COMPARATIVE POLITICS DEMOCRACY TRANSFORMED?
COMPARATIVE POLITICS Comparative politics is a series for students and te...
75 downloads
1084 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
COMPARATIVE POLITICS DEMOCRACY TRANSFORMED?
COMPARATIVE POLITICS Comparative politics is a series for students and teachers of political science that deals with contemporary issues in comparative government and politics. As Comparative European Politics it has produced a series of high quality books since its foundation in 1990, but now takes on a new form and new title for the new millennium—Comparative Politics. As the process of globalization proceeds, and as Europe becomes ever more enmeshed in world trends and events, so it is necessary to broaden the scope of the series. The General Editors are Max Kaase, Vice President and Dean of Humanities and Social Sciences, International University, Bremen, and Kenneth Newton, Professor of Government, University of Southampton. The series is published in association with the European Consortium for Political Research. OTHER TITLES IN THIS SERIES Coalition Governments in Western Europe Edited by Wolfgang C. Müller and Kaare Strøm Parties without Partisans: Political Change in Advanced Industrial Decomcracies Edited by Russell J. Dalton and Martin P. Wattenberg Political Institutions: Democracy and Social Choice Josep M. Colomer Mixed-Member Electrol Systems: The Best of Both Worlds? Edited by Matthew Soberg Shugart and Martin P. Wattenberg Divided Government in Comparative Perspective Edited by Robert Elgie Political Parties: Old Concepts and New Challenges Edited by Richard Gunther, José Ramón Montero, and Juan J. Linz Social Movements and Networks: Relational Approaches to Collective Action Edited by Mario Diani and Doug McAdam Extreme Right Parties in Western Europe Piero Ignazi
Democracy Transformed? Expanding Political Opportunities in Advanced Industrial Democracies
edited by Bruce E. Cain, Russell J. Dalton, and Susan E. Scarrow
Great Clarendon Street, Oxford OX2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Bangkok Buenos Aires Cape Town Chennai Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi São Paulo Shanghai Taipei Tokyo Toronto Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York © the several contributors 2003 The moral rights of the authors have been asserted Database right Oxford University Press (maker) First published 2003 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose this same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available Library of Congress Cataloguing in Publication Data Democracy transformed? : expanding political opportunities in advanced industrial democracies / edited by Bruce Cain, Russell J. Dalton, and Susan Scarrow. p. cm.—(Comparative politics) Includes bibliographical references and index. 1. Democracy—History—20th century—Congresses. 2. Elections—History— 20th century—Congresses. I. Cain, Bruce E. II. Dalton, Russell J. III. Scarrow, Susan E. IV. Comparative politics (Oxford University Press) JC421.D46373 2003 321.8—dc22 2003066203 ISBN 0–19–926499–6 1 3 5 7 9 10 8 6 4 2
Preface and Acknowledgements By some accounts, the reforms of democratic institutions in the United States during the last three decades of the twentieth century were as extensive as the populist reforms that transformed American democracy during the first three decades of the century. The folklore of comparative politics suggested that this pattern was being repeated across the advanced industrial democracies. Our goal in this project was to chart the reforms of democratic institutions that were actually occurring, and to assess their cumulative implications for the workings of the democratic process. This is by no means a simple task, and we see this book as an initial step in what should be a continuing research programme on this topic. To move towards this research goal, the project brought together scholars who focus on different aspects of democratic politics to assess the evidence of change and address each other's findings and conclusions. The University of California's Irvine Center for the Study of Democracy first convened a planning meeting in Palm Springs, California, in January 2001 to discuss how to frame the project. This meeting developed the research template presented in Chapter 1, especially Table 1.1, as the framework for the project. We also decided to focus on institutional changes in the established democracies of the Organization for Economic Cooperation and Development (OECD) since 1960. We wanted to include as many of these countries as possible in each chapter, relying on archived data, published sources of information, and e-mail queries to various country experts. We are especially grateful to Martin Wattenberg, who participated in this first planning meeting and the development of the project's theoretical framework, and to Andrew Drummond, who helped administer this planning meeting. We next assembled at the Institute of Governmental Studies at the University of California at Berkeley during the 2001 American Political Science Association meetings in San Francisco. Here we presented some initial empirical findings, expanded the project to include new members, and finalized our overall theoretical and empirical framework. With a generous grant and continuing intellectual support from the Institute for European Studies at UC Berkeley, we organized the project's major research conference at Berkeley in spring 2002. The project members presented the basic empirical findings at this conference. Ironically, some might say, at the very place where the 1960s Free Speech Movement stimulated some of the processes that produced the democratic changes that were the focus of the project, we took stock of how democratic institutions had changed and adapted over this four-decade span. We also gained from the critical comments of several colleagues who served
vi
PREFACE AND ACKNOWLEDGEMENTS
as commentators at this conference, and their advice and insights have improved our final analyses. Our thanks to these discussants: Henry Brady, Giuseppe DiPalma, Nelson Polsby, Austin Ranney, and Ekkehard Schroeder; and to David Meyer, who provided insightful comments on later material. In addition, at various points in the project Andrew Drummond and Nhu-Ngoc Ong provided valuable research assistance. We also owe a great debt to the many people around the world who collected data upon which this book relies. It is a testament to the research development of modern political science that large-scale data collection projects—such as national election studies, the Katz and Mair collection of party data, and the comparative manifesto data—are now available for comparative research. Many of our chapters quite openly build upon these data sources, and we gratefully acknowledge the efforts of others in compiling these data. We also wish to thank the many people around the world who answered the e-mail queries of various contributors to this book and generously shared information with us. The Internet has not only contributed to greater efficiencies in the world economy; it has also facilitated broad comparative scholarly work such as found in this volume. The Rockefeller Foundation generously provided support to host a final project conference at the Villa Serbelloni in Bellagio, Italy, in September 2002. This meeting refined the presentation of the empirical evidence and focused on the fundamental question of the implications of these patterns for the functioning of the democratic process in contemporary democracies. The photograph on the book jacket was taken at that conference, on the same balcony where Queen Padamé and Anakin Skywalker were married. The hospitality of the Villa and its staff provided a week during which we could reflect on the changes transforming our political systems and the implications of these changes. This conference led to more extensive discussion of the implications of each empirical chapter, and the conclusions of the project in Chapter 11. This conference included several external discussants to critique our findings and our theorizing about democratic politics. We greatly appreciate the commentaries from John Aldrich, Hans-Dieter Klingemann, and Gillian Peele; their advice improved our final presentations in this book. In addition, Joanne Caddy shared the related findings from the OECD's programme on public governance; this was a valuable addition to our discussions. Henrik Bang presented his research on cultural governance that enriched our discussion of the theoretical aspects of democratic political change. Henrik was not able to contribute to the final project volume, but his published findings appear in our references. We owe a special debt of gratitude to Marc Levin at the Institute of Governmental Studies. Marc was a masterful administrator of the project. He managed two project meetings at Berkeley with exceptional success, and supervised the complicated arrangements of the Bellagio meeting with equal skill. Indeed, one of the most memorable parts of this research enterprise was the memorial to the
PREFACE AND ACKNOWLEDGEMENTS
vii
9/11/2001 terrorist attacks that Marc arranged on the anniversary date in Bellagio. Our thanks to Marc for all he has done for the project and its participants. We also want to thank the fine people at Oxford University Press in the United Kingdom, who moved this project to publication with ease and grace, especially Dominic Byatt for his invaluable support and guidance and Michael James for contributions to the manuscript. The bulk of the financial support for this project was generously provided by the Institute for European Studies at UC Berkeley. In addition, the Rockefeller Foundation, the Center for the Study of Democracy (UC Irvine) and the Institute of Governmental Studies (UC Berkeley) furnished additional support for the project. We are greatly indebted for all this support. Bruce E. Cain Russell J. Dalton Susan E. Scarrow May 2003
This page intentionally left blank
Contents List of Figures List of Tables Notes on Contributors Foreword by Austin Ranney 1. New Forms of Democracy? Reform and Transformation of Democratic Institutions Russell J. Dalton, Susan E. Scarrow, and Bruce E. Cain I.Electoral Change 2. Expanding the Electoral Marketplace Russell J. Dalton and Mark Gray 3. Making Elections More Direct? Reducing the Role of Parties in Elections Susan E. Scarrow 4. Political Parties and the Rhetoric and Realities of Democratization Miki Caul Kittilson and Susan E. Scarrow 5. Changing Party Access to Elections Shaun Bowler, Elisabeth Carter, and David M. Farrell II.Change in Non-electoral Institutions 6. Towards More Open Democracies: The Expansion of Freedom of Information Laws Bruce E. Cain, Patrick Egan, and Sergio Fabbrini 7. Trends in Decentralization Christopher Ansell and Jane Gingrich 8. Reforming the Administrative State Christopher Ansell and Jane Gingrich
xi xii xiv xvii 1 23 44 59 81 115 140 164
x
CONTENTS
9. Participation, Representative Democracy, and the Courts Rachel A. Cichowski and Alec Stone Sweet III.The Consequences of Political Reform 10. A Second Transformation of Democracy? Mark E. Warren 11. Democratic Publics and Democratic Institutions Russell J. Dalton, Bruce E. Cain, and Susan E. Scarrow References Index
192 223 250 276 305
List of Figures 2.1. Expanding electoral markets 2.2. Total number of elections by year, 1960–2001 2.3. Growing share of the population eligible to vote in national elections, 1960–2001 3.1. Number of referendums in 18 OECD democracies, 1945–2002 3.2. Number of referendums in 17 OECD democracies (minus Switzerland), 1945–2002 3.3. Number of referendums in US states, 1945–99 3.4. Number of state referendums in Germany and Australia, 1945–99 4.1. Saliency of democratic discourse in party platforms 4.2. Models of party governance 6.1. Enactment of national freedom of information laws 6.2. Sources of appeals to CADA, 1986–2000 6.3. Subjects of FOIA stories in major US newspapers, Oct. 1999–Sept. 2000 9.1. Civil rights claims as a percentage of the US Supreme Court's case load, by term 1933–90 9.2. Civil rights claims as a percentage of the Canadian Supreme Court's case load, 1960–90 9.3. Total number of cases before the Japanese Supreme Court, 1960–91 10.1. Normative dimensions of democracy 11.1. Inequality in participation by education group
27 31 32 50 51 52 52 63 66 118 129 135 199 200 209 230 263
List of Tables 1.1. 2.1. 2.2. 2.3. 3.1. 4.1. 4.2. 4.3. 4.4. 5.1. 5.2. 5.3. 5.A1. 5.A2. 5.A3. 6.1. 6.2. 6.3. 6.4. 6.5. 6.6. 6.7.
Potential changes in democratic institutions and processes The institutional structure of nationwide electoral options Elections per five-year period, 18 OECD nations, 1960–99 Increase in the number of electoral decisions over time Availability of referendums and personal elections, 1970:2001 Saliency of democratic themes in party platforms, national averages, 1960–98 Degree of inclusiveness of party selectorate in legislative candidate selection, 2000 Degree of inclusiveness of national party leadership selectorate, 2000 Guaranteed delegates from party sections on parties' national leadership bodies, 1960:1990 Indicators of electoral system change Change in the electoral environments of OECD countries, 1960s–2000s: summary measures Assessing the effects of electoral law change on measures of party system change, 1960s–1990s Requirements for ballot access Requirements for media access Requirements for state subventions Exemptions to FOI laws in G7, Australia, New Zealand, and Sweden United States FOIA request, denial, and appeal statistics, 2000 Reasons for denials of FOIA requests, 2000 Departments with notably long delays in responses to FOIA requests, 2000 CADA appeals rulings by level of government, 2000 Appeals rulings by CADA, 2000 Reasons for denial per statutory exemption, CADA, 2000
16 28 30 34 48 61 70 72 74 83 90 94 97 101 106 123 125 126 127 128 130 130
LIST OF TABLES
6.8. 6.9. 6.10. 6.11. 7.1. 7.2. 7.3. 7.4. 7.5. 7.6. 8.1. 8.2. 8.3. 8.4. 8.5. 8.6. 9.1. 9.2.
xiii
Geographical origin of FOI requests to the EU, 1994–96 132 Subjects of FOI requests to the EU, 1994–96 133 Categories of applicants under EU FOI laws, 1994–96 133 Disposition of FOI requests to EU, 1994–96 134 Expected reform paths of OECD countries, by institutional type 143 Creation of new levels of sub-national government 144 Decentralization of responsibility to sub-national government 145 Increases in sub-national organizational/fiscal autonomy 146 Major OECD education decentralization reforms 157 Percentage of decisions taken at each level of government in public lower secondary education, 1998 158 Selected New Public Management reforms aimed at access, accountability, and transparency 167 Ombudsman offices in advanced industrial democracies 170 Trends in ombudsman complaint loads: number of cases 172 Powers of the ombudsman, 1999 173 Laws on administrative procedures 175 New participatory initiatives across the advanced industrial democracies 182 Review powers held by courts, by country 201 Total number of individual rights claims pursuant to Article 234, EU Treaty, by country and policy domain, 1961–97 204 9.3. Total number of rights claims brought before the European Court of Human Rights, by country, 1961–97 206 9.4. Important statutes and case law relating to standing rules for citizen environmental litigation in the United States and Canada, 1970s–1990s 210 9.5. Important statutes and case law relating to standing rules for citizen environmental litigation in European countries, 1960s–1990s 212 9.6. Important statutes and case law relating to standing rules for citizen environmental litigation in Australia and Japan, 1960s–2000 214 10.1. Venues of democracy 226 11.1. Dahl's democratic principles 258
Notes on Contributors Christopher K. Ansell is an Associate Professor of Political Science at the University of California, Berkeley, where he teaches organization theory, public administration, and political sociology. His current interests include European integration, regulation and risk analysis, and multi-level governance; and he is presently working on a project on European food safety regulation. He is the author of Schism and Solidarity in Social Movements: The Politics of Labor in the French Third Republic (2001). Shaun Bowler is Professor of Political Science at the University of California, Riverside. His interests include the comparative analysis of democratic institutions in relation to citizen participation. His publications include Demanding Decision (1998) and Alternative Electoral Systems and Minority Representation (2003); and he co-edited Demanding Choices (2001), Elections in Australia, Ireland and Malta under the Single Transferable Vote (2000), and Citizens as Legislators (1998). Bruce E. Cain is Robson Professor of Political Science at the University of California, Berkeley, and Director of the Institute of Governmental Studies. He was a Rhodes Scholar (1970–2) at Trinity College, Oxford. In 1976 he received his Ph.D. from Harvard University. His writings include The Reapportionment Puzzle (1984), and co-authorship of The Personal Vote (1987) and Congressional Redistricting (1991). He has also co-edited numerous books, including Developments in American Politics (2002), Constitutional Reform in California (1995), Racial and Ethnic Politics in California (1998), and Voting at the Political Fault Line: California's Experiment with the Blanket Primary (2002). Cain has served as a polling consultant for state and Senate races to Fairbank, Canapary, and Maulin (1985–6); redistricting consultant to national, state, and local governments; and Special Master for the Arizona State Legislative Redistricting in 2002. He received the Zale Award for Outstanding Achievement in Policy Research and Public Service in March 2000, and was elected to the American Academy of Arts and Sciences in April 2000. Elisabeth Carter completed her doctorate at the University of Manchester and now holds a research position in European politics at Keele University in Staffordshire, England. Her main research interests include European political parties (in particular those of the extreme right), party systems, and electoral systems. Her work on right-wing extremism and electoral systems has appeared in West European Politics and Representation.
NOTES ON CONTRIBUTORS
xv
Rachel Cichowski is Assistant Professor of comparative law in the Department of Political Science with a joint appointment in the Law, Societies and Justice programme at the University of Washington. Her visiting research positions include Visiting Research Fellow at the European University Institute, Florence (1998–9) and Visiting Fellow at the Max Planck Institute, Bonn (2000). Her primary research interests include comparative law and politics, empirical democratic theory, and European integration. Her publications include State of the EU: Law, Politics and Society (2003) and her research is also published in numerous edited volumes and journals, including Comparative Political Studies, Journal of European Public Policy, and Women & Politics. Russell J. Dalton is Professor of Political Science and Director of the Center for the Study of Democracy at the University of California, Irvine. He has been a Fulbright Professor at the University of Mannheim and held a German Marshall Fund Fellowship. His scholarly interests include comparative political behaviour, political parties, social movements, and empirical democratic theory. His recent books include Democratic Challenges, Democratic Choices (2003), Citizen Politics (2002), and The Green Rainbow: Environmental Interest Groups in Western Europe (1994). He is co-author of Critical Masses: Citizens, Environmental Destruction, and Nuclear Weapons Production in Russia and the United States (1999), and co-editor of Parties without Partisans (2001) and Germans Divided (1996). Patrick Egan is a doctoral student in the Department of Political Science and a researcher at the Institute of Governmental Studies at the University of California, Berkeley. His research interests include public policy, political behaviour, and political regulation. His dissertation focuses on the relationship between public opinion and social policy in the United States. Sergio Fabbrini is Professor of Political Science at Trento University (Italy) and of Comparative Political Institutions at the Ph.D Program, Sienna University (Italy). He has published seven books, one co-authored book, seven edited or co-edited books, and one hundred scientific articles and essays in the field of comparative politics, transatlantic relations, American and Italian politics, political theory, and public policies. He is editor of Nation, Federalism and Democracy (2001), The EU: Institutions and Actors of a Supranational System (2002), and The Europeanization of Italy (2003). David Farrell is a Jean Monnet Professor of European politics at the University of Manchester and chair of the London-based McDougall Trust. A co-editor of the journals Party Politics and Representation, Professor Farrell has written widely on the study of parties, campaigns, and electoral systems. His most recent books include Electoral Systems (2001), Do Political Campaigns Matter? (co-edited, 2002), and Political Parties in Advanced Industrial Democracies (co-edited, 2002). He is currently completing two projects, one on the representative role of Members of the European Parliament (with Roger Scully) and the other on Australia's electoral systems (with Ian McAllister).
xvi
NOTES ON CONTRIBUTORS
Jane Gingrich is a doctoral candidate in political science at the University of California, Berkeley. Her research interests include public sector reform, comparative social policy, and West European politics. Mark M. Gray is an Assistant Research Professor at the Center for Applied Research in the Apostolate, Georgetown University, Washington, DC. His research interests focus on political behaviour, elections, and the role of religion in politics. His doctorate ‘In the Midst of Fellows’ (2003) estimates household influences on voter turnout in America. His research on partisan influences on public policy appeared in Parties without Partisans (2000) and his research on explaining declines in voting turnout has appeared in Comparative Political Studies. Miki Caul Kittilson is Assistant Professor of Political Science at the University of Texas, San Antonio. Her research interests include comparative political participation, parties, and women in politics. Her research has been published in journals such as the Journal of Politics, Comparative Political Studies, and Party Politics. Her current research examines declining electoral participation across long-established democracies. Susan E. Scarrow is Associate Professor of Political Science at the University of Houston. Among her main research interests are the development and impact of political parties in established democracies. Her publications include Parties and Their Members (1996), Perspectives on Political Parties (2002), edited special issues of European Journal of Political Research and Party Politics, and many journal articles and book chapters on political parties, German politics, electoral systems, and direct democracy. Alec Stone Sweet is Official Fellow, Chair of Comparative Government, Nuffield College, Oxford, and Senior Fellow, Schell Center for International Human Rights, the Yale Law School. He works in the fields of comparative and international politics, and comparative and international law. He is the author of Governing with Judges (2000) and On Law, Politics, and Judicialization (2002), and the co-editor of European Integration and Supranational Governance (1998), The European Courts and the National Courts (1998), The Institutionalization of Europe (2001), and The Politics of Delegation to NonMajoritarian Institutions in Europe (2002). Mark E. Warren is Professor of Government at Georgetown University, Washington, DC. He is author of Democracy and Association (2001), editor of Democracy and Trust (1999), and author of articles and book chapters in the areas of democratic theory and European political thought.
