading to a large, unsustainable debt overhang. These political imperatives are accentuated in a compound republic because such a republic faces agency problems with incomplete contracts as taxing, spending and regulatory functions, especially for shared rule, remain undefined. Intergovernmental bargaining on fiscal coordination leads to higher transactions costs. The failure of collective action to enforce fiscal discipline at the national level arises from the ‘tragedy of the commons’ or ‘norm of universalism’ or ‘pork-barrel politics’, but these problems are not unique to federal systems. Legislators in both federal and unitary countries, in their attempt to avoid deadlocks, trade votes and support each other’s projects by implicitly agreeing that ‘I’ll favor your best project if you favor mine’ (Inman and Rubinfeld, 1991). Such behaviour leads to national overspending and higher debt. It also leads to regionally differentiated bases for federal corporate income taxation and, thereby, loss of federal revenues through these tax expenditures. Such tax expenditures accentuate national fiscal deficits. During the first 140 years of US history, the negative impact of ‘universalism’ was kept to a minimum by two fiscal rules: the Constitution formally constrained the federal spending power to narrowly defined areas and an informal rule was followed to the effect that the federal government could only borrow to fight recession or wars (Niskanen, 1992). The Great Depression and the New Deal led to abandonment of these fiscal rules. Inman and Fitts (1990) provide empirical evidence supporting the working of ‘universalism’ in post-New Deal eras. To overcome difficulties noted above with national fiscal policy, solutions proposed include ‘gate-keeper’ committees (Weingast and Marshall, 1988; Eichengreen et al., 1999); party discipline within legislatures (Cremer, 1986); constitutionally imposed or legislated fiscal rules (Niskanen, 1992; Poterba and von Hagen, 1999); executive agenda setting (Ingberman and Yao, 1991); market discipline (Lane, 1993); and decentralization when potential inefficiencies of national government democratic choice outweigh economic gains from centralization. Given that the potential for errant fiscal behaviour by national and sub-national governments can complicate the conduct of fiscal policy, what institutions are necessary to guard against such an eventuality? As discussed below, mature federations emphasize intergovernmental coordination through executive or legislative federalism as well as fiscal rules to achieve a synergy among policies at different levels. In unitary countries, the emphasis traditionally has been on using centralization or direct central controls. These controls typically have failed to achieve a coordinated response due to intergovernmental gaming (Shah, 2007). Moreover, the national government completely escapes any scrutiny except when it seeks international help from external sources such as the International Monetary Fund (IMF). However, external help creates a moral 111

Anwar Shah

hazard in that it creates bureaucratic incentives on both sides to ensure that such assistance is always in demand and utilized. An outstanding example of such moral hazard comes from Pakistan; the country has been under an IMF structural adjustment programme for more than 50 of its 60-plus years of existence. In mature federations, fiscal policy coordination is exercised through both executive and legislative federalism as well as formal and informal fiscal rules. In recent years, legislated fiscal rules have come to command greater attention in both federal and unitary countries. These rules take the form of budget balance controls, debt restrictions, tax or expenditure controls, and referenda for new taxes and spending. For example, the EU, in its goal of creating a monetary union through the provisions of the 1992 Maastricht Treaty, established ceilings on national deficits (< 3% of gross domestic product – GDP) and debts (< 60% of GDP), and supporting provisions that there should be no bailout of any government by member central banks or by the European Central Bank. The EU is also prohibited from unconditionally guaranteeing the public debt of a member state (Pisani-Ferry, 1991). These provisions were subsequently strengthened by the Growth and Stability Pact (GSP) 1997 (legislated fiscal rules adopted by the European Parliament). The GSP established medium-term objectives of budgets being in balance or in surplus and associated monitoring mechanisms. It provided for remedial action within 10 months or face a mandatory non-interest-bearing deposit of 0.5% of GDP – refundable if excessive deficit is eliminated within two years. These guidelines were violated by France and Germany, but the EU’s Council of the Economics and Finance Ministers (ECOFIN) held the European Commission’s adverse ruling against the two countries in abeyance. Most of old Europe is not in compliance. Expectedly, these guidelines have been relaxed in the face of financial crises in Portugal, Ireland, Italy, Greece and Spain (PIIGS). The EU has announced a revamped GSP with sanctions (0.2% of GDP) and a Euro Plus Pact for binding fiscal rules. It has also specified excessive imbalance and permanent crisis mechanism procedures under a European Stability Mechanism and has set up an Enhanced European Stabilization Facility to assist member states. Needless to say, these mechanisms have already proven ineffective in dealing with Europe’s financial and fiscal crises. The EU Central Bank has been active in bailing out PIIGS. An overall conclusion from the EU experience is that the convergence criteria proved helpful but could not discipline large members. It should be noted, however, that in the EU (Wierts, 2005), sub-national governments’ contributions to consolidated public-sector deficits and debts were relatively smaller as compared to the central governments in most EU countries – federal and unitary. Most mature federations also specify no bailout provisions in setting up central banks, with the notable exceptions of Australia until 1992 and Brazil until 1996. In the presence of an explicit or even implicit bailout guarantee and preferential loans from the banking sector, the printing of money by sub-national governments is possible, thereby fuelling inflation. EU guidelines provide a useful framework for macro coordination in federal systems, but such guidelines may not ensure macro stability as the guidelines may restrain smaller jurisdictions with little influence on macro stability but may not restrain heavyweights. Proper enforcement of guidelines may require a fiscal coordinating council. Recent experiences with fiscal adjustment programmes suggest that while legislated fiscal rules are neither necessary nor sufficient for successful fiscal adjustment, they can help forge sustained political commitment to achieve better fiscal outcomes, especially in countries with divisive political institutions or coalition regimes. For example, such rules can help sustain political commitment to reform in countries with proportional representation (Brazil) or multi-party coalition governments (India) or in countries with separate legislative and executive functions (United States and Brazil). Fiscal rules in such countries can help restrain pork-barrel politics and, thereby, improve fiscal discipline. Von Hagen (2005) concluded 112

Fiscal federalism and territorial finance

that budgetary institutions matter more than fiscal rules. The EU fiscal rules may have encouraged European countries to strengthen budgetary institutions, which, in turn, had welcome effects on fiscal discipline and fiscal outcomes in the new Europe, but old Europe seems not to have paid attention to these rules, as revealed by the recent financial and fiscal crises.

Concluding remarks Most federal countries do not conform to an ideal form of multi-order governance – a compound republic of civic republics and a centre – with independent self-financing governments with complete contracts with citizens as to their mandates. Such an ideal system is more in tune with democratic accountability and also the needs of an information age requiring fewer orders but smaller and leaner governments that are closer to the people. Instead, most federal countries have opted for a compound republic of states and a centre where the centre typically has access to revenue means that exceed its expenditure needs, and vice versa for states and their local governments. In such a setting, territorial finance assumes critical importance in the federal compact. Division of a fixed fiscal pie creates incentives for universalism and pork-barrel politics to lay claim to a greater share of this pie by various orders of government. This encourages Leviathan tendencies as politicians and bureaucrats race to create bureaucratic empires. Institutions of accountability become accommodating ‘team players’ to facilitate fiscal profligacy rather than restraint as government agents have incomplete contracts and citizens have limited access to information to hold them to account. Further, the transactioncosts for citizens to stop this runaway train becomes prohibitive due to weak countervailing institutions. Needless to say, mature federations have worried about these problems and have tried to circumvent these with various rules and restraints. Typically, they do better than centralized unitary republics but worse than decentralized unitary republics that have tried to mimic a model of the compound republic of independent civic republics and a centre as done by the Scandinavian countries.

References Bahl, Roy and Bayar Tumennasan (2002) ‘How Should Revenues from Natural Resources be Shared in Indonesia’, Andrew Young School of Policy Studies Working paper no. 02–24. Bishop, Grant, Baoyun Qiao and Anwar Shah (2010) Sharing of Revenues from Non-Renewable Natural Resources, Working Paper, World Bank Institute, Washington, DC. Boadway, Robin (1992) The Constitutional Division of Powers: An Economic Perspective (Ottawa: Economic Council of Canada). Boadway, Robin and Anwar Shah (2009) Fiscal Federalism: Principles and Practice of Multi-order Governance (New York: Cambridge University Press). Breton, Albert (1995) Competitive Governments (Cambridge: Cambridge University Press). ——(2006) ‘Modeling Vertical Competition’, in Ehtisham Ahmad and Giorgio Brosio (eds) Handbook of Fiscal Federalism (Cheltenham, UK and Northampton, MA, USA: Edward Elgar), 86–105. Buchanan, James (1965) ‘An Economic Theory of Clubs’, Economina Vol. 32, No. XX: 1–14. Buchanan, James M. (1997) Post-Socialist Political Economy: Selected Essays (Cheltenham, UK and Lyme, NH, USA: E. Elgar). Cremer, Jacques (1986) ‘Cooperation in Ongoing Organizations’, Quarterly Journal of Economics 101 (1): 33–49. Dollery, Brian and Joe Wallis (2001) The Political Economy of Local Government (Cheltenham: Edward Elgar). Dong, Qiu (2007) ‘China Fiscal Federalism’, Working Paper, World Bank Institute, Washington, DC. Eichengreen, Barry J., Ricardo Hausman and J. von Hagen (1999) ‘Reforming Budgetary Institutions in Latin America: The Case for a National Finance Council’, Open Economies Review 10(4): 415–42. Fox, William (2007) ‘Fiscal Federalism in the US: A Structure in Continuing Transition’, in Anwar Shah (ed.) The Practice of Fiscal Federalism (Montreal and Kingston: McGill-Queen’s University Press), 349–69.

113

Anwar Shah

Frug, Gerald E. and David J. Barron (2008) City Bound: How States Stifle Urban Innovation (New York: Cornell University Press). Hartwick, John M. (1978) ‘Investing Returns from Depleting Renewable Resource Stocks and Intergenerational Equity’, Economics Letters Vol. 1(1): 85–88. Horn, Murray (1997) The Political Economy of Public Administration (Cambridge: Cambridge University Press). Iaryckzower, M., S. Saiegh and M. Tommasi (2000) ‘Coming Together: The Industrial Organization of Federalism’, Paper Presented at the 2000 Meeting of the Latin American Studies Association, Miami, 16–18 March. Ingberman, D. and D. Yao (1991) ‘Presidential Commitment and the Veto’, American Journal of Political Science 35: 357–89. Inman, Robert (2006) ‘Why Federalism?’ Wharton School, University of Pennsylvania, Philadelphia, unpublished paper. Inman, Robert and M.A. Fitts (1990) ‘Political Institutions and Fiscal Policy: Evidence from the US Historical Record’, Journal of Law, Economics and Organization 6: 79–132. Inman, Robert P. and Daniel L. Rubinfeld (1991) ‘Fiscal Federalism in Europe: Lessons from the United States Experience’, National Bureau of Economic Research Working Paper Series (US), No. 3941, December. ——(1997) ‘The Political Economy of Federalism’, in Dennis C. Mueller (ed.) Perspectives in Public Choice: A Handbook (Cambridge: Cambridge University Press), 73–105. Kenyon, Daphne A. and John Kincaid (eds) (1991) Competition Among States and Local Governments: Efficiency and Equity in American Federalism (Washington, DC: Urban Institute Press). ——(1996) ‘Fiscal Federalism in the United States: The Reluctance to Equalize Jurisdictions’, in Werner W. Pommerehne and George Ress (eds) Finanzverfassung im Spannungsfeld zwischen Zentralstaat und Gliedstaaten (Baden-Baden: Nomos Verlagsgesellschaft), 34–56. Kincaid, John (1967) Municipal Perspectives in Federalism, unpublished paper, cited in Ann O. Bowman and Robert C. Kearney, State and Local Government (Boston, MA: Houghton Mifflin Co., 1990). ——(1990) ‘From Cooperative to Coercive Federalism’, Annals of the American Academy of Political and Social Science Vol. 509 (May): 139–52. Lane, Timothy (1993) ‘Market Discipline’, Staff Papers, International Monetary Fund 40 (March): 53–88. Longo, Carlos Alberto (1994) ‘Federal Problems with VAT in Brazil’, Revista Brasileira de Economia 48: 85–105. McKenzie, K.J. (2006) ‘Fiscal Federalism and the Taxation of Nonrenewable Resources’, in R. Bird and F. Vaillancourt (eds) Perspectives on Fiscal Federalism (WBI Learning Resources Series, World Bank Institute). McLure, Charles E., Jr. (1994) ‘The Sharing of Taxes on Natural Resources and the Future of the Russian Federation’, in Christine Wallich (ed.) Russia and the Challenge of Fiscal Federalism (Washington: The World Bank), 181–217. Niskanen, William A. (1992) ‘Case for a New Fiscal Constitution’, Journal of Economic Perspectives (US) Vol. 6 (Spring): 13–24. Oates, Wallace (1972) Fiscal Federalism (New York: Harcourt Brace Jovanovich). ——(1999) ‘An Essay in Fiscal Federalism’, Journal of Economic Literature 37 (3): 1120–59. Pisani-Ferry, J. (1991) ‘Maintaining a Coherent Macro-economic Policy in a Highly Decentralized Federal State: The Experience of the EC’, OECD Seminar on Fiscal Federalism in Economies in Transition, 2–3 April. Poterba, James and Jurgen von Hagen (eds) (1999) Fiscal Institutions and Fiscal Performance (Chicago: University of Chicago Press). Qiao, Baoyun and Anwar Shah (2006) ‘Local Government Organization and Finance in China’, in Anwar Shah (ed.) Local Governance in Developing Countries (Washington, DC: World Bank), 137–68. Rao, M.G. (2007) ‘Fiscal Federalism in India: Emerging Challenges’, in A. Shah (ed.) The Practice of Fiscal Federalism (Montreal and Kingston: McGill-Queen’s Press), 151–77. Rezende, F. (2007) ‘Fiscal Federalism in the Brazilian Federation’, in A. Shah (ed.) The Practice of Fiscal Federalism (Montreal and Kingston: McGill-Queen’s Press), 73–97. Riker, William (1964) Federalism: Origin, Operation, Significance (Boston, MA: Little, Brown). Salmon, Pierre (2006) ‘Horizontal Competition Among Governments’, in Ehtisham Ahmad and Giorgio Brosio (eds) Handbook of Fiscal Federalism (Cheltenham, UK and Northampton, MA, USA: Edward Elgar), 61–85. Salomao, Miguel (2000) ‘Fiscal Federalism and Tax Competition in Brazil’, paper presented at the First Conference in Cooperative Federalism, 9–11 May, Brasilia, DF. 114

Fiscal federalism and territorial finance

Schmitt, N. (2005) ‘Swiss Confederation’, in J. Kincaid and G. Alan Tarr (eds) Constitutional Origins, Structure and Change in Federal Countries (Montreal and Kingston: McGill-Queen’s University Press), 347–80. Seabright, Paul (1996) ‘Accountability and Decentralization in Government: An Incomplete Contract Model’, European Economic Review 40(1) (January): 61–89. Shah, Anwar (2005) ‘On Getting the Giant to Kneel: Approaches to a Change in the Bureaucratic Culture’, in Anwar Shah (ed.) Fiscal Management (Washington, DC: World Bank), 211–29. ——(2006) ‘Fiscal Decentralization and Macroeconomic Management’, International Tax and Public Finance 13 (4): 437–62. ——(2007) ‘A Practitioner’s Guide to Intergovernmental Fiscal Transfers’, in Robin Boadway and Anwar Shah (eds) Intergovernmental Fiscal Transfers (Washington, DC: World Bank). ——(2007) ‘Institutional Arrangements for Intergovernmental Fiscal Transfers and A Framework for Evaluation’, in Robin Boadway and Anwar Shah (eds) Intergovernmental Fiscal Transfers (Washington, DC: World Bank), 293–317. ——(ed.) (2007) The Practice of Fiscal Federalism: Comparative Perspectives (Montreal and London: McGill-Queen’s University Press). ——(2010) ‘Adapting to a Changing World: Reflections on the Reform of Local Governance for the Next Decade’, in Antti Moisio (ed.) Local Public Sector in Transition: A Nordic Perspective (Helsinki: Prime Minister’s Office, Finland), 185–220. Shah, Anwar and Furhawn Shah (2007) ‘Citizen-Centric Local Governance: Strategies to Combat Democratic Deficit’, Development 50: 72–80. Shah, Anwar and Sana Shah (2006) ‘The New Vision of Local Governance and Evolving Role of Local Governments’, in Anwar Shah (ed.) Local Governance in Developing Countries (Washington, DC: World Bank), 1–44. Steytler, N. (2005) ‘Republic of South Africa’, in J. Kincaid and G. Alan Tarr (eds) Constitutional Origins, Structure and Change in Federal Countries (Montreal and Kingston: McGill-Queen’s University Press), 311–46. Tiebout, Charles (1956) ‘A Pure Theory of Local Expenditures’, Journal of Political Economy 64(5): 416–24. Von Hagen, Jurgen (2005) ‘Budgetary Institutions and Public Spending’, in Anwar Shah (ed.) Fiscal Management (Washington, DC: World Bank), 1–29. Weingast, Barry (1995) ‘The Economic Role of Political Institutions: Market Preserving Federalism and Economic Growth’, Journal of Law, Economics, and Organization 11(1): 1–31. ——(2006) ‘Second Generation Fiscal Federalism: Implications for Decentralized Democratic Governance and Economic Development’, discussion draft, Hoover Institution, Stanford University. Weingast, Barry and W. Marshall (1988) ‘The Industrial Organization of Congress: Why Legislatures, Like Firms, are not Organized Like Markets’, Journal of Political Economy 96: 132–63. Wierts, Peter (2005) ‘Federalism and the EU: Fiscal Policy Coordination within and between EU Member States’, International Conference on Federalism, Brussels, 1–5 March. Williamson, Oliver (1985) ‘The Economic Institutions of Capitalism’ (New York: Free Press).

115

8 Federalism and public policy Do federalism, regionalism and hybridity make any difference? Evidence from environmental policy Sonja Wälti

Introduction Comparative federalism traditionally has focused on understanding and comparing how various multi-tiered contexts operate institutionally and constitutionally. In recent years, federal theorists and policy scholars have become increasingly interested in how federal, decentralized, regionalized, or hybrid political systems solve policy problems. What instruments do they apply to tackle problems? Are they better at tackling some than others? Are they more or less able to govern the environment, healthcare, or fiscal matters than hybrid or unitary countries? As the federal-unitary distinction became more permeable, these questions become not only more interesting but also more relevant. Yet, the lack of predictive ‘federal theory’ gets in the way of generating hypotheses regarding the policy performance of various multi-tiered arrangements. Actor-centred approaches, such as multilevel governance, have helped to overcome some of these struggles by offering a generic lens on federal, decentralized and regionalized polities, and by drawing our attention to governmental and non-governmental actors within the framework of formal and informal multilevel institutions. Environmental policy research provides fertile ground to revisit the study of federalism and decentralization with this new insight in mind and thereby refine comparative federalism and its predictive capabilities when it comes to assessing the policy performance of multi-tiered contexts. Inspired by actor-centred institutionalism, multilevel governance and veto point theory, it suggests examining multi-tiered polities in terms of offering multiple venues to various categories of actors. Venues can serve as both (pro-active) access and (reactive) veto points, and their availability may vary from policy to policy and from policy stage to policy stage. In the first three sections, I outline how constitutional federalism, economic federalism and intergovernmental relations and management have contributed to explaining policy performance. Thereafter, I explain why and how an actor-centred approach to comparative federalism has helped to bring additional insight into explaining policy outputs and outcomes in multi-tiered systems. Lastly, I conclude about expected contributions and further research. 116

Federalism and public policy

Constitutional federalism Comparative federalism traditionally characterized multilevel governance arrangements from a constitutional point of view by examining the distribution of powers, resources and organization across levels of government (e.g. Watts, 1999). While comparative federalism has not traditionally focused on policy performance, some theoretical claims can nevertheless be derived. Elazar’s federal principle of combined self-rule and shared rule (see Elazar, 1973) meets particular challenges when dealing with problems of varying territorial scales. For example, the primary target groups of environmental policies that aim to provide national public goods – producers and consumers – are local. They inevitably become the dual subjects of the federation and the constituent units. Many such policies require shared authority but are in turn subject to performance impediments stemming from shared authority. Legal and administrative approaches to federalism have pointed to significant variance in the allocation of powers, resources and organizational means across levels of government that may explain policy performance. These institutional differences affect how well they cope with problems (see Rabe and Lowry, 1999, on environmental policies). In dual (also called ‘competitive’) federal arrangements, which characterize namely AngloSaxon federations, national and sub-national jurisdictions see their authority contained in separate ‘watertight compartments’. Dual multilevel arrangements have the advantage of being ‘direct’: once decided at one level of government, policies can also be implemented by that level (Salamon, 2002: 29). When that is the case, implementation is swift and in line with relevant policy goals. However, with many environmental problems not contained within jurisdictional boundaries (Oates and Portney, 2001), the dual nature of federal polities is but an ideal-type. In reality, different governmental levels often enjoy competing authority over environmental policy matters. Thus, the downside of dual arrangements is the potential for conflict, stalemate and inaction, as is arguably the case of Canada’s environmental policies (Rabe and Lowry, 1999; Weibust, 2009). The involved governmental levels can conveniently shift the blame for mounting problems and, unless the problem or political pressure is unusually high, neither level of government may initiate centralization to resolve problems. Cooperative federal patterns, such as the ones that predominate in Germany and Switzerland, are often thought to experience difficulties in swiftly adopting central policies and thus solve national public goods problems. Unlike dual federal systems, even when they are ready, they cannot pre-empt sub-national-level powers but instead have to compromise on joint policies. Once adopted, policies may experience multiple barriers to implementation in cooperative federal arrangements due to their tendency to delegate policy implementation to sub-national agencies. Implementation research has identified this aspect of bureaucratic discretion as the ‘principal-agent problem’, denoting the difficulty the principal has to oversee its agent when chains of command are long and fragmented. Implementation may take longer and be spottier due to the multiple opportunities for polices to be thwarted to suit regional and local interests. By contrast, delegated implementation is also said to make the bureaucracy responsive and accountable to local concerns and thus improve policy outcomes. This accountability effect is likely more pronounced in the case of particularly visible (including many environmental) policies, meaning policies with outputs and/or outcomes that can be immediately monitored by the public (Wilson, 1989). While traditional comparative federalism, by stressing constitutional (and thus legal and organizational) aspects, has limited itself to the study of federal countries, the insight gained can be extended to multilevel polities in general. By emphasizing a federal system’s regulatory regime rather than its polity (Kelemen, 2004: 160), any multilevel system appears to function with separate or concurrent powers, resources can be more or less decentralized, and organizational capabilities are conceivably more or less developed. The flexibility and adaptability of multilevel 117

Sonja Wälti

regimes makes them promising candidates to respond swiftly and effectively to problems characterized by natural rather than institutional boundaries. Multilevel regimes may also find it easier to cope with the ever-growing complexity of supra-national governance levels. When focusing on constitutional and legal arrangements, a more serious limitation in explaining policy performance is the inability to capture the full complexity of the policy making process. The focus of constitutional federalism is typically on ‘high politics’ (Wright, 1990). When paid attention to at all, policy outcomes are depicted largely as the result of an historic struggle over authority and are seen as correlated with the distribution of organizational, administrative and fiscal competencies and capacities.

Economic federalism When asking how policies are or should be governed across levels of government, economic – also referred to as functional or fiscal – theory of federalism (see Peterson, 1995) offers another possible starting point to explain how the territorial division of power may affect policy performance. At its core, economic federalism claims that the territorial distribution of power in federal systems ensures efficient policies because (and if) functions can be allocated at the appropriate level of government to suit a particular problem and jurisdictions can compete over providing the most advantageous service mix for the revenues taxpayers are called to provide (see Oates, 1999). Economic federalism also argues that if the principles of fiscal federalism are followed, the territorial distribution of power tends toward preserving markets and limiting the growth of government (see Weingast, 1995). However, many policy outcomes suggest that the distribution of authority by levels of government may come at a ‘price’ (see Peterson, 1995). Environmental policy makes for an illustrative example: economic federalism suggests that if environmental spill-overs across jurisdictional boundaries occur, as is often the case with air and water pollution, higher levels of authority are best suited to govern those areas (Rose-Ackerman, 1995; Oates and Portney, 2001). If national public goods or large-scale spill-over problems, such as air quality, public land, rare species, are at stake, the central government will be the best policy provider, whereby ‘best’ means most efficient (Pareto-optimal), i.e. maximizing aggregate social welfare. By the same token, global commons and spill-overs call for supranational policies. Only environmental problems of a local nature, such as household waste management or stationary water and soil pollution, could and should be left to local governance levels. These efficiency considerations have driven the development of environmental policies in many multilevel contexts, most notably the United States and the European Union (EU). The difficulty has been that many areas of environmental policy, especially those affecting production and product standards, are intrinsically linked with domestic and international economic, agricultural and trade policies. Many environmental problems also pertain to health and safety hazards. Therefore, the reality of environmental policy often is more centralized than economic federalism tends to predict. The question is then: does this benefit the environment or is it detrimental to it? Economic federalism, which is primarily normative rather than predictive, would suggest that policy performance is affected by the allocation of resources and functions between levels of government. Countries that succeed in regulating problems by the functionally appropriate level of government can be expected to develop more successful policies. Aside from allocating functions to the appropriate level, economic federalism also highlights that for policies to be successful the distribution of resources must conform to the principle of ‘fiscal equivalence’. The level in charge of a certain function also needs to have the relevant resources to act on its responsibilities and to be accountable to the electorate for services provided. These principles of economic federalism can be applied to both federal and non-federal countries. 118

Federalism and public policy

In principle, the prevailing pattern of territorial power distribution in a country should not matter in how effectively it addresses problems, so long as the appropriate level enjoys powers and resources. Comparative policy research has time and again concluded that federal, regionalized and highly decentralized countries do not perform differently from unitary and highly centralized countries when it comes to governing many policies, including environmental policies (Wälti, 2004). When differences are apparent, economic federalism does not explain them well. In fact, in the environmental policy field there is even counter-intuitive evidence that more localized environmental problems such as water quality are less effectively addressed in multilevel polities, while problems suffering from externalities and national public goods seem to fare better when multiple levels are involved (Ringquist, 1993). Federal and regionalized countries that are characterized by significant decentralization of policy powers and accountability may take longer to adopt central policies when needed because of a more cumbersome multilevel decision-making process. Yet, when they do manage to adopt such policies, multilevel systems may be more likely to effectively address local problems than centralized ones because they can rely on local governments that are accustomed to tailoring policies and are held accountable for their success. Local accountability is particularly important in environmental policy matters because outcomes are often felt most vividly at the local level. A caveat to be kept in mind here is that in many federal regimes, local governments are actually rather weak, often weaker than in unitary or non-federal states. Also, unitary systems often have an easier time governing based on ‘functional overlapping (competing) jurisdictions’ (Frey and Eichenberger, 1999) because local spheres of authority are set up more flexibly. It would be worth investigating whether hybrid systems could be particularly well situated to enjoy the best of both worlds when it comes to environmental policies by combining the advantages of decentralization with a tradition of strong, functional local governance. The performance of different multilevel governance arrangements may vary most notably when it comes to addressing regional environmental problems, which are neither national nor local in scope. Multilevel polities may be well equipped to solve problems at the sub-national level, but only so long as they do not span multiple jurisdictions (Oates and Portney, 2001), as often happens with sub-national environmental challenges such as the management of resources and large ecosystems (Gerber et al., 2009). Multilevel regimes, which are characterized by a fluid patchwork of overlapping jurisdictions, could generally be expected to do better in the face of often spatially and functionally fluid and overlapping environmental problems. By contrast, the more locally accountable multilevel polities may be particularly effective at matching people’s preferences and willingness to pay for services they receive, including hazardous waste management or the monitoring of local water and soil quality (see Livingston, 1987; Breton, 2000).

Inter-jurisdictional competition: a special case Fiscal federalism traditionally has paid special attention to inter-jurisdictional competition by arguing that federal and decentralized polities are particularly efficient because they allow consumers and producers to (re-)locate in the jurisdiction that best matches their needs (including environmental quality) for the least tax burden (e.g. Weingast, 1995). However, the efficiencyenhancing effect of competition hinges on the congruence of each jurisdiction’s service and tax powers. If there is no such congruence one jurisdiction can free-ride on the services of others or offload negative externalities to them for free. Hence, as spill-overs between jurisdictions grow, as is the case in many fields of environmental policy, the assumption that inter-jurisdictional competition benefits policies becomes questionable (Rose-Ackerman, 1995; Oates and Portney, 2001). Indeed, in the presence of significant spill-overs inter-jurisdictional competition may 119

Sonja Wälti

become ruinous. Some jurisdictions may relax environmental requirements to attract economic development, jobs and tax revenues. In fact, knowing that other jurisdictions are doing the same, they may each go beyond what is optimal for their own jurisdiction. The dynamics of this cut-throat inter-jurisdictional competition have become known as ‘race to the bottom’ (Stewart, 1977; Vogel, 1995), a term that is confusing because competitive dynamics may not cause a net worsening of the environment but simply depress an otherwise improving trend (Wälti, 2009). The pessimistic outlook for multilevel governance arrangements to develop and sustain regional, national and international environmental policies due to inter-jurisdictional competition has seen many revisions. Most importantly, from the viewpoint of economic federalism, industries may be only moderately responsive to environmental regulation. Considerations such as the presence of skilled labour, general tax conditions, the proximity to markets and other production sites probably outweigh their attention to environmental requirements (Jaffe et al., 1995; Braun, 1998: 257). The relocation into another jurisdiction also entails significant moving costs. The salience of inter-jurisdictional competition depends on the environmental problem structure. Environmental problems involving large-scale spill-overs are especially prone to competition if not regulated centrally. There is also less hope for environmental problems resulting from stationary production processes than there is for problems resulting from mobile products (Lowry, 1992; Scharpf, 1999: 91–101). The voter-taxpayer and consumer can more immediately affect the environmental quality of products such as cars by choosing to buy those that are more fuel efficient or by requesting that the home jurisdiction mandate them to be fuel efficient. If a jurisdiction responds to such a request by regulating a product such as cars, both that jurisdiction and the affected industry will want the new fuel-efficiency standards imposed at a larger scale and industry wide. This is especially true if the new demands emanate from a large market, such as California in the United States or Germany in the EU. This ‘California effect’ is further enhanced if the resulting increases in production costs and potential job losses occur in other (car-producing) jurisdictions. In the case of environmental problems that emanate from production processes, which is more common for point-source pollution, the same dynamics do not play out (Scharpf, 1999: 98).

Intergovernmental relations and management The study of intergovernmental relations and management focuses on the vertical as well as horizontal relations between levels of government. The study of ‘intergovernmental relations’ is usually associated with a primary focus on policy making, while reference to ‘intergovernmental management’ pays closer attention to policy implementation (Agranoff, 1994). Over time, attention to territorial governments has given way to a broader consideration of the way networks of actors operate within multilevel governance and government arrangements. Thanks to the study of intergovernmental relations, aside from federal-state and state-state relations, direct federal-local and federal-state-local relations shifted into view (Wright, 1990). In addition, the primary focus on public entities gave way to a broader consideration of public and private relations. While institutional and legal constraints play a considerable role in the discussion of intergovernmental relations, actors of all sorts take centre stage in accounting for policy performance. (In this regard, the study of intergovernmental relations in the United States features significant parallels with the study of ‘multilevel governance’ in Europe.)

Policy diffusion, transfer and convergence The consideration of intergovernmental relations has drawn attention to specific inter-jurisdictional dynamics, policy innovation and diffusion being the most prominent. Policy diffusion takes 120

Federalism and public policy

place when jurisdictions emulate one another’s practices. In environmental policy, diffusion typically involves borrowing other jurisdiction’s policy instruments or administrative practices (e.g. Jordan et al., 2003). Policy transfer, although denoting a more active horizontal exchange of practices, is often used indiscriminately. ‘Pioneers’ are thought to adopt certain practices, which neighbours then emulate and adapt (Andersen and Liefferink, 1997; Liefferink et al., 2009). ‘Laggards’ take especially long or never adopt the new practice. Neighbourhood plays a crucial role in diffusion, whereby jurisdictions in close geographic, linguistic, political, or economic proximity may more easily follow one another’s lead. Although the micro-processes in policy diffusion are still being researched, the diffusion of ideas and practices within policy communities involving both public and private actors are thought to be very important (Jordan et al., 2003: 18–19). Environmental policy innovation and diffusion have received attention not only at the domestic (e.g. Kern, 2000; Sapat, 2004; Daley and Garand, 2005), but also at the international levels (e.g. Liefferink and Andersen, 2002; Tews et al., 2003; Albrecht and Art, 2005; Busch and Jörgens, 2005; Jörgens, 2005; Holzinger et al., 2008). As a result of policy diffusion and transfer, policies in different jurisdictions converge over time until a new wave of innovation takes place followed again by convergence. Innovation and diffusion patterns can differ greatly but are often S-curve patterned. A few pioneers start a new trend, which is followed by others in an accelerated pattern before tapering off, while the laggards finally follow suit or are forced into compliance. Aside from neighbourhood effects, policy convergence may also be due to similarities in problems or similarities in internal (domestic) conditions. Or convergence may be the result of overarching national or international policies, which happen to lead to similar responses at the national or sub-national level. In environmental policy matters, all three of these trends undoubtedly coexist, making it particularly hard to distinguish between them, despite the promising advances in modelling and testing for such patterns (see Daley and Garand, 2005; Holzinger et al., 2008). Convergence is most pronounced at the policy level (whether there is a policy or not) and less so at the instrumental level (what policy instruments or settings prevail) (Sommerer et al., 2008). Jurisdictions converge most when it comes to trade-related policies. Not surprisingly, the obligation to adopt a policy due to harmonization requirements significantly drives convergence. The EU’s increasing influence on environmental policy matters, in both member and future member states, seems to have become the most important predictor of a country’s willingness to embrace stricter environmental policies (Zürn and Jörges, 2005; Liefferink et al., 2009). However, other than EU-related environmental regimes also foster convergence. Inter-jurisdictional communication within policy networks seems to play a central role, thus corroborating that policy innovation, transfer and diffusion may happen via these channels (Sommerer et al., 2008). The horizontal and vertical paths that policy innovation and diffusion follow have been captured in the so-called ‘vertical interaction model’, which postulates that policy diffusion is often accelerated (and occasionally decelerated) by vertical transfers: innovations are uploaded and subsequently funnelled downward within a multilevel governance arrangement. Multi-tiered systems are often credited with a high capacity to innovate because new ideas and practices can emerge and be tested in independent territorial policy laboratories without putting the entire system at risk. The sheer presence of multiple governance levels likely fosters environmental policy innovation. Although it is difficult to make clear predictions, it seems plausible that multilevel polities may have a greater capacity to innovate due to their ability to problem solve across different policy fields. However, it is unclear whether the subsequent transfer and diffusion of such innovation takes place regardless of a particular federal, regional, or hybrid multilevel format. It seems plausible, for example, that asymmetric federal, regional, or 121

Sonja Wälti

hybrid systems are less prone to the horizontal transfer and diffusion of policies and practices because not all constituent units are expected or equipped to adopt them.

Actor-centred federalism The various theories of federalism outlined so far each have significant shortfalls. They typically are not designed to explain policy performance but tend to focus on macro-political processes. When they are applied to explain policy performance, relevant explanatory variables tend to be depicted as (constitutionally or legally) constant. Such applications also quickly lose sight of the complexity of the policy process, at best considering either policy making or policy implementation as relevant. When the full complexity is considered – such as in the case of Braun’s distinction between the ‘right to decide’ and the ‘right to act’ (Braun, 2000) – the central focus tends to shift back to considerations about the formal distribution of powers and resources. A powerful alternative is to take an actor-centred stance. Actor-centred approaches to the study of territorial divisions and power and their impact on policy performance help pool and probe underlying political and actor-related dynamics to explain the emergence and success of certain policies. They are also particularly effective in examining how institutional and actor-related factors jointly affect policy performance. The basic idea to do so is inspired by ‘actor-centred institutionalism’ (see Scharpf, 1997), which depicts policy outputs and outcomes as driven by more proximate actor-related variables, which are in turn impacted by more remote institutional factors. Actor-centred institutionalism assumes actors are motivated by interests, which may also include ideational and ideological components. In the neo-institutionalist tradition, institutions encompass formal as well as informal rules. Other, not specifically federal, theories have proceeded in a similar fashion, motivated by similar quests to explain policy variation and success: thus, Tarrow argues that social movements are successful and shape their repertoire of action depending on the political opportunity structures (POS) they face (Tarrow, 1998). POS can be open or closed in many different regards. Over time, Tarrow added an emphasis on the importance of perceived opportunity structures and also stressed that opportunities may include not only stable institutional structures but also situational components. Federalism can work both ways. It may add to the openness of POS by providing multiple venues for movements. On the other hand, there may also be significant territorial variance in the openness of state and local opportunity structures. POS theory is not primarily focused on explaining policy outputs and outcomes but rather on the institutional receptiveness towards agents of change. In a similar vein, Tsebelis’s Veto Player Theory (Tsebelis, 2002) explains policy performance (e.g. the ability to practise fiscal restraint) by the existence, number and constellation of veto players and the configuration of their preferences. Although the focus of this theory is decidedly on the actor end of the spectrum, relying extensively on game-theoretical models, like Scharpf, Tsebelis demonstrates that institutional structures (such as federalism) affect policy performance by shaping the constellation of players as well as their preferences. Squarely focused on policy continuity and change, Baumgartner and Jones’s Punctuated Equilibrium Framework also depicts institutional variables as enabling or impeding (Baumgartner and Jones, 1993). With regard to the US political system, Baumgartner and Jones have argued that long periods of policy continuity (‘stasis’) are punctuated with brief moments of rapid change because crucial institutional components that commonly act as impediments to change – such as vertically and horizontally overlapping powers – can offer multiple access points for policy entrepreneurs in moments of crisis. In such moments of crisis, institutional factors help actors propel particular policy issues into the macro-political arena, creating new policy images that may bring about change. 122

Federalism and public policy

In pursuing an actor-centred approach to federalism, we can learn a great deal about the connection between federalism and policy performance by calibrating successful actor-centred theories to the study of federalism. In this sense, an actor-centred approach to federalism does not constitute an alternative to existing theories but rather invites the analyst to (re)visit them from a particular angle and using a particular lens, which I propose to do here using environmental policy as a reference point. I will focus on various types of actors and, when results are available, review how they have been able to affect environmental policy performance.

Political parties and factions Environmental policies have been shown time and again to be sensitive to politics. Environmental movements and the emergence of green parties or green factions within established parties were instrumental in the development of environmental policies (Neumayer, 2003; Scruggs, 2003). Environmental policy expansion and performance also seems to have benefitted from institutional contexts characterized by accommodative bargaining arrangements (Crepaz, 1995). Yet, the theoretical frameworks to study federal polities that have been outlined above pay surprisingly little attention to multilevel politics. In the early stages of environmental policy formation, the presence of multiple levels of government brought greater opportunities for green movements and parties to emerge and become a significant force (Weidner, 1995). Multilevel party politics is naturally more significant in multilevel polities where jurisdictions provide ample opportunities for party politics. Green parties and forces have had an easier time emerging and sustaining themselves in proportional electoral systems such as Germany and Switzerland, than in contexts of majoritarian politics as is prevalent in the United States and Canada. This may be one of the most significant differences explaining the EU’s surpassing the United States in environmental matters in recent years. Even when green forces are present, the lack, or weakness, of a country-wide and vertically integrated party system, as is the case in Canada, for example, can also curb efforts to expand environmental policies. Political parties may also be a vector when it comes to vertical as well as horizontal policydiffusion processes. Aside from the proximity of jurisdictions in terms of their exposure to a problem or in terms of their geographical situation, they may look to one another or not for ideas depending on whether the same majorities or similar coalitions are in power (e.g. Stokes Berry and Berry, 1999). Such dynamics play a significant role in German environmental policy, for example. However, except for the EU, there is little evidence to suggest that the way political parties manoeuvre multilevel polities has an impact much beyond domestic boundaries.