Foreword Since the 1960s a number of political commentators around the world have hailed what they call ‘the triumph of democracy.’ The signal event, of course, was the dissolution of the Soviet Union in 1991 followed by the partial (Russia and Ukraine) or complete (Estonia, Latvia, and Lithuania) democratization of some of the former USSR's constituent republics. The democratic tide has also surged in Eastern Europe (Hungary, Poland, the Czech Republic), Latin America (Bolivia, Paraguay, Nicaragua), and Africa (Botswana, Ghana, Namibia, South Africa). Freedom House's annual categorizations of the world's nations as Free (fully democratic), Partly Free, or Not Free confirm the trend: in 1984, only 32 percent of the nations were classified as Free, while in 2002, 45 percent—including well over half of the world's population—were rated Free.1 The authors and editors of this book, however, do not hail (or jeer) these developments. Rather, they are concerned with what is happening to the institutions, especially of representative government, that have traditionally implemented democracy in the advanced industrial democracies in which they have long been established. Accordingly, they focus their research and commentary on what is happening in eighteen ‘large, continuously democratic OECD member nations’: Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Ireland, Italy, Japan, the Netherlands, New Zealand, Norway, Sweden, Switzerland, the United Kingdom, and the United States. In some instances the authors cite the works of others, and in some they report the results of their own studies. On the basis of this impressive body of empirical data, the editors, Bruce E. Cain, Russell J. Dalton, and Susan E. Scarrow, and their twelve contributors report a major decrease in popular confidence in many of the traditional institutions of representative democracy and the people who operate them in the eighteen countries. The evidence they cite includes the general decline in voter turnout in elections; the shrinking of dues-paying membership in most major political parties; the growing scepticism about and declining esteem for politicians and political parties; and the widespread and growing general dissatisfaction with the traditional institutions and processes of representative government. Many other commentators have pointed (generally with alarm) to the same trends, but Cain, Dalton, and Scarrow point out that these developments have been accompanied by others that, at least at first glance, seem quite inconsistent: the general increase in public interest in politics and increasing public participation in other forms of political action, such as signing petitions and joining in the activities of citizen
1
Freedom House, Freedom in the World 2001–2002.
xviii
FOREWORD
interest groups. What is happening, they believe, is not the popular rejection of democracy's basic principles but rather a growing desire for new institutions that will permit and encourage ordinary citizens to participate directly and unmediated in government's decision-making processes. This leads these authors to suggest that there are not one but three modes of democracy that should be used to analyze changes. They are (1) Schumpeterian democracy, with its elected elite policy-makers held accountable by the people in periodic regular elections; (2) direct democracy, as manifested mainly in national, regional, and local referendums; and (3) advocacy democracy, in which citizens and public interest groups directly participate in policy-formation and/or administration by pressing their groups' causes. By thus broadening the concept of democracy the book's editors and authors seek a more comprehensive understanding of recent changes in democracy not usually considered by devotees of either the Schumpeterian competing-elites model or the Rousseauian town-meeting model. The book's middle chapters (2–10) present cross-national surveys of material from the eighteen nations bearing on a wide variety of topics including the expansion of the number and variety of elections and its impact on the variety and levels of citizen participation; changes in the rules of political parties governing the methods of choosing party leaders and candidates; the increase in the number and role of national, regional, and local referendums; the increase in the number of freedom-of-information acts and access to formerly secret government records; changes in the structures and procedures of administrative agencies providing easier citizen access to and participation in administrative decisions; the devolution of legislative and administrative powers from the national level to regional and local levels; and the growth of judicial concern with protecting human rights against national governments. In sum, Democracy Transformed? makes a number of significant contributions to the study of democracy in the modern world. Its conceptual innovations, particularly in setting forth the proposition that ‘advocacy democracy’ is a third mode, deserving attention comparable to that devoted to representative democracy by such contemporary theorists as Robert Dahl and Benjamin Barber, broaden and deepen the discussion. Its surveys of the democratic developments in the eighteen OECD advanced industrialized nations constitute the most comprehensive as well as the most recent compilation of data about what is happening to traditional democratic institutions in the modern world. And, as the standards of political science require, the authors tell us what is not understood as well as what is, thereby suggesting a useful path for future research on democracy and its current development. Anyone interested in the theory and practice of modern democracy should read this book. Austin Ranney University of California, Berkeley
1 New Forms of Democracy? Reform and Transformation of Democratic Institutions Russell J. Dalton, Susan E. Scarrow, and Bruce E. Cain Democracy in the full sense of the word will always be no more than an ideal; one may approach it as one would a horizon, in ways that may be better or worse, but it can never be fully attained. (Vaclav Havel addressing a joint session of the US Congress, 21 February 1990) Over the past quarter century, citizens and political elites in advanced industrial democracies have displayed a growing willingness to question whether a fundamental commitment to the principles and institutions of representative democracy is sufficient to sustain the legitimacy and effectiveness of current mechanisms of self-government. In most of these nations, turnout in elections has declined, as has party membership and various forms of electoral participation (Gray and Caul 2000; Blais 2000; Wattenberg 2002). In addition, the public is increasingly sceptical of politicians, political parties, and political institutions (Dalton 2004; Pharr and Putnam 2000; Norris 1999a). These signs point to a spreading dissatisfaction with the institutions and processes of representative democracy. These trends are often concomitant with increasing demands for political reforms to expand citizen and interest group access to politics in new ways, as well as to restructure the process of democratic decision-making. Public interest in politics is generally growing across the advanced industrial democracies (Dalton and Wattenberg 2000: 57; Putnam 2002).1 Although electoral participation is generally declining, participation is expanding into new forms of action. Today, more people are signing petitions, joining citizen interest groups, and engaging in unconventional forms of political action (Inglehart 1997; Dalton 2002: ch. 4; Norris 2002: ch. 10). The enormous expansion of public interest groups, social movements, and non-governmental organizations (NGOs) creates new opportunities for action.
2
DALTON, SCARROW, AND CAIN
Reforms to implement these new demands for citizen participation can have multiple goals and multiple effects: expanding access to politics, increasing citizen control of elites, taking decisions out of the hands of elites, prompting more responsive policy-making, and engaging new groups into the political process. And, given this expanded menu of participation choices, it is perhaps not surprising that some traditional forms of electoral participation are declining. These trends suggest that the public's preferred mode of democratic decision-making is moving toward new forms of more direct involvement in the political process. Public opinion surveys routinely find that large majorities favour shifting political decision-making from elites to the citizens themselves (Dalton, Buerklin, and Drummond 2001; Bowler, Donovan, and Tolbert 1998). Reflecting these tendencies, contemporary publics and political groups seemingly place greater reliance on referendums as a tool for policy influence and agenda setting (Butler and Ranney 1994; Gallagher and Uleri 1996). For example, as Italian parliamentary politics struggled with scandals and immobilism during the 1990s, referendum usage increased dramatically in that country. Growing interest in the processes of deliberative or consultative democracy is another indication of this trend (Fishkin 1995; Elster 1998). There are also regular calls for greater reliance on citizen advisory committees for policy formation and administration, especially at the local level where direct involvement is possible. Now a chorus of voices is calling for democracies to reform and adapt to changing political conditions and a changing public. Benjamin's Barber's (1984) ‘strong democracy’ and Robert Dahl's (1989) discussion of ‘transformative democratic reform’ raise deeper questions about how democratic institutions can involve the public more directly (also see Habermas 1992a,b). Mark Warren (2001: 226) writes, ‘Democracy, once again in favor, is in need of conceptual renewal. While the traditional concerns of democratic theory with state-centered institutions remain importantly crucial and ethically central, they are increasingly subject to the limitations we should expect when nineteenth-century concepts meet twenty-first century realities’. The pragmatic American political analyst, Dick Morris (2000), similarly observes: ‘The fundamental paradigm that dominates our politics is the shift from representative to direct democracy. Voters want to run the show directly and are impatient with all forms of intermediaries between their opinions and public policy.’ Ralf Dahrendorf (2000: 311) has summarized the mood of the times: ‘representative government is no longer as compelling a proposition as it once was. Instead, a search for new institutional forms to express conflicts of interest has begun.’ The same themes are heard in government reactions to these trends. In 1999 the OECD held a symposium: ‘Government of the Future: Getting from Here to There’. The symposium report noted that technological advances and a more knowledgeable citizenry create more demands on contemporary governments. In response, the OECD began a dialogue about how its member states could reform their governments to create new connections to the public (OECD 2000a).
NEW FORMS OF DEMOCRACY?
3
Building on this experience, the OECD conducted several studies to assess existing opportunities for citizen access to information and consultation, active participation in policy-making, and ‘best practice’ reforms for expanding these opportunities. The report of these activities begins: New forms of representation and public participation are emerging in all of our countries. These developments have expanded the avenues for citizens to participate more fully in public policy making, within the overall framework of representative democracy in which parliaments continue to play a central role. Citizens are increasingly demanding more transparency and accountability from their governments, and want greater public participation in shaping policies that affect their lives. Educated and well-informed citizens expect governments to take their views and knowledge into account when making decisions on their behalf. Engaging citizens in policy making allows governments to respond to these expectations and, at the same time, design better policies and improve their implementation. (OECD 2001b: 9) As this suggests, public officials within the OECD clearly recognize these new citizen demands for greater access, transparency, and accountability. Similarly, the European Union recently issued a white paper on the need to increase citizen involvement in the policy process, to help citizens hold their political leadership to account for the EU's decisions and to guarantee that the EU's system will be opened up to greater public scrutiny and debate (Commission of the European Union 2001). In summary, there is an apparently growing consensus on the need for political reform to adapt democracy to new conditions; and the cumulative experiments in democratic reform suggest that important new developments may be at hand. The most avid proponents of such reforms conclude that we may be experiencing the most fundamental transformation of the democratic process since the creation of mass democracy in the early twentieth century. Yet cycles of democratic reform are a recurring theme in history, and pressures for change in one direction often wane as new problems and new possibilities surface (see next section). So in discussing the impact of reform we must go beyond the rhetoric and ask whether these changes are really transforming the foundations of the democratic process or merely accommodating popular pressures without altering the basic nature of representative democracy. This book begins an assessment of the degree of democratic institutional change. A distinguished international team of scholars examines the institutional reforms that are occurring in contemporary democracies. We assess the extent of change in several areas of potential reform, and evaluate how established political actors, especially the core actors of representative democracy, are responding to these new structures. Finally, we consider the apparent and potential effects of these reforms on the workings of the democratic process. Certainly a single book cannot fully and definitively assess the extent and implications of the current democratic transformation. But we can begin to develop an impression of how democracy's future is developing, and of whether contemporary
4
DALTON, SCARROW, AND CAIN
reforms are moving us closer towards Vaclav Havel's democratic ideal or in a different direction. This is the goal of our volume.
THE FIRST WAVE OF DEMOCRATIC REFORM At one level, there is nothing new about the call to inject more democracy into the institutions of representative government. The history of modern democracies has witnessed repeated waves of debate about the nature of the democratic process, some of which have produced institutional reform (Huntington 1981, 1993; Lipset 1981). These debates have revolved around two questions. First, how does one ensure that democratic processes best represent the views of the governed? This involves the sub-questions of who should be decision-makers and how they should be selected and also raises the more normative question of whether decisions are legitimate if the process for reaching them is not broadly inclusive and equitable (Held 1996). Thus, a first theme in the democratization process is whether to expand and improve the process of representative government as the ‘democratic compromise’ between the rulers and the ruled. The second question concerns the desirability of direct citizen participation in government as an alternative (or corrective) to representative democracy. Scholars and practitioners repeatedly debate whether it is possible to design procedures that expand citizen participation without overly impairing the efficiency of government. Direct democracy is understandable in small settings such as the New England town hall or the Swiss canton. It is less certain how direct democracy can be applied to large-scale national policy decisions and debates over complex policy matters. Thus, a second theme in the democratization process involves a search for new procedures or technologies that allow more direct citizen input into the political process. These questions about how to realize democracy are as old as efforts to institutionalize popular government. The patrician framers of the then radically democratic United States Constitution fundamentally rejected the idea of pure democracy. Madison, Hamilton, and others argued that representative structures were better, in part because they made it possible to ‘refine and enlarge the public views, by passing them through the medium of a chosen body of citizens, whose wisdom may best discern the true interest of their country, and whose patriotism and love of justice will be least likely to sacrifice it to temporary or partial considerations’ (Federalist 10). Similar views were stated by Edmund Burke in his famous ‘Address to the Electors of Bristol’ in 1774 and by Abbé de Sieyès in describing the events of the French Revolution. In short, this logic called for a system of limited representative government as the basis of the democratic process. Of course, these views were challenged by others at the time. Jefferson, Payne, and other early American democrats expressed greater faith in the ability of the average citizen to make political decisions. Moreover, Jefferson stressed the
NEW FORMS OF DEMOCRACY?
5
educational element of democracy: by participating in the decisions affecting their lives, citizens learned how to become better democrats. From this tradition began the mantra of populist democratic reformers for the next two centuries: the cure for the ills of democracy is more democracy. Belief in the innate superiority of representatives lost ground during the nineteenth century. By the end of the century an increasing number of reformers questioned both the notion that representatives ought to act as trustees and the idea that public decisions were improved by legislators' independence from the wishes of those who elected them. This change came about because legislators were increasingly seen as captives of parties or of special interests; many people felt that politicians ignored the wishes of their voters for selfish reasons, not because they were honestly working for the common good. This distrust of elected officials stimulated the first waves of pressure for democratic reform in the late nineteenth and early twentieth centuries that defined the Progressive era. Coupled with this dissatisfaction with political elites was a growing trust in the competence of the broader electorate. In the United States, this wave of democratic discourse had multiple effects (Goodwyn 1976). Progressive-era reforms introduced a variety of changes to strengthen the electoral process, such as the introduction of the secret ballot in the late 1800s or the extension of the franchise to women in the early 1900s. Other reforms, such as the introduction in 1913 of the direct primary and direct elections of US senators, expanded the citizens' role in the electoral process as a corrective to the corruption of machine politics and special interests. These reforms sought to ensure that democratic elections were more representative, so that the public's wishes would be translated more clearly and directly into governmental action.2 Second, many of these same American reformers developed an interest in direct democracy through their new-found attention to constitutional and legislative referendums in Switzerland (Sullivan 1892; McCrackan 1893; Oberholtzer 1900; Wuarin 1895). Populist and later Progressive reformers promoted devices to give citizens a more direct say in politics and portrayed these changes as tools for wresting power back from corrupt politicians and from the powerful economic interests that dominated them (Cree 1892; Croly 1914). These arguments led to the widespread adoption of legislative and constitutional referendums and initiatives at the state level in the United States. Between 1900 and 1920, twenty-two American states adopted the referendum and/or the initiative into their state constitutions (Cronin 1984: 52). Similar arguments were used a little later in the twentieth century in Canada, where populists promoted direct democracy as a means of shaking off the shackles of entrenched interests (Laycock 1990: 37–40). During this same period there were similar moves to expand democracy in Europe, both by improving the methods of representation and by using direct democracy to circumvent representation. A broadening franchise rapidly expanded democratic participation in the late nineteenth century and early twentieth century. Even after the ‘democratizing’ First Reform Bill of 1832, only 650,000 voters
6
DALTON, SCARROW, AND CAIN
registered for the British election of 1833, at a time when the population of United Kingdom exceeded 16 million (Bentley 1987: 88). By 1921 Britain had nearly universal adult male suffrage. In the wake of expanded manhood suffrage, female suffrage spread across a host of European nations in the first quarter of the twentieth century. At about the same time, many countries witnessed debates about which electoral system would most fairly represent public opinion in elections, debates that eventually led to the adoption of proportional representation electoral systems for most of Europe's lower legislative chambers. These reforms were intended to strengthen the process of representative democracy so that participation yielded equitable and representative results. Europeans, like Americans, also were intrigued by the new Swiss institutions of direct democracy (see Dicey 1894; Deploige 1898; Rittinghausen 1877). Europeans held different views about the benefits of direct democracy, however. British advocates of such reforms, for example, were motivated by the fear that the parliamentary system was inadequate to deal with complicated issues on which parties were divided, such as home rule for Ireland. British interest in introducing the referendum as a ‘people's veto’ became strongest around 1911, in conjunction with the constitutional showdown between the House of Commons and the House of Lords. In light of the rising power of the House of Commons, some viewed the referendum as a potential new check on the government and on the political party that controlled it (see Belloc and Chesterton 1911; Lecky 1896; Selborne 1911). Some continental Social Democrats saw direct democracy as a logical extension of socialist calls for true popular government, and in this spirit the German Social Democratic Party's 1875 Gotha Programme and 1891 Erfurt Programme contained calls for direct popular votes on legislation. This party's long tradition of support for direct democracy bore fruit in 1919, when Social Democrats successfully backed the introduction of referendums into Germany's new (‘Weimar’) constitution. However, the European left did not unanimously support direct democracy. Some leftists opposed the referendum because they feared that ‘the people’ would be more conservative than their elected representatives (an argument made by Henry Maine as early as 1886; see Delbrück 1914; Kautsky 1898; MacDonald 1909). Despite these disagreements on both sides of the Atlantic about exactly what direct democracy would achieve, the first decades of the twentieth century saw a growing acceptance of referendums as a device for ensuring that major political decisions had broad popular support. This was one reason that the victorious Western Allies promoted the use of plebiscites to settle issues of national self-determination at the end of the First World War (Suksi 1993). Several of the new constitutions written in the wake of the First World War also included provisions for referendums. The twentieth-century advocates of a new science of politics added their voices to these debates about institutional reform. They argued that the defects of representative democracy could not be remedied by entrusting more decisions to citizens (Goodnow 1900; Bruce 1924; Lowell 1913). These scholars questioned the claim that the cure for democracy was more democracy. Instead, they suggested that more decisions should be taken by professional policymakers who could be
NEW FORMS OF DEMOCRACY?
7
insulated from parties and electoral politics. Such insulation might include expanding merit-based civil services or establishing life tenure for judges. These analysts were well-aware of the problems of representative democracy, but they rejected the idea that more democracy could solve them, arguing that many issues were just too complicated for the mass of citizens to be able to make informed decisions. Some of those who hailed the virtues of non-partisan professional policy-makers in Third Republic France helped establish the school that became the training ground for generations of French administrators and political leaders: the Ecole Libre des Science Politiques (Quagliariello 1996). The various waves of democratic reform had begun to recede by the 1920s, and were in full ebb by the 1930s. The vulnerability of new democracies to populist and fascist movements in the inter-war era and the undemocratic use of plebiscites in this period prompted many observers to revise their ideas about the desirability of allowing ‘the masses’ greater direct access to government. In particular, the failure of the citizenry was widely (if unfairly) cited as the reason for the collapse of democracy in Weimar Germany (Schiffers 1971). The abuse of the referendum by the Third Reich further worsened the image of direct democracy. The fear of ‘too much democracy’ and an unpredictable mass public was thus very prominent as the Second World War came to an end (Kornhauser 1959). As a result, by the middle of the twentieth century many democratic politicians and scholars returned to a more Madisonian view that saw representative institutions as having an elevating influence, and there was more wariness of giving ‘the masses’ too much direct say. This attitude was best exemplified by the (West) German constitution of 1949, which reacted against perceived deficiencies of the Weimar constitution both by eliminating provisions for direct democracy, such as the national referendum, and by restricting the process of representative democracy by eliminating direct elections of the federal president. Among academic theorists and constitutional scholars, corporatist and pluralist paradigms stressed the role of bargaining among groups and group leaders, and downplayed a direct governance role for individual citizens. The most notable exception during this period is France—but it may be the exception that ‘proves’ the generalization. President de Gaulle resurrected the use of national referendums during the early years of the Fifth Republic, but this reinforced many observers' doubts about direct democracy. De Gaulle's referendums were more reminiscent of the Napoleonic plebiscites of the nineteenth century than the model of direct democracy. These referendums were dictates from on high that the public could (and should) endorse. One plebiscite even went so far as to unconstitutionally revise the constitution.