Interest and advocacy groups The presence of multiple levels of government also increases opportunities for interest and advocacy groups, be it as veto players or as agents of policy change. The assumption has long been that corporatism may hinder environmental efforts by providing business interests with veto opportunities to block unwanted policies. Yet, empirically corporatism seems to have had a favourable effect on environmental policy performance (Crepaz, 1995; Neumayer, 2003; Scruggs, 2003). However, comparative research shows that in fact this holds true only for multi-tiered contexts (Wälti, 2004). This result suggests that multilevel governance structures turn accommodative bargaining arrangements into access points for environmental advocacy groups. Similar dynamics have been uncovered beyond the nation-state, where non-state actors drive the development of international environmental regimes, in some cases (e.g. forest 123

Sonja Wälti

certification) even in the absence of strong state control and relying solely on the market (Cashore et al., 2007). It is conceivable that these dynamics can play out in multilevel polities as well as in less institutionalized and more loosely connected multilevel regimes, such as the EU or international governance (see Kelemen, 2004). Indeed, the latter may even be more receptive to accommodative bargaining precisely due to the weakness of politically and electorally accountable jurisdictions. The interplay of interest groups at various levels of government is also at the core of Peterson’s ‘legislative federalism’ (Peterson, 1995), which according to him comes into play in policy areas that are characterized by high local stakes, for example when economic benefits are concentrated in the hands of locally important industries, as is the case in mining (see also Ringquist, 1993). Such industries tend to affect policies by exerting strong influence at the state level. Attention to interest groups also led to a refined understanding of environmental economics, by paying increased attention to the ‘political economy of environmental protection’ (Oates and Portney, 2001). The focus on actors defending business interests likely provides the most value added in revisiting the theory on interstate competition. Vogel (1995) was one of the first to point out that interstate environmental competition had less to do with consumer or producer preferences than with whether or not environmental advocacy groups are able to convince one or the other jurisdiction to press ahead in tightening environmental regulations. Especially when this happens in a large market such as California in the United States or Germany in the EU, producers’ preferences are subsequently affected in that they would rather see uniform, albeit stricter, standards applied nation-wide and will typically rally to support an upward levelling of the playing field. Environmental problems are particularly sensitive to this dynamic because political elites can often respond to diffuse environmental interests (Pollack, 1997) by imposing concentrated costs, for example on certain industries. What is more, this constellation of diffuse benefits and concentrated costs often fuels ‘entrepreneurial politics’ (Wilson, 1989: 72), which may more easily translate into innovation in multilevel contexts, most notably by multiplying opportunities for policy entrepreneurs to initiate and press for new ideas. These dynamics can play out in multilevel polities as well as in multilevel regimes, as the experience of the EU demonstrates (Vogel, 1997). In policies involving the trade of goods, it is not unheard of for innovative industries with strong market power to go as far as advocating stricter product standards so as to get a head start in a market in which competitors will struggle to catch up. These dynamics cannot be explained by market models underlying economic federalism, but rather by an actor-centred understanding of multilevel governance arrangements (Scharpf, 1999).

Constituent units Constituent units can be understood in the same manner. Thus, the strategic behaviour of jurisdictions within a federation or regionalized system impacts policy performance in unique ways, just as political parties/factions or interest/advocacy groups do. For example, in environmental matters, regardless of the path by which pioneering jurisdictions press ahead with their environmental policies, they subsequently have an intrinsic interest in levelling the playing field by uploading their comparatively more stringent policies and thus ultimately subjecting other jurisdictions to them, as studies on interstate competitions at the domestic and international level have shown (Scharpf, 1999: 84–120; Liefferink and Andersen, 2002; Vogel, 2004). In many areas of environmental policy, this has resulted in an intergovernmental dynamic of pioneers ‘ratcheting up’ environmental policies rather than in a ‘race to the bottom’, as predicted by inter-jurisdictional competition. 124

Federalism and public policy

Once again, the study of interstate competition benefits greatly from an actor-centred approach. While there is some empirical evidence that firms are sensitive to environmental standards and may even shift production accordingly (Henderson, 1996; Braun, 1998; Levinson, 1999), there is none confirming that, in response, jurisdictions undercut one another’s environmental policies (see List and Gerking, 2000; Fredriksson and Millimet, 2002; Arts et al., 2008). The contrary seems to hold true: if anything, intergovernmental environmental relations have been characterized by a ‘race to the top’, though there is evidence that only some jurisdictions participate in what Fredriksson and Millimet (2002) call ‘yardstick competition’. Also, jurisdictions may not actually respond to the threat – perceived or real – of relocating industries. With an eye on environmentally sensitive voter-taxpayers who are willing to pay a price for environmental quality (e.g. Cragg and Kahn, 1997), governments may prefer to resist business interests (Revesz, 1992). Indeed, taxpayers may even consider moving to high-quality areas just as they do to access transportation or secure good schools for their children. As a result, jurisdictions may ratchet up pollution control and thus become a centre for high-tech, low-pollution industries to suit the needs and skills of their pollution-sensitive population. In the context of the United States, this has become known as the ‘California effect’ (Vogel 1995). The framing of issues also plays a role. When priority is given to investments and job creation, authorities may be inclined to give into deregulation to meet lower standards in neighbouring states. By contrast, when sustainable growth is in demand, authorities may be receptive to pro-environmental regulatory pressure. More informal and loosely coupled multilevel regimes that are composed of single-purpose jurisdictions may be expected to be more responsive to inter-jurisdictional environmental competition than multilevel or hybrid polities, which have to accommodate diverse political demands and policy needs. Whether that competition is beneficial or detrimental hinges on their ability to allocate powers and resources efficiently between levels of authority, as discussed under economic federalism above. These findings confirm that ultimately actor-related dynamics in conjunction with institutional structures are more convincing in explaining policy performance than the underlying market forces. The study of constituent units as strategic actors pursuing their interests within various institutional contexts will require a more differentiated approach, further distinguishing between executive, legislative and administrative actors, and fully exploiting developments and findings in the study of intergovernmental relations and management. Such an actor-centred approach to the study of intergovernmental relations and management, as they pertain to the explanation of policy continuity and change, recast institutions as providing ‘opportunity structures’ to thwart as well as to enable policies. These opportunity structures are not constant; they shape different stages of the policy process in unique ways. In one case, the setting of the environmental policy agenda may provide multiple access points to a plethora of contestants, whereas decision making may be concentrated in the hands of few actors with powerful means to veto unwanted policies. Then, yet further into the policy process, the implementation of the presumably few environmental policy successes emerging from this constellation may be facilitated by intergovernmental accommodation.

Conclusion and directions for future research The most notable contribution of an actor-centred approach to comparative federalism is its ability to shed new light on some of the existing micro- and meso-theories of federal, regionalized and hybrid forms of government and governance. Another important contribution is to allow more systematically for broader comparisons, not only involving federal as well as nonfederal countries, but also looking beyond the domestic context at any situation involving 125

Sonja Wälti

multiple governance and governmental levels. The invitation to examine domestic and international findings indiscriminately, indeed to connect and compound those findings, opens up exciting opportunities for the advancement of knowledge and the connection of previously separate academic communities. The following consequences of this trend in environmental policy seem particularly noteworthy and point to future avenues of research. The state-centredness of comparative environmental policy research is complemented by the study of international regimes, which are responsible for an increasing number of policy initiatives affecting national and sub-national environmental policies. Especially when studying the effectiveness of international regimes, we may draw on the longstanding tradition of domestic implementation research, which has always paid significant attention to multilevel dynamics. Levels beyond the nation-state may receive further attention, while at the same time additional sub-national levels – local, regional, or city levels – shift into view. The exploration of connections and interactions between the supranational and the local level are without a doubt particularly promising in environmental policy matters, and their study promises to help improve policies. By shifting our focus from multilevel government, which federalism scholars have long favoured, to a broader consideration of various public and private actors, new actors and processes catch our analytic attention. Unconventional avenues for comparative research open up, such as a more systematic comparative understanding of the United States and the EU, and possibly their juxtaposition with regional entities in other parts of the world. These comparisons will help point out what type of multilevel arrangements may be particularly effective. They also foster the diffusion of best practices. The salience and particularity of the environmental policy field is that it adds another dimension to the multilevel mix, namely the natural (earth system) level (Winter, 2006). Like no other policy field, environmental policies govern territorially complex natural systems such as urban or rural areas, riverbeds, watersheds, coastal zones, mountain ranges, island systems, or climatic zones. These are often functional spaces with no, few, or overlapping institutional boundaries. What is more, these functional spaces are increasingly connected by virtue of international law and policies: small island states form cooperatives, cities join forces in networks, and rainforest zones team up across seas. Multilevel governance promises to take them into account. By fostering flexibility and creativity in developing solutions to ever evolving problems and across ever shifting territories, multilevel governance may well promise better environmental policies.

References Agranoff, Robert (1994) ‘Comparative Intergovernmental Relations’, in Randall Baker (ed.) Comparative Public Management: Putting U.S. Public Policy and Implementation in Context (Westport, CT: Praeger), 165–83. Albrecht, Johan and Bas Art (2005) ‘Climate Policy Convergence in Europe: An Assessment Based on National Communications to the UNFCCC’, Journal of European Public Policy Vol. 12, No. 5: 885–902. Andersen, Mikael Skou and Duncan Liefferink (eds) (1997) European Environmental Policy: The Pioneers (Manchester and New York: Manchester University Press). Arts, Bas, Duncan Liefferink, Jelmer Kamstra and Jeroen Ooijevaar (2008) ‘The Gap Approach: What Affects the Direction of Environmental Policy Convergence?’ in Katharina Holzinger, Christoph Knill and Bas Arts (eds) Environmental Policy Convergence in Europe: The Impact of International Institutions and Trade (Oxford: Oxford University Press), 196–226. Baumgartner, Frank and Bryan Jones (1993) Agenda and Instability in American Politics (Chicago: University of Chicago Press). Braun, Boris (1998) ‘Locational Response of German Manufacturers to Environmental Standards’, Tijdschrift voor Economische en Sociale Geografie Vol. 89, No. 3: 253–63. 126

Federalism and public policy

Braun, Dietmar (ed.) (2000) Federalism and Public Policy (Aldershot: Ashgate). Breton, Albert (2000) ‘Federalism and Decentralization: Ownership Rights and the Superiority of Federalism’, Publius: The Journal of Federalism Vol. 30, No. 2: 1–16. Busch, Per-Olof and Helge Jörgens (2005) ‘The International Sources of Policy Convergence: Explaining the Spread of Environmental Policy Innovations’, Journal of European Public Policy Vol. 12, No. 5: 860–84. Cashore, Benjamin, Graeme Auld, Steven Bernstein and Constance McDermott (2007) ‘Can Non-state Governance ‘Ratchet Up’ Global Environmental Standards? Lessons from the Forest Sector’, Review of European Community & International Environmental Law Vol. 16, No. 2: 158–72. Cragg, Michael and Matthew Kahn (1997) ‘New Estimates of Climate Demand: Evidence from Location Choice’, Journal of Urban Economics Vol. 42, No. 2: 261–84. Crepaz, Markus (1995) ‘Explaining National Variation of Air Pollution Level: Political Institutions and their Impact on Environmental Policy-Making’, Environmental Politics Vol. 4: 391–414. Daley, Dorothy and James C. Garand (2005) ‘Horizontal Diffusion, Vertical Diffusion, and Internal Pressure in State Environmental Policymaking, 1989–98’, American Politics Research Vol. 33, No. 5: 615–33. Elazar, Daniel J. (1973) ‘First Principles’, Publius: The Journal of Federalism Vol. 3, No. 2: 1–10. Fredriksson, Per G. and Daniel L. Millimet (2002) ‘Strategic Interaction and the Determination of Environmental Policy across U.S. States’, Journal of Urban Economics Vol. 51, No. 1: 101–22. Frey, Bruno S. and Reiner Eichenberger (1999) The New Democratic Federalism for Europe: Functional, Overlapping and Competing Jurisdictions (Cheltenham: Edward Elgar). Gerber, Jean-David, Peter Knoepfel, Stéphane Nahratha and Frédéric Varonec (2009) ‘Institutional Resource Regimes: Towards Sustainability Through the Combination of Property-rights Theory and Policy Analysis’, Ecological Economics Vol. 68, No. 3: 798–809. Henderson, J. Vernon (1996) ‘Effects of Air Quality Regulation’, American Economic Review Vol. 86, No. 4: 789–813. Holzinger, Katharina, Christoph Knill and Bas Arts (eds) (2008) Environmental Policy Convergence in Europe: The Impact of International Institutions and Trade (Cambridge: Cambridge University Press). Jaffe, Adam B., Steven Peterson, P.R. Portney and R.N. Stavins (1995) ‘Environmental Regulation and the Competitiveness of U.S. Manufacturing: What Does Evidence Tell Us?’ Journal of Economic Literature Vol. 33, No. 1: 132–63. Jordan, Andrew, Rüdiger K.W. Wurzel and A.R. Zito (2003) ‘‘New’ Instruments of Environmental Governance: Patterns and Pathways of Change’, Environmental Politics Vol. 12, No. 1: 1–24. Jörgens, Helge (ed.) (2005) ‘Diffusion and Convergence of Environmental Policies in Europe’, European Environment, Special Issue 15(2). Kelemen, R. Daniel (2004) The Rules of Federalism: Institutions and Regulatory Politics in the EU and Beyond (Boston, MA: Harvard University Press). Kern, Kristine (2000) Die Diffusion von Politikinnovationen (Opladen: Leske + Budrich). Levinson, Arik (1999) ‘NIMBY Taxes Matter: The Case of State Hazardous Waste Disposal Taxes’, Journal of Public Economics Vol. 74: 31–51. Liefferink, Duncan and Mikael Skou Andersen (2002) ‘Strategies of the ‘Green’ Member States in EU Environmental Policymaking’, in Andrew Jordan (ed.) Environmental Policy in the European Union (London and Sterling, VA: Earthscan), 63–80. Liefferink, Duncan, Bas Arts, J. Kamstra and J. Ooijevaar (2009) ‘Leaders and Laggards in Environmental Policy: A Quantitative Analysis of Domestic Policy Outputs’, Journal of European Public Policy Vol. 16, No. 5: 677–700. List, John A. and S. Gerking (2000) ‘Regulatory Federalism and Environmental Protection in the United States’, Journal of Regional Science Vol. 40, No. 3: 453–71. Livingston, Marie Leigh (1987) ‘Evaluating the Performance of Environmental Policy: Contributions of Neoclassical, Public Choice, and Institutional Models’, Journal of Economic Issues Vol. 21, No. 1: 281–94. Lowry, William R. (1992) The Dimensions of Federalism: State Governments and Pollution Control Policies (Durham and London: Duke University Press). Neumayer, Eric (2003) ‘Are Left-wing Party Strength and Corporatism Good for the Environment? Evidence from Panel Analysis of Air Pollution in OECD Countries’, Ecological Economics Vol. 45: 203–20. Oates, Wallace E. (1999) ‘An Essay on Fiscal Federalism’, Journal of Economic Literature Vol. 37, No. 3: 1120–49. Oates, Wallace E. and Paul R. Portney (2001) From Research to Policy: The Case of Environmental Economics (Washington, DC: Resources for the Future). Peterson, Paul E. (1995) The Price of Federalism (Washington, DC: Brookings Institution). 127

Sonja Wälti

Pollack, Mark A. (1997) ‘Representing Diffuse Interests in EC Policy-Making’, Journal of European Public Policy Vol. 4, No. 4: 572–90. Rabe, Barry G. and William R. Lowry (1999) ‘Comparative Analyses of Canadian and American Environmental Policy: An Introduction to the Symposium’, Policy Studies Journal Vol. 27, No. 2: 288–306. Revesz, Richard L. (1992) ‘Rehabilitating Interstate Competition: Rethinking the ‘Race-to-the-bottom’ Rationale for Federal Environmental Regulation’, New York University Law Review Vol. 67, No. 6: 1210–54. Ringquist, Evan J. (1993) Environmental Protection at the State Level: Politics and Progress in Controlling Pollution (Armonk, NY and London: M.E. Sharpe). Rose-Ackerman, Susan (1995) Controlling Environmental Policy: The Limits of Public Law in Germany and the United States (New Haven: Yale University Press). Salamon, Lester M. (2002) ‘The New Governance and the Tools of Public Action: An Introduction’, in Lester M. Salamon (ed.) The Tools of Government: A Guide to the New Governance (Oxford: Oxford University Press), 1–47. Sapat, Alka (2004) ‘Devolution and Innovation: The Adoption of State Environmental Policy Innovation’, Public Administration Review Vol. 64, No. 2: 141–51. Scharpf, Fritz W. (1997) Games Real Actors Play: Actor-Centered Institutionalism in Policy Research (Boulder and Oxford: Westview). ——(1999) Governing in Europe: Effective and Democratic? (Oxford and New York: Oxford University Press). Scruggs, Lyle A. (2003) Sustaining Abundance: Environmental Performance in Industrial Democracies (Cambridge: Cambridge University Press). Sommerer, Thomas, Katharina Holzinger and Christoph Knill (2008) ‘The Pair Approach: What Causes Convergence of Environmental Policies?’ in Katharina Holzinger, Christoph Knill and Bas Arts (eds) Environmental Policy Convergence in Europe: The Impact of International Institutions and Trade (Oxford: Oxford University Press). Stewart, Richard B. (1977) ‘Pyramids of Sacrifice? Problems of Federalism in Mandating State Implementation of National Environmental Policy’, Yale Law Journal Vol. 86: 1196–272. Stokes Berry, Francis and William D. Berry (1999) ‘Innovation and Diffusion Models in Policy Research’, in Paul A. Sabatier (ed.) Theories of the Policy Process (Boulder, San Francisco, Oxford: Westview Press), 169–200. Tarrow, Sidney (1998) Power in Movement (New York: Cambridge University Press). Tews, Kerstin, Per-Olof Busch and H. Jörgens (2003) ‘The Diffusion of New Environmental Policy Instruments’, European Journal of Political Research Vol. 42, No. 4: 569–600. Tsebelis, George (2002) ‘Veto Players: How Political Institutions Work’ (Princeton: Princeton University Press). Vogel, David (1995) ‘The California Effect’, in David Vogel (ed.) Trading Up: Consumer and Environmental Regulation in the Global Economy (Cambridge: Harvard University Press), 248–70. ——(1995) Trading Up: Consumer and Environmental Regulation in the Global Economy (Cambridge: Harvard University Press). ——(1997) ‘Trading Up and Governing Across: Transnational Governance and Environmental Protection’, Journal of European Public Policy Vol. 4, No. 4: 556–71. ——(2004) ‘Trade and the Environment in the Global Economy: Contrasting European and American Perspectives’, in Norman J. Vig and Michael G. Faure (eds) Green Giants? Environmental Policies of the United States and the European Union (Cambridge, MA: MIT Press), 231–52. Wälti, Sonja (2004) ‘How Multilevel Structures Affect Environmental Policy in Industrialized Countries’, European Journal of Political Research Vol. 43, No. 4: 597–632. ——(2009) ‘Intergovernmental Management of Environmental Policy in the United States and the European Union’, in Miranda Schreurs, Henrik Selin and Stacy D. VanDeveer (eds) Transatlantic Environment and Energy Politics: Comparative and International Perspectives (Aldershot: Ashgate), 41–56. Watts, Ronald L. (1999) Comparing Federal Systems (Kingston, Ont.: McGill/Queen’s University Press). Weibust, Inger (2009) Green Leviathan: The Case for a Federal Role in Environmental Policy (Aldershot: Ashgate). Weidner, Helmut (1995) 25 Years of Modern Environmental Policy in Germany: Treading a Well-Worn Path to the Top of the International Field (Berlin: Wissenschaftszentrum Berlin für Sozialforschung). Weingast, Barry R. (1995) ‘The Economic Role of Political Institutions: Market-Preserving Federalism and Economic Growth’, Journal of Law, Economics, and Organization Vol. 11, No. 1: 1–31. Wilson, James Q. (1989) Bureaucracy: What Government Agencies Do and Why They Do It (New York: Basic Books). 128

Federalism and public policy

Winter, Gerd (ed.) (2006) Multilevel Governance of Global Environmental Change: Perspectives from Science, Sociology and the Law (Cambridge: Cambridge University Press). Wright, Deil S. (1990) ‘Federalism, Intergovernmental Relations, and Intergovernmental Management: Historical Reflections and Conceptual Comparisons’, Public Administration Review Vol. 50, No. 2: 168–78. Zürn, Michael and Christian Jörges (eds) (2005) Law and Governance in Post-national Europe (Cambridge: Cambridge University Press).