THE SECOND WAVE OF DEMOCRATIC REFORM The democratic tide shifted again as a new wave of democratic debate and rhetoric grew in the last third of the century. The stimulus for change often appeared first among university students and young professionals who pressed the boundaries of
8
DALTON, SCARROW, AND CAIN
the conventional system of representative democracy (Altbach and Laufer 1972). The Free Speech Movement in the United States was an early example of the rejection of traditional political processes and a search for alternative methods of political expression and influence. These sentiments soon broadened to include social protests over race, urban issues, and the Vietnam War. A strikingly parallel wave of student unrest spread across Europe in the late 1960s and the early 1970s: the Provo movement in Dutch politics in the late 1960s, the Alternativ and APO movements in Germany, and the RAF assaults on the Italian government during the 1970s. In France, the revolts of May 1968 marked the apogee of the student movement. Moreover, embedded within the French student movement were clear challenges to the established system of representative democracy. Although these dramatic protests subsequently waned, the new challenges to democracy that they embodied are still affecting advanced industrial democracies. Calls for reform in recent decades seem to emanate from a complex mix of needs and motivations. One factor may simply be the underlying logic of democracy. Participation and consensusbuilding are essential characteristics of the democratic process. Once these values become accepted, there may be an inevitable pressure to expand these processes to allow greater citizen access and ensure the effectiveness of democratic participation. For instance, after becoming German Chancellor in the late 1960s, Willy Brandt challenged Germans to ‘risk more democracy’. Democratic expectations may also expand to include other domains, such as Dahl's (1989) emphasis on democratic reforms in the economic sphere. In other words, the logic of democracy may generate its own expectations for the expansion of the democratic process. Relatedly, the modernization process in advanced industrial democracies also may contribute to calls for democratic reform (Luhmann 1998; Lipset 1981). The dramatic growth of public interest groups in the United States and citizen action groups in Europe has introduced new actors and new styles of action into the democratic process. From the 1960s onwards there was a general expansion of new forms of direct action, used by student protestors but also by environmentalists and other social movements, and eventually by the Gray Panthers, teachers, neighbourhood associations, and a wide spectrum of society. The increasing social and political diversity of interests active in the democratic process may generate pressures to formalize the rules as well as to assure the transparency and legitimacy of the process. In addition, the increasing skills and resources of average citizens may stimulate new demands and expectations about how citizens will influence the decisions affecting their lives (Inglehart 1990; Dalton 2002: ch. 2). Thus, advanced industrial democracies are experiencing a ‘participatory revolution’ in which new forms of political action compete with the traditional participation style of representative democracy.3 Another explanation of the changes focuses on Schumpeter's (1943) model of elite competition. The expansion of political actors stimulates a search for new opportunities for political access and influence. Elites try to change the rules of the
NEW FORMS OF DEMOCRACY?
9
democratic game to produce a fairer process or more transparent rules, or even to manipulate the rules to their advantage. For example, the electoral primary was introduced by a conservative party in Iceland to create a new basis for attracting voters; in response, this innovation was gradually accepted by the other parties. From this perspective, the changing patterns of democratic institutional structures are part of an ongoing process of political competition between contending forces in a democracy, perhaps accelerated by an increasing number and diversity of actors. A contrasting view suggests that the congestion of the governance process may stimulate institutional change. In line with the governmental overload approach (for example, Crozier, Huntington, and Watanuki 1975), the claim is made that the demands for access and influence now exceed the ability of democratic governments to respond. This might arise either from increases on the input side or because of restrictions on throughput by party or parliamentary elites (as posited by the cartel thesis of party action). Such a development would stimulate attempts to develop new access points through processes of institutional reform and restructuring. Governments themselves decide to delegate authority to administrators or the courts, which leads citizens and elites to try to influence these new decision-making centres. Finally, as documented in the chapters that follow, democratic innovation diffuses across the advanced industrial democracies. Innovations such as freedom of information laws or the office of ombudsman have quickly spread across the OECD nations. In part this is because information-sharing is more common in the modern age. But such patterns also arise because international institutions such as the OECD and the European Union encourage their member states to discuss common problems and share common solutions. Undoubtedly such exogenous factors also contribute to the democratization wave of the late twentieth century. Our project focuses on assessing the degree of institutional change in contemporary democracies and the implications of these changes, rather than testing alternative causal explanations for these processes. The analytic chapters demonstrate, however, that a mix of factors is needed to explain the patterns we uncover. But whatever the sources of change, we are particularly interested in the question of how these reforms are affecting the democratic process.
THREE MODES OF DEMOCRACY The contemporary pressures for democratic reform, as in earlier democratization waves, first appeared in efforts to improve the process of representative democracy, in which citizens elect elites who then deliberate on policy matters and ultimately make the policy choices for the nation. US presidential elections are an especially striking example of the trends to reform representative processes. In a thirty-year span there was a dramatic shift toward expanding citizen influence by selecting candidates through primary elections. In 1968 the Democratic Party held seventeen
10
DALTON, SCARROW, AND CAIN
presidential primaries and the Republicans sixteen; in 2000 there were Democratic primaries in forty states and Republican primaries in forty-three. In addition, first the Democratic Party and then the Republican Party instituted reforms intended to ensure that the convention delegates were more representative of the party's supporters (Shafer 1983). Public funding of presidential elections sought to limit the influence of money and thereby ensure the equality of citizens. More recently, reformers have championed devices such as term limits and campaign finance reform as the new populist causes to remove the influence of special interests from the democratic process (Carey, Niemi, and Powell 2000). If Dewey and Truman were brought back to observe the modern presidential election process, they would hardly recognize the system as the same one that nominated them in mid-twentieth century. The institutionalized system of party government in Europe restrained some of the populist reforms that occurred in the United States, but there are parallels in many nations. On a limited basis, some political parties have experimented with, or even adopted, closed primaries as a means of selecting parliamentary candidates (Scarrow, Webb, and Farrell 2000; Bille 2001; Hopkin 2001). There is some evidence that party members are wielding greater influence in the selection of party candidates (Farrell and Webb 2000). Other reforms apparently increase the number of electoral choices available to voters by transforming appointed positions into elective offices (Scarrow 2001). Suffrage was again expanded, this time to include voters aged 18–20. In summary, these reforms are expanding the realm of representative democracy, and thus implicitly accepting the ultimate premise of democracy by representation. Second, this democratization wave stimulated calls for new forms of direct democracy that bypass (or compliment) the processes of representative democracy. With direct democracy, citizens both participate in the discussion and deliberation about policies, and then make the final policy choice: it is unmediated participation in both policy formation and policy decision. For example, referendum usage has increased in the United States and other democracies. The Initiative and Referendum Institute calculates that there were 118 statewide referendums in the US during the 1950s; this figure increased to 378 referendums in the 1990s. Several other nations have amended laws and constitutions to provide greater opportunities for direct democracy at both national and local levels (Scarrow 2001). For instance, Britain had its first national referendum in 1975, Sweden's constitutional reform introduced the referendum in 1980, and Finland adopted the referendum in 1987. In these and other cases, the referendum won a new legitimacy as a basis for national decision-making, one that runs strongly counter to the ethos of representative democracy. There is also mounting interest in expanding direct democracy into new institutional forms, such as citizen juries and methods of deliberative democracy (Matthews 1999; Fishkin 1995). Third, this democratization wave expanded the participation repertoire to include a new style of advocacy democracy in which citizens directly participate
NEW FORMS OF DEMOCRACY?
11
in the process of policy formation or administration (or participate through surrogates such as environmental groups and other public interest groups), although the final decisions are still made by elites. In other words, participation of this type seeks to influence the process rather than make outright decisions, as is done with referendums and initiatives. For example, the principle of ‘maximum feasible participation’ became the watchword of the Great Society social service reforms in the United States in the 1960s. As a result, citizen consultation and public hearings are now embedded in an extensive range of legislation, giving citizens new access points to government policy formation and administration (Ingram and Smith 1983; Berry, Portney, and Thomson 1993). Congressional hearings and state government meetings are public events in the US, and legislation such as the 1972 Federal Advisory Committee Act even extended open-meeting requirements to advisory committees. This yields, for example, the contrast between the system of Environmental Impact Reviews (EIRs) and public hearings now required by US environmental policy on the one hand and the traditional closed system of British environmental policy-making protected by the Official Secrets Act on the other. In 1970 only a handful of nations had freedom of information laws; now they are almost universal in the OECD nations. It is not difficult to guess what Edmund Burke would think about the ‘government in the sunshine’ provisions that have been enacted in recent years. Advocacy democracy also includes the new legal rights that citizen groups and individuals have won that give them access to information and influence. The judicialization of the policy process enables citizen groups in most Western democracies to develop class action suits on behalf of the environment, women's rights, or other public interests (Stone Sweet 2000). Virtually every public interest can be translated into a rights appeal, which provides a new avenue of action through the courts. This is a new form of policy participation that was seldom available to individual citizens or interest groups a generation ago.
BRITAIN AS AN ILLUSTRATION Britain's experience in the later twentieth century illustrates the processes we are studying. In 1960 Britain was considered a highly successful democracy. Indeed, the support of the British people for their government and their nation was one factor that sustained them through the Second World War and was a hallmark of British democracy (Almond and Verba 1963). At the same time, however, citizens' actual access to the democratic process was limited in terms of both the processes of representative democracy and direct and advocacy democracy. For most British citizens, their participation focused on electoral involvement. But actual input through the process of representative government was quite limited: a single vote for a Member of Parliament every five years, and perhaps single votes for the local and county councils. Participation beyond elections was quite limited.
12
DALTON, SCARROW, AND CAIN
Gradually pressures mounted for greater democratic access. In response, various reforms increased the role of elections. For instance, the creation of the Scottish Parliament and the Welsh National Assembly in the 1990s expanded citizens' electoral input, as did the new direct elections to the European Parliament and the Local Government Act of 2000 allowing cities to institute directly elected mayors. New parties emerged to represent regional interests and for short periods new parties such as the Social Democrats and Greens appeared on the electoral stage. Other reforms sought to increase the citizens' role within the political parties, such as the restructuring of the Labour Party nominating process in the 1980s and the increasing formalization of participatory rights for individual members in all the major parties. Direct forms of citizen access also expanded during this period. In 1975 Britain held the first national referendum in its history, which was followed by several regional referendums on devolution. Even if referendums are used infrequently, their existence has shifted the content of political debate in Britain, where now the public rather than Parliament is seen as the source of political sovereignty. There are also clear prospects for further national referendums (such as on membership of the eurozone). Protest and other forms of direct political action increased; in place of the deference of British political culture arose a culture that tolerated and encouraged elite-challenging activities. Membership in the European Union brought new rights of legal standing and citizen rights that were lacking under Britain's unwritten constitution. No longer was parliament supreme, because the public could turn to the European Court of Justice to challenge parliamentary sovereignty. And Britain finally initiated a Freedom of Information Act in 2000, which lifted some of the veil of secrecy that protected governments under the Official Secrets Act. Thus Britain developed new channels for political access as well as participation in elections. Progress along each dimension is uneven, and likely follows a non-linear course. But gradually Britain has moved from a starting point of limited public access in 1960 towards more opportunities for citizen involvement in the process through representative institutions, direct democracy, and advocacy democracy channels. Our estimate of the degree of change along these dimensions is imprecise as we begin this study, but even at the outset we can see that access has increased on all dimensions. The guiding hypothesis of this study is that the changes illustrated by the British example are far from unique. Although other nations may have different starting points, we posit that they, too, have followed a similar trajectory in developing both representative and direct channels. Determining the degree of change along these three dimensions and the relationship between dimensions is one of the main empirical goals of this project.
THE CROSS-NATIONAL PATTERN In Britain as in most other advanced industrial democracies, each wave of democratic reform has brought forth multiple claims about the merits of democratization,
NEW FORMS OF DEMOCRACY?
13
only some of which are realized. Calls for democratic reform generally favour changes to strengthen both the system of representative government and direct democracy. Thus, the American populists of the early 1900s advocated both electoral reform and institutional changes that took decisions away from the electoral arena. Similarly, many democratic reformers of the late twentieth century pressed for electoral reforms as well as new methods of direct democracy and advocacy. These processes are often intertwined as part of the democratic passion of reformers. The recent democratic reform wave contains all three elements, but we believe it is important to realize the important theoretical and political differences between them. In addition, history shows that the democratic process is dynamic and complex—so that the consequences of institutional reforms are not always as predicted. Some scholars—including some contributors to this collection—doubt the Jeffersonian logic that the cure for democracy is more democracy. These analysts argue that overly extensive reforms may undermine the democratic process by making governing so complex that accountability and transparency are diluted. Meny and Surel (2002), for instance, claim that constitutionalist expansion of democratic institutions and processes actually reduces the space left for the public (see also Bobbio 1987). Another argument holds that, by eroding governability, democratic reforms may make government less effective (Huntington 1981). One of the strongest critical statements comes from Hibbing and Theiss-Morse (2002), who claim that Americans' current disenchantment with politics actually reflects a public preference for fair politicians to make policy without bothering the citizen to become engaged. This is a position we find inconsistent with their own evidence, and which we will demonstrate to be inconsistent with the processes of institutional change in advanced industrial democracies. We must also realize that actors in the democratic process adapt to changes in the political rules of the game, and these adaptations may undermine reforms or exchange one set of problems for another. Treschel and Sciarin (1998) suggest that the Swiss referendum process pushes elites into consensual decision-making that goes beyond the normal democratic tenets of majority rule. Similarly, in the midst of his current efforts to reform campaign finance laws in the United States, Sen. John McCain has openly acknowledged that if the reforms were passed they would have effect for a limited period until special interests adapted to the new system, and then a new round of reform would be needed. In short, because of the complexity of the democratic process, history teaches us to be cautious in making predictions about the long-term consequences of institutional reforms. The following chapters assess the degree of institutional change that has actually occurred in the democratic institutions and processes of advanced industrial democracies. Are the initial examples of reform cited in the literature typical of a general pattern of institutional change, or is it their exceptionality that leads to their visibility in the literature? In addition, how have the three broad currents of democratic reform—the strengthening of representative democracy, the expansion
14
DALTON, SCARROW, AND CAIN
of direct democracy, and the introduction of new forms of advocacy democracy—shaped this reform wave? To what extent has the rhetorical shift towards greater direct citizen governance been matched by change in practices or in outcomes? In examining contemporary democratic institutional reforms our project focuses on three principles that tap the foundations of democracy, whether through representative, direct, or advocacy channels. Our analyses focus on three key principles of democratic politics. 1.
2.
3.
Access. We first ask whether access to the political process has expanded over time. Access to decision-makers and participation in collective decisions is key to the democratic process; and the processes of representative democracy, direct democracy, and advocacy democracy can provide additional channels of access. Thus we examine whether the opportunities for citizen access have expanded within each domain, and whether citizen usage of these opportunities has increased. Transparency. We are also concerned that citizens and other democratic participants have sufficient knowledge of the policy choices and the methods of policy-making. Better information about government policy, for example, empowers citizens regardless of which means of access they use. In addition, open government should increase the accountability of political elites as policy is made ‘in the sunshine’ instead of behind closed doors. Accountability. Finally, for democracy to be meaningful, public preferences should have a clear impact on policy outcomes. This is the essence of the democratic ideal. Many of the reforms we discuss are aimed at increasing the political accountability of elites, whether in party and parliamentary politics or in the administration of public policy.
These three principles are our initial yardsticks for measuring the effect of recent institutional reforms in changing the functioning of the democratic process. They enable us to assess whether contemporary democracies are moving towards the democratic ideal or experiencing a transformative set of reforms that are reshaping the nature of the democratic process.
OUR RESEARCH FRAMEWORK Pressures for democratic reform are clearly visible in most advanced industrial democracies, but often the discussion of these phenomena is narrowly focused on a single institutional reform or single nation. Systematic cross-national analysis is clearly needed. Moreover, a systematic analysis can show whether reformist pressures for greater direct democracy manifest themselves across most advanced industrial democracies. Our project thus focuses on the set of eighteen large, continuously democratic OECD member nations.4 These nations provide a reasonable approximation of ‘advanced industrial democracies’ that should experience the reformist pressures
NEW FORMS OF DEMOCRACY?
15
most strongly if social modernization is a major source of change. In addition, these nations display substantial variation in their party systems, electoral experiences, constitutional structures, and other political factors. Finally, empirical data sources are more accessible for this set of nations because social science research is well-developed in them. Our study includes Australia, Austria, Belgium, Britain, Canada, Denmark, Finland, France, Germany, Ireland, Italy, Japan, the Netherlands, New Zealand, Norway, Sweden, Switzerland, and the United States. Because we are interested in large-scale social and political trends, we track political changes over the democratic wave of the late twentieth century. The evidence of change first became apparent during the late 1960s and continued to the end of the century. Therefore, to the extent feasible, most analyses adopt a longitudinal research design, tracking political changes from 1960 to 2000. One of our primary theoretical goals is to develop a more comprehensive framework concerning the range of potential institutional reforms and then systematically to study developments across the advanced industrial democracies. The pressures for political reform can take many shapes and involve different aspects of the democratic process. Thus, this project examines a number of potential institutional reforms, ranging from the processes of representative democracy to those that provide citizens with direct access to policy-makers and administrators. Table 1.1 summarizes this research framework.5 A first major divide distinguishes between representative and direct democracy within the electoral arena. We divide our investigation into reforms that focus on representative democracy—inter-party, extra-party, and intra-party reforms—and changes that involve principles of direct democracy, primarily through referendums and initiatives. Another category involves the new forms of advocacy democracy that are emerging in advanced industrial democracies.
Electoral Politics: Representative and Direct Democracy Citizen access to democracy through elections is measured by the ‘amount of electing’: the number of electoral choices that voters can make. Current debates about democracy deficits often focus on the public's limited amount of electing. Some critics suggest that the institutional structures of representative democracy were consciously designed to limit the demands being placed on the voters and the ability of voters to influence political outcomes. According to this logic, an increase in the amount of electing would expand citizens' access to their government and thus the potential for political participation. In addition, a growth in referendums and direct democracy elections may reshape the democratic process and alter the behaviour of citizens and their representatives (Bowler and Donovan 2001; Scarrow 1999). In Chapter 2, Russell Dalton and Mark Gray track the changes in the total ‘amount of electing’ that has occurred in advanced industrial democracies. They establish whether citizen access has increased through an increasing number of
16
DALTON, SCARROW, AND CAIN
Table 1.1.Potential changes in democratic institutions and processes Representative democracy Inter-party competition Introduction of elections for more offices. Are parties competing for more offices over time? Are the barriers for electoral access changing in ways that may encourage or hinder new challengers?
Intra- or extra-party competition Changes in media access or campaign funding that benefit or exclude certain political parties. Opportunities for competition within parties, such as introduction of primaries, extension of voting rights for party members. Examples: introducExamples: directly tion of primaries in elected mayors in Britain and Germany, Finland, party memexpansion of the plu- bership ballots in ral executive in US Britain and Germany, greater role for party states. members and voters within parties; nonpartisan elections as in US states and localities.
Advocacy democracy Direct democracy
Direct access to policy process New opportunities New access to inforfor citizens to decide mation on governpolicy through elec- ment policy actions; tions or other forms also new access to of direct democracy additional governmental bodies.
Direct engagement in policy process Revision of administrative and judicial processes to incorporate citizen advisory boards, public hearings, legal review, and other direct input.