129

9 The ‘new regionalism’ and the politics of the regional question John Agnew

‘New regionalism’ refers to theoretical and policy perspectives developed since the 1980s that directly relate the relevance of regions as sub-national or city-regional units of economic and political account to the economic restructuring of national economies as a result of globalization and supranational regionalization (as with the European Union (EU), North American Free Trade Agreement (NAFTA), etc.). Part of the motivation for the adoption of regions (as opposed to classes or ethnic groups) as the motif for this restructuring lies in seeing how wealth is distributed and redistributed across places (through incentive and training programmes, for example). This is because people’s social lives are anchored geographically and their welfare thus rests, at least to some degree, on staying put rather than moving elsewhere. Another part lies in the popular idea that global uneven development is operating increasingly in relation to regions and localities rather than states as these states are ‘hollowed out’ economically and the ‘regulatory’ response to this process is now best organized at the regional scale. There are therefore both normative and positivist elements to the ‘new regionalism’. It is about the seemingly increasing economic disparities between sub-national regions as a result of enhanced economic competition but it is also concerned to argue that economic processes are increasingly organized with reference to the regional scale rather than that of the national state as they presumably were in the recent past. In this chapter I begin by briefly surveying the origins and development of those perspectives in geography, economics, sociology and political science broadly consonant with the above statements. I then turn to the diverse and distinctive understandings of ‘regions’ found in this literature, particularly those that emphasize global cities (and their regions) versus those that focus on more ‘traditional’ sub-national regions or metropolitan/hinterland differences. Finally, I discuss the assumptions about ‘the regional question’ that the new regionalism and many of its most vociferous critics tend to reflect, suggesting that its proponents’ general tendency to economic determinism and their critics’ often relational rather than territorial understanding of the nature of regions are both problematic when it comes to grasping the real course of regionalism around the world in recent years. I claim that the regional question is in fact much more of a political question than either most new regionalists or some of their critics are inclined to admit, both in the sense of having a powerful bottom-up political element to it in calls for more local/ regional policy levers and increased devolution of powers, and in the sense of being a chosen rather than an automatic response to the exigencies of globalization and economic regionalization at 130

The politics of the regional question

the supranational level. From this perspective, the new regionalism in the world as opposed to that in the academy has political as opposed to strictly economic roots, and the economic effects of globalization are always mediated politically in terms of whether a real regionalism rather than a theoretically imagined one actually emerges.

What is the ‘new regionalism’? At the risk of being overly schematic, it seems that there are three crucial ideas underpinning much of what has become known as the ‘new regionalism’. The first is that capitalism relies on uneven development to work its wonders by keeping up the rate of capital accumulation through constant disinvestment and reinvestment and this requires a constant ebb and flow of capital between economic sectors and the various regions in which those sectors are differentially embedded. At any particular moment, different ‘spatial divisions of labour’ (Massey, 1984) incorporate different geographical areas or regions into their webs in distinctive ways. Thus, corporate headquarters and associated services (law firms, advertising agencies, etc.) are located in metropolitan regions, whereas various ‘stages’ of production (headquarters, assembly, etc.) and the labour forces they require, are located in hinterland regions with different ratios of skilled to unskilled labour and other economic factors. The ‘regional problem’, in a political sense, arises when decline sets in within the secondary or subordinate regions and governments are forced to act to try and reverse this trend. This was the history of so-called regional policy in many countries in the aftermath of the Great Depression of the 1930s. The second idea is related to this last point. From the 1930s until the 1970s across the industrialized world a so-called Keynesian welfare-state model of social and economic regulation both stabilized the capitalist business cycle and created (and legitimized) a ‘mixed’ ownership national economy with active government intervention both to stimulate employment (and consumption) and provide direct financial support to the unemployed, the elderly and poorer people of whatever age (e.g. Jessop, 2002). Though specific policies differed between countries, one overall trend was that national governments were reinforced at the expense of lower tiers of government and regional policy was a top-down process of offering subsidies to industries and underwriting individual and household welfare in poorer areas. Beginning in the 1960s, this ‘system’ began to break down under the dual impact of increased international competition as a result of the globalization generated by multinational businesses and the spread of neo-liberal ideology, giving precedence to markets over state economic intervention. Different accounts give differing emphases to these two elements. This is where the third idea comes in. In response to post-Keynesian conditions the ‘state’ itself has been undergoing restructuring, one important component of which has been the progressive creation of a more complex governance system in which states are ‘re-scaled’ to respond more flexibly to increasingly volatile external economic conditions (Loughlin, 2009). To Brenner (2004), for example, the regional is longer a ‘residual’ as it could perhaps be construed in an earlier period as national levels of government superseded more localized ones, but is now crucial to the patterning of state intervention. In other words, new economic conditions call forth new modes of territorial organization. In a recent update he notes that: ‘the reorientation of state spatial strategies from nationally redistributive modalities towards urban-centric, competitivenessoriented forms of locational policy still appears quite pervasive across the EU’ (Brenner, 2009: 129). Economic change triggers a state response. To others, however, the directing and regulatory role of the state itself is in question as global city-regions or mega-regions undermine its relevance for economic growth (e.g. Scott, 1998). Certainly, across a significant number of countries since the 1980s levels of government and methods of governance other than those associated with central government and centralization 131

John Agnew

of power have undergone something of a renaissance. Many new regionalists insist on seeing these substantively different trends (uneven devolution here, as in the UK and Spain, fiscal decentralization there, as in Italy), with distinctive roots in different places, as reflecting a universal ‘scalar turn’ within the borders of ‘the post-national state’ as a direct response to the pressures of economic globalization (Brenner, 2009). Of course, some countries have effectively experienced little if any such re-scaling beyond the merely nominal (e.g. France, Sweden). A more pluralist interpretation may be in order. Keating’s (1998) analysis of the new trends towards regionalism in Western Europe from the 1980s, for example, while explicitly deriding the idea that it is simply a recapitulation of an older provincialism, nevertheless insists that its various manifestations have no singular causation but arise from a confluence of three different if interdependent tensions inherent in time and place. If the first conforms closely to the functional restructuring for economic reasons to which much of the literature on the new regionalism remains entirely beholden, the others relate to an institutional restructuring with roots in ideological desire, claims to institutional efficiency, and/or regionalist/nationalist movements and, finally, bottom-up mobilization on behalf of regions to make demands of central governments seen as too favourable to other regions or neglectful or exploitative of the region in question. In a similar vein, some students of regionalism in the United States have noted how political-administrative and social more than singularly economic pressures lie behind efforts to create metropolitan-wide governments in the face of historic jurisdictional fragmentation and reanimate an increasingly centralized federalism in response to cultural and political anxieties not addressed to local satisfaction by the federal government. Notwithstanding the varying interpretations that can be put on the ‘nature’ of the new regionalism, particularly monist versus pluralist, there is undoubtedly something to the phenomenon that the literature has tried to capture. However, the phenomenon is much more of a hybrid response to calls for administrative and political decentralization in the face of a widely perceived over-centralization of states and a rapid reorganization of socio-economic life than a singularly economic-based trend forcing the emergence of a new administrative regionalism. Keating perhaps best captures the moment and the phenomenon in its entirety. Even though he restricts his comment to Western Europe, I think that the claim can be made more generally, when he writes: The political, economic, cultural, and social meaning of space is changing in contemporary Europe … new types of regionalism and of region are the product of a decomposition and a recomposition of the territorial framework of public life, consequent on changes in the state, the market, and the international context. (Keating, 1998: 383)

Regions in the new regionalism Three understandings of regions tend to prevail in the ‘new regionalism’. The first and most dominant among European scholars is of newly empowered sub-national regions emerging into political-administrative prominence as a result of state re-scaling in the aftermath of economic restructuring, such as the relative decline of manufacturing industries in some regions and the rise of finance and other services in others since the 1970s (e.g. Brenner, 2004; MacLeod, 2001). From this perspective, regions in the administrative form of such entities as England’s Regional Development Agencies are regarded as new ‘drivers’ of economic development. Local infrastructural regeneration and enhanced regional competitiveness become the objects of concern. Though such initiatives are characteristic of many European countries (and elsewhere around 132

The politics of the regional question

the world), what remains in doubt is the degree to which these arrangements map onto and reinforce possible patterns of competitive economic activity, could ever serve as catalysts for regional economic development, and have had much of an impact at all in chronically underperforming regions (MacLeod and Jones, 2007: 1182). Perhaps most important of all, from the viewpoint of the practical initiation of the new regionalism, it is by no means clear that the pressure to create new regions is a direct response to new competitive economic conditions either on the part of central governments (except insofar as it could be a ‘sop’ to declining regions) or as a result of local demand for a new administrative tier geared towards attracting inward investment from outside (Cox, 2009). The second body of research and political argument associated with the new regionalism as I have defined it is that which has emerged largely in the United States and Canada and is devoted to linking together the reorganization of social-service delivery and the productive competitiveness of metropolitan regions on the one hand, and changing economic circumstances on the other (e.g. Cox, 1998; Swanstrom, 2001; Jonas and Pincetl, 2006). Implicit in this perspective on the new regionalism is the long and established history of federalism and municipal territorial fragmentation at the metropolitan level in North America, particularly in the United States. From this perspective, local and regional development interests and local politicians focus their energies on managing the local public economy (through consolidating municipal jurisdictions in an historically fragmented metropolis to achieve economies of scale in public goods, imposing rules on building and infrastructure at the metropolitan scale, ‘jumping’ to higher levels such as that of the states or the federal government to bypass political barriers at the metropolitan level), in order to increase the prospects for capital accumulation in an economic context where increasingly mobile capital threatens the returns on historically ‘fixed’ capital in particular places (e.g. Harvey, 1985; Cox, 1998, 2010). The main difference between this rationale for a ‘new regionalism’ and that noted above is that here the focus is on the crucial role of local actors (particularly businesses) in organizing political responses to economic challenges rather than on the notion that the central state is the unique author of new regional arrangements. The emphasis on the territorial politics of local-regional-national relations reflects the view that regionalism is the joint outcome of ‘threats and opportunities external to state space’ and ‘of territorial-distributional politics internal to the state’ (Jonas, 2011: 8). The third notion of the new regionalism builds on ideas about ‘world cities’ and ‘industrial districts’ to argue that under globalization economic competition increasingly depends on geographic access to agglomeration economies that accrue regionally in some places but not in others (e.g. Scott, 1998; Amin and Thrift, 1994; Florida, 1995). As such considerations rise in importance for the location of leading industries, the global city-region (Scott) or mega-region (Florida) is the emerging unit upon which the world economy is being reorganized. From this viewpoint, big cities and their immediate hinterlands form the new regions of the new regionalism. There is neither the focus on regions as the product of some regionalization scheme to counter the depredations of globalization characteristic of the first perspective, nor the emphasis on the bottom-up activities of various actors and interests locked in conflict of the second. Rather, the emerging urban-based regions are seen as heralding a fundamental economic shift with political institutions and processes as assets or barriers to this as the case may be. More or less explicit in this perspective is the view that countries are becoming less central to economic development as competition across the global urban hierarchy replaces interstate economic competition as the main driver of economic growth (McCann and Acs, 2010). In this usage regions are no longer sub-divisions of states as in the other two perspectives but emergent entities that can sprawl across state borders and are defined more by their relations with one another than by their nesting within contiguous territory. These ‘polycentric agglomerations of cities and their lower density 133

John Agnew

hinterlands’ (Florida et al., 2008: 461), are seen as leading to the increasing redundancy of states as economic actors and thus underline the fatuity of attempts, as in the first perspective on the new regionalism, of using regions to staunch the flow of capital and people from declining to flourishing city regions. These three different conceptions of regions in the new regionalism are obviously not readily conjoined. As a result, writing as if there is a singular ‘new regionalism’ is inherently problematic. What they all do share is a tendency to ‘functionalize’ the political as an addendum or supplement to the more profound and directing role of the economic. They reflect understandings of regionalism that effectively depoliticize it when to the extent that it exists in practice it has explicit political roots. None of the three perspectives gives much purchase on the actual course of regionalism. They imagine different regionalisms that could be in the offing as a result of the economic challenges they investigate but none of these actually corresponds to actually existing regionalism in the world as such.

The regional question as a political question The gap between this sort of theorizing about regionalism and the practice of regions as effective political units can be put down to the desire to place the understanding of regions in overarching narratives such as those associated with globalization or Europeanization (Entrikin, 2008). There is certainly something to this. The confusion between regionalization, in the sense of the ascription of powers to new regional agencies as a result of central state initiatives, and regionalism, the increased importance of regions of some sort or another in people’s political and economic life, may also account for part of the problem, particularly in relation to the first perspective (Loughlin, 2000). However, neither of these addresses what to me is the key defect of all varieties of the ‘new regionalism’. This is their reduction of regions and regionalism to end products in a functionalist logic in which economic causes (of various sorts: capital accumulation, inter-firm competition, implanting agglomeration economies, etc.) produce political effects among which are jurisdictional redefinitions, the (presumed) emergence of new governance arrangements at a city-region scale, and regionalist policy initiatives to counter current trends in globalization. Yet overwhelmingly, regionalism as an approach to recognizing and reordering the organization of social and economic life in terms of regions is the result of political initiation by political movements, private interests and/or technocratic agencies. It never simply ‘happens’ to ‘help’ out the reorganization of the world economy or whatever. Indeed, in a searching empirical study across numerous studies and many countries, Treisman (2007) can find only very limited and weak evidence that regional devolution actually produces what economic rationalists would think of as functionally more propitious economic outcomes than centralized administration does. In other words, it does not do what they want. If it were the result of an economic rather than a political imperative its accomplishments would be singularly bad. The regional question, then, is always a political question. After saying more precisely what I mean by this, I turn to criticisms that have been directed at the ‘new regionalism’ (particularly in its first manifestation) from an explicitly ‘political’ standpoint, and then lay out an understanding of regionalism that emphasizes its political origins but that differs from that of other critics of the new regionalism. By a political question, I mean that regionalism is always the fruit of several essentially political processes which often work together but which can be seen analytically and sometimes practically as having distinctive effects. The first is the necessity in reasonably democratic states for governments (and political parties) to be responsive to the demands of electoral constituencies with identities and interests connected to the places where they live and which are 134

The politics of the regional question

often framed in regional terms. One response can be to create institutionalized regional agencies or governments that have powers to spend (and, sometimes, to tax) in pursuit of regionally distinctive goals. However, these are usually the result of bottom-up pressures to cede or share powers from the centre down (e.g. Rokkan and Urwin, 1981; Paasi, 2001; Keating, 1998). The second is the fact that many states are multinational or multiethnic in which different groups are concentrated in different geographical regions within the state. Central governments must frequently attend to the interests of these groups and this can create pressures for devolution in certain circumstances, particularly when regionalist parties arise to make up for the ‘neglect’ of ‘their’ region by mainstream ones (e.g. Tronconi, 2005). Regional policies are frequently inspired by such motivations. ‘They are usually not about either facilitating economic efficiency or redressing economic grievances’ (MacLaughlin and Agnew, 1986: 251). They are about favouring some groups in order to buy their political loyalty or obligation. Even in states without major ethnic-geographical divisions, patronage-based distribution of expenditures to support given political parties can be aided by regional mechanisms when votes are geographically concentrated. Italy’s ‘southern policy’ for many years, for example, followed this logic (e.g. Tarrow, 1977). Third, states, particularly in Europe, are now part of supranational arrangements that facilitate the emergence of regional consciousness and pressures to institutionalize this in the face of the decline of old-style regional ‘stabilization’ policies and the explicit focus of the EU, for example, on regional ‘cohesion’ across its member states even as it adopts a neo-liberal narrative about the necessity of becoming more ‘competitive’ globally through lifting labour-market regulations and reducing the ‘role’ of government (e.g. Agnew, 2001; Murphy, 2008; Faludi, 2011). Finally, various levels of government, from the most local to the national, are faced with challenges concerning immigration, education, pollution, etc., that simply cannot be addressed adequately at a single level. Some of this reflects differences in local social and economic circumstances but some of it also relates to institutional blockages at other levels, leading to increased ‘polyphony’ rather than ‘dualism’ in the allocation of policy arenas across governmental levels (Agnew, 2011). The recent ‘turn’ to regionalism and the emergence of ‘new’ regional arrangements in various countries, therefore, cannot be adequately understood in the economic-functionalist terms of any of the varied strands of the ‘new regionalism’. Most critics of the new regionalism, while generally agreeing with my diagnosis, have tended to see the ‘political’ somewhat differently and in more limited terms from how I have articulated it above. Some tend to see the problem in terms of the ‘territorial’ framing of regions that ‘fixes’ them in space rather than sees them as provisional outcomes of the workings of power networks over space (e.g. Allen and Cochrane, 2007). What this comes down to is much the same as the fourth of my points about regionalism as a political question: the governance of regions, and its spatiality, now works through a looser, more negotiable, set of political arrangements that take their shape from the networks of relations that stretch across and beyond given regional boundaries. The agencies, the partnerships, the political intermediaries, and the associations and connections that bring them together, increasingly form regional ‘assemblages’ that are not exclusively regional, but bring together elements of central, regional and local institutions. (Allen and Cochrane, 2007: 1163) However, that is about it, even though Allen and Cochrane are sure that ‘the language of territorial politics is not only stubborn, but equally that it cannot simply be wished away by some conceptual wand, since it is itself a powerful political construction’ (Allen and Cochrane, 2007: 1162, my emphasis). The ‘new’ world is such, however, that the territorial reification of regions is 135

John Agnew

increasingly passé as we move beyond multi-level governance based on territorialized regions towards relatively open regional assemblages. Yet, and this is my main issue with what is otherwise a sophisticated argument about the multi-scalar and networked constitution of contemporary English regionalization, this is to ignore the four political sources of regionalism I have outlined previously – demand for responsiveness to local economic circumstances, multi-ethnicity, regional cohesion and regional social problems – which usually have quite definite territorial shapes to them. Other critics focus more on the ‘imaginaries’ that sustain discourses about regions and why considerations of ‘the economic’ have tended to be so dominant in them. The idea of the region has been a ‘resource’ in response to particular general political dilemmas seen as arising from globalization and Europeanization rather than a response to popular demand or functional necessity. To Lagendijk (2007: 1202), for example, the EU has been a particularly important ‘mediator’ of ‘genre chains that link ‘competitiveness’ aims to conditions of sustainability and spatial-social cohesion (‘balanced development’)’, which privilege the idea of the activist/resilient region as a way of squaring the circle of the EU’s dual aims of increased economic growth and maintaining local communal bonds. In this light, the politics of regionalism has been constituted as a strategy within the broader framework of EU policies (and those of other actors operating with broad geographical scope) which attempt to respond positively to the exigencies of uneven economic development. They are thus ‘accidents’ more than deliberate constructions reflecting specific geographical origins. While this argument draws attention to the significance of the wider contemporary framework in which debate about regions and regionalism is embedded (particularly that of the EU), it bypasses all but the third of the points I have made above. This conception of the politics of regionalism ends up being an entirely top-down affair. A third criticism is addressed not to the first two perspectives in the new regionalism but to the third: that of so-called city-regions and mega-regions. The basic elements in the critique are that the world in which cities exist has been seen in this perspective entirely as a world economy in which states (of whatever efficacy) are largely (and increasingly) absent and that the cities themselves have been represented overwhelmingly as nodes for global law firms, segmented labour markets and other economic functions rather than ‘meaningful built environments’ (Therborn, 2011: 272). The world of city-regions thus reduces to the world of ‘business geography’ in which a global scale of business transactions and investments determines everything else. However, as Therborn (2011: 280) points out, as of 2005 (before various national governments came to the rescue of their city-based banking and insurance sectors in 2008–09), the proportion of gross domestic product (GDP) accounted for by governments in the major developed capitalist countries was either much the same or greater than it had been in 1974–76. Since 2008 these proportions have undoubtedly increased. Moreover, national politics has lain behind the rise of finance as a dominant sector in many world cities. Beginning in the 1980s, many national governments adopted the neo-liberal economic policies that have essentially licensed and, as we have seen in recent years, underwritten, the ‘booms’ that led to the hubris about the rise of world cities and the demise of nation-states. Of course, others did not (China, for example), suggesting that political contingencies at the level of the nation-state still heavily condition the likelihood that ‘mega-regions’ or city-regions will ever amount to anything as politically autonomous units. Today they remain aggregations of populations on maps and clusters of lights in satellite images on clear nights. Yet, some big cities have become increasingly metropolitan in their territorial organization and in the case of London, for example, have acquired a power in relation to the wider national territory that suggests something of an historic shift. What seems undeniable is that the rise of global city-regions does not necessitate a parallel demise of the nation-states in which they are embedded. 136

The politics of the regional question

A political conception of regionalism as it is found today in the world rather than in the academic literature must have a number of components. First of all, regionalism must be distinguished clearly from the usage of regions as economic territories or places with the politicaladministrative character of the regions simply read of from their putative economic determinants. Now if one’s primary interest lies in explaining which industries cluster where, and even addressing how local/regional institutional arrangements either encourage or discourage this, then there is no problem (e.g. Storper, 1997). It is, rather, the claim that a politically responsive regionalism of whatever variety is underwritten by a singular economic determination (whatever its precise nature) that is inherently problematic. Second, it is important to distinguish between regionalism as a category of analysis and as one of practice. This distinction between a concept as a sociopolitical practice and as a theoretical term, made most recently by Brubaker and Cooper (2000) in their discussion of the concept of ‘identity’, is important because otherwise the two types of usage can become totally confused. This has happened to ‘regionalism’ in the new regionalism literature where the absence of much evidence for regionalism in practice in many places has excited little attention while much has been made of how regionalism can be construed analytically as necessarily and ideally following from the broad economic trends flowing from the contemporary evolution of the world economy and global capitalism. Third, it is useful to place the discussion of regionalism within the wider debate in geography and other fields about geographic scale and networks (Jonas, 2006; MacLeod and Jones, 2007; MacKinnon, 2011). Regions are a type of place or geographic area that never exist in isolation and are defined as much by their co-existence with other places as by their boundedness and presumed internal homogeneity or functional ties. There is no reason to see them in either/or terms as entirely either territorial (the traditional sense of regions) or relational (in the sense that their openness mitigates against the identities and interests that they can define for their residents and others). Indeed, from a political viewpoint it is precisely their capacity to elicit group identities and interests and affect socio-economic practices that makes them of any concern whatsoever (e.g. Agnew, 1995; Oakes, 2000; Harvey et al., 2011). This means, however, and this is a fourth point, that regions in the context of regionalism need not always actually exist in practice. They only come about and matter if there are meaningful efforts and outcomes in terms of realizing them as political projects of one sort or another (e.g. Hooghe et al., 2008). Fifth, as such, regionalism is not a fixed or recurring feature of even the places where it has been most developed. It can fade, as with Quebec regionalism in Canada or Irish republicanism in Northern Ireland, for example. The conditions under which it flourishes are historically and geographically specific (Keating, 1997). Grievances can be overcome and new patterns of authority and patronage established with previously despised seats of power. The discourse of regionalism on a wider scale is also historically contingent, as is evident with recent trends in the EU to downplay a prior emphasis on ‘the Europe of the regions’ for a much weaker redistributive regime of territorial politics, as the EU has become more neo-liberal and less corporatist in its economic orientation (Elias, 2008). Finally, we should be wary of writing off the nation-state as an increasing irrelevance in a world seemingly undergoing fundamental territorial/spatial reorganization. The totalistic rhetoric of much of the new regionalism literature has an apocalyptic tone to it. Presumably having a simple deterministic message helps get one’s publications into print. Be that as it may, the world is not all motion and no fixity or forever in flux. All that is solid is not suddenly melting into air. State borders have if anything ‘hardened’ in recent years. To the extent that regionalism occurs, it is because of negotiations in and around long-established governmental institutions (Agnew, 1995; Loughlin, 2009). The national landscape of power has deeper and more persistent roots than the new regionalists and some of their critics have imagined could ever be the case. 137

John Agnew

Conclusion It is the politicization of particular regional spaces that begets regionalism. To understand regionalism, therefore, requires understanding this politicization. The confusion about this evident in the literature labelled the ‘new regionalism’ has been the primary topic of this chapter. Seemingly any manifestation of regions as economic or other units is held to signal regionalism. I have detailed some of the strands in this literature, identifying their differences but also noting the degree to which they share a number of crucial assumptions, particularly an all-too typical economic functionalism that has come to hold sway over much of what goes for regional studies in fields such as geography and sociology. Some useful criticisms from a political viewpoint tend, however, to be partial rather than complete in emphasizing the cross-scale and networked nature of territorial politics, discourses of regionalism as responses to policy dilemmas at national and supranational levels, and the neglect of the centralized state. My main thrust has been to argue for regionalism as a political phenomenon that can be understood satisfactorily only in terms of the pluralist vision enunciated first by Keating but elaborated on here in terms of my posing of the regional question as a political one and in terms of six components of a political conception of regionalism: regionalism versus functional economic regions; regionalism as category versus as practice; regions and geographic scale; regions as parts of political projects; the historical rise and fall of regionalisms; and regions and the persistence of statehood. The empirical evidence about the incidence and causes of regionalism, at least in Europe and North America, suggests strongly that what we are seeing today is the creation of various types of ‘hybrid’ state with the co-existence in varied ways between national and local/regional modes of government (and governance involving private and public but non-state actors) which simply cannot be explained by some singular theoretical logic based on the dominance of economic processes over all others (Loughlin, 2009; Agnew, 2011).

References Agnew, J.A. (1995) ‘The Rhetoric of Regionalism: The Northern League in Italian Politics, 1983–94’, Transactions of the Institute of British Geographers Vol. 20, No. 1: 156–72. ——(2001) ‘How Many Europes? The European Union, Eastward Enlargement and Uneven Development’, European Urban and Regional Studies Vol. 8, No. 1 : 29–38. ——(2011) ‘Dualisme contre polyphonie dans la gouvernance territoriale contemporaine’, in G. Bettoni (ed.) Gouverner les territoires: antagonismes et partenariats entre acteurs publics (Paris: IGPDE, Ministry of Finance). Allen, J. and A. Cochrane (2007) ‘Beyond the Territorial Fix: Regional Assemblages, Politics and Power’, Regional Studies Vol. 49, No. 9: 1161–75. Amin, A. and N. Thrift (1994) ‘Living in the Global’, in A. Amin and N. Thrift (eds) Globalization, Institutions, and Regional Development in Europe (Oxford: Oxford University Press). Brenner, N. (2004) New State Spaces: Urban Governance and the Rescaling of Statehood (Oxford: Oxford University Press). ——(2009) ‘Open Questions on State Rescaling’, Cambridge Journal of Regions, Economy and Society Vol. 2, No. 1: 123–39. Brubaker, R. and F. Cooper (2000) ‘Beyond ‘Identity’’, Theory and Society Vol. 29, No. 1: 1–47. Cox, K.R. (1998) ‘Spaces of Dependence, Spaces of Engagement and the Politics of Scale, or, Looking for Local Politics’, Political Geography Vol. 17, No. 1: 1–23. ——(2009) ‘‘Rescaling the State’ in Question’, Cambridge Journal of Regions, Economy and Society Vol. 2, No. 1: 107–21. ——(2010) ‘The Problem of Metropolitan Governance and the Politics of Scale’, Regional Studies Vol. 44, No. 2: 215–27. Elias, A. (2008) ‘Whatever Happened to the Europe of the Regions? Revisiting the Regional Dimension of European Politics’, Regional and Federal Studies Vol. 18, No. 5: 483–92. Entrikin, J.N. (2008) ‘Introduction’, in J.N. Entrikin (ed.) Regions (Farnham: Ashgate).

138

The politics of the regional question

Faludi, A. (2011) ‘La cohesion territoriale à la croisée des chemins’, in G. Bettoni (ed.) Gouverner les territoires: antagonismes et partenariats entre acteurs publics (Paris: IGPDE, Ministry of Finance). Florida, R. (1995) ‘Toward the Learning Region’, Futures Vol. 27: 527–36. Florida, R., T. Gulden and C. Mellander (2008) ‘The Rise of the Mega-Region’, Cambridge Journal of Regions, Economy and Society Vol. 1: 459–76. Harvey, D. (1985) ‘The Geopolitics of Capitalism’, in D. Gregory and J. Urry (eds) Social Relations and Spatial Structures (London: Macmillan). Harvey, D.C., H. Hawkins and N.J. Thomas (2011) ‘Regional Imaginaries of Governance Agencies: Practising the Region in South West Britain’, Environment and Planning A Vol. 43, No. 3: 470–86. Hooghe, L., A.J. Schakel and G. Marks (2008) ‘Appendix A: Profiles of Regional Reform in 42 Countries’, Regional and Federal Studies Vol. 18, No. 2: 183–258. Jessop, B. (2002) The Future of the Capitalist State (Cambridge: Polity). Jonas, A.E.G. (2006) ‘Pro Scale: Further Reflections on the ‘Scale Debate’ in Human Geography’, Transactions of the Institute of British Geographers Vol. 31, No. 3: 399–406. ——(2011) ‘Region and Place: Regionalism in Question’, Progress in Human Geography 35. Jonas, A.E.G. and S. Pincetl (2006) ‘Rescaling Regions in the State: The New Regionalism in California’, Political Geography Vol. 25, No. 5: 482–505. Keating, M. (1997) ‘The Invention of Regions: Political Restructuring and Territorial Governance in Western Europe’, Environment and Planning C Vol. 15, No. 3: 383–98. ——(1998) The New Regionalism in Western Europe: Territorial Restructuring and Political Change (Cheltenham: Edward Elgar). Lagendijk, A. (2007) ‘The Accident of the Region: A Strategic Relational Perspective on the Construction of the Region’s Significance’, Regional Studies Vol. 41, No. 9: 1193–208. Loughlin, J. (2000) ‘Regional Autonomy and State Paradigm Shift in Western Europe’, Regional and Federal Studies Vol. 10, No. 1: 10–34. ——(2009) ‘The ‘Hybrid’ State: Territorial Governance in Western Europe’, Perspectives on European Politics and Society Vol. 10, No. 1: 51–68. McCann, P. and Z.J. Acs (2010) ‘Globalization: Countries, Cities and Multinationals’, Regional Studies Vol. 45, No. 1: 17–32. MacKinnon, D. (2011) ‘Reconstructing Scale: Towards a New Scalar Politics’, Progress in Human Geography Vol. 35, No. 1: 21–36. MacLaughlin, J.G. and J.A. Agnew (1986) ‘Hegemony and the Regional Question: The Political Geography of Regional Industrial Policy in Northern Ireland, 1945–72’, Annals of the Association of American Geographers Vol. 76, No. 2: 247–61. MacLeod, G. (2001) ‘New Regionalism Reconsidered: Globalization and the Remaking of Political Economic Space’, International Journal of Urban and Regional Research Vol. 25, No. 4: 804–29. MacLeod, G. and M. Jones (2007) ‘Territorial, Scalar, Networked, Connected: In What Sense a ‘Regional World’?’ Regional Studies Vol. 41, No. 9: 1177–91. Massey, D. (1984) Spatial Divisions of Labour (London: Macmillan). Murphy, A.B. (2008) ‘Rethinking Multi-level Governance in a Changing European Union: Why Metageography and Territoriality Really Matter’, GeoJournal Vol. 72, No. 1: 7–18. Oakes, T. (2000) ‘China’s Provincial Identities: Reviving Regionalism and Reinventing “Chineseness”’, Journal of Asian Studies Vol. 59, No. 3: 667–92. Paasi, A. (2001) ‘Bounding Spaces in a Mobile World: Deconstructing ‘Regional Identity’’, Tijdschrift voor Economische en Sociale Geografie Vol. 93, No. 1: 137–48. Rokkan, S. and D.W. Urwin (eds) (1981) The Politics of Territorial Identity (London: Sage). Scott, A.J. (1998) Regions and the World Economy: The Coming Shape of Global Production, Competition and Political Order (Oxford: Oxford University Press). Storper, M. (1997) The Regional World: Territorial Development in a Global Economy (New York: Guilford). Swanstrom, T. (2001) ‘Regionalism Reconsidered’, Journal of Urban Affairs Vol. 23, No. 2: 467–78. Tarrow, S. (1977) Between Center and Periphery: Grassroots Politicians in Italy (New Haven: Yale University Press). Therborn, G. (2011) ‘End of a Paradigm: The Current Crisis and the Idea of Stateless Cities’, Environment and Planning A Vol. 43, No. 1: 272–85. Treisman, D. (2007) The Architecture of Government: Rethinking Political Decentralization (Cambridge: Cambridge University Press). Tronconi, F. (2005) ‘Ethnic Identity and Party Competition. An Analysis of the Electoral Performance of Ethnoregionalist Parties in Western Europe’, World Political Science Review Vol. 2, No. 2: 137–63. 139

10 Economic regionalism in federal and hybrid systems of government Pieter van Houten

Introduction Economic factors play an important role in the territorial politics of federal and regionalized states. This chapter focuses on one specific set of economic developments.1 I use the term economic regionalism2 to denote a process by which the economic importance of the sub-national regional level has increased in recent decades in the context of structural changes to capitalist economies. More specifically, it refers to ‘the emergence of new localized production systems of specialized industrial agglomerations, as part of a more general ‘resurgence’ of regions and cities as the loci of contemporary economic development and governance’ (Asheim et al., 2006: 1). These developments are sometimes referred to as ‘new regionalism’ (e.g. MacLeod, 2001). This term, however, is more usefully employed by Keating et al. (2003: 6) to describe the ‘rise of new territorial frameworks for action below the state, often linked into transnational networks’, with economic regionalism (as defined here) being one of the contexts in which these frameworks develop. In this chapter, I focus primarily on the political and institutional implications of this context for federal and hybrid systems of government. The literature describing and analyzing the increased economic importance of the regional level is diverse, although most prominent in economic geography, and has expanded rapidly in recent years (Cruz and Teixeira, 2010; Asheim et al., 2011). The phenomenon has also featured prominently in policy debates (Asheim et al., 2006), as exemplified by its inclusion in reports and activities by international organizations. For example, the World Bank emphasized it in one of its recent World Development Reports (World Bank, 2009), the European Union’s (EU) discussions about regional policy reforms are informed by findings from this literature (Barca, 2009), and the Organization for Economic Co-operation and Development (OECD) has focused extensively on the role of regions in economic development (e.g. OECD, 2005, 2011). One of the arguments of this chapter, however, is that – despite this extensive literature and attention – the political and institutional effects of economic regionalism have not been sufficiently investigated. By the late 1990s, the ‘resurgence of regional economies’ was widely asserted. The first part of the chapter summarizes the alleged sources of this resurgence and some of the ongoing debates about the role and position of regions in the contemporary global economy. The second part discusses the possible political and institutional implications of economic regionalism for federal 140

Economic regionalism

and hybrid systems of government. I focus on the potential effects of these economic dynamics on political regionalism (the emergence or resurgence of regional movements or parties demanding more regional powers), the content and process of regional development policies, and the general territorial structures of states. This chapter reviews some claims and evidence about these effects (relying mostly on examples from Western Europe) and indicates several remaining questions.

The nature and sources of economic regionalism Economic regionalism is an umbrella term for a variety of claims that sub-national territories – or, at least, some sub-national territories – have gained increased economic importance in recent decades. From different disciplinary angles, these claims have been made in literatures in economics, geography, sociology and political science.3 The terms used to describe the economically relevant sub-national units are manifold, and include industrial districts, clusters, regional economies, local or territorially embedded production systems and regional innovation systems. Not surprisingly, definitions of these sub-national economic units vary, but there are some common, or at least widely shared, elements. These include the dominance of one or a few industries or sectors, the presence of a variety of small and medium-sized firms, complex patterns of competition and coordination between these firms, and spatial proximity of the firms (with a significant proportion of economic relations occurring between economic actors within this space). For example, Zeitlin (2007: 223) defines an industrial district as ‘a geographically localized productive system based on an extended division of labor between small and medium-sized firms specialized in distinct phases of complementary activities within a common industrial sector’. Alternatively, in more expansive definitions, Porter (1998: 197–98) defines clusters as ‘geographic concentrations of interconnected companies, specialized suppliers, service providers, firms in related industries, and associated institutions (for example, universities, standards agencies, and trade associations) in particular fields that compete but also cooperate’, while Whitford and Potter (2007: 498) speak of ‘territorially embedded networks of firms enmeshed in a tangle of cooperative, competitive and social relations’. The paradigmatic examples of successful regional economies are the industrial districts in the centre, north and north-east of Italy (the so-called ‘Third Italy’), where clusters of tightly connected small firms carved out globally dominant positions, mostly in design-based industries related to fashion goods (such as clothing, leather goods and furniture), but also in machinery and metalworking sectors. These regional economies have been widely analysed (for references, see Zeitlin, 2007; Whitford and Potter, 2007), and some of the used concepts, in particular the adaptation of Marshall’s notion of ‘industrial districts’ to modern economic agglomerations, are directly based on these analyses. Other frequently mentioned examples of modern industrial districts and clusters include parts of southern Germany (in the regions of Baden-Württemberg and Bavaria), especially in the machine and tools industries, but also in the media industry; technology districts in southern Paris and metalworking districts in the Rhône-Alpes region in France; the software and other technological industries in Silicon Valley and the media industry clusters around Los Angeles in the United States; machine industry districts near Osaka in Japan; and – when applying the notion of ‘clusters’ more broadly – financial districts such as the City of London and service-sector clusters as in the Randstad in the Netherlands. Many other, if less prominent, examples in advanced industrial states could be added to this list (e.g. Crouch et al., 2001, 2004). Economic regionalism also occurs in non-Western and developing countries. For example, Zeitlin (2007: 233) mentions the proliferation of local industrial clusters in the Pearl River and Yangtze delta regions in China, and some of the ‘specialized economic zones’ in 141

Pieter van Houten

India and China exhibit similar features. Scott and Garofoli (2007) analyse regional economic clusters in ‘emerging economies’ such as China, Brazil, Vietnam, Morocco and South Africa. Spatially concentrated clusters of industries and territorial variation in economic activity are not new phenomena (Zeitlin, 2007). Marshall’s (1927) original analysis of ‘industrial districts’ and its application to British industrial history testifies to this, while, for instance, Herrigel (1996) demonstrates this for Germany. However, and ignoring considerable variation across countries, it can be said that advanced industrial economies became more centralized and homogeneous in the period after the Second World War through the implementation of Keynesian economic and regional development policies, the growth of welfare states and the centralization of state structures (Keating, 1998). It is against the background of these more territorially homogeneous economies and policies that we can speak of the ‘economic resurgence of regions’ from the 1970s on. Most analyses point to two related and reinforcing developments in global capitalism as the context for this resurgence. The first of these developments is the alleged crisis of an economic system organized around mass production (‘Fordism’) and large, vertically integrated firms, and its gradual replacement – at least in advanced industrial states – by post-Fordist arrangements characterized by ‘flexible specialization’ (Piore and Sabel, 1984). In this changed context, firms and industries need to be flexible and innovative to deal with evolving market opportunities, consumer demands and technological change. The second development is globalization (e.g. Storper, 1997; Scott, 1998), reinforced for European countries by European integration (van Houten, 2003; Keating et al., 2003). Globalization, applied here primarily in its economic manifestations of increases in capital and product flows beyond national markets, creates both opportunities and challenges for firms and regional economies, and makes innovation potential and flexibility crucial requirements for economic success. What makes the sub-national level and various forms of regional economic clusters so important and potentially successful in this context of post-Fordist production and globalization? The basic claim of the analyses of the economic resurgence of the regional level is that – contrary to the common intuition that globalization makes the specific location of economic activity less important – geographical proximity and clustering provides crucial conditions for flexibility, specialization and innovation. In some industries (such as in high technology and design sectors), smaller and vertically disintegrated firms have the potential to adapt quickly to changes in technology and market conditions. Such firms, in turn, need to rely on various ‘external economies’ (Whitford and Potter, 2007) – that is, goods and factors that individual firms cannot provide themselves and need to obtain from the (spatially) concentrated network or environment in which they are embedded. Some accounts focus primarily on economic goods and factors such as labour skills, technological knowledge and the availability of specialized suppliers, all of which are likely to be subject to economies of scale (e.g. Krugman, 1991). While not denying the importance of these factors, many other analyses emphasize the role of socio-cultural goods and factors, partly taking their cues from Marshall’s focus on the role of ‘industrial atmosphere’ in districts. These factors – denoted by terms such as ‘conventions’ and ‘untraded dependencies’ (Storper, 1997), ‘social capital’ (cf. Putnam, 1993) or simply ‘culture’ (Keating et al., 2003) – refer to issues such as tacit knowledge about technology and production methods, trust among economic actors as a result of informal practices, repeated interactions over time, and regional cultural features and shared identities. Some analyses emphasize the role of ‘local collective competition goods’ (Crouch et al., 2001) and the ‘regulation of local economic systems’ (Scott, 1998). These include some of the factors mentioned above, but also point to the importance of more formal ‘intermediate’ institutions that can include business associations, unions and local or regional governments, and which are influenced by public policies and national institutions. As Zeitlin summarizes, a successful district or cluster: 142