Examples: referen- Examples: public acdums and initiatives; cess to information, ‘government in the policy juries. sunshine’. Decentralization of decisionmaking to more accessible local and regional bodies.
Examples: German administrative reforms, US advisory process, public interest lawsuits.
NEW FORMS OF DEMOCRACY?
17
electoral choices and whether any changes are concentrated on representative or direct channels of participation. The chapter concludes by considering whether changes in the amount of electing and the specific types of elections are altering the relationship between citizens and their government in significant ways. Direct democracy normally occurs through the processes of referendums and initiatives. In Chapter 3, Susan Scarrow focuses on the development of citizen access through these forms of direct democracy. She also examines changes in the processes of representative democracy through the introduction of non-partisan elections. Scarrow considers whether the expansion of direct democracy by referendums signifies a challenge to the system of representative democracy and parties or whether it merely moves competition between party and group interests to a new venue. Partisan politics is at the heart of many current debates about reforming representative institutions. There are frequent claims that democratic electoral politics needs to become more open and inclusive, despite frequent observations that democracy is based on these internally undemocratic institutions (Schattschneider 1942; Michels 1958). Thus, contemporary parties face growing demands for more internal participation opportunities (intra-party reforms). At a different level, the degree of inclusion can be judged by the ease of access for new party contenders. Cartel parties supposedly monopolize campaign funding and government support to strengthen their own position, and presumably exclude new contenders (Katz and Mair 1995). In addition, the spiralling costs of political campaigns and the need for professional staffs, pollsters, and campaign consultants may exclude new parties (and the interests they represent). In both of these instances, expanding political access presumably improves the process of representative democracy by ensuring that political parties are more representative of their supporters and the public at large. In Chapter 4, Miki Caul Kittilson and Susan Scarrow examine several features of democratization within parties. They first examine changes in the attention devoted to democracy in party manifestos to see whether the rhetorical commitment to democracy has increased. They also use party statutes and formal procedures to assess the balance of power within parties. Some reforms have expanded access within political parties—how directly can the rank and file participate in party decision-making? Other reforms address the accountability of parties—the rules of inclusion and decision-making within the parties. The extent of these changes can potentially redefine the linkage between citizens and political parties. Shaun Bowler, Elisabeth Carter, and David Farrell examine another aspect of citizen access: whether participation in the electoral process has become easier or more difficult for new political parties (Chapter 5). They identify barriers that might facilitate or discourage the participation of new parties in elections, such as the rules governing ballot access, as well as media access and campaign funding. Then they assess the changes in these barriers over time. The chapter also considers whether any observed institutional changes have had demonstrable impact on the number of new and small parties in these electoral systems.
18
DALTON, SCARROW, AND CAIN
Advocacy Democracy A third potential area of institutional change involves expanding citizen access through non-electoral channels that involve advocacy democracy. The debates over the expansion of ‘government in the sunshine’ in the United States and the abandonment of the Official Secrets Act in Britain illustrate attempts to increase the transparency of democratic governments. Such reforms offer the potential for new citizen awareness of, and involvement in, policy planning and administration. One of the prerequisites of democracy is the ability of citizens and other political actors to make informed political choices: the transparency dimension. Bruce Cain, Patrick Egan, and Sergio Fabbrini examine the passage of freedom of information laws across the advanced industrial democracies (Chapter 6). They examine the political characteristics that lead to these laws' passage, and, with a series of case studies, they investigate whether citizens are utilizing this potential to learn about their government. In Chapter 7, Christopher Ansell and Jane Gingrich describe the devolution of government programmes, budgets, and decision-making to lower levels of government. The decentralization trend increases the importance of pre-existing forms of local and regional government and creates new local and regional administrative and legislative institutions. The chapter documents the broad sweep of decentralization reforms in the OECD countries that may increase both citizen access (by creating new arenas for action) and the accountability of government (by moving policies closer to the citizens). In Chapter 8, Ansell and Gingrich describe how administrative reforms are opening this aspect of the policy process to the citizenry and interest groups. For instance, the diffusion of ombudsman offices provides one new means of access. Then the authors examine the extent of administrative reform across these nations; they ask whether they represent significant new forms of access and whether they influence the transparency and accountability of government. Rachel Cichowski and Alec Stone Sweet then examine the changing relationship between citizens and the courts. For instance, new definitions of legal standing and citizen rights mean that almost any issue can be judicialized in the United States, giving citizens a potentially powerful way of influencing public policy. Chapter 9 tracks how citizen access to the political process is expanding over time in many nations, providing new means of influencing outcomes and new forms of political engagement.
The Consequences of Political Reform Our goal is to assemble primary evidence of how modern democracies are actually changing in their institutions and governing practices in response to pressures for democratic reform. But this is only a starting point. There are deep debates in the theoretical literature about whether populist reforms in democratic institutions are feasible or desirable—whether they will move us towards the democratic horizon
NEW FORMS OF DEMOCRACY?
19
or away from it. For example, there are ongoing debates on the actual extent of the purported reforms and whether they represent real institutional changes or merely symbolic attempts to defend the status quo (for example, Bang 2001; Bobbio 1987). In addition, there are intense debates on whether contemporary publics are sufficiently sophisticated and engaged to support an expansion of the democratic process, or whether reform will actually marginalize the citizenry (Piven and Cloward 1989; Bowler, Donovan, and Tolbert 1998). By bringing together experts from many fields, our project is uniquely situated to address these questions. In Chapter 10, Mark Warren reviews the evidence from the perspective of normative democratic theory. Given the principles of democratic politics, what are the implications of our findings for ideal definitions of democracy? What does this comparison tell us about our definitions of democracy and our present political systems? Warren asks whether there are signs of a second democratic transformation in the evidence assembled here. Chapter 11 examines how the various aspects of institutional change discussed in this volume are interrelated. Are these reinforcing processes, or does change in one area act as a safety valve so that reforms in other areas are less likely? In addition, by comparing evidence across various domains we can begin to determine how institutional and societal factors affect the processes of reform. Finally, we discuss the implications of these collective findings for the nature of modern democracy. As Gramsci once wrote, the old order is dying but the new order is not yet apparent. Our project can outline whether a new democratic order is developing in the advanced industrial democracies and, if so, the institutional forms that this takes. Thus, we conclude by considering whether the observed reforms are strengthening or transforming the democratic process.
CONCLUSION Our project posits a tension between existing institutions and norms of democracy and developing public preferences for a more participatory democratic process. The democratization wave of the late twentieth century attempted to alter this balance—and our research determines the degree of this shift. We do not anticipate that participatory reforms will erase parties or other established institutions of representative democracy. Rather, we ask whether the workings of the democratic process are broadening to include new institutional structures. If so, then it becomes important to investigate how the established political actors, especially the traditional actors of representative democracy, are responding to these new structures. Their responses will shape both the rate of institutional change and the exact trajectory of the trend. In summary, the rise of citizen movements and existing evidence of institutional reform suggest that the pressures for democratic reform are real and significant. There is a potential for a significant transformation in the democratic process, but the
20
DALTON, SCARROW, AND CAIN
breadth of these changes and their policy and institutional implications remain uncertain. This project assesses whether this potential for reform is truly being realized, and asks what the broader implications are for the nature and practice of democracy. In short, we want to know whether contemporary democracies have been transforming the representative institutions that were created nearly a century ago.
NOTES 1 2
3 4
5
The contrary evidence tends to emphasize the US experience (for example, Putnam 2000; Hibbing and TheissMorse 2002), but even in the US the increase in non-electoral forms of participation has been generally documented (Dalton 2002: ch. 4; Verba, Schlozman, and Brady 1995: 72). At the same time, these movements frequently were exclusionary in their definition of ‘the people’. Some of the nineteenth-century electoral reforms, for example, aimed at restricting the electoral participation of foreigners and immigrants. Other democratic reforms increased political inequality by increasing access in ways that require higher levels of political skills and sophistication. Thus the effect of democratic reform is not always egalitarian. While we link the beginnings of this democratization wave to groups that often had a New Left orientation, in the 1990s conservative populists raised additional questions about the failures of representative democracy, albeit from a contrasting perspective. We exclude Spain, Portugal, and Greece from our analyses because of their relatively recent transitions to democracy and their lower level of economic development than that of most other OECD nations. We exclude Iceland and Luxembourg because of their small size, although available data for these nations are reported in some chapters. Our thanks to Martin Wattenberg for his contributions in developing this framework and for his early collaboration in this project.
I Electoral Change
This page intentionally left blank
2 Expanding the Electoral Marketplace Russell J. Dalton and Mark Gray
Democracy exists only where … the people determine directly their own destiny, or where through direct primaries, the referendum, the initiative and the recall they are able to enforce their will upon their representatives. It follows that parliamentary government is that form of the state in which party exerts its will through representatives of the people; democracy is that form in which this is done through the people themselves. (Charles Seymour and Donald Page Frary, How the World Votes) The critical difference between authoritarian states and democratic regimes is that in the latter residents1 can freely and fairly select representatives, and in some cases directly decide public policies through the referendum and initiative. Modern representative democracies are thus built upon the electoral process. The voters’ choice between competing parties or candidates is still the primary basis of public access and influence in contemporary democracies. Moreover, participation in elections is typically the most common form of citizen political engagement. This book examines multiple means of citizen access to politics; but, however important other forms of access and participation are, they lack the quality of collective representation and democratic legitimacy that elections possess. To paraphrase Schattschneider, ‘democracy without elections is unthinkable’ in the modern nation-state. Elections are also important for what they decide. Electoral outcomes determine who manages the affairs of government and who makes public policy. The selection of leaders and the ability to ‘throw the rascals out’ at the next election are the public's ultimate instruments of non-violent power. Political elites may not always act as they promise, but the selection of a government provides some popular control over them. Moreover, campaign debates and electoral discourse are among the prime mechanisms for ensuring openness and accountability in the political process. Indeed, the competitive nature of electoral politics encourages elites and citizens to discuss the policies of the government and the behaviour of potential policy-makers. If we are concerned with citizen access and influence within a democratic process, then we must begin with a study of elections.
24
RUSSELL J. DALTON AND MARK GRAY
The development of elections in democratizing states has been well studied (LaPalombara and Weiner 1966; Diamond, Linz, and Lipset 1988), but few have examined how long-term advanced industrial democracies have continued to modify their election procedures to allow for expanded citizen access and influence. Although the importance of elections to the democratic process is widely acknowledged, most Western democracies have historically resisted the expansion of the public's voting choices beyond a few key offices. Even after the franchise became virtually universal in European democracies, national elections typically remained limited to a single choice for the national legislature. This pattern was repeated at regional or local elections, giving democratic citizens a mere handful of voting choices during a full electoral cycle. The United States is an extreme exception to this pattern, as we discuss below, but even in the United States the democratic system began with only a single direct vote for one's representative at the national level (senators and the president were elected indirectly). In short, if democracy is measured by the amount of electing that gives citizens access and influence over political elites, these quantities historically have been quite limited in Western democracies. There is evidence, however, that the new populist pressures described in Chapter 1 are expanding the electoral marketplace (Scarrow 2001). There is growing evidence that the number of elected offices is increasing. The decentralization of governments and attempts to create new governing structures—from local neighbourhood associations to regional governments—creates new arenas for elections. A recent Assembly of European Regions (1998: 3) report notes: ‘During the last few years, Europe has witnessed an important phenomenon: the development of regionalization and regionalism as the driving force of strengthening democracy in Europe.’ There are also frequent claims that new forms of direct democracy are expanding the marketplace, especially at the sub-national level. The creation of supranational governance in Europe also eventually led to new international elections (a quarter-century after the European Community was created without any direct citizen input). This chapter examines the amount of electing being done in advanced industrial democracies and assesses whether the total amount is changing over time. If elections are the most common means of citizen access, have more opportunities and choices emerged in recent years? We have collected a unique new database to examine the different ways that the electoral marketplace may expand—from local contests to supranational elections—within the set of advanced industrial democracies from 1960 to 2000. These data enable us to track how the amount of electing is changing and to identify the sources of any changes. More important, we examine what conditions seem most conducive to the expansion of electoral choices and discuss the implications of our findings.
EXPANDING THE ELECTORAL MARKETPLACE
25
THE AMOUNT OF ELECTING The debate about the nature of citizen access and influence through electoral politics is an ongoing theme in democratic theory. Many early political theorists believed that an exclusive or excessive reliance on elections and processes of representation undermined the very tenets of democracy. Voters had political power only on the day their ballots were cast, and then waited in political servitude until the next election: typically a period of four or five years. Thus Jean Jacques Rousseau warned that the instant a people allows itself to be represented it loses its freedom. More recently, contemporary theorists have continued to criticize the limits of traditional representative democracy (Barber 1984; Bobbio 1987; Dahl 1989; Warren 2001). The citizens’ control of political elites was further limited by the small number of opportunities they had to make electoral choices. One of the first to highlight this point was Ivor Crewe, who contrasted the amount of electing in the United States to the limited opportunities in most other nations: No country can approach the United States in the frequency and variety of elections, and thus in the amount of electing. No other country elects its lower house as often as every two years, or its president as frequently as every four years. No other country popularly elects its state governors and town mayors, or has as wide a variety of nonrepresentative offices (judges, sheriffs, attorneys general, city treasurers, and so on) subject to election. Only one other country (Switzerland) can compete in the number and variety of local referendums, and only two (Belgium and Turkey) hold party ‘primaries’ in most parts of the country. Even if differences in turnout rates are taken into account, American citizens do not necessarily vote less often than other nationalities; most probably, they do more voting. (Crewe 1981: 262) Citizen access and influence is severely restricted when voters in most nations are normally allowed to cast only a handful of votes in a four- or five-year electoral cycle. Crewe was writing more than two decades ago, but this general cross-national pattern has continued. Dalton (2002: 38) recently illustrated this gap by comparing the experience of an American and British voter: ‘between 1995 and 2000 a resident of Oxford, England could have voted four times; a resident of Irvine, California, could have cast more than fifty votes in just the single year of 2000.’ In recent decades, institutional reforms have focused on this ‘democracy deficit’ within the electoral arena. For instance, Scarrow (2001) found that two-thirds of the OECD nations had changed their institutional structures to provide citizens with more direct access. Efforts to decentralize government also create new sub-national governing units and more settings for electing elites and making other voting choices (see Chapter 7). The expanding powers of the European Union (EU) have prompted a debate about the democracy deficit within the EU, leading to direct elections of the European Parliament (EP) and continuing calls
26
RUSSELL J. DALTON AND MARK GRAY
for even more citizen access and influence (Reif 1984; Blondel, Sinnott, and Svensson 1998). In addition, demands for more direct citizen influence through referendums, initiatives, and other forms of direct policy mandates are now a rallying point in the demands for greater citizen influence (see Chapter 3). However, the literature's examples of expanding electoral choices sometimes rely on specific examples that may be atypical of overall developments. For instance, the 1994 poll of members of Germany's Social Democratic Party to select the party's candidate for chancellor was a significant innovation, but it was not repeated in the following election. One can note other counter-examples, although these are less prominently featured in the literature: for example, the shift from bicameral to unicameral national legislatures in Sweden and Denmark, or the diminished role of party members in the selection of Dutch parliamentary candidates. In addition, analyses of the expanding institutional opportunities for participation does not mean that these new structures are used,2 and new opportunities may be offset by other changes with obverse effects. The prime goal of this chapter is to provide the evidence of whether the amount of electing is systematically changing in these nations. There are four general ways in which the electoral marketplace may expand (Fig. 2.1). One option is to vote more often on an electoral choice. This involves changes in the electoral cycle to lessen the length of a term, such as the recent reduction of the French president's term from seven years to five, or the introduction of primary or run-off elections. A second way to expand the electoral marketplace is to add more levels of electoral decision-making. The development of the European Union is the most obvious example, where this led to both direct election of European Parliament delegates and to various national referendums on EU issues. The expansion of sub-national governance is another example, such as through the creation of new regional governments in Belgium (1992), France (1986), and Italy (1970). Third, the electoral arena can expand to allow more voters access to the electoral process. This might involve changes in suffrage laws, such as the lowering of the voting age in most OECD nations during the 1970s or the revision of registration or voting procedures to facilitate electoral participation. Finally, these processes can combine to provide more choices to voters at election time. For instance, the introduction of referendums and initiatives gives citizens new matters to decide. The lengthening of the ballot by adding new elected offices is another example, such as the introduction of directly elected mayors or election of other previously appointed offices (Scarrow 2001). In the Irvine, California, example cited above, voters not only make decisions on president, senators, members of Congress, and members of the state legislature, in 2000 they also voted on two judicial offices, four junior college school district members, four city government officials, four water district officials, and more than thirty initiatives and referendums.3 In short, lengthening the ballot may be one of the major means of expanding the electoral marketplace.
EXPANDING THE ELECTORAL MARKETPLACE
27
Fig. 2.1.Expanding electoral markets
We use this framework of the expanding electoral marketplace to structure the empirical analyses of this chapter. There is obviously some overlap between these types of electoral change; but each expands citizen access through electoral channels, albeit with different implications for the democratic process. Shortening the electoral cycle, for instance, decreases the period of political servitude before the public can reassess the course of government. The introduction of more levels of electoral choice gives citizens more direct input on a range of government policy-making and administration. Multiple voting choices at various levels allow citizens access to these separate governing processes, but this also complicates the process of assigning responsibility to government outcomes. This pattern of complex government is now becoming the norm in contemporary democracies. We first map how the electoral marketplace is changing, and then discuss the implications of what we find.
EXPANDING THE ELECTORAL MARKETPLACE Although elections are highly structured and institutionalized activities, counting the total amount of electing in crossnationally comparable form is a challenging task.
28
RUSSELL J. DALTON AND MARK GRAY
Complete records of national legislative and executive elections are readily accessible. It is more complicated to collect regional and local data in comparable terms, and the amount of electing often varies across locales within a nation. An additional complication is the ad hoc election that can take multiple forms, such as referendums, legislative byelections, or other special elections. Since we desired data for eighteen advanced industrial democracies across a fortyyear time span (1960–2000), the magnitude of this data collection task was extensive. Differences in the incidence and complexity of elections across and within nations is largely determined by institutional structures. The existence of a separately elected executive, bicameralism, federalism, European Union membership, and constitutional provisions for referendums define the potential for electoral choices. As an introduction to these institutional arrangements, Table 2.1 presents some of the most prominent features of the electoral structures in our set of nations and the significant changes that have occurred over time. Where the voters are making electoral choices, this is denoted by a black box; if this option was instituted during the time frame of our research, the date of implementation is noted in the table. (We do not list the lower house of the national parliament, because this is
29
EXPANDING THE ELECTORAL MARKETPLACE
Table 2.1.The institutional structure of nationwide electoral options Nation Australia Austria Belgium Britain Canada Denmark Finland France Germany Ireland Italy Japan Netherlands New Zealand Norway Sweden Switzerland United States
Federal System ■ ■ ■ (1992)
Elected president ■
Elected upper house ■ ■ (partial)
■
■
■ ■
■ ■ (1965) ■
■
■ ■
⊏ (1971) ■ ■
National referendum ■ ■ ■ (1950) ■ (1975) ■ (1972) ■ ■ (1994) ■ (sub-national) ■ ■ ■ (2001) ■ ■ ■ ■ (sub-national)
EU election ■ (1995) ■ (1979) ■ (1979) ■ ■ ■ ■ ■ ■
(1979) (1995) (1979) (1979) (1979) (1979)
■ (1979) ■ (1995)
■ Indicates that a structure exists, an accompanying date indicates when this was first introduced if during the 1960–2000 time frame. ⊏ Represents an electoral opportunity that ended during the time frame, followed by the date it changed.
directly elected in all eighteen nations.) The table first documents the diversity of institutional arrangements that exist across these nations. Half have elected regional governments within a federal system, but only four have elected presidents or an elected upper house. National referendums exist in most nations, but are significantly lacking in the United States that otherwise offers its voters a virtually overwhelming number of electoral choices. It is also clear that institutional reforms are systematically expanding the electoral marketplace across these various arenas (also see Scarrow 2001). For example, seven nations have a federal system and elections for regional governments; one of these elections began during the time frame of our study.4 One nation instituted direct election of the president (and Finland moved from an electoral college to direct election), and three introduced national referendums for the first time. Susan Scarrow (Chapter 3) documents even wider institutional reforms to allow referendums at the local level in this set of nations. The European Union also has had a broad cross-national impact: eleven nations have experienced their first EU election (either for the European parliament or for a referendum on EU entry) since 1979. In fact, the only instance of reform in the opposite direction is Sweden, which in 1971 abolished its indirectly elected upper house in favour of a unicameral Riksdag. Even this case should be seen as strengthening the electoral process, since it created a directly elected parliament without the possibility that an indirectly elected chamber could influence legislation. Thus, the answer to one of our questions is already apparent. In overall terms, contemporary publics are making more choices at more levels. The expansion of regional governments and especially the European Union are calling voters to the polls to make new electoral choices. These institutional changes have the potential to increase significantly the actual amount of electing, especially in European nations that originally had limited electoral opportunities for the voters. When we combine the multiple electoral alternatives represented in Table 2.1, we can determine whether citizens are going more often to the polls for nationwide elections. Table 2.2 presents the number of nationwide ‘mass elections’ per five-year period for our set of nations.5 By the term ‘mass election’ we mean any nationwide election in which the general public can participate; this excludes internal party primaries or candidate selection processes or sub-national elections that are held at different times in different areas.6 Five years is an arbitrary span of an electoral cycle that does not exactly match all nations, but when cumulated across nations this comparison should describe the overall trend. We count six general types of elections: national legislative elections, national executive elections where they apply, coordinated nationwide regional/provincial elections, nationwide municipal elections, referendums, and elections to the European Parliament.7 The frequency of elections is affected by many factors. For instance, the instability of governments can lead to early elections and thus shorten the electoral cycle. Changes in the length of electoral cycles can also increase the number of
30
RUSSELL J. DALTON AND MARK GRAY
trips to the polls voters make. In addition, changes in the nature of the electoral system may affect the frequency of election, such as the multiple shifts between proportional representation (PR) and plurality electoral systems (some with two tours) in France or the introduction of the multiple-member proportional (MMP) system in New Zealand and Japan. In a few instances, governments have attempted to decrease the number of trips to the polls by consolidating different elections on a single day, such as the Swedish consolidation of local and national elections. When two or more of these elections occur on the same date (concurrent elections of offices or referendum votes), we count this as a single trip to the polls.8 In other words, at this point we are simply counting trips to the polls and not the number of choices being made in the voting booth. Thus, the American long ballot election is treated as one event, just as the single vote for a Member of Parliament in British national elections. Or referendum elections with multiple measures on the ballot, such as the Italian election of June 1995 or June 1997, are counted as a one election at this point in our analysis. Table 2.2 illustrates the pattern over time for our set of nations. The table tracks an increase in the number of nationwide elections; there were sixty-two elections for this set of nations in 1960–4, and this increases to eighty-one in 1995–9. Table 2.2.Elections per five-year period, 18 OECD Nations, 1960–99
Australia Austria Belgium Britain Canada Denmark Finland France Germany Ireland Italy Japan Netherlands New Zealand Norway Sweden Switzerland United States TOTAL
*
1960–4
1965–9
1970–4
1975–9
1980–4
1985–9
1990–4
1995–9
Total
3 3 1 2 3 5
4 3 2 2 3 5
4 5 2 5 3 5
3 4 3 4 3 6
3 3 2 3 3 4
3 4 4 2 3 6
3 6 3 2 3 6
5 6 4 2 3 4
28 34 21 22 24 41
Time trend 0.071 0.381* 0.369* −0.119 0.000 −0.012
4 5 4
3 9 4
3 5 2
4 3 4
4 5 5
4 7 4
5 4 4
5 5 4
32 43 31
0.238* −0.202 0.083
1 1 4 1
4 1 5 1
4 2 4 4
3 4 4 3
6 3 5 4
5 5 4 5
4 6 4 4
3 5 4 2
30 27 34 24
0.238 0.726* −0.047 0.310
3
5
3
3
4
3
4
6
31
0.202
2 3 14
3 2 9
3 3 16
3 2 19
2 2 14
3 2 14
3 3 17
3 3 15
22 20 118
0.071 0.023 0.429
3
2
3
2
3
2
3
2
20
−0.047
62
67
76
77
75
80
84
81
602
Note: The time trend is calculated by regressing the five-year totals for elections by the time variables (coded 1 = 1960–4 and 8 = 1995–9); this is an unstandardized regression coefficient. Coefficients significant at the 0.05 level.
EXPANDING THE ELECTORAL MARKETPLACE
31
At the same time, we find significant variation between nations in the time trend. The number of nationwide trips to the polls has grown in several nations, such as Austria, Belgium, Finland, and Italy. The expanded use of referendums is the major source of this trend in Italy; in Belgium it is the creation of a federal system; in Austria and Finland the introduction of EU elections on separate dates has increased the number of elections. A few nations have consolidated their elections to decrease the voters’ trips to the polls, such as Sweden's introduction of concurrent national and regional elections in 1971, the increasing tendency to hold concurrent upper and lower house elections in Australia, or the use of concurrent election dates as Belgium regionalized—although these changes in legislative elections can be counterbalanced by more referendum elections or new EU elections. In a several cases there are no institutional changes that affect the frequency of nationwide elections over this time span; the most prominent examples are Canada, Japan, Norway, and the United States. Table 2.2 also describes considerable cross-national variation in the frequency with which voters make a trip to the polls. Residents of Norway, for instance, normally make two trips to the polls during an electoral cycle (national and municipal elections). Swiss voters, in contrast, are typically asked to make more than a dozen trips to cast nationwide ballots during a full electoral cycle. To better illustrate the overall trends, Fig. 2.2 presents the average number of elections per year for seventeen nations (excluding Switzerland because of its exceptional record).9 The number of trips that voters make to the polls for national elections has steadily increased across the set of advanced industrial democracies. In the first period (1960–4), the average voter made just under three trips (2.8 nationwide elections) to the polls across these five years; but in 1995–9 this Fig. 2.2.Total number of elections by year, 1960–2001
Note: Figure entries are the total number of elections in Table 2.2 occurring in each year for 17 OECD nations (excluding Switzerland). Source: Compiled by the authors.
32
RUSSELL J. DALTON AND MARK GRAY
had increased to nearly four separate elections (3.9). A single additional trip to the polls at the national level does not appear to be a dramatic increase in the amount of electing, especially given the expanded institutional options described in Table 2.1. But at this point we are counting only trips to the polls for nationwide elections, not subnational elections or the number of voting choices that citizens make in the ballot booth. A third way to expand the electoral marketplace is to increase the number of voters eligible to participate. Perhaps the most systematic evidence of change comes from a lowering of the voting age requirements for national elections. All but one of these eighteen nations (Japan) has taken steps to introduce more ‘choosers’ into the electoral process. In 1960, almost every nation defined the eligible voting age as 21; by 1999 virtually every nation had lowered the legal threshold to age 18.10 Of course, there were other, more dramatic, and politically significant expansions of the franchise over this period, such as the enfranchisement of African-Americans and Australian aborigines in the 1960s and suffrage for Swiss women in 1971. The public's increasing access to elections can be summarized by tracking the percentage of the voting-age population (age 18 and over) that is registered to vote (in the United States and other open-registration systems this is the voting age population, VAP). Even if we exclude Switzerland because it initially did not grant women's suffrage, the crossnational average of the eligible voting-age population was around 60 per cent in the early 1960s and close to 80 per cent by the end of the 1990s (Fig. 2.3). This increase occurs because of the lowering of the
EXPANDING THE ELECTORAL MARKETPLACE
33
Fig. 2.3.Growing share of the population eligible to vote in national elections, 1960–2001
Note: Table entries are the percentage of the adult population (age 18 and over) who are registered to vote in 17 OECD nations (excluding Switzerland); each point represents the percentage eligible among those nations having an election in a specific year. Source: Compiled by the authors from IDEA statistics (IDEA 2002). voting age and because of the expansion of registration access and efficiency across contemporary democracies. Finally, we expect the largest expansion of the electoral marketplace has been the culmination of these separate processes, producing more choices for citizens to make about democratic governance. This comes from increasing both the levels at which choices are made and the number of items on the ballot, either by electing formally appointed offices or adding referendums and other policy choices. The introduction of a directly elected mayor of London can expand citizen just as a referendum on EU membership does. And concurrent elections should be counted separately to accurately measure the total amount of electing. Counting all electoral choices across time and across nations is a daunting task, and creates multiple measurement problems. To estimate the broad trends across time, we selected a specific locale in each nation and asked how many choices a citizen could make during a ‘normal’ electoral cycle in the early 1960s as compared with an electoral cycle at the end of the 1990s.11 Our analyses are based on mass elections for residents living in the capital city or the largest population centre; the choice depended on whether the capital had some extra-legal status that necessarily made it atypical of the options available in other nations (such as the District of Columbia in the United States). We realize that any single city in a nation has its unique features and any single electoral cycle can have a unique mix of elections, but this decision mechanism allows us to collect data systematically across time for a roughly comparable set of locales. We welcome future efforts to expand this measurement of the electoral marketplace to other cities and cross-national comparisons to evaluate the trends presented here. Table 2.3 provides examples of the full expansion of the electoral marketplace across time. The table includes all local, regional, national, and supranational voting opportunities for citizens of the chosen city in each nation; and multiple items on the same ballot are counted separately. For instance, the June 1997 referendum election in Italy included nine different propositions; in Table 2.2 this is counted as a single election, here it is counted as nine choices. Even more dramatic is the long ballot that Americans face. The expansion of sub-national governments and the European Union elections also increases to the total amount of electing in Table 2.3, even if these elections occur concurrently with other voting choices. The only substantial decrease in the number of vote choices is for New York City, although this pattern reflects the unique circumstances of the way state constitutional changes were presented in a set of referendums.12 The totals at the bottom of the table illustrate how access and influence through the electoral process expanded during the latter part of the twentieth century. There is considerable variation across nations, but the overall pattern is clear. In the early 1960s the median number of choices was 4.0 during a normal electoral cycle.13 By the end of the 1990s this had expanded to a median of 6.5 choices per cycle. A small portion of the overall increase in the amount of electing comes from expansion of elected offices or new levels of government. The total number of
34
RUSSELL J. DALTON AND MARK GRAY
Table 2.3.Increase in the number of electoral decisions over time Nation Australia
Early 1960s 6
Late 1990s 8
Austria Belgium Britain
6 4 3
12 6 6
Canada
5
5
Denmark
8
7
Finland France Germany Ireland Italy Japan Netherlands
4 5 4 2 3 6 3
5 7 6 9 23 6 9
New Zealand
4
6
Norway Sweden
2 3
2 5
Switzerland United States
15 53
38 30
TOTAL MEDIAN
136 4.0
190 6.5
Election cycles Sydney: 1961–3, 1998–2001 Vienna: 1962–6, 1995–9 Brussels:1961–5, 1995–9 London: 1959–64, 1997–2001 Toronto: 1958–62, 1997–2000 Copenhagen: 1960–4, 1998–2001 Helsinki: 1962–6, 1995–9 Paris: 1962–7, 1993–7 Bonn: 1961–5, 1998–2002 Dublin: 1961–5, 1992–7 Rome: 1960–3, 1996–2000 Tokyo: 1960–3, 1993–6 Amsterdam: 1963–7,1994–8 Wellington: 1960–3, 1996–9 Oslo:1961–5, 1997–2001 Stockholm: 1960–4, 1994–8 Bern: 1963–7, 1995–9 New York City:1960–3, 1997–2000
Source: Compiled by the authors. See chapter Appendix for additional detail.
within-nation elected offices stayed virtually constant across these eighteen nations, and the European parliament elections account for about a quarter of the increase. The greatest expansion comes from the growing use of referendums, which accounts for about three-quarters of the overall increase. Referendum usage clearly grew in Ireland and Italy, which went from no referendums in the first period to six and eighteen respectively in the second period, but usage also grew in five other nations (Australia, Austria, the Netherlands, Norway, and Sweden). In summary, if elections are the measure of democracy, then our analyses indicate the pitifully poor record of advanced industrial democracies by the mid-twentieth century. The history of democratic expansion up until 1960 still left the typical citizen with very limited access to the electoral marketplace—casting barely a single vote a year for all levels of democratic government. One might claim that the historical criticisms of representative democracy were accurate descriptions of how elections were used to limit citizen access. This criticism was even more powerful since other forms of direct citizen action and involvement were also relatively underdeveloped at this time (Barnes, Kaase, et al. 1979). Since 1960 there has been a significant expansion of the electoral marketplace on almost every dimension. More people have access to the polls, vote more often, at more levels of government. If the electoral marketplace is an appropriate
EXPANDING THE ELECTORAL MARKETPLACE
35
measure of democracy, there has been substantial progress over this period even though the total amount of electing in most nations remains quite limited.
THE ELECTORAL MARKETPLACE AND DEMOCRACY Elections are where most citizens participate in the democratic process and attempt to influence politics, and our findings demonstrate an expansion of the electoral marketplace over the latter part of the twentieth century. Even by 1960, advanced industrial democracies allowed their citizens to make only a modest number of voting choices; the electoral marketplace has grown considerably since then. Although the number of trips to the polls has grown slowly and not uniformly, the different voting choices that citizens make—the amount of electing—has increased substantially over this period. Moreover, the barriers to participation in elections have steadily decreased, such as through the lowering of the voting age and provisions to each registration in formerly restrictive systems such as the United States. On average, contemporary publics now have more opportunities to make electoral input on more offices at more levels of government. This enlargement of the electoral marketplace should mean increased access and influence on politics for the public, and thus represents an expansion of the democratic process. The creation of new venues for citizen input, such as through the creation of local governments or a new supranational European Union, provide new avenues for interests that might not be well represented in other electoral forums. These multiple levels of governance give the public more access to policy-making and administration than might have occurred without direct citizen involvement. The growing use of policy referendums marks a further extension of the public's direct access and influence within the policy process. In our group of nations, the number of national referendums nearly doubled between the 1960s and the 1990s, and indirect evidence points to a parallel increase in sub-national referendums (see Chapter 3). This expansion of the electoral marketplace should also increase citizen access to information about government and increase the transparency of government policy-making. Samuel Popkin (1991) described elections as civics lessons for the public as they learn and debate the actions of government: more elections increases this exposure to politics. By discussing more issues and more levels of government, the civics lessons have expanded. This is the appropriate point at which to note that the pattern of electoral decentralization and the growth in the amount of electing has been uneven across the nations we surveyed. This is partially due to the individual institutional history or constitutional rules of each nation, as well as cultural traditions. As J. A. Chandler (1993: 188) noted, ‘there are far more differences than similarities in these systems, and … each one is significantly and demonstrably unique’. Thus the amount of electing during a recent electoral cycle ran from only two votes in Oslo,
36
RUSSELL J. DALTON AND MARK GRAY
Norway, for instance, to several dozen in New York City and Bern, Switzerland. Differences in electoral access across the advanced industrial democracies is extremely large. One might fairly ask whether these new forms of electoral access and influence are equal in weight to the key votes on national legislatures and executives that are the basis of representative democracy. The vote for the US president, for instance, carries much more political weight than a new voting choice for an elected school district member. Karlheinz Reif (1984) distinguished between first-order national elections, such as presidential or national legislative elections, and second- or third-order elections, such as regional or EU elections. Most of these new votes are for second-order or third-order elections. Nevertheless, the European Union elections illustrate how the expansion of the electoral marketplace can have effects that make even second- and third-order elections significant expansions of the democratic process. It is often claimed that European Parliament elections are relatively hollow electoral exercises because the Parliament wields little influence in EU politics. But even in this instance the expansion of the electoral marketplace can have benefits for democracy in addition to the public's direct representation in EU institutions. Since European Parliament elections often are treated by politicians and elites as a referendum on the national government, they provide another means for influencing national policy-making during the multi-year time span between national elections. European Parliament elections also provide another setting to inform and mobilize citizens about political matters, both about specific issues facing Europe and issues in their own nation. The European Parliament is important as a recruitment and training venue for political elites, especially as the EP has become a path to membership in national parliaments and other party offices. The European Parliament also offers alternative representation and access routes for political groups that might not gain national visibility, and provides a forum where transnational networks might form. For instance, the different PR threshold in EU elections allows parties to be represented which otherwise would have had difficulty winning representation in the national legislature. This applied both to the early emergence of Green parties in the European Parliament and to the ability of the French National Front to win representation in Brussels while fairing poorly in the two-tier system of National Assembly elections. The creation of the European Parliament and elected Members of the European Parliament also affects national political parties, which develop new policies on EU issues and have new internal representatives of EU interests. Thus, to the extent that democracy is measured by the amount of citizen access and influence through the electoral process, it seems clear that there has been a major and politically significant expansion of the electoral marketplace over the latter part of the twentieth century. Yet it is not clear that these institutional changes have had uniformly positive consequences for the functioning of the democratic process, and we conclude by raising some questions about them.