Economic regionalism

must develop a set of coordination and governance mechanisms capable of checking opportunistic behavior without stifling fluid cooperation among decentralized economic actors. Crucial in this regard are institutions for the resolution of disputes and the provision of collective services beyond the capacity of the individual small and medium-sized firms to supply for themselves, such as training, research, market forecasting, credit, and quality control. (Zeitlin, 2007: 226) Some brief examples illustrate the various ways in which this general logic plays out. In the famous industrial districts in northern Italy, the formation and reproduction of dense clusters of small firms have been facilitated by the vertical division of labour – rather than direct competition – between many of these firms, the availability of skilled workers trained by technical schools, the presence of a network of local banks with close ties to the community, and various associations involving local and regional government actors (Storper, 1997; Whitford and Potter, 2007). In addition, extended family structures and close interpersonal ties have played an important role in providing stable supplies of labour and facilitating coordination within networks. In Silicon Valley links to universities have played an obviously crucial role in the emergence and success of the technology cluster. However, social networks between skilled workers and entrepreneurs, deregulated labour markets, the presence of venture capitalists providing investment and financial knowledge, and formal business associations working with local and regional governments have also been important in sustaining the regional economy (Saxenian, 1994; Whitford and Potter, 2007). In Baden-Württemberg, some similar factors are in play, but the crucial role of policies and regulations provided by the regional government, and public-private bodies dealing with training and technology transfer, stand out in comparison to other regions (Cooke, 1997). If successful, a spatially concentrated economy generates a ‘localized virtuous circle that links together intra-regional divisions of labor, specialization, social learning, innovation, and economic growth’ (Scott, 1998: 160). According to various studies (e.g. Porter, 2003; Whitford and Potter, 2007: 503; Spencer et al., 2010), regions with well-developed industrial districts or economic clusters indeed perform well. However, only certain regions will benefit from economic regionalism (Zeitlin, 2007). Only some regions will have – or be able to shape – the economic, social and institutional conditions indicated above, and develop the appropriate ‘development coalition’ (Keating et al., 2003). This is acknowledged in most writings on the subject but is worth emphasizing, as it has implications for an analysis of the political consequences of economic regionalism. This section only scratches the surface of the vibrant literature on spatially concentrated economies. It ignores many interesting issues of debate in this literature, such as the historical roots and antecedents of industrial districts (Zeitlin, 2007), challenges to clusters as a result of the further intensification of globalization (Whitford and Potter, 2007), the role of cities and city-regions in the territorialization of economic development (Scott, 2001; Le Galès, 2002; Rodríguez-Pose, 2008), conditions for learning and innovation in regional economies (Morgan, 1997; Cooke et al., 2007; Asheim et al., 2011), the role of larger firms and branches of multinational companies in economic clusters (Crouch et al., 2001, 2004; Whitford and Potter, 2007), the relevance of transnational networks for regional economies (Saxenian, 2006), and the design and effects of policies to create and sustain economic clusters (Casper, 2007). However, this basic overview suffices for the main focus of this chapter – namely, the political and institutional consequences of economic regionalism.

Political and institutional effects of economic regionalism If there is some truth to these analyses about the increased economic importance of the sub-national level (as the amount of theoretical and empirical research certainly suggests), then we would 143

Pieter van Houten

expect this to have political and institutional consequences. Given that policies and the provision of certain public goods can facilitate the emergence and sustainability of sub-national economic clusters, and that these would need to be geared towards specific local circumstances, there is an incentive to give appropriate sub-national levels of government the responsibilities and powers to provide these policies. Thus, there are reasons to think that economic regionalism will induce or reinforce decentralization, or at least create pressures and demands for this, especially in the advanced industrial states which became relatively centralized in the post-war period. Moreover, the economically relevant sub-national territories will often not coincide with existing political and administrative regions, cities or localities. This can create further pressures for changes in the territorial organization of states. There is no shortage of claims about the possible political and institutional impact of the economic resurgence of regions and urban areas. At its most extreme, Ohmae (1995) argues that the economic importance of (sub- and transnational) regions makes states increasingly irrelevant and likely to disappear. In a more nuanced claim, Scott (1998: ch. 8) argues that a wide variety of new governance arrangements, which involve public authorities and which he terms ‘regional directorates’, will emerge at the sub-national level to deal with evolving economic challenges and opportunities. Similarly, Le Galès and Trigilia (2004: 343) state that the ‘territorialization of economic activities requires more local governance’, and Keating (1998: 74) indicates that ‘there is a search for new mechanisms for managing the impact of economic change on territories, focused more on the contribution of regions themselves and less on the directing and planning policies of the state’. More abstractly, a strand of literature argues that the sub-national economic dynamics resulting from processes of flexible specialization and globalization have induced a ‘rescaling’ of the state (e.g. Brenner, 2004, 2009; Lobao et al., 2009). As concrete manifestations of this, these authors point to changes in the content of planning policies and the emergence of new urban governance arrangements. However, some of these claims have a functionalist flavour (Keating, 1998) and the empirical record is ambiguous. States certainly have not disappeared yet; Scott (1998) himself admits that his claim about institutional changes is largely speculative and not yet realized, which is still a valid assessment; and the institutional and governance changes at the regional and urban level mentioned by the ‘state rescaling’ literature have often been rather limited (Cox, 2009). This is not to deny that there have been significant changes to territorial politics and the territorial organization of states (as the chapters in this volume make clear), but other factors may have been the main drivers of this. So what exactly have been the empirical political and institutional consequences of economic regionalism? Despite the large volume of work on these economic dynamics, this question has received surprisingly limited attention. In this section, I suggest some possible and tentative answers to this question, and indicate some specific issues that would benefit from further research. This section is divided into two parts. The first focuses on the possible effects of the renewed economic importance of regions on regional political mobilization, while the second evaluates its institutional impact.

Economic regionalism and political mobilization Political regionalism, defined here as political movements and parties demanding more powers (ranging from forms of autonomy to independence) for sub-national regions, has been on the rise in advanced industrial states in recent decades. What role, if any, have economic developments enhancing the importance of the regional level played in this? Economic regionalism seems to have played a significant role in the rise of new regionalist political movements. The most salient new movement in Europe in recent years has occurred in 144

Economic regionalism

northern Italy, where the Northern League emerged as a significant political party in the early 1990s, has sustained itself since then, and has participated in the government coalitions led by Berlusconi. The party’s programme has oscillated over time (from demands for federalism to secession and back to federalism again), but the common threads have been demands for more powers for the northern regions, a reduction in financial flows to the south and protection from an inflow of immigrants. The rise and success of this party can be directly linked to economic regionalism. Importantly, the party ‘has long seen itself as the voice of the small-business sector in the industrial districts of northern Italy’ (Agnew, 2002: 164). Voting analyses indicate that the party’s support reflects this focus. In Veneto, a region known for its various successful industrial districts and one of the strongholds of the Northern League, the party’s support in the 1990s was strongest in areas where small firms are prevalent and where it managed to tap into voters’ ‘increasing anger at the state’s failure to provide what they perceive as necessary services to local industries at a time when international competition in their sectors is increasingly intense’ (Agnew, 2002: 153). Similarly, in the Lombardy region, another of the party’s strongholds, the party’s support is strongest in areas where industrial districts developed in recent decades. New forms of political regionalism, although in a relatively weak form, also emerged in Germany in the 1980s and 1990s, as political actors in the southern German regions of Bavaria and Baden-Württemberg voiced demands for a more ‘competitive’ form of federalism, with more powers and autonomy for the regions (Jeffery, 1999). These regions recently have been economically highly successful, which is partly attributable to the type of districts and clusters discussed in this chapter, and a desire to keep its regional economies globally competitive undoubtedly played a role in the demands for federalism reforms emanating from these regions (Deeg, 1996). Similarly, French regions such as Rhône-Alpes, which host successful economic clusters of smaller and globally competitive firms, have also voiced some demands for more control over economic and industrial policies. Aside from these examples, however, it is not clear that there is a large upsurge in new regionalist movements; instead, political regionalism has been reinforced in regions where it was already present (van Houten, 2003). Many advanced industrial states have some (relatively) successful economic clusters, but not all regions in which these clusters or districts are located have witnessed the rise of political regionalist movements. The strongest movements can be found in territories such as Catalonia, the Basque Country, Quebec, Flanders and Scotland, where political mobilization for more regional autonomy or independence has a long history, which predates the economic regionalist dynamics discussed here. However, the nature of these movements arguably has evolved in recent decades, and changes in the economic position of these regions in the global economy, including an increased importance of networks of small and medium-sized firms, appear to have played a role in most of these cases. Keating (1998) argues that these movements’ demands for independence and autonomy have been replaced slowly by a ‘post-sovereign’ focus on improving regional economic and social performance, although the insistence of the current Scottish government led by the Scottish National Party (SNP) on independence as its ultimate goal qualifies this to some extent. There is no doubt, however, that some assertive regions have emphasized the importance of the conditions that make spatially concentrated economic clusters successful. In turn, economic success, partly resulting from this emphasis, has fuelled political demands for more powers and autonomy. Flanders and Catalonia provide good examples of these developments (Keating et al., 2003). An important qualification to these points about the link between economic and political regionalism is that other factors also determine the rise and development of political regionalism. This applies both to the older (where, in particular, the presence of politically mobilized regional or nationalist identities are important) and newer regionalist movements. For example, 145

Pieter van Houten

the Northern League’s anti-immigrant stance and larger developments in the Italian party system have both been significant factors in the party’s fluctuating fortunes. Moreover, there is a more general link between the economic position of regions and political regionalism. Other things being equal, regions that are relatively rich (within their countries) are more likely to give rise to regional parties and regionalist activity than poorer regions (Sorens, 2005; van Houten, 2007). The presence of regional industrial districts or clusters can be, and in some cases is, one reason for relative regional wealth, but other possible reasons include the presence of successful traditional industries, multinational companies, regional resource endowments, or tourism income. In addition, the ‘mechanism’ by which relative economic strength translates into political regionalism is often a desire to reduce public financial flows to poorer parts of the country. This is the case for ‘ethnoregionalist’ movements in regions such as Flanders and Catalonia, and plays an important role in newer instances of regionalism in Italy and Germany, where the Northern League appeals for the reduction of transfers to southern Italy, and the southern German regions try to reduce transfers to the northern city states and the eastern regions. Thus, the underlying motive linking economic to political regionalism seems to be largely distributive rather than a general concern about improving economic development (Keating, 2001: 378; Cox, 2009). This is likely to have implications for the response to regionalist demands and activities by other political actors, and for the likelihood that political regionalism attains its goals.

Economic regionalism and the territorial organization of the state To what extent has economic regionalism induced changes in the distribution of powers across territories in federal and regionalized states? Such changes can be the outcome of successful political mobilization by regional economic and political actors, but can also result from ‘top down’ initiatives by national or supra-national actors. Thus, while there is some overlap with the topic of the previous subsection, this topic requires discussion in its own right. A direct effect of economic regionalism can be expected in the area of regional development policy. In many countries, the stimulation of local and regional districts and economic clusters has become an increasingly important objective of these policies, as a response to the success of some regions and the advice of consultants and international economic organizations (Asheim et al., 2006). This required an adjustment of existing policies, which were often geared towards the equalization of territorial development and the stimulation of traditional industries (Keating, 1998; Brenner, 2004). In order to tailor these policies towards the needs of specific territories, more involvement of local and regional levels of government in their design and implementation seems appropriate. Economic regionalism has undoubtedly induced changes to the content of regional development policies. As McDonald and Belussi (quoted in Whitford and Potter, 2007: 499) note, ‘the widespread occurrence of the phenomenon of geographical concentrations of firms has led local, regional, national and international governmental agencies to create and implement policies to boost competitiveness and to help regional economies to develop by encouraging firms to form and build up regional districts/clusters’. Regional development policies have become more supply-side oriented, focusing on issues such as training, infrastructure provision, and research and development stimulation rather than ‘Keynesian’ policies evolving around direct subsidies and attempts to harmonize economic development across territories (Keating, 1998). This trend is visible in all states, and is in Europe reinforced by some of the competition and cohesion policies of the EU. Beyond these general trends, the specific policies and institutional arrangement for policy making are manifold and vary by country (Keating, 1998: 157–60), and are impossible to discuss in detail here. However, the following general developments and examples give some flavour of the changes that have occurred. 146

Economic regionalism

One significant development is the proliferation of public-private initiatives at the regional level aimed at providing governance mechanisms for regional economies. Baden-Württemberg, which established a network of centres for technology advice and technology transfer coordinated by the public-private Steinbeis Foundation (Sturm, 1995), is often mentioned as a particularly successful example, and other regions have tried to emulate this. The Regional Development Agencies established in England in 1998 can also be seen as an example (though less successful and now abandoned) of such bodies. Scott (1998: 145–48) mentions further examples from Europe and the United States, while Keating et al. (2003) discuss arrangements in selected regions in Belgium, France, Spain and the United Kingdom (for other examples, see Crouch et al., 2001, 2004). To some extent these developments have led to more decentralization and regional autonomy. If these bodies have some power and regional governments participate in them, then the power of regions seems to have increased. More broadly, local and regional governments and agencies appear to have gained more authority in regional policy and related policy areas in many countries. This is certainly the case in formerly unitary states such as France and Spain. For example, in Spain regional governments increasingly have taken responsibility for science, technology and innovation policies (Sanz-Menéndez and Cruz-Castro, 2005). It can also be observed in federal states such as Germany, where regional governments gained more authority and initiative in industrial and regional policy in the course of the 1980s and early 1990s (Deeg, 1996). Decentralization and increases in sub-national powers are, however, not the only institutional consequences of the changes to regional development policies. The other, and arguably more pervasive, effect has been higher complexity in the policy-making process, with involvement and (attempts at) coordination between political actors and agencies from different levels of government (Le Galès and Trigilia, 2004). For example, in Germany the changes in regional development and innovation policies appear to have increased the need for and occurrence of coordination between federal, regional and local government agencies (Wilson and Souitaris, 2002; Fürst, 2006). In a very different institutional context, Pike and Tomaney (2009: 30) argue that the UK’s attempt to change and upgrade its regional development policy reveals ‘a picture of complexity, experimentation, fragmentation and incoherence in the UK’s state rescaling of the governance of economic development’. Similar experiences of higher levels of complexity and coordination, albeit it with considerable variation in effectiveness, can be observed in other countries (see, for example, the case studies in Keating et al., 2003). In the European context, the role of the EU and its regulations and policies have further contributed to the development of ‘multi-level governance’ (Hooghe and Marks, 2001). The influence of the EU in stimulating multi-level and multi-layered coordination mechanisms is particularly noticeable in the area of regional policies. Benz and Eberlein (1999) discuss how this played out in Germany and France in the 1990s, and this trend in EU member states has continued. Thus, the developments described here, which have been induced partly by the territorialization of economic development, have led to a more ‘hybrid’ system of policy making (Loughlin, 2009). More general claims that economic regionalism, as a significant aspect or stage of contemporary capitalism, has led to the transformation or ‘rescaling’ of states are harder to evaluate empirically. Some of these claims refer to long-term developments that may not have occurred yet, while others are focused on more subjective factors related to imaginations and perceptions of the nature of the state and its manifestations on different scales (Agnew, 2002; Brenner, 2004), which may not be directly reflected in institutional changes. As far as concrete institutional changes are concerned, states in the developed world have, on average, become more decentralized in recent decades, although there is significant variation between countries (Ansell and Gingrich, 2003; Hooghe et al., 2010; Loughlin et al., 2011). The clearest examples are formerly unitary 147

Pieter van Houten

states such as Belgium, Spain, Italy and, to a lesser extent, France. In federal states such as the United States, Canada and Germany, which were already significantly decentralized, further decentralization is harder to detect, but no trend of centralization has occurred. It is certainly possible that economic developments, including the increased relevance of territorialized economic activities, have played a role in this. Rodríguez-Pose and Ezcurra (2011) note that economic justifications for decentralization have become more prominent. It is noteworthy, however, that analyses of the decentralization trends in advanced democracies (e.g. Ansell and Gingrich, 2003; Hooghe et al., 2010) do not mention economic factors as driving forces. Instead, their focus is on concerns about administrative efficiency, expanding venues for democratic processes, and ‘pressures from below’ (from regional and local political actors) as causes. These factors, in turn, can be affected by economic regionalism – as discussed above for the case of pressures from below – and such indirect effects on the decentralization of state powers deserve further research. It is also worth noting that a trend of fiscal decentralization (an increase in taxing and spending powers of sub-national levels of government), which one might expect to be particularly sensitive to economic developments and which has frequently been linked to globalization, is not clearly evident in advanced industrial states (Garrett and Rodden, 2003). While more research on these issues is needed, this suggests the need for some caution with bold claims on the effects of economic regionalism on state decentralization. A final empirical issue to consider is whether the governance of cities and urban regions has changed. Clustering of economic activities often occurs in or around cities, and globalization and post-Fordism are argued to give rise to changes in urban governance (e.g. Scott, 2001; Brenner, 2004). Salient changes in urban governance in Western Europe include the creation of the Greater London Authority in the United Kingdom, and the direct elections of city mayors in Italy, but a general trend of power transfers to cities is not apparent. Brenner (2004) and Le Galès (2002) document a multitude of changes to planning and urban development policies, with implications similar to the regional development policies discussed above (including increasing forms of ‘multi-level governance’). However, more direct institutional changes, such as the creation of new urban governments or giving existing city authorities more power and authority, are less evident. As Le Galès (2002: 245) indicates, ‘[a]ttempts at creating metropolitan governments have, in most Western countries, ended in failure and contested experiences’, mainly due to opposition from entrenched political interests. The failed attempts to reform the urban administrative frameworks in the Dutch Randstad are a good example of this (Salet, 2006). Thus, economic regionalism has induced some decentralization and a considerable increase of ‘hybridity’, but its effect on more general state transformation is less pronounced, or at least harder to detect. I want to suggest here two possible explanations for why the – at first sight, theoretically compelling – logic linking the increased economic importance of the sub-national level to state transformation and decentralization is, at a minimum, incomplete. First, even if economic regionalism induces some pressures and political mobilization for state transformation (as discussed in the previous subsection), other factors influence whether such transformation will occur. The specific political context is a particularly important factor. For example, recent episodes of decentralization in Spain, Italy and the United Kingdom all occurred in particular political constellations: in Spain, minority governments found themselves forced to rely on support from regionalist parties, in return for which they transferred some powers and tax money to the regional level; in Italy, the Berlusconi governments needed the Northern League for a parliamentary majority, which induced them to propose decentralization reforms; and in the UK, the Labour Party’s dependence on support in Scotland and Wales, which was threatened by regionalist parties, was one of the reasons for implementing devolution in the late 1990s. The process of institutional reforms in these countries would have been 148

Economic regionalism

different if electoral politics had played out differently. Furthermore, for ‘bottom-up pressures’ to lead to decentralization, it may be necessary for sub-national political actors – in particularly regional governments – to cooperate and form a common front. For instance, the regions in southern Germany have not been able to build a sufficiently large coalition of regional governments, which explains the failure of their demands to reform the German federal system (Rodden, 2006). There are several reasons why the formation of a common ‘sub-national front’ may be difficult. One reason is that the focus on distributive issues of many regionalist movements and parties (see above) may undermine efforts to form broad coalitions of regions. In addition, the logic of economic regionalism itself may complicate the formation of such coalitions. Although some authors claim that the economic resurgence of the sub-national level can benefit all or most regions (e.g. Porter, 1998) and gives regional political actors an incentive to cooperate (e.g. Pasquier, 2011), one of the defining features of these economic developments is competition between regions, in which – as indicated in the first section above – only some regions are likely to be successful. This is widely recognized in the literature (e.g. Storper, 1997: ch. 10; Scott, 1998: ch. 8; Keating, 1998, 2001; Keating et al., 2003; Brenner, 2004: ch. 4), but the political implications, including for the possibility of political cooperation between regions, are usually neglected (for an exception, see Keating, 2001). Territorial inequalities within countries have increased in recent decades (Rodríguez-Pose and Gill, 2004), and the increased importance of economic clusters and districts may exacerbate some regional disparities (Rodríguez-Pose and Crescenzi, 2008). This can further intensify inter-regional competition and territorial distribution issues, and make building and sustaining a coalition in favour of decentralization difficult. The general implication of this first possible explanation is that we need to formulate and test more fully developed arguments about the interaction between economic developments – such as economic regionalism and globalization – and political factors in causing ‘authority migration’ (Gerber and Kollman, 2004). Second, the alleged link between economic regionalism and state decentralization assumes that economic actors involved in spatially concentrated economies (firms and companies and their managers and employees) want more power for sub-national authorities in order to get a better provision of goods and policies. However, it is possible that some of the public goods, investments and regulatory regimes that would benefit economic clusters, as well as possible measures to moderate and compensate for competition between regions and clusters, are better provided by national or supra-national levels of government. Thus, we need further research on the exact implications of the new geographical patterns of economic development on support for (de)centralization and on the geographic distribution of political and policy preferences more generally (cf. Rodden, 2010).