EXPANDING THE ELECTORAL MARKETPLACE
37
The expansion of the amount of electing places new demands on contemporary publics. Not only might they be called on to elect the chief executive or a representative to the national legislature, but now decisions must also be made on a plethora of other offices and referendum issues. Recent Italian referendums illustrate the new questions that voters are asked to decide. In the June 1997 referendum the ballot included measures dealing with the special powers of the Ministry of the Treasury over private organizations, conscientious objectors to military service, rules for hunters, two issues of judicial reform, the competencies of the agricultural ministry, and two referendums in Rome dealing with the privatization of local firms. Just two years earlier Italians had been asked to decide on television ownership rules, television broadcasting policy, the hours that stores may remain open, the commercial activities that municipalities may pursue, several issues of labour union reform, the regulations for administrative elections, and residency rules for Mafia members. Even before the expansion of the electoral marketplace, voting researchers debated whether citizens were knowledgeable enough to make reasonable decisions on the ballot choices they faced (for instance, Converse 1964; Delli Carpini and Keeter 1996; cf. Dalton 2002; Sniderman, Brody, and Tetlock 1991). One might ask whether an Italian college student majoring in political science, or even Italian political science professors, could make reasonable choices confronted by the flurry of referendums and elections that occurred during the late 1990s. The same question applies to the American, Swiss, and other electorates who face an increasing number and diversity of electoral choices. It seems plausible that the expansion of education and the increasing availability of political information through the media makes it easier for voters to make decisions on first-order elections, but does this carry over to the election of low-visibility local elections, complex referendums, and supranational institutions? More democratic choices may not be the same as better democratic choices. Furthermore, voters confront an increasingly complex framework of governance, which has potentially diverse effects. On the one hand, citizens have more opportunities to vote on more levels of government and more issues. On the other hand, this pattern of complex governance may obfuscate political responsibility. Different levels of government and different branches of government frequently point the finger at each other when times are bad. The desire to ‘vote the rascals out’ becomes more difficult to fulfil when different branches of government blame one another for poor policy performance. This pattern is well-known in the current US system of divided government and a federal structure, but it can also be seen in other advanced industrial democracies. Divided government between the Bundestag and Bundesrat has become a more common feature of German politics, reinforced by its federal system. The increase in split-ticket voting and more electoral arenas suggest that this is a generally increasing pattern in these nations (see, for example, Dalton and Wattenberg 2000). The debates about the democracy deficit in the European Union raise similar issues. As the EU assumes an
38
RUSSELL J. DALTON AND MARK GRAY
increasing policy role within Europe, this raises questions of the relative issue competency of EU and national governments. European politicians have quickly learned to blame Brussels when policies can be linked to EU actions. Taken together, these patterns of increasingly complex governance raise new challenges for voters. As Powell (2000) has recently argued, if elections are to serve as instruments of democracy, then there must be clear means by which voters can hold governments accountable. Powell was concerned with the difficulties in ensuring electoral accountability in national governments under different electoral systems, but this problem is greatly magnified within a system of increasingly complex governance. Complex governance and divided government may complicate the electorate's ability to use the vote to direct the course of government. Perhaps the greatest irony is that the expansion of the electoral marketplace has apparently been greeted by fewer voters willing to turn out at each election (Gray and Caul 2000). Turnout is down in virtually all advanced industrial democracies over the time period of our study. Although there are multiple causes for this decline, our findings suggest that the increasing number of elections may be one factor contributing to the trend. Expansion of the electoral marketplace creates more opportunities for access, but it also diminishes the weight of each single election as the number of elections multiplies. With many electoral opportunities, voters may feel less duty to vote in every contest. In addition, the increasing number of elections multiplies the demands being placed on voters to become informed on the issues of each election. In short, the motivation to participate in any single election may decrease as the number of electoral opportunities increases (Wattenberg 2002; Jackman and Miller 1995: 482). One graphic illustration of this syndrome comes from the 1999 elections to the Scottish Parliament in which turnout was considered low (58 per cent). A study by the Scottish Executive concluded that voter fatigue was one of the major reasons for the low turnout. It cited the views of one middle-class Scottish voter to illustrate this feeling: We just seem to have had so many elections recently where we have been expected to go along and vote. You get fed up after a while and even if you would normally vote, you start to wonder whether you can be bothered to do it again.14 This was a historic election, the first time in over 300 years that Scots had been able to vote for a their own legislature. In the previous five years there had been fewer than one election a year. Yet some Scots apparently felt voter fatigue. One wonders what happens when voters are asked to make dozens of choices in a single year or several visits to the polls each year, as frequently happens in the United States. In separate analyses, Mark Gray (2003) has systematically examined this fatigue hypothesis. He began with the pooled time series model of turnout he had previously developed with Miki Caul (Gray and Caul 2000). To this model he added a measure of the number of trips to the polls that voters make. Gray finds
EXPANDING THE ELECTORAL MARKETPLACE
39
that the number of national elections held in the previous five years has a strong negative impact on voting turnout even while controlling for the other standard predictors of voting. His model predicts that five additional trips to the polls over a five-year span reduces turnout by 4 per cent. Given the expansion of the amount of electing we have demonstrated in this chapter, the negative impact on turnout becomes clear. We do not mean that institutional changes that increase opportunities for electoral participation actually decrease total involvement in the democratic process. Many more ballots are cast today during a full electoral cycle—the total amount of electing has increased substantially—even if the greater number of elections might decrease involvement in any single election.15 Moreover, the politicization produced by more frequent elections may encourage a general increase in political interest in advanced industrial democracies (Dalton and Wattenberg 2000: ch. 3) and the expansion of political participation in other forms, such as citizen groups, contacting politicians, and direct forms of political action. In most advanced industrial democracies, contemporary publics are more involved in politics than they were a generation ago. Many of the subsequent chapters in this volume demonstrate how institutional reforms create additional channels for access and influence.
Appendix: APPENDIX: ELECTIONS INCLUDED IN TABLE 2.3 Nation Australia
Belgium
1960s National House, Senate, state legislative (2), mayor, local council Nationalrat, President (2), Landtag/municipal, Bezirksvertretungen, 1 referendum Chamber of Representatives, Senat, Province, commune
Britain
House of Commons, London Council, Borough
Canada Denmark
House of Commons, Ontario legislature, mayor, municipal council, school board trustee Folketinget, Amtsrad, communal, parish council, 4 referendums
Finland France
Eduskunta, President (2), Kommuner National Assembly (2), President (2), Commune
Germany Ireland
Bundestag (2), Landtag, communal Dail, local (presidential election was 1966)
Italy
House of Deputies, provincial, municipal
Japan
House of Representatives, House of Councillors, Governor, Prefect Assembly, Mayor, Municipal Assembly Tweede Kamer, province, municipal council
Austria
Netherlands New Zealand Norway Sweden Switzerland United States
House of Representatives, local, 2 referendums Stortinget, local Riksdag, county, municipal National Council, Council of States, canton, local, 11 referendums President, House of Representatives (2), US Senate, GovernorLt. Governor, State Senate (2), State Assembly (2), Mayor, City Council (2), Council President, Borough President, Comptroller (2), Attorney General, District Attorney, Supreme Court (3), Civil Court (2), Municipal Court (2), Court of Appeals (2), state and city referendums (26)
1990s National House, Senate, state legislative (2), mayor, local council, 2 referendums. Nationalrat, President, Landtag/municipal, Bezirksvertretungen, Euro-parliament, 7 referendums Chamber of Representatives, Senat, French/Flemish Council, Brussels Region Council, commune (in 2000), Euro-parliament House of Commons, Euro-parliament, London mayor, London Assembly (2), Borough Council House of Commons, Ontario legislature, mayor, municipal council, school board trustee Folketinget, Amtsrad, communal, parish council, 2 referendums, Euro-parliament Eduskunta, President (2), Kommuner, Euro-parliament National Assembly (2), President (2), Commune, Paris Mayor, Euro-parliament Bundestag (2), Landtag, communal, mayor, Euro-parliament Dail, Local, Euro-parliament, 6 referendums (presidential election was 1997) House of Deputies, provincial, municipal, regional, Europarliament, and 16 national referendums, 2 local referendums House of Representatives, House of Councillors, Governor, Prefect Assembly, Mayor, Municipal Assembly Tweede Kamer, province, municipal council, neighbourhood council, 4 local referendums, Euro-parliament House of Representatives (2), local, 3 referendums Stortinget, local Riksdag, county, municipal, 1 referendum, Euro-parliament National Council, Council of States, canton, local, 34 referendums President, House of Representatives (2), US Senate (2), Governor-Lt. Governor, State Senate (2), State Assembly (2), Mayor, City Council (2), Borough President, Comptroller (2), Attorney General, Supreme Court, Civil Court (2), Public Advocate, Community School Board, state and city referendums (8)
Note: The count does not include by-elections, optional candidate preference votes in PR electoral systems, or primaries run internally by political parties. We include multiple votes, such as in the German 2-vote ballot, as separate choices. In Vienna, Paris, and Oslo, the regional government level is identical to the city government; but in other municipalities in Austria, France, and Norway there would be an additional election. Canada has a three-year cycle for municipal elections and two of these elections fell within each of the national legislative cycles we selected; we therefore counted only one of these municipal elections in each cycle.
40
RUSSELL J. DALTON AND MARK GRAY
NOTES 1
‘Residents’ is used here rather than ‘citizens’ because historically and currently non-citizens can legally cast ballots in some elections once they have established residency. Even the United States has some history of non-citizen voting at the State and territorial levels in the nineteenth century (Keyssar 2000). Furthermore, the current United States voter registration system is largely based on the honour system of the person filling out
EXPANDING THE ELECTORAL MARKETPLACE
2 3
4 5
6
7
41
the form in claiming citizenship. In theory and practice, any resident can vote without showing proof of citizenship or even age. Since 1994, European Union member states have allowed non-citizens to run as candidates and vote in municipal and EU elections. For instance, Scarrow (2001) notes that reforms have allowed for the direct election of mayors in Britain, but very few municipalities have pursued this option, and often the initial discussions have led to its rejection. London is the most noticeable example of the introduction of a directly elected mayor, and is an atypical case. In 1997 there were over 87,000 governmental units in America, including the federal government, 50 states, 3,043 counties, 19,372 municipalities, 16,629 townships, 13,726 local school districts, and 34,683 special districts—each of which had some constitutional or statutory power to make policies and most had elections (US Census 2001). The same data source indicates that there were more than 500,000 elected officials in the United States in 1997. By comparison, we estimate that there are fewer than 25,000 elected officials in Britain. In addition, Italy created new elected regional governments in 1970, France in 1986, and Britain in 1999. Because these are not formally federal systems, they are not included in Table 2.1. We also collected election data for each country on the basis of a sample capital city (or the largest city in those cases where the administration of the capital is exceptionally different from other cities). See Table 2.3. We counted trips to the polls for every election, from local councils to parliament or presidential contests, with the exception of American primary elections or in any country where the contest is a special by-election to fill an elected position open under extraordinary circumstances (for example, the death of a representative). Under this expanded measure (see Gray 2003), eleven of the eighteen nations made more trips to the polls in the 1990s than in the 1960s. On average, the number of trips to the polls increased by four over this period. Three nations (Australia, Canada, and France) had fewer trips to the polls in the 1990s than in the 1960s, by an average of three elections. Four nations—Britain, Denmark, Germany, and Japan—had no change in the number of elections. We do not include these results in the analysis because the addition of the local contests within the cities sampled is not always representative of other localities within the same nation. In these analyses we do not count separate sub-national elections that are held on different days in different areas, such as local and state elections or sub-national referendums. This decision was guided by the difficulty in assembling a reliable and cross-nationally consistent database over time. Adding these local and regional elections would increase the number of total votes and probably strengthen the upward trend. Adding local and regional elections would also change the relative ranking of nations in their total trips to the polls. Table 2.3 attempts to count all elections for a single locale in each nation to provide such evidence. We do not count primaries because they are ‘selections’ and not ‘elections’, that is, the outcome selects a candidate rather than ensuring a government position. In practical terms, only the Unites States uses institutionalized government-run primaries that might be included under the election heading on other definitions. As we will show below, an increase in the number of levels of government will also increase the frequency of elections. Martin Wattenberg (2002: ch. 8) illustrates the hyperactivity of American democracy by citing media reports that, when Vice-President Cheney lived in Dallas between 1996 and the summer of 2000, he was called to the polls for sixteen separate elections (most of which he missed). This is a common pattern in
42
8
9
10 11
12
13 14
RUSSELL J. DALTON AND MARK GRAY
the United States, leading to calls to consolidate federal, state, and local elections on a fewer number of dates; similar efforts to develop concurrent elections are evident in other nations. The procedures for concurrent elections vary across nations. In some nations the multiple offices or electoral choices appear on the same ballot, as in the United States. In other nations separate ballots are used and sometimes even different voting booths, even if the polling station is at the same location and the elections are concurrent. We include all of these concurrent elections as a single election in Table 2.2 and Fig. 2.2. We exclude Switzerland because its large number of referendums distorts the image of the overall number of elections in most other nations. With Switzerland included, however, the trend is roughly comparable but with a higher intercept. In addition, we extended the analyses back to 1945 to track the immediate post-war elections. The number of elections is slightly higher in the 1945–9 and 1950–4 periods, as new party systems in several nations became established and then declind in the 1960–5 period. To be consistent with other analyses in this project, we present the data beginning in 1960. The additional time points are available from the authors. The exceptions are Austria at age 19 in 2000 and Japan, which kept the limit at age 20 since the 1949 election. We selected the capital city as an initial locale in each nation. In several cases the capital city has a special political status that makes it exceptional. For instance, the District of Columbia and the Australian Capital Territory both have a special status that limits the role of their local governments. In such cases we selected the largest other city in the nation. The list of cities and the time frames for the two electoral cycles are presented in Table 2.3. We exclude party primaries or candidate selection processes or elections where participation is limited by party membership. We also exclude by-elections, preference votes (such is in single transferable vote, STV, system), or other special elections that represent a second vote for a single office. In most cases, our definition of a normal electoral cycle is the length of the standard national legislative term. We are attempting to estimate how many voting choices a voter has, for example, in the four years between parliamentary elections in Germany or the three-year cycle in Australia. In the United States the cycle was determined by the presidential elections. Another complication is early elections, and we selected cycles to represent a ‘normal’ full period. Furthermore, in some cases other elected offices have different cycles, such as the longer term of the French president or local elections in Ireland. Thus, the results in Table 2.3 should be seen as illustrative rather than as a definitive measure of the amount of electing. New York is one of the states (and cities) that make constitutional and charter changes by referendum. In some cases a list of proposed changes is offered to voters and they have been able to select one-by-one those they support or oppose. In the most recent time period, however, these choices have often been bundled into one choice, which substantially reduced the number of referendums listed on the ballot. We use the median instead of the mean in order to lessen the influence of an outlier on the results, such as the Swiss and US statistics. This is from a report by the Central Research Unit of the Scottish Executive, ‘Assessment of the voter education campaign for the Scottish Parliaments elections’,http://www.scotland.gov.uk/cru/kd01/assess01.htm
EXPANDING THE ELECTORAL MARKETPLACE
43
15 The evidence is less clear on the implications of this pattern for the absolute number of people who vote in any election during a full cycle. It may be that fewer people are voting in more elections, or that more people are voting but in different elections (some may be attracted to national politics, and somewhat different voters to local elections, and so forth). We suspect it is the latter pattern, but we cannot demonstrate this with the data at hand.
3 Making Elections More Direct? Reducing the Role of Parties in Elections Susan E. Scarrow In contemporary democracies, political parties and elections are virtually synonymous. While partisan elections are certainly the current norm in the political realm, we nevertheless can distinguish different degrees of party involvement in these elections. It is possible to conduct democratic elections without parties, and there are many examples from outside the governmental realm of organizations that do just that: voluntary associations, even large national ones like the American Sierra Club or trade unions, routinely conduct internal leadership elections without the aid of formally organized parties. In the nineteenth century, many writers considered elections without party organization to be ideal for the public arena as well (Scarrow 2002), and even today a few democracies use non-partisan procedures to elect some public officials. And electoral choices over policy—through referendums and initiatives—often involve collective decisions unmediated by parties. This chapter investigates these differences, looking in particular at changes in the extent to which political parties occupy central mediating roles in electoral decision procedures. Is there evidence that elections are becoming what Peter Mair (2000) labels more ‘populist’ institutions, ones in which the declining role of parties means that citizens gain more direct opportunities to participate in the political process? Many of the reasons why we would expect changes of this sort are cited elsewhere in this volume (see especially Chapter 1), but it is worth repeating some of them to highlight how they apply to the question of party mediation in elections. To begin with, where de-alignment, or even alienation, from established political parties creates a climate of distrust of representative institutions, we expect that citizens may look more favourably on electoral procedures that minimize the party role and give citizens a more direct say. At the same time, dominant political parties may allow or even promote such procedures as a way of showing their responsiveness to public pressures, and may do so even while acknowledging that such devices limit parties’ control over the political arena (see, for instance, on Austria, Pelinka and Greiderer 1996: 31; on Germany, Scarrow 1997).
MAKING ELECTIONS MORE DIRECT? REDUCING THE ROLE OF PARTIES IN ELECTIONS
45
Second, changing political environments stimulate issues that cut across the traditional political cleavages around which party systems formed. As a result, parties have greater incentives to turn to referendums to settle issues which otherwise might fragment party unity (Gallagher 1996: 249). Third, changing technological and communications environments make it easier to ensure that citizens have the information they need to make political decisions without the aid of partisan cues. Innovations like direct mail advertising and Internet information delivery overcome some of the traditional objections to using ballots to settle specific policy issues (Arterton 1987; Budge 1996). These arguments provide good reasons to suspect that there is a widespread pattern of countries making more use of this type of nonparty election. The sections that follow consider whether the predicted changes have indeed occurred, as well as the question of whether the observed reforms have increased citizens’ direct access to, and influence on, the political realm.
NON-PARTY ELECTIONS Non-party elections can be divided into two general types: referendums and other types of issue elections on the one hand, and elections to choose public officials on the other. The first type is generally considered an example of direct democracy, because it asks citizens to settle a policy issue directly. The second type is a species of representative democracy (citizens chose the decision-makers), although it can be characterized as a more ‘direct’ form of representation than partisan elections in that it reduces the mediating element of party-directed choice. Of the two types of non-party elections, referendums are by far the more widely used. In referendums, party labels do not appear on the ballot because voters are being asked to decide about one or more specific policy propositions. Of course, the absence of party labels does not mean that parties necessarily stay out of the fray, which is one reason the term ‘non-party’ election is preferable to the more common contrast between ‘partisan’ and ‘non-partisan’ elections. Indeed, parties or party leaders often play a major role in the decision to hold a referendum, in the campaigns that precede issue elections, and in the calculations voters make when casting their votes (van der Eijk, Franklin, and Marsh 1994; Budge 1996). This likelihood is explicitly recognized by Britain's recent political finance legislation, which regulates political parties’ spending in referendum campaigns along with that of other groups. Nevertheless, although party campaigns and party cues may be central to the outcome of referendum votes, referendum ballots generally force citizens to cast their verdict at the polling place without the aid or encumbrance of party labels. Indeed, some argue that politicians often promote the use of referendums for precisely this reason, in hopes either of gaining a supra-party mandate on a particular issue or of defusing an issue on which their own party is divided.
46
SUSAN E. SCARROW
Non-party elections also can be used to select public officials. One defining feature of such elections is that their ballots do not specify candidates’ party affiliations: voters are asked to choose between individuals without being offered the mediating cues of party labels. However, the absence of a party label on the ballot does not by itself make an election a non-party event. For an election to be a truly non-party affair, it is also necessary that the offices in which elected representatives will serve are not organized along overtly partisan lines. The British experience makes this distinction clear: prior to 1970 parliamentary election ballots did not list candidates’ party affiliations. Despite this omission, no one would assert that British elections were ‘non-party’ events before this time, because the elected Members sat in the House of Commons as partisans, the viable candidates were chosen by local party branches, they campaigned as party representatives and with the explicit intent of getting voters to associate their name with a party label, and they expected that voters would reward or punish them on the basis of their national party's performance. In true non-party elections, candidates would campaign on their own records and ideas, and would not be able to point to party labels as shorthand cues. True non-party elections to select public officials are quite rare in established democracies, as will be detailed below. To the extent that they are found, it is usually in contests to select local or regional government officials, or to select officials for non-legislative posts. Even here, however, closer examination may show that ostensibly non-party elections are to some degree effectively partisan. For instance, even where party labels are absent on ballots and in the organization of the assembly, candidates nevertheless may need strong party ties and party backing in order to be serious contestants. Party considerations may also be very evident in a formally non-partisan local assembly if members regularly use the body as a stepping stone to a career in more overtly partisan politics. Despite these caveats, there are enough examples of non-party elections to indicate that they are a viable option. Where non-party elections are used, they clearly reduce the role political parties play in selecting candidates and shaping voters’ decisions. In this sense we can say that they offer voters more direct (less mediated) political choices. Several types of changes might increase the political importance of either of these two categories of non-party election. To begin with, citizens could be given more opportunities to use direct democratic procedures to resolve specific issues. This is a question not just of how easy or hard it is to call a referendum, but also of the policy issues for which they may or must be employed, who may initiate their use, and whether the verdicts of referendums are binding or merely advisory for elected representatives. To take the last issue first, if referendums are legally defined as solely consultative procedures, citizens cannot use a popular vote to override legislative autonomy, though legislators facing re-election may find it difficult to ignore popular verdicts. Another issue involving the relationship between partycontrolled legislatures and direct democracy involves the question of what
MAKING ELECTIONS MORE DIRECT? REDUCING THE ROLE OF PARTIES IN ELECTIONS
47
may be voted on: are voters asked to pronounce on potential legislation or are they permitted only to give a verdict on laws that already have been adopted by a legislature (abrogative referendums)? Finally, the question of who may call a referendum is crucially important in determining the extent to which direct democracy empowers citizens to take charge of policy decisions: citizens have the greatest opportunity to use this electoral tool if they can petition to force a vote on proposed or existing legislation (the right of initiative) (Bogdanor 1994: 24–33; Uleri 1996: 8–14). Countries could increase the political importance of non-party issue elections by moving along one or more of these dimensions, that is, by expanding the range of issues which can be put to a vote, by making it easier (or possible at all) for citizens themselves to initiate the use of direct democratic procedures, or by giving citizens tools to write and make binding laws of their own instead of just to advise or to abrogate existing legislation. Another way to boost the importance of non-party elections is to increase the number of offices that are filled by elections which do not use party labels, and in which parties have no right of nomination. This could be done by instituting non-party elections for positions that were previously filled by appointment or by attempting to expand opportunities for independent candidates in electoral procedures that have been dominated by parties. The sections that follow look at changes in each of these realms of non-party elections.
THE TRANSFORMATION OF ELECTORAL DEMOCRACY? Referendums In recent decades the idea of using referendums to settle policy issues has gained increasing popular support in many established democracies. Procedural opportunities for referendum usage have expanded, as has use of these procedures. These trends have been particularly evident at the sub-national level. As of 2001, thirteen of the OECD's eighteen established democracies offered some options for national legislative referendums. In addition, fifteen of the countries provided for optional or mandatory referendums on constitutional changes (see Table 3.1).1 As Table 3.1 also shows, in most countries these referendum options are well-established. This table provides summary scores from 0 to 4 showing approximate changes in the degree of institutional recognition for various devices that lower the party mediation in politics. Zero indicates non-availability, while higher scores indicate the prevalence and nature of procedures which provide the greatest scope for citizen involvement. As columns 1 and 2 of this table show, between 1970 and 2001 only two countries added provisions for national-level legislative referendums, while an additional three countries added provisions for constitutional referendums.
48
SUSAN E. SCARROW
Table 3.1.Availability of referendums and personal elections, 1970:2001a National constitutional referendum Australia 4:4 Austria 3:3 Belgium 1:1 Canada 0:3 Denmark 4:4 Finland 1:1 France 3:3 Germany 0:0 Ireland 4:4 Italy 2:2 Japan 4:4 Netherlands 0:0 New Zea1:1 land Norway 1:1 Sweden 0:1 Switzerland 4:4 United 0:1 Kingdom United States 0:0 Number 3 adding availability a
National legislative referendum
Local refer- Directly Directly No. areas endum elected head elected may- changed of state or
Score 1970:2001
3:3 3:3 0:0 3:3 3:3 3:3 3:3 0:0 3:3 4:4 0:0 0:2 3:4
1:1b 1:3 0:2 3:3 1:1 0:2 0:1 0:4c 0:0 0:2 4:4 0:2 4:4
0:0 4:4 0:0 0:0 0:0 2:4 4:4 0:0 4:4 0:0 0:0 0:0 0:0
2:2 0:2 0:0 3:3 0:0 0:0 0:0 1:4 0:4 0:4 4:4 0:0 3:4
0 +2 +1 +1 0 +2 +1 +2 +1 +2 0 +2 +2
9:9 11:15 1:3 9:12 8:8 6:10 10:11 1:8 11:15 6:12 12:12 0:4 11:13
3:3 3:3 4:4 0:0
1:1 0:1 4:4 3:4
0:0 0:0 0:0 0:0
0:0 0:0 0:0 0:1
0 +2 0 +3
5:5 3:5 12:12 3:6
0:0 2
4:4d 9
4:4 1
3:3 6
0
11:11
Note: For explanation of specific changes, see chapter Appendix. Coding. Coding to highest category; lower options may be possible. Change is attributed to year when constitutions or legal codes are altered to formally recognize the possibility of using these instruments, even if implementation legislation was not adopted simultaneously. For countries lacking a single constitutional document, constitutional changes are attributed to the year in which practice shifted. Codes in italics where availability expanded. Constitutional referendum. 0: none. 1: optional, called by legislature/executive/head of state. 2: optional, called by petition. 3: mandatory for some revisions. 4: mandatory for all revisions. Legislative referendums direct decisions on municipal ordinances (excluding for changing names and/or boundaries of political units). 0: none. 1: nonbinding referendums called by council/legislature/ executive/head of state. 2: non-binding referendums called by petition. 3: binding referendums called by council/legislature/executive/head of state. 4:binding referendums called by petition.
b c d
Direct election of executives (mayors, prime ministers) and heads of state. Larger cities only, if small municipalities differ. 0: never. 1: seldom. 2: sometimes. 3: usually. 4: always. Available in some states. Available in one state in 1970; in all states in 1998. 79% of municipalities in 1970; 90% in 1993. Sources: General: Batley and Stoker (1991); Council of Europe (1993); Gunlicks (1981); Humes and Martin (1969); Luthardt (1994); Norton (1994). Austria: Pauger (1983); Pelinka (1994); Pernthaler (1980); Welan (1986). Belgium: personal communication from Kris Deschouwer. Canada: Boyer (1992); Higgins (1986). Italy: Hine (1993); Vandelli (1992). Netherlands: Lucardie and Voerman (1997); personal communication from Paul Lucardie. New Zealand: Mulgan (1994); Simpson (1992). Sweden: Holberg and Stjernquist (1988); personal communication from Jan Teorrell. United Kingdom: Alderson (1975). United States: Klevit (1972); Renner and DeSantis (1993).
MAKING ELECTIONS MORE DIRECT? REDUCING THE ROLE OF PARTIES IN ELECTIONS
49
So while the availability of referendum opportunities at the national level is relatively high, this is not the result of a recent shift towards letting citizens use tools of direct democracy. Nor do the tools that are available generally provide for citizens to act independently of the decisions of party-controlled legislatures. Few countries allow any direct democratic procedures by petition, which means that elected representatives still retain control of the political agenda despite growing opportunities for direct decision-making. There are a few exceptions: Italy provides for an abrogative referendum by petition, while Switzerland allows citizens to petition to put issues to a binding vote. In 1993 New Zealand introduced the possibility of citizen-petitioned non-binding referendums, and in 2001 the Netherlands authorized citizen-petitioned non-binding referendums. Except for these cases, the use of direct democracy at the national level is either mandatory, as in the case of some constitutional changes, or is at the discretion of partisan legislative or executive gatekeepers. Only in the case of New Zealand was there a shift from already authorized legislative referendums towards increased citizen control by enabling citizens to petition to hold a referendum. Thus, the national record in this area does not suggest that citizens are gaining significant new powers to govern directly at the national level, without party mediation. In contrast, the record of institutional change at the local level is much more striking. As the third column of Table 3.1 indicates, by 2001 sixteen of the eighteen countries made some provision for local-level referendums to decide issues other than, or in addition to, changes in government boundaries. Of these, nine countries—half the total set—had added or expanded these opportunities since 1970. If we are looking for institutional change in the availability of direct decision-making procedures, we find them at the local level. Significantly, however, this is not because there has been greater willingness to introduce these procedures locally but rather because in many countries the local level has caught up to national norms. To what extent have citizens and legislatures taken advantage of these devices for soliciting public opinion? Here we find modest support for the idea that the referendum is catching on as a major alternative to representative forms of decision-making. Two previous studies of referendum use through the mid-1990s provided only scant evidence of an upsurge in referendum use up to this point (Butler and Ranney 1994a; Gallagher and Uleri 1996). Although the Butler and Ranney study has frequently been cited for its conclusion that ‘worldwide the use of referendums has increased’ (1994: 4), less attention has been paid to the second part of its conclusion: that this increase all but disappeared if US states and Switzerland were excluded. Similarly, Gallagher (1996) found no increase in usage in Europe during 1970–95 as compared with 1945–69, despite the fact that referendums were very much in the news in the early 1990s as several European countries used them to ratify constitutional changes related to European integration. Even faced with these findings, all three authors agreed that popular sentiment in favour of referendums was indeed increasing by the 1990s, and that
50
SUSAN E. SCARROW
Fig. 3.1.Number of referendums in 18 OECD democracies, 1945–2002
Sources: Butler and Ranney (1994: Appendix A); Centre d'études et de documentation sur la démocratie directe (2002). it therefore seemed likely that future studies would register an increase in referendum use. Updated figures for the relevant set of countries do suggest that since these authors wrote the growth in referendum usage has continued (see Fig. 3.1; see also Chapter 2). As in the past, much of the increase comes from Switzerland, which has held more referendums than the seventeen other countries combined. In Switzerland referendum usage at the federal level has jumped dramatically in recent years, growing from twenty-three in the 1960s to seventy-eight in the 1970s to a record eighty-eight in the 1990s. This growth was sparked in part by a huge increase in the use of citizen-initiated referendums: a record 527 petitions were submitted in the 1991–2000 decade, almost twice the number in any previous decade (Centre d’études et de documentation sur la démocratie directe 2002). Even if we exclude Switzerland, the growth in referendum usage from the 1960s to the 1990s is notable (see Fig. 3.2). Some of the most recent growth has indeed been driven in part by a surge of constitutional referendums held in conjunction with European Union expansion. In some cases, a single expansion proposal, such as the Maastricht Treaty or the Treaty of Amsterdam, generated optional or obligatory referendums in several countries. As Fig. 3.2 makes clear, however, the increased referendum use of the 1990s is not merely a by-product of European integration, nor is it an exclusively European phenomenon, given that countries like New Zealand and Australia also contribute to this increase. Figures 3.1 and 3.2 show that, outside of Switzerland, referendums are more likely to be held on changes to constitutional and/or electoral law than on other topics of ‘ordinary’ legislation.
MAKING ELECTIONS MORE DIRECT? REDUCING THE ROLE OF PARTIES IN ELECTIONS
51
Fig. 3.2.Number of referendums in 17 OECD democracies (minus Switzerland), 1945–2002
Source: As of Fig. 3.1. There is evidence that similar changes in referendum usage are taking place at the sub-national level. Of the seven federal or quasi-federal states in the set of countries considered here (Australia, Austria, Belgium, Canada, Germany, Switzerland, United States), provincial/state-level referendums are now common in only four: Australia, Germany, Switzerland, and the United States. In all four, sub-national referendums seem to be increasing since the 1970s. For instance, in the United States the number of state-level referendums resulting from ballot initiatives steadily increased from 88 in the 1960s to a twentieth-century high of 378 in the 1990s, eclipsing the previous high of 293, recorded in the wake of populist reforms in 1911–20 (see Fig. 3.3). Similarly, in Australia the number of state referendums went from two in the 1980s to eight in the 1990s, though most of these are clustered at the beginning of the decade, not the end. Germany also shows a big change in this area. Between 1970 and 1995 provisions for calling legislative referendums were added to the constitutions of all the German states, and these constitutional changes were followed by a surge in the use of state-level referendums (Scarrow 1997). In the 1990s there were nineteen state-level referendums in Germany, compared with a single state-level referendum in the 1980s and eleven in the 1970s. Although six of the votes in the 1990s might be dismissed as the extraordinary products of German unification (four were on adopting constitutions in eastern states, two were connected with the proposed fusion of Brandenburg and Berlin), this increase nevertheless represents a big jump in referendum usage (Mehr Demokratie e.V. 2002) (see Fig. 3.4). Germany also offers some evidence about changing practices at the local government level. Here the use of local initiatives and referendums has increased
52
SUSAN E. SCARROW
Fig. 3.3.Number of referendums in US states, 1945–99
Source: Initiative and Referendum Institute (2002). Fig. 3.4.Number of state referendums in Germany and Australia, 1945–99
Source: Centre d'études et de documentation sur la démocratie directe (2002). since the beginning of the 1990s, an increase that was above all the result of legal changes in the late 1980s and early 1990s that first made it possible for citizens to use these procedures in most of the German states. Local communities held more than 800 referendums in the 1990s, indicating that referendums and ballot initiatives are becoming an accepted feature of the landscape of German local politics. In a country with over 14,000 local political units, however, such a usage rate
MAKING ELECTIONS MORE DIRECT? REDUCING THE ROLE OF PARTIES IN ELECTIONS
53
hardly constitutes a ‘democratic revolution’; but it suggests that citizens are not reluctant to employ these unfamiliar devices (for details see Wollmann 2001). Thus, at both state and local levels the German case provides the clearest evidence that rule changes can lead to changes in democratic practices, with direct votes becoming a more important part of the political landscape. But there is little to suggest that the German case of changing rules is typical of broad trends. Even if it were, we should not overstate the change: the increased referendum use in countries like Australia and Germany is impressive in terms of percentages but is slight in terms of overall number. More generally, while the evidence does suggest that referendum use is expanding, it is important to note that in most countries that growth builds on what was an initially very low figure (see also Chapter 2). Even with a dramatic doubling in the frequency of referendum use, in most countries the referendum remains a device for resolving occasional policy questions rather than being a central or dominant political procedure. Even if referendums were free of partisan influences—which they seldom are—the increase in referendum usage is certainly not enough to diminish the importance of political parties, though it may increase the importance of some single-issue groups. After we have examined these usage tallies, it is worth noting that usage alone may not be sufficient to measure the political importance of such devices. Once referendums become an accepted way of settling issues, those who lose in the legislative arena can threaten to ‘go public’ by calling for a popular vote on a particular issue; this threat by itself may change the dynamic of policy negotiations whether or not a referendum is held. For instance, we certainly would judge referendums as of relatively minor importance in British political life if we looked only at usage. No national referendum has been held in Britain since the first one in 1975 (on whether to stay in the European Economic Community). Four regional referendums have been held since then (two on Scottish devolution, one on whether to create a Welsh Assembly, one on the Northern Ireland peace accords). Although we might make something of the fact that three of these were held in the late 1990s, this record of five referendums in eighteen years hardly suggests that direct democracy plays a big role in British political life. Yet a reading of British newspapers over the past several years would show that the question of when and how to hold a referendum on British participation in the European single currency has been one of the key issues of British politics. Despite the contentiousness of the policy decision, no British party disputes that the question of whether to join the single currency ought to be settled by a referendum. This attitude marks a big shift in political attitudes since the 1960s, when questions were still raised about whether the referendum had a legitimate place in British constitutional practice (Bogdanor 1981). As this example suggests, the frequency of referendums may not be the only way to judge the political importance of these devices. Because the option to
54
SUSAN E. SCARROW
invoke the threat of a referendum also may tell us something about the role of direct democracy in political life, the institutional change in itself may mark a shift in political attitudes and practices regardless of whether there are great shifts in the rate of usage. This increased prominence of non-party elections is as likely to change the dynamics of party competition as it is to stimulate non-partisan political organizing. Still, the formally non-partisan nature of the ballot, and in some cases the desire to build cross-party coalitions, may lead to the formation of referendum-specific nonparty pressure groups.
Non-party Election of Ofce-holders What about the other type of non-party elections, those to select public officials? Few countries have national elections that are conducted on a non-party basis. Of the world's democracies, only three ‘micro-states’—Micronesia, the Marshall Islands, and Palau—employ non-partisan procedures for selecting the main chambers of their national assemblies (Lutz forthcoming). As is documented elsewhere in this volume, in most countries access to ballots for national legislatures tend to be written in terms of nominees of parties, whether new or old, and to make little or no provision for non-party candidates (see Chapter 5). This is somewhat different at the local level, where at least nominally independent candidates and groups are more common in some countries. Even so, few countries seem to use election procedures at the local level that formally preclude the use of party labels (though the lack of good comparative data on local government elections makes this difficult to pin down). The United States and Canada seem to be exceptional in having many cities that organize local elections on a formally non-party basis. In both countries, the origins of these non-party rules go back to efforts to weaken parties at the beginning of twentieth century. Although the move towards non-party, ‘professional’ local government is nothing new in the United States, there nevertheless is some evidence that the use of non-partisan election procedures may have expanded in this country in recent decades, where the proportion of municipalities without party affiliation on the ballot increased from 50 per cent in 1971 to almost 75 per cent in 1991—and to over 80 per cent of the cities of over 50,000 (Klevit 1972: 26; Renner and DeSantis 1993: 68). In some US states judges, including state supreme court justices, are elected on non-party ballots or face a non-partisan retention election after serving an initial, appointed term in office: this is probably the highest level of office which is elected in non-party contests in any of the established democracies. Of course, what is most distinctive is not that the US states use non-party elections to fill these posts, but that they use any kind of election at all: in most countries, judges are appointed by elected officials or by a panel from within the legal system. In the US there has been a slight move away from partisan judicial elections since 1970, when they were used by sixteen of the thirty-eight sates (42 per cent) that relied on elections to select or re-select most members of the state judiciary. In contrast, by
MAKING ELECTIONS MORE DIRECT? REDUCING THE ROLE OF PARTIES IN ELECTIONS
55
2001 only twelve of the thirty-nine states (31 per cent) that held judicial elections used partisan ballots (Council of State Governments 1971; 2001). Thus, in the United States non-party elections remain an important feature of local politics, and the principle of at least nominally non-partisan elections still finds strong and perhaps growing support in some political arenas. However, countries other than the United States and Canada have shown little interest in organizing elections which formally exclude political parties. In short, even if we conclude that there have been some shifts that allow citizens a more direct say in political processes, this will not be because political parties are being formally excluded from elections for public office.
A CHANGING PARTY ROLE IN PARTISAN ELECTIONS? What about the role of parties in partisan elections? Is there any evidence that these elections are changing in ways that give citizens a greater say over which candidates get elected? As noted above, one way this might happen is by introducing procedures which allow voters more choice between individuals, not just between party labels. In fact, in all but three of the countries examined here, ballots for national legislatures do now offer voters the chance to express support for an individual candidate rather than (or sometimes in addition to) a political party. However, in most countries these rules have not changed much in recent decades, and in New Zealand, one of the few places where they have changed, the move was away from voting for individual candidates towards the adoption of closed party lists (see Chapter 5). Another way to reduce the importance of parties in channelling voters’ choices is to separate elections for legislatures and executives, thus enabling voters to express a direct say over which individual occupies the top spot in a government and giving voters greater opportunity to split their votes between representatives of different parties. Once again, when examining the record of institutional change we see that there has been more movement in this direction at the local than at the national level. In this period six of the eighteen countries in this set introduced the direct election of mayors in at least some jurisdictions (see Table 3.1). This is a strong pattern of change. What is not clear is why this change has come about and, in particular, whether it reflects popular pressure to give voters more say. Experiences in Britain cast an interesting light on this. Here the Local Government Act of 2000 authorized municipal councils to call referendums on whether to introduce mayoral elections. By early 2002, twenty-three towns and cities had held such referendums, most of which were called at the behest of the local council. (Only in one case was a vote called by popular petition, an option the law provided for.) Turnout in most of these referendums was low, and only eight of the twenty-three votes supported the change to a directly elected mayoral system (Parker 2002). Apparently this reform lacked strong popular support even
56
SUSAN E. SCARROW
though it was proposed by the government as a way of strengthening citizens’ links to the political realm. Nevertheless, this British evidence of citizen indifference does not in itself negate the importance of the shift towards direct election of local executives. At the least, the change can have some profound political consequences because it opens up the possibility that there will be divided party control of local government. It also opens the way for politicians to build local careers based on their own popularity rather than on their standing with national party leaders. Both phenomena have been observed in Germany since this country's expanding use of direct mayoral elections, with some of Germany's directly elected mayors avoiding party labels and campaigning as independents or at least distancing themselves from their parties once in office. The incentives to distance themselves from partisan politics increase when mayors find themselves without a sympathetic partisan majority in the city council, something that occurred in several major cities very soon after the institutional set-up was changed (Wehling 1993:32; Wehling 1994:186). Similar patterns are seen in Britain. For instance, Ken Livingston was elected mayor of London in that city's first mayoral election, much to the displeasure of leaders of his own Labour Party. In some of the subsequent British mayoral elections, candidates not affiliated with any of the mainstream parties have done surprisingly well, prompting some to call for a re-examination of the institution. These and other experiences in both countries seem to justify von Arnim's (2002) description of direct mayoral elections as devices that weaken party dominance at the local level.