Conclusion Certainly not all contemporary economic activity and development takes place in territorially concentrated clusters of small and medium-sized firms, and traditional industries, multinational companies and global networks remain important components of the global capitalist system. There is no doubt, however, that the regional economic level has become increasingly significant in recent decades. This has given rise to a rich literature, to which analysts of territorial politics should pay attention. This chapter has, however, argued that the political and institutional consequences of these economic developments remain to be precisely identified. Economic regionalism plays a role in the rise and transformation of political regionalism, has contributed to changes in the content and organization of regional and urban development policies, and may 149

Pieter van Houten

have played some role in more general changes in the territorial organization of states. Overall, economic regionalism has undoubtedly contributed to making federal and regionalized systems of government increasingly ‘hybrid’. However, other factors have also played a role in these developments, and the exact effects of economic regionalism are not fully clear. Identifying these effects should be a key focus for future research. Quantitative analyses face several challenges, including getting indicators for the extent and success of economic regionalism in specific territories. In an analysis of regional autonomy demands in six Western European countries, I found that various indicators of regional competitiveness and exposure to globalization have no statistically significant effect (van Houten, 2003), but this is a limited analysis, and much more work can be done. Other possibilities include qualitative ‘paired comparisons’ (Tarrow, 2010) of politically assertive regions with and without strong economic regionalism, or of economically dynamic regions with and without political movements or decentralization reforms. Such detailed comparisons may also give clues about possible indirect effects of economic regionalism on state transformation. A different focus for further research would be to explore the opposite causal relation: how the territorial organization of a state affects the emergence and performance of sub-national economic clusters and districts. Zeitlin (2007: 227–31) and Keating et al. (2003: 183) discuss some of these effects, but further research is possible. Finally, the extent to which economic regionalism is occurring and is impacting state organization in emerging economies and developing states deserves further attention (building on, for example, Scott and Garofoli, 2007), with large and rapidly growing states such as Brazil, India and China as particularly interesting cases.

Notes 1 For surveys of other aspects of the political economy of federalism, see Oates, 1999; Rodden, 2006. 2 This term is not widely used in the literature. Its most frequent use is in reference to the development of supra-national economic blocs (e.g. Gordon, 1961; Katzenstein, 1996). In contrast, this chapter focuses exclusively on sub-national regions. 3 For useful overviews, see Storper, 1997: ch. 1; Asheim et al., 2006; Zeitlin, 2007; Whitford and Potter, 2007.

References Agnew, John A. (2002) Place and Politics in Modern Italy (Chicago: University of Chicago Press). Ansell, Christopher and Jane Gingrich (2003) ‘Trends in Decentralization’, in Bruce E. Cain, Russell J. Dalton and Susan E. Scarrow (eds) Democracy Transformed? Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press). Asheim, Bjørn, Philip Cooke and Ron Martin (2006) ‘The Rise of the Cluster Concept in Regional Analysis and Policy: A Critical Assessment’, in Bjørn Asheim, Philip Cooke and Ron Martin (eds) Clusters and Regional Development: Critical Reflections and Explorations (London: Routledge). Asheim, Bjørn, Helen Lawton Smith and Christine Oughton (2011) ‘Regional Innovation Systems: Theory, Empirics and Policy’, Regional Studies Vol. 45, No. 7: 875–91. Barca, Fabrizio (2009) An Agenda for a Reformed Cohesion Policy: A Place-Based Approach to Meeting European Union Challenges and Expectations (Brussels: European Union), ec.europa.eu/regional_policy/archive/ policy/future/barca_en.htm. Benz, Arthur and Burkard Eberlein (1999) ‘The Europeanization of Regional Policies: Patterns of Multi-Level Governance’, Journal of European Public Policy Vol. 6, No. 2: 329–48. Brenner, Neil (2004) New State Spaces: Urban Governance and the Rescaling of Statehood (Oxford: Oxford University Press). ——(2009) ‘Open Questions on State Rescaling’, Cambridge Journal of Regions, Economy and Society Vol. 2, No. 1: 123–39. Casper, Steven (2007) Creating Silicon Valley in Europe: Public Policy Towards New Technology Industries (Oxford: Oxford University Press). 150

Economic regionalism

Cooke, Philip (1997) ‘Regions in a Global Market: The Experiences of Wales and Baden-Württemberg’, Review of International Political Economy Vol. 4, No. 2: 349–81. Cooke, Philip, Carla de Laurentis, Franz Tödtling and Michaela Trippl (2007) Regional Knowledge Economies: Markets, Clusters and Innovation (Cheltenham: Edward Elgar). Cox, Kevin R. (2009) ‘“Rescaling the State” in Question’, Cambridge Journal of Regions, Economy and Society Vol. 2, No. 1: 107–21. Crouch, Colin, Patrick le Galès, Carlo Trigilia and Helmut Voelzkow (eds) (2001) Local Production Systems in Europe: Rise or Demise? (Oxford: Oxford University Press). ——(2004) Changing Governance of Local Economies: Responses of European Local Systems (Oxford: Oxford University Press). Cruz, Sara C.S. and Aurora A.C. Teixeira (2010) ‘The Evolution of the Cluster Literature: Shedding Light on the Regional Studies-Regional Science Debate’, Regional Studies Vol. 44, No. 9: 1263–88. Deeg, Richard (1996) ‘Economic Globalization and the Shifting Boundaries of German Federalism’, Publius Vol. 26, No. 1: 27–52. Fürst, Dietrich (2006) ‘The Role of Experimental Regionalism in Rescaling the German State’, European Planning Studies Vol. 14, No. 7: 923–38. Garrett, Geoffrey and Jonathan Rodden (2003) ‘Globalization and Fiscal Decentralization’, in Miles Kahler and David A. Lake (eds) Governance in a Global Economy: Political Authority in Transition (Princeton: Princeton University Press). Gerber, Elisabeth R. and Ken Kollman (2004) ‘Introduction – Authority Migration: Defining an Emerging Research Agenda’, PS: Political Science and Politics Vol. 37, No. 3: 397–401. Gordon, Lincoln (1961) ‘Economic Regionalism Reconsidered’, World Politics Vol. 13, No. 2: 231–53. Herrigel, Gary (1996) Industrial Connections: The Sources of German Industrial Power (Cambridge: Cambridge University Press). Hooghe, Liesbet and Gary Marks (2001) Multilevel Governance and European Integration (Lanham, MD: Rowman and Littlefield). Hooghe, Liesbet, Gary Marks and Arjan H. Schakel (2010) The Rise of Regional Authority: A Comparative Study of 42 Democracies (London: Routledge). Jeffery, Charlie (1999) Recasting German Federalism: The Legacies of Unification (London: Pinter). Katzenstein, Peter J. (1996) ‘Regionalism in Comparative Perspective’, Cooperation and Conflict Vol. 31, No. 2: 123–59. Keating, Michael (1998) The New Regionalism in Western Europe: Territorial Restructuring and Political Change (Cheltenham: Edward Elgar). ——(2001) ‘Governing Cities and Regions: Territorial Restructuring in a Global Age’, in Allen J. Scott (ed.) Global City-Regions: Trends, Theory, Policy (Oxford: Oxford University Press). Keating, Michael, John Loughlin and Kris Deschouwer (2003) Culture, Institutions and Economic Development: A Study of Eight European Regions (Cheltenham: Edward Elgar). Krugman, Paul (1991) Geography and Trade (Leuven: Leuven University Press). Le Galès, Patrick (2002) European Cities: Social Conflicts and Governance (Oxford: Oxford University Press). Le Galès, Patrick and Carlo Trigilia (2004) ‘Conclusions’, in Colin Crouch, Patrick Le Galès, Carlo Trigilia and Helmut Voelzkow (eds) Changing Governance of Local Economies: Responses of European Local Systems (Oxford: Oxford University Press). Lobao, Linda, Ron Martin and Andrés Rodríguez-Pose (2009) ‘Rescaling the State: New Modes of Institutional-Territorial Organization’, Cambridge Journal of Regions, Economy and Society Vol. 2, No. 1: 3–12. Loughlin, John (2009) ‘The “Hybrid” State: Reconfiguring Territorial Governance in Western Europe’, Perspectives on European Politics and Society Vol. 10, No. 1: 49–66. Loughlin, John, Frank Hendriks and Anders Lidström (eds) (2011) The Oxford Handbook of Local and Regional Democracy in Europe (Oxford: Oxford University Press). MacLeod, Gordon (2001) ‘New Regionalism Reconsidered: Globalization and the Remaking of Political Economic Space’, International Journal of Urban and Regional Research Vol. 25, No. 4: 804–29. Marshall, Alfred (1927 [1919]) Industry and Trade, 4th edn (London: Macmillan). Morgan, Kevin (1997) ‘The Learning Region: Institutions, Innovation and Regional Renewal’, Regional Studies Vol. 31, No. 5: 491–503. Oates, Wallace E. (1999) ‘An Essay on Fiscal Federalism’, Journal of Economic Literature Vol. 37, No. 3: 1120–49. Ohmae, Kenichi (1995) The End of the Nation-State: The Rise of Regional Economies (New York: Free Press). 151

Pieter van Houten

OECD (2005) Building Competitive Regions: Strategies and Governance (Paris: Organization for Economic Co-operation and Development). ——(2011) Regions and Innovation Policy (Paris: Organization for Economic Co-operation and Development). Pasquier, Roman (2011) ‘Cities, Regions and the New Territorial Politics’, in Erik Jones, Paul M. Heywood, Martin Rhodes and Ulrich Sedelmeier (eds) Developments in European Politics 2 (Basingstoke: Palgrave). Pike, Andy and John Tomaney (2009) ‘The State and Uneven Development: The Governance of Economic Development in England and Post-Devolution UK’, Cambridge Journal of Regions, Economy and Society Vol. 2, No. 1: 13–34. Piore, Michael J. and Charles F. Sabel (1984) The Second Industrial Divide: Possibilities for Prosperity (New York: Basic Books). Porter, Michael E. (1998) On Competition (Cambridge, MA: Harvard Business School Press). ——(2003) ‘The Economic Performance of Regions’, Regional Studies Vol. 37, No. 6/7: 549–78. Putnam, Robert D. (1993) Making Democracy Work: Civic Traditions in Modern Italy (Princeton: Princeton University Press). Rodden, Jonathan (2006) ‘The Political Economy of Federalism’, in Barry R. Weingast and Donald A. Wittman (eds) The Oxford Handbook of Political Economy (Oxford: Oxford University Press). ——(2010) ‘The Geographic Distribution of Political Preferences’, Annual Review of Political Science Vol. 13: 321–40. Rodríguez-Pose, Andrés (2008) ‘The Rise of the “City-region” Concept and its Development Policy Implications’, European Planning Studies Vol. 6, No. 8: 1025–46. Rodríguez-Pose, Andrés and Riccardo Crescenzi (2008) ‘Mountains in a Flat World: Why Proximity Still Matters for the Location of Economic Activity’, Cambridge Journal of Regions, Economy and Society Vol. 1, No. 3: 371–88. Rodríguez-Pose, Andrés and Roberto Ezcurra (2011) ‘Is Fiscal Decentralization Harmful for Economic Growth? Evidence from the OECD Countries’, Journal of Economic Geography Vol. 11, No. 4: 619–43. Rodríguez-Pose, Andrés and Nicholas Gill (2004) ‘Is There a Global Link between Regional Disparities and Devolution?’ Environment and Planning A Vol. 36, No. 12: 2097–117. Salet, Willem (2006) ‘Rescaling Territorial Governance in Randstad Holland: The Responsiveness of Spatial and Institutional Strategies to Changing Socio-Economic Interactions’, European Planning Studies Vol. 14, No. 7: 959–77. Sanz-Menéndez, Luis and Laura Cruz-Castro (2005) ‘Explaining the Science and Technology Policies of Regional Governments’, Regional Studies Vol. 39, No. 7: 939–54. Saxenian, AnnaLee (1994) Regional Advantage: Culture and Competition in Silicon Valley and Route 128 (Cambridge, MA: Harvard University Press). ——(2006) The New Argonauts: Regional Advantage in a Global Economy (Cambridge, MA: Harvard University Press). Scott, Allen J. (1998) Regions and the World Economy: The Coming Shape of Global Production, Competition, and Political Order (Oxford: Oxford University Press). ——(ed.) (2001) Global City-Regions: Trends, Theory, Policy (Oxford: Oxford University Press). Scott, Allen J. and Gioacchino Garofoli (eds) (2007) Development on the Ground: Clusters, Networks and Regions in Emerging Economies (London: Routledge). Sorens, Jason (2005) ‘The Cross-Sectional Determinants of Secessionism in Advanced Democracies’, Comparative Political Studies Vol. 38, No. 3: 304–26. Spencer, Gregory M., Tara Vinodrai, Meric S. Gertler and David A. Wolfe (2010) ‘Do Clusters Make a Difference? Defining and Assessing Their Economic Performance’, Regional Studies Vol. 44, No. 6: 697–715. Storper, Michael (1997) The Regional World: Territorial Development in a Global Economy (New York: The Guildford Press). Sturm, Roland (1995) ‘Regions in the New Germany’, in Michael Keating and John Loughlin (eds) The Political Economy of Regionalism (London: Frank Cass). Tarrow, Sidney (2010) ‘The Strategy of Paired Comparison: Toward a Theory of Practice’, Comparative Political Studies Vol. 43, No. 2: 230–59. van Houten, Pieter (2003) ‘Globalization and Demands for Regional Autonomy in Europe’, in Miles Kahler and David A. Lake (eds) Governance in a Global Economy: Political Authority in Transition (Princeton: Princeton University Press). ——(2007) ‘Regionalist Challenges to European States: A Quantitative Assessment’, Ethnopolitics Vol. 6, No. 4: 545–68. 152

Economic regionalism

Whitford, Josh and Cuz Potter (2007) ‘Regional Economies, Open Networks and the Spatial Fragmentation of Production’, Socio-Economic Review Vol. 5, No. 3: 497–526. Wilson, David and Vangelis Souitaris (2002) ‘Do Germany’s Federal and Land Governments (Still) Co-ordinate Their Innovation Policies’, Research Policy Vol. 31, No. 7: 1123–40. World Bank (2009) World Development Report 2009: Reshaping Economic Geography (Washington, DC: The World Bank). Zeitlin, Jonathan (2007) ‘Industrial Districts and Regional Clusters’, in Geoffrey Jones and Jonathan Zeitlin (eds) The Oxford Handbook of Business History (Oxford: Oxford University Press).

153

This Page intentionally left blank

Part II

Case studies by region: North America

This Page intentionally left blank

11 The United States of America From dualistic simplicity to centralized complexity John Kincaid

Since 1789, US federalism has moved toward centralization and, since the late 1960s, toward coercive federalism – namely, a greater concentration of powers in the federal government and a greater ability of that government to impose its will on the states and their local governments with little or no consultation with state and local officials (Kincaid, 1990, 1993b, 2011). Political and social forces have propelled centralization because proponents of change seek to implement their ideas nationwide, not in a few states. The large geographic and population sizes of the United States, its large number of constituent states (i.e. 33 by 1859, 48 by 1912 and 50 by 1959), and its regionalized cultural and socio-economic diversity have contributed to policy nationalization through federal action rather than state-by-state action or interstate compacts, uniform state laws and reciprocity. Many reforms start in the states – which US Supreme Court Justice Louis D. Brandeis called ‘laboratories of democracy’ (New State Ice Company v. Liebmann 1932) – but are nationalized as soon as politically feasible. Nevertheless, the implementation of federal policies by state and local governments is largely cooperative and the states, as constitutional polities, still legislate independently in many fields. Centralization has not increased at a constant pace. It advanced slowly from 1789 to about 1887, picked up speed during the Progressive era (1890–1920), accelerated during President Franklin D. Roosevelt’s (1933–44) New Deal, and leaped into high velocity after 1968. A major impediment to centralization for the first 180 years of US history was de facto bi-communalism whereby the South, in a manner similar to such regions as Catalonia and Québec, defended states’ rights against national power. Complementing this bi-communalism was a nationally confederated party structure rooted in county governments which restrained federal power by allowing state and local officials to control the electoral fortunes of federal officials. The social-reform and equality revolutions of the 1960s, especially the black Civil Rights Movement, along with other social and political changes, dismantled the confederated party system, precipitated the South’s collapse as states’ rights defender, and equated states’ rights with tyranny (Riker, 1964). Since then, no state or region has fervently defended states’ rights, and the more nationalized political parties largely ignore federalism. Barriers to centralization, therefore, are weak. 157

John Kincaid

The historical thrust for centralization has been, essentially, to liberate persons from the tyranny of places – namely, perceived suppressions of individual rights by both the constituent states and the marketplace, from the liberation of slaves in slave-holding states in 1863 to contemporary battles over gay rights in the 50 states. Because the states, not the federal government, historically exercised almost all of the powers affecting daily life, the states infringed on citizens’ rights much more than did the federal government. Hence, the federal government came to be seen as the primary force for liberation. Similarly, with the rise of a national industrial economy in the late 19th century, the federal government came to be seen as the only effective countervailing power (Galbraith, 1952) to big business. Furthermore, as expressed in the Declaration of Independence of 1776, Americans have been committed mainly to universal individual rights, not to communal rights. Indeed, the first changes to the US Constitution were 10 amendments (ratified in 1791) protecting primarily individual rights and, secondarily, states’ rights against federal infringement.