CONCLUSION The findings in this chapter provide rather mixed support for those who are looking for signs of radical transformation in the ways that elections are used to give citizens a direct say in political decision-making. On the one hand, institutional changes have been occurring, and it is significant that the movement is almost entirely in the expected direction: countries are expanding opportunities for citizens to directly decide issues, and they are giving citizens more chances to directly select executive leaders rather than having these leaders selected by assemblies. Looking again at the last two columns of Table 3.1 we see that eleven of the eighteen countries have made at least one institutional change that provides possibilities for more non-party elections or for executive elections in which voters have more opportunities to choose between personalities instead of party labels. At least as significantly, the one move in the opposite direction (removing mentions of local referendums in the revised English local government act of 1972) was itself reversed in 2000 by new legislation that provides for local government referendums on one specific subject (the direct election of mayors). And, as the final column of this table also makes clear, many of the countries that did not
MAKING ELECTIONS MORE DIRECT? REDUCING THE ROLE OF PARTIES IN ELECTIONS
57
make such institutional changes were those, like Switzerland and the United States, that started out scoring relatively high in terms of the availability of ‘non-mediated’ elections. This kind of shift is consistent with what Peter Mair (2000) labels a move towards ‘populist’ democracy, in which parties remain important as governors but lose much of their importance as mediators and bridges between society and the state. Yet, despite these very real signs of institutional change, it is important not to overstate the magnitude of the shift. Usage of these new opportunities remains relatively rare. The addition of new forms of direct election undoubtedly contributes to the increase in voters’ chances to express their views by casting ballots (see Chapter 2). But outside Switzerland and some US states, referendums still play only a very limited role in the political life of established democracies. To the extent that they are used, they are most often called at the behest of elected representatives, not as a result of a popular initiative, not least because few countries provide for citizen-petitioned votes. And when they are held, participation often is lower than participation in national legislative elections. It is harder to gauge the popularity of directly elected executives from figures on electoral participation. Nevertheless, the experience in Britain, where municipalities have been slow to exercise the option of switching to directly elected mayors, at least seems to suggest that voters here have been less interested in making this change than national political leaders had assumed. Thus, the evidence of institutional change and institutional usage suggests that we are far from witnessing a radical move away from parties or party-dominated political processes. And indeed, even if direct democracy became much more common, it would not necessarily undermine the strength of parties. Budge (1996) convincingly argues that traditional juxtapositions between direct democracy and party democracy are misleading. Instead, he sees the two as complementary: direct democracy can increase popular support for party-dominated representative structures without threatening party dominance, and high party involvement will improve the choices made using direct democratic procedures. Whether or not we agree with Budge on this point, it certainly is clear that most of the institutional moves towards direct democracy are designed to preserve a strong role for parties. Most notably, the spread of referendum procedures has not been accompanied by a similar spread of provisions for popular initiatives, the devices that might more easily slip beyond the control of party leaders. In short, there seems to be growing acceptance of the idea that voters deserve to have a direct say on some occasions, and this has led to the introduction of new opportunities for citizen participation. Use of these new opportunities—and even the threat to use them—has affected and will continue to affect the outcomes of important political discussions. Nevertheless, acceptance of the idea that there is a place for direct citizen control is still coupled with wariness about giving citizens too much say too often. As a result, we are seeing not a tidal wave of revolution but rather a slow trickle of change in ideas about what democratic opportunities should
58
SUSAN E. SCARROW
be available, and whether parties should have an exclusive role in structuring the choices which voters are offered.
Appendix A: APPENDIX: EXPLANATION OF CHANGES LISTED IN TABLE 3.1 Country Austria Belgium
Date 1984 1994 1995
Canada Finland
1992 1990 1991
France
1992
Germany
1980s/1990s 1980s/1990s
Ireland Italy
2001 1990
Netherlands
1993 1994 2001
New Zealand
1980s 1993
Sweden
1977
United Kingdom
1979 1972 1974 1998 2000
Change Local referendums (binding) authorized Direct election of mayors authorized Local initiatives (consultative) authorized Federal referendums legally recognized Local referendums (consultative) authorized Direct election of president replaces electoral college Local referendums (consultative) authorized Direct election of mayors introduced in most states Local referendums (binding) authorized in most states Direct election of mayors introduced Local referendums (consultative) authorized Direct election of mayors introduced Local referendums (consultative) authorized Citizen-petitioned local and national (consultative) referendums authorized Rural towns introduce directly elected executives Citizen-petitioned national referendums (consultative) authorized Local referendums (consultative) authorized Constitutional referendum authorized Mention of local referendums removed from Local Kingdom Govt. Act National referendums introduced Direct election of London mayor introduced Direct election of mayors added as option; decided by local referendums
Source: See Table 3.1.
NOTES 1
The absence of constitutional recognition for using referendums does not mean that such devices could not be used, only that the results would not be legally binding. In some countries, local governments have held such advisory referendums even in the absence of formal institutional provisions for referendums.
4 Political Parties and the Rhetoric and Realities of Democratization Miki Caul Kittilson and Susan E. Scarrow Discussions of a presumed shift toward more participatory styles of politics often focus on ‘unconventional’ participation and on ways citizens may circumvent traditional partisan channels of representative democracy. However, political parties remain central players in the democratic processes of the OECD nations (Dalton and Wattenberg 2000; Webb, Farrell, and Holliday 2002). Party actions help shape the nature of participation opportunities and even influence the level of participation within any given society (Rosenstone and Hansen 1992; Gray and Caul 2000). Thus, if we are to have a clear picture of the ways that political participation takes place in democracies, it is essential to include developments shaped by, and occurring within, the political parties themselves. Of course, even within democracies, parties differ widely in the extent to which they encourage participation and distribute responsibility for decision-making among their supporters. There is nothing wrong with this: as E. E. Schattschneider (1942: 60) reminded us long ago, democracy exists in the competition between parties, not within them. So parties that operate on a non-democratic basis are not per se a threat to democracy. Yet this does not mean that it is not possible or perhaps even desirable to have democracy within parties, only that intra-party democracy is not essential to guarantee the democratic life of the broader polity. And indeed, since Schattschneider made his observations, social and political transformations appear to have significantly altered citizens’ views about how political processes should operate. Rising levels of education, information, and the subsequent processes of cognitive mobilization (Dalton 1984; Inglehart 1990) apparently have increased citizens’ reluctance to merely choose among prepackaged party platforms and party candidates, and citizens are now more interested in having opportunities to participate directly in political decision-making processes. The main purpose of this book is to investigate the extent to which democratic institutions and practices are changing through these apparent new understandings of politics. In this chapter we consider some of the roles that political parties may play in this process while operating as intermediaries between changing citizen preferences and state structures. To begin with, we look at the extent to which parties
60
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
express an interest in democratizing reforms. In doing this, we not only look at the values that parties proclaim in their campaign documents but also assess parties’ internal constitutions. We view these documents as, in part, public relations exercises that reveal something about how particular parties would like to be perceived. Finally, we look at these constitutions from the perspective of political practice, asking whether parties’ increased interest in the rhetoric of democracy translates into meaningful changes in how the parties run themselves.
PARTY MANIFESTOS AND THE PROMINENCE OF DEMOCRATIC THEMES Much of the research reported in this volume looks for evidence of changing institutions and practices. It is also important to look at the other side of this equation by examining indicators of the attitudinal changes that ostensibly promote institutional reforms. The most common way to do this is through surveys that ask citizens to evaluate various forms of self-government. We can also do this more indirectly by looking at the rhetoric that politicians use to justify current practices and to back calls for reform. Since politicians are both opinion leaders and opinion followers, political rhetoric can provide useful clues about which principles they think will resonate widely. So our initial question concerns the extent to which party rhetoric suggests that parties have become more concerned about ‘selling’ themselves to voters along the democratic dimension. We answer this question by examining the cross-national, longitudinal data assembled in the Comparative Manifestos Project. The Comparative Manifestos Project assembles party election platforms across twenty-five nations from 1945 to 1998, and through content analysis measures each document's emphasis on several different policy categories (Budge et al. 2001). The project's common framework and cross-temporal nature make this data-set ideal for examining changing party rhetoric. The data do not allow us to say whether parties have implemented their promises once elected. However, if we do not see much change in the extent to which parties emphasize democratic themes, we should not be surprised if we do not see parties promoting major institutional changes of this sort once they are elected. To what degree do party manifestos highlight democratic themes, and how has this changed since the 1960s? We examine the proportion of the election platform dedicated to ‘Favorable mentions of democracy as a method or goal in national and other organizations; involvement of all citizens in decision-making; as well as generalized support for the country's democracy’. A great variety of issues might be coded as a ‘democratic theme’. For example, in an excerpt from the British Liberal Party/Social Democratic Party Alliance programme of 1983, the following were coded under this category: ‘Because we are not prisoners of ideology we shall listen to the people we represent and ensure that the good sense of the voters is allowed to illuminate the corridors of Westminster and Whitehall.’ Later in the
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
61
programme a different issue falls under this same democratic category: ‘. . . and reform trade unions to make them genuinely representative institutions. We propose … an Industrial Act to provide for the introduction of employee participation at all levels.’ In other words, this is a rather catch-all theme, and change could stem from a variety of sources of shift in emphasis. Nevertheless, if parties clearly shift their attention to the kind of democratizing themes that are of most interest to us, this would likely show up in these data—and the absence of change would suggest that the use of democratic rhetoric has not changed. Table 4.1 displays the average percentage of party platforms devoted to democratic themes in eighteen nations. National averages simplify presentation, and more detailed scores by party are in Appendix A. In order to focus on trends, the first four columns represent decade averages, from 1960 to 1990. Often a certain election platform will show a spike in emphasis on democratic themes as a party intensifies its focus or perhaps as other themes lose ground. By using decade averages, we downplay such short-term variation to focus on long-term trends. The final column represents the percentage point change from 1960 to 1990. Overall, there are only modest changes in democratic rhetoric. The overall net change from the 1960s to 1990s—across all nations— turns out to be nearly
62
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
Table 4.1.Saliency of democratic themes in party platforms, national averages, 1960–98 Country Australia Austria Belgium Canada Denmark Finland France Germany Ireland Italy Japan Netherlands New Zealand Norway Sweden Switzerland United Kingdom United States Average
1960s 2.0 13.1 2.2 1.8 3.1 3.2 4.7 2.7 2.4 1.5 6.5 5.0 0.6 2.7 2.5 4.9 1.9 0.5 3.4
1970s 6.9 9.4 5.1 1.8 2.9 5.3 5.9 2.8 4.8 2.5 4.6 7.4 1.7 2.3 5.8 6.4 4.6 1.3 4.5
1980s 0.6 7.0 3.0 2.1 2.2 7.8 4.2 2.6 1.7 3.1 3.3 3.3 2.8 2.0 1.9 6.6 3.9 2.4 3.4
1990s 1.9 4.8 5.1 2.9 1.8 2.8 4.2 2.9 4.2 3.9 2.3 4.2 4.4 2.9 3.2 6.9 2.3 1.8 3.5
% Change, 1960s–1990s −0.1 −8.3 +2.9 +1.1 −1.3 −0.4 −0.5 +0.2 +1.8 +2.4 −4.2 −0.8 +3.8 +0.2 +0.7 +2.0 +0.4 +1.3 0.1
Note: Entries represent mean percentage election platform dedicated to ‘Favorable mentions of democracy as a method or goal in national and other organizations; involvement of all citizens in decision-making; as well as generalized support for the country's democracy’. Source: Comparative Manifestos Project (Budge et al. 2001).
zero (+0.1). Of the eighteen nations featured in Table 4.1, eleven show an increase from the 1960s to the 1990s. Out of the 125 parties in the data-set, sixty parties increase and forty-four parties decrease their discussion of democracy. For the most part, gains and losses are quite small. The largest increases occurred in New Zealand (+3.8), Belgium (+2.9), Italy (+2.4), Switzerland (+2.0), Ireland (+1.8), US (+1.3 ), and Canada (+1.1). Only three national party systems see an average decline: Austria (−8.3), Denmark (−1.3), and Japan (−4.2). The remaining national party systems experience very little change (Sweden, UK, Norway, Germany, Australia, Finland, France, and the Netherlands). Only in rare instances do parties dedicate more than 10 per cent of their platform to democratic themes, and where there is change over time it is usually in the single digits or less. These data also demonstrate a great deal of variation across nations, parties, and time. For instance, Austria shows a spike in democratic themes in the 1960s, followed by a consistent drop into the 1990s. The high in the 1960s is largely driven by the Austrian Freedom Party, which focused an extraordinary 27 per cent of its platform on democratic themes in the 1960s, yet dedicated barely 5 per cent of its platform to the same issues by the 1990s, reflecting the party's rightward shift.1 In other instances attention to democracy climbed in the 1970s or 1980s, only to fall by the 1990s. France also shows a spike in the 1970s, driven partly by the French Socialist Party (PS). This was the period when the PS committed itself to the policy of ‘autogestion‘, or worker self-management (see Appleton 2001). Similarly, Australian and British parties stressed the language of democratization to a higher degree in the 1970s, yet attention to this theme sank over the next two decades. At the party level, we see examples like the Irish Fine Gael, which began with 2 per cent in the 1960s, climbed to 6.5 per cent in the 1970s, sank back down to 1 per cent in the 1980s, and again rose to 5 per cent in the 1990s. The Swedish Social Democratic Party appears to have experienced a similar apex in the 1970s. Only in the cases of the Danish Conservative (KF) and the US Democratic parties do we observe steady gains over all four decades, and the gains are quite small in both cases. Of course, these changes may not be randomly distributed across parties. Previous research suggests that ideology may strongly affect a party's disposition towards democracy. Social Democratic and Green/New Left parties should be more likely to stress democratic themes. We expect rightist and ultra-right parties to devote the least attention to democratic themes (Kitschelt and McGann 1997; Kitschelt 1989). Thus, to determine whether democratic discourse is more salient on the left, we examine ‘families’ of parties across the ideological spectrum. Figure 4.1 displays the average attention to democracy in party platforms by party type. The results support our expectations in one sense: By far, New Left parties devote the greatest amount of attention to democracy, averaging 6 per cent in the 1990s, almost double the amount of any other party family. Yet differences among the remaining party family types are less clear. New Left parties are trailed
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
63
Fig. 4.1.Saliency of democratic discourse in party platforms
Note: Entries represent decade averages based upon mean percentage election platform dedicated to ‘Favourable mentions of democracy as a method or goal in national and other organizations; involvement of all citizens in decisionmaking; as well as generalized support for the country's democracy’. Source: Comparative Manifestos Project (Budge et al. 2001). by the Socialist/Labour parties, which average almost 4 per cent of their platform dedicated to democratic ideas. Strikingly, religious parties also devote nearly 4 per cent of their platform to democratic themes, and Liberal parties 3 per cent. Most surprising of all is that ultra-right parties average 3 per cent, a higher degree of democratic discourse than among Centre, Ethnic, Conservative or Progress Parties. Within each ‘family type’ there is considerable variation. For instance, among Green parties in the 1990s, the Finnish Greens dedicated 15 per cent of their platform to democratic themes, while the Belgian, French, and German Greens gave only 3 per cent of their attention. Similarly, among religious parties, the Austrian ÖVP devoted less than 2 per cent, while the Belgian Christian parties dedicated approximately 7 per cent. Conversely, Conservative parties tend to be fairly consistent in their lack of attention to democratic themes across nations. In summary, we do see that New Left parties emphasize democratic themes in their rhetoric, which is what we would expect given their ideological commitment to inclusion of under-represented groups and participatory norms. Yet ideology has its limitations as an explanatory variable, since attention to democracy is spread across the ideological spectrum. In addition, the trends within individual countries suggest that the saliency of democracy in party platforms follows national patterns. For example, increases in democratic discourse are generally
64
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
apparent across the entire Belgian system over this time span. Similarly, the Japanese parties display a common pattern of emphasizing democratic themes in the 1960s and 1970s, but backing off from these topics in the 1990s. The results suggest that when one party in a system intensifies its focus on democracy other parties follow suit. The rhetoric of party platforms also suggests that, while some parties consider democratization to be a vote-winning theme, few make it a paramount issue. Although there has been some shift towards paying greater attention to democratic themes, particularly evident with the advent of New Left parties, it is certainly not a major priority nor one that has assumed a much more dominant position since the 1960s. This is a good reminder of the broader political context within which the changes studied in this book have occurred. Issues of democracy and participation certainly have been prominent in the political discourse of recent decades, but they have by no means been the only theme or even a primary one. The next sections look at more evidence of party rhetoric, namely, at parties’ formal structures and at the ways parties may have tried to alter these structures to appeal to citizens who seem to be interested in exercising more direct control over political institutions.
PARTY CONSTITUTIONS AND POLITICAL IDEALS Political scientists have devoted considerable effort to abstracting models of party organization based on their observation of party practices. Among the best known of these are the cadre and mass (Duverger 1954), catch-all (Kirchheimer 1966), and, most recently, the cartel model (Katz and Mair 1995). But these four types by no means exhaust the list.2 As Katz and Mair (1995) explain, one of the fundamental ways in which parties conform to these various models is in their (usually implicit) visions of the proper roles of groups and individuals in the processes of self-government. According to them, each vision of society promotes a different view of the roles party leaders and party supporters should play in defining the party's message. For instance, mass party structures spring from a view of society as composed of self-aware and self-interested groups, and such parties position themselves as group representatives, drawing support from pre-defined social segments, mobilizing a natural constituency, and articulating its demands. Such parties may integrate representatives of their constituent groups into structures of party governance. In contrast, catch-all parties correspond to a vision of society as composed of atomized individuals; such parties try to invent loose communities of interest by ‘selling’ their politics to citizen consumers and by helping create a package of policy concerns that unite successful coalitions. According to Katz and Mair (1995), leaders of such parties feel compelled to offer members and other supporters more apparent say in party decisions because citizens want to feel more involved in governing processes. However, the authors assert, party leaders make these concessions in ways that really expand leaders’ own power.
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
65
In making this point, Katz and Mair are reminding us to differentiate between the rhetoric of party constitutions—what they consider to be part of the ‘official story’ (Katz et al. 1992)—and the reality of power distribution within these organizations. Clearly this is an important distinction, and any understanding of politics requires us to uncover where the power really lies. But this does not mean that we should ignore the rhetoric. At the least, it is a useful indicator of changing attitudes concerning the appropriate relations between party leaders, groups, and individuals. For instance, there is a politically significant difference between the British Conservative Party of the 1960s, in which party notables picked the party leader behind closed doors, and the Conservative Party of the 1970s through the mid-1990s, under whose rules Margaret Thatcher won and lost the party leadership on the basis of votes by Conservative members of the House of Commons. There is also a real difference between this latter party and the Conservative Party of 2002, which asked all dues-paying members to choose between the two finalists selected by the parliamentary party. Though members’ choice of candidates was limited under party rules, their formal inclusion into the process signals a rapidly changing understanding of the political importance of intra-party democracy. However we evaluate the power transfers involved, such rule changes seem to reflect a shift in parties’ understanding of public expectations, and as such they are important because of what they tell us about political norms. Whether parties actually deliver on their promises of internal democratization is another important matter, one that can be examined separately. Because party constitutions are documents that are prepared for public consumption, we can analyse them as an indirect indicator of public expectations. Party organizers and activists are often keenly aware of the symbolic value of party rules and about the political messages they send. This is one reason why we witness debates over whether places should be reserved on party executive committees for members of sub-groups that are considered politically relevant (women, ethnic or religious minorities). Such debates are clearly about the distribution of power within the party, but they also reveal assessments of how much electoral support is to be gained from symbolically including or excluding such groups. Of course, parties do not necessarily feel compelled to practise what they preach. Many parties do not consider it necessary to adopt internally