The compound Republic According to The Federalist (especially numbers 51 and 62), the US Constitution created a compound republic. In Federalist 23, Alexander Hamilton contended that ‘the circumstances of our country are such, as to demand a compound instead of a simple, a confederate instead of a sole government’. Like a chemical compound, the Constitution combined simpler elements – namely, confederalism and unitarism – so as to create a new form of government later called ‘federalism’. The compound republic is similar to the concept of hybridity elaborated in this volume, especially Chapter 1 by John Loughlin, except that American federalism was established as a unique and seemingly singular principle, and this system became the exemplary template of a federal system against which others were later measured. Contemporary hybrids, therefore, often include elements of the American federal principle. In Federalist 39, James Madison detailed the ways in which the new government was ‘partly federal, and partly national’. In The Federalist, the word ‘federal’ means what is now called ‘confederal’ (Diamond, 1974). The Senate, for example, is a ‘federal’ part of the compound; the House of Representatives is a ‘national’ part. The president is a ‘national’ officer elected by ‘federal’ means (i.e. the electoral college). The singular innovation of the Constitution, said Hamilton, was to preserve the federal form while also giving the new government of the federation the authority to legislate for individuals. This means the authority to levy taxes, conscript individuals into military service, regulate commerce and enforce terms of treaties. The Federalist is optimistic about the stability of this compound. Indeed, after 225 years, the US Constitution is the world’s oldest written national constitution, with only 27 amendments, none of which altered the federal system’s basic structure, although several amendments, especially XIV (1868), XVI (1913), and XVII (1913), facilitated shifts of power from the states and to the federal government. However, The Federalist hints at tensions between the compound’s principles that could alter the relative status of the two principles over time. Even the principal writers of The Federalist disagreed on their relative weight. Hamilton championed the national principle; Madison more often defended the federal principle. For example, Hamilton believed that the federal government has virtually unlimited tax powers; Madison believed the federal government’s tax powers are confined to its sphere of delegated powers (Kincaid, 2012). Hamilton proved to be more prescient.

The bi-communal constraint on federal power The United States is not usually viewed as having been bi-communal, but it was significantly bi-communal from its founding in the 1780s until the 1960s. The union’s deepest, most enduring 158

The United States of America

culturally based territorial cleavage has been between the North and the South. As Hamilton noted in Federalist 13, if the 13 states did not unite, they would likely form two confederacies. The New England states along with New York and New Jersey would form a northern confederacy; the states below New Jersey would forge a southern confederacy. Only Pennsylvania’s allegiance was uncertain. Pennsylvania ‘may deem it most consistent with her safety to have her exposed side turned towards the weaker power of the southern, rather than towards the stronger power of the northern confederacy’, he commented. ‘This would give her the fairest chance to avoid being the Flanders of America’. The founders recognized the sharp socio-economic and cultural border between North and South, free states and slave states, Puritans and Cavaliers. The foundation of this bi-communalism was African slavery, which generated southern white construction of a distinct culture that ardently defended slavery and resisted Lockean liberalism, capitalism, human equality, natural freedom, and the individualistic natural-rights principles of the Declaration and the Constitution (e.g. Cash, 1941; Carpenter, 1990).

Antebellum and reconstruction eras Before the Civil War (1861–65), the South demanded parity in the US Senate by coupling the admission of free states into the union with slave states – a pattern that resurfaced in 1959 when Alaska and Hawaii joined the union after southerners opposed Hawaii’s admission. Many white southerners portrayed themselves as Cavaliers – descendants of England’s Norman conquerors – arrayed against the North’s ‘Yankee race’ of Puritan Roundheads. Southerners constructed a distinct, even national, identity during the antebellum decades and spoke increasingly of ‘disunion’ (Varon, 2008). As a prominent southern political writer, J.B. DeBow, asserted in 1861, the framers of the US Constitution foolishly tried to ‘erect one nation out of two irreconcilable peoples’ (quoted in McPherson, 2008). ‘The great wonder’, editorialized Virginia’s Richmond Dispatch in 1863, ‘is not that the two sections have fallen asunder … but that they held together so long’. The US Constitution is not overtly bi-communal. Instead, the founders made concessions to the southerners’ slave interests and met southern demands for political parity with the North through the US Senate and the electoral-college mechanism of electing presidents. These arrangements gave the South enough weight in Congress, especially the Senate, and sufficient influence over the presidency and, thereby, the Supreme Court to stave off northern efforts to expand federal power in directions opposed by southerners for 72 years. Nine southerners occupied the presidency for 52 of those 72 years. The South’s influence on the Supreme Court was amply reflected in the Court’s disastrous attempt to resolve the slavery conflict in 1857 (Dred Scott v. Sandford, 1857). Chief Justice Roger Taney, a southerner and slave owner, was joined by four other southern justices and two northern justices in upholding slavery and authorizing its spread throughout the union. The two dissenters were northerners. By the 1830s, it became evident that the constitutional system might be unable to manage the North–South conflict. In an effort to fix the federal system, US Senator John C. Calhoun of South Carolina proposed to make the Constitution formally bi-communal by creating a two-headed North-South presidency, requiring legislation to be enacted by majority votes of northern and southern members of Congress, and allowing states to nullify federal laws they deemed to be unconstitutional (Read, 2008). Calhoun’s views were an anathema to northerners, who held an individualistic, majoritarian view of the Constitution. No constitutional change occurred between 1804 and 1865; instead, unable to resolve the sectional conflict, the country plunged into a civil war that took more than 750,000 lives. Thus, the United States exemplified the instability and conflict that can plague a bi-communal federation (Duchacek, 1988). 159

John Kincaid

The war produced constitutional change, the most important being the 14th Amendment (1868), which, while recognizing dual national and state citizenship, prohibits any state from abridging ‘the privileges and immunities of citizens of the United States’, depriving ‘any person of life, liberty, or property, without due process of law’, and denying ‘to any person within its jurisdiction the equal protection of the laws’. Congress can enforce this amendment. This marked a significant change, but it did not have a nationalizing impact until the 1950s because Congress largely ignored it and the US Supreme Court employed it to protect the rights of corporations, not individuals, against state action. In a major concession to the South, the Court upheld state laws requiring race segregation (Plessy v. Ferguson, 1896). The post-war decade (1865–75) saw a tremendous expansion of federal power as the federal government ‘reconstructed’ the South under military rule. The federal government also attacked Mormon polygamy. Pre-war attempts to outlaw polygamy were blocked by southerners in Congress who feared that such legislation might legitimize federal abolition of slavery. After the South’s defeat in 1865, the Republican-controlled federal government launched a constitutionally dubious legal, economic and quasi-military war on Utah’s Mormons which ended in 1890 when the church abandoned polygamy (Kincaid, 2008). The federal government also unleashed its legal and military forces on the western Indian nations, which were genocidally decimated by 1898. These events echoed the federal government’s campaign to Americanize Louisiana after purchasing it in 1803, by suppressing manifestations of Catholic, Spanish, French and Creole cultures. The federal government has thus been a tool for wiping out territorially based expressions of religious, ethnic and linguistic communalism. The federal government never sought to destroy any of the states; even the southern states were left intact after the Civil War. Instead, it sought to prevent any state or other legally recognized political jurisdiction from constructing a distinct communal identity. By the 1890s, the federal government had succeeded in this enterprise everywhere except in the South, where Reconstruction failed to convert southerners fully to American (i.e. northern) values. The final campaign against the South did not start until the 1950s. After the Civil War, the South re-built a distinct culture without slavery but with white supremacy, race segregation and blacks’ exclusion from politics and government. The South became solidly Democratic as opposed to the predominantly Republican North. The confederated county-based party system, which first emerged in the late 1820s, became more fully institutionalized after the war and, along with the South, blocked expansions of federal power detrimental to the states. In the South, strong white Democratic Party organizations became rooted in county courthouses. In the North, as a result of massive immigration following the Civil War and lasting until federal law stopped it in 1924, powerful urban-county Democratic Party organizations, called ‘political machines’, emerged in many northern cities where they battled strong state Republican organizations. These county-based party organizations exercised substantial control over the electoral fortunes of members of the US House of Representatives, over state legislative appointment of US senators and, through the electoralcollege system, over the fates of presidential candidates. National Democratic and Republican party organizations were extremely weak and could not control who entered Congress or the White House under their party banner. The combination of a solidly Democratic South seeking to prevent any federal intervention into its way of life and of northern political machines seeking to preserve their prerogatives restrained federal intrusions into state and local affairs and sustained a dualistic simplicity in which the federal and state governments were said to occupy separate spheres of sovereign power. 160

The United States of America

Progressive era The Progressive reform movement, which emerged in the late 1880s, battled the corruption associated with the party system as well as injustices of the new urban-industrial economy. Progressives expanded federal power over the economy through such laws as the Interstate Commerce Act of 1887, Bankruptcy Act of 1898 (the first permanent exercise of this power even though it was clearly delegated to Congress in 1788), Sherman Anti-Trust Act of 1890, and Pure Food and Drug Act of 1906. Progressives expanded federal power despite a generally hostile Supreme Court because they garnered support from some southern white Populists and many northern and western state and local party bosses to pass legislation through Congress and elect sympathetic presidents. For example, Theodore Roosevelt, a Republican president (1901–09), was a leading Progressive whose famous 1910 New Nationalism speech summarized the Progressive agenda. ‘I do not ask for overcentralization’, declared Roosevelt, but: The national government belongs to the whole American people, and where the whole American people are interested, that interest can be guarded effectively only by the national government … The New Nationalism … is impatient of the bitter confusion that results from local legislatures attempting to treat national issues as local issues. (Quoted in Wright, 1929: 529) Progressives secured four amendments to the US Constitution, all of which laid foundations for expanded federal power. The amendments were achieved through alliances the Progressives built with other social movements. The first was the 16th Amendment (1913) allowing the federal government to collect personal and corporate income taxes. This amendment was ratified by an alliance of southern Populists and western Progressives. It did not become a potent power, however, until the Second World War when the federal government increased tax rates substantially and expanded the tax base to virtually all income-earning residents. After the war, the federal government used this power to expand its influence over state and local governments through grants-in-aid. The second was the 17th Amendment (1913), requiring the election of US senators by the voters of each state rather than appointment by the state legislatures. This amendment, opposed by most southern states, was ratified by an alliance of northern and western Progressives. The nationalizing impact of this amendment was minimized for about 55 years, however, because state and local party organizations controlled the electoral fortunes of senators until the late 1960s. The 18th Amendment (1919) instituted alcoholic beverage prohibition. The amendment was ratified by 92% of the states within 13 months because most state legislatures (the ratifying instruments) were dominated by rural members who saw drinking as an urban immigrant evil. This nationalizing amendment failed spectacularly. Although it was repealed in 1933 and alcoholic beverage regulation was restored to the states, the federal government’s fight against rising organized crime during Prohibition legitimized expanded federal criminal-justice and regulatory roles that found new outlets after repeal. Furthermore, since the early 1980s, the federal government has expanded intrusions into state alcoholic beverage policy making. The fourth change was the 19th Amendment (1920) giving women the vote. This amendment was opposed by most southern states. Many Progressives believed that women voters would combat corruption. This amendment is neutral with respect to federalism, but it has been somewhat nationalizing because women support national social welfare and education policies more than do men. Since the 1980 presidential election, moreover, women have usually voted more heavily for Democrats than Republicans in federal elections. Initially, these changes had minimal impacts on the balance of state-federal power because the South and the confederated party system defended state and local prerogatives from federal 161

John Kincaid

overreach. Consequently, there was no rush toward nationalizing legislation in the Senate after 1913. The era, though, established precedents that served as foundations for future centralization. Additionally, the admission of 12 western states into the union after the Civil War reduced the 11 states of the southern confederacy to 23% of the Senate and counteracted the country’s bicommunalism. To restrain federal power, southern members of Congress had to find allies in the North and West, where many Republicans were abandoning Progressivism for laissez faire capitalism.

New Deal era One indicator of the minimal operational centralization resulting from Progressive reforms is that the United States became a world power with public functions still financed mainly by local property taxes. From 1789 to about 1842, states were the major fiscal actors. They financed infrastructure, development and government operations from asset income derived from canal tolls, bank-stock dividends and land sales, as well as indirect business taxes. By 1841, state debt stood at US$193 million compared to $25 million for local governments and $5 million for the federal government. From 1842 to the mid-1930s, local governments were the major fiscal actors, relying heavily on property taxes. By 1902, property taxes (mostly local) accounted for 42% of all federal, state and local revenues (Wallis, 2000). Even by 1932, local governments accounted for more than half of all government revenue. In 1927, federal spending amounted to only 31% of all own-source government expenditures, compared to 52% for local governments and 17% for the states. Of federal spending in 1932, 22% went to veterans’ benefits, 19% to the post office, 17% to defence and only 5% to state and local aid (US Bureau of the Census 1975). The Great Depression (1929–41) facilitated Franklin D. Roosevelt’s landslide election to the presidency in 1932 by a coalition of southern and northern Democrats. Democrats also captured both houses of Congress. The New Deal (1933–41) and the Second World War (1941–45) are commonly seen as great centralizing periods, which they were relative to the past; yet, the policies of those periods hardly disturbed state and local powers, primarily because Roosevelt (1933–45) refused to attack the state and local party bosses who provided essential electoral support for his Democratic coalition (Kincaid, 1987). Although Roosevelt needed the southern Democrats in order to occupy the White House, they – along with conservative Republicans from northern rural and small-town districts – frustrated many of his legislative goals. The New Deal’s major welfare legacy, the Social Security Act of 1935, was, therefore, a paradigm of what soon came to be called ‘cooperative federalism’. Also, despite expanded federal regulation of the economy and Keynesian policy making, Congress maintained dual state and federal regulation and taxation of banking, telecommunications, securities and other industries; countermanded the US Supreme Court by preserving exclusive state regulation of insurance; and interfered little with state and local regulation of small businesses and occupational licensing and with such traditional state functions as transportation, education and criminal justice. Federal cash grants to state and local governments, which first began in the 1880s but remained tiny until the 1930s, increased by 282% from 1932 to 1942. Few conditions (i.e. rules and regulations) accompanied the money, and there was little federal capacity to audit expenditures. The period was a fiscal nirvana for state and local officials. Quite important also for preserving state and local powers, the New Deal did not champion nationalization of the US Bill of Rights under the 14th Amendment, nor did it disturb southern race segregation. The New Deal’s signature federalism case (United States v. Darby Lumber Co.) in 1941 sounded the death knell for the 10th Amendment to the US Constitution (which reserves all residual powers to the states or the people) by declaring it ‘a truism’, but the ruling had little practical impact on federalism for two decades. The federal Fair Labor Standards Act of 1938 162

The United States of America

(FLSA), upheld in Darby, applied to private-sector employees, not to state and local government employees. Hence, many state and local bosses supported the FLSA. They never imagined Congress would apply the FLSA to employees of ‘sovereign’ state and local governments three decades later (Kincaid, 1993a). This new cooperative federalism, however, laid foundations for coercive federalism. For one, local governments, especially big cities (most controlled by Democrats), gained a seat at the intergovernmental table as the ‘third partner’ in the federal system. This undercut the states as the Constitution’s authorized partner in cooperative federalism. Cities also received direct federal aid that bypassed their state capitol. Second, the federal government became the dominant fiscal power by relying heavily on income taxes. The federal share of all government spending increased to 50% by 1940 and to 77% in 1958, fluctuating thereafter in the 62–70% range. Third, under pressure from Roosevelt, who proposed increasing the Supreme Court’s size so that he could appoint justices sympathetic to his legislation, the Court executed an historic reversal of its commerce clause (US Constitution, Art. I, Sec. 8, Cl. 3) jurisprudence in 1937, giving Congress virtually unfettered authority to regulate the economy. This reversal had little effect on state and local prerogatives until the 1960s, when Congress used the commerce clause to expand federal regulation into social, environmental and consumer affairs historically controlled by states and localities (somewhat like the European Union’s use of its economic powers to change member-state practices in such areas). Fourth, the New Deal and, later, the Great Society reforms of Democratic President Lyndon B. Johnson (1963–69) highlighted weaknesses in the ability of state and local governments to solve the problems of an urban-industrial society, cope with national crises and ensure justice. Americans began to view the federal government as the system’s most progressive partner – an engine for social change and personal benefits (e.g. Medicare and Medicaid enacted in 1965).1 The corruption of many northern urban bosses and the odious practices of southern white supremacists discredited states’ rights, prompting William H. Riker’s declaration that ‘if in the United States one disapproves of racism, one should disapprove of federalism’ (Riker, 1964: 155). Fifth, the New Deal’s reaffirmation of the Progressive idea that the federal Constitution is a ‘living document’ that judges and legislators must adapt to modernity, coupled with the view of the federal system as being one government serving one people under conditions of shared responsibilities (Grodzins and Elazar, 1974), fostered a pragmatic federalism that motivated interest groups and public officials to forum shop and place policy objectives above constitutional niceties. Constitutional questions of which government is authorized to do what were replaced by political questions of which government can do what – a shift that almost always favours national power, in part because of the federal government’s ability to engage in deficit spending. The federal government has run deficits in 41 of the 46 years of coercive federalism.

Rise of coercive federalism Coercive federalism arose from massive efforts to liberate persons from the tyranny of state and local jurisdictions and the marketplace. Since the colonial era, states and localities had governed the key aspects of life. By the 1960s, they came to be viewed as oppressive of blacks, Hispanics, women, gays and other ‘minorities’, as well as lax in fostering social welfare. Despite the federal government’s genocidal policies toward Indians and the US Supreme Court’s dismal humanrights record, by the 1950s the federal government assumed the role of champion of human and civil rights. This campaign was launched in 1954 when the US Supreme Court reversed its 1896 Plessy decision by ruling unanimously that state laws mandating race segregation violated 163

John Kincaid

the 14th Amendment (Brown v. Board of Education of Topeka, 1954). This ruling generated massive, even violent, resistance in the South, which triggered massive federal intervention. Bi-communalism and the cooperative federalism that had emerged to accommodate it during the Progressive and New Deal expansions of federal power collapsed, symbolically, in August 1968 during the Democrats’ tumultuous presidential nominating convention in Chicago. Street protestors and inside reformers drove the state and local bosses out of the nominating system and reoriented the party’s representative base from places to persons by mandating proportional representation of blacks, women, young adults and other minorities, and by emphasizing primary elections to choose party candidates for general elections. Republicans were more resistant to such reforms. Consequently, while federalism declined as an issue for Democrats, Republicans still defended federalism, at least rhetorically, a position that fitted Richard M. Nixon’s 1968 southern-state presidential election strategy. The Chicago convention dramatically highlighted the confluence of political, social and cultural forces unleashed during the late 1950s and early 1960s. The pre-eminent struggle for the liberation of persons from the tyranny of places was the black Civil Rights Movement. In order to redress movement grievances, the federal government penetrated deeply into historic realms of state and local authority. Social movement demands for equality assaulted the legitimacy of the territorial diversity, especially bi-communalism, which had sustained federalism and state and local powers. Reformers asked, for example, why a woman should have the right to an abortion in one state but not in another. These movements fostered public awareness of many other issues, such as environmental degradation, that cross-cut state and local boundaries. More and more public matters were said to have negative interstate externalities requiring remedial action by the federal government. By 1962, moreover, 90% of US households had a television – a potent medium for social movements and for focusing public attention on the federal government in Washington, DC. Television also changed political campaigns, reducing the role of local political party foot soldiers in mobilizing voters and vastly increasing the clout of national interest groups eager to fund costly television advertising for increasingly entrepreneurial candidates. The US Supreme Court played a pivotal role, too. Under the doctrine of ‘selective incorporation’, it nationalized much of the US Bill of Rights via the 14th Amendment so as to protect individuals’ rights against state-local as well as federal infringements. More than half (56%) of the Court’s rulings incorporating provisions of the Bill of Rights were issued during the 1960s. In addition to race desegregation, the Court massively expanded federal power to reform state and local governments in the name of rights. Although this form of judicial activism substantially abated after the Court triggered a political firestorm by voiding state anti-abortion laws in 1973 (Roe v. Wade), the federal courts had achieved unprecedented levels of rights protection and intervention into state and local affairs, from which there has been little retreat, except in criminal rights. The Court’s ‘one person, one vote’ rulings in 1964 shifted representation in both the US House of Representatives and the state legislatures from places to persons. Before 1964, most election districts conformed to county and municipal boundaries, thereby emphasizing the legislative representation of local government jurisdictions rather than individuals and social groups. Counties, moreover, were the key power centres of the parties. Such districting, however, produced malapportionment (i.e. districts having different population sizes). The Court ruled that the 14th Amendment required election districts to have equal populations. The full impact of the new apportionment system was felt in the early 1970s, at which time Congress became more individualistic and atomistic as members became more attentive to interest groups representing nationally organized groups of persons and less attentive to hometown state and local government 164

The United States of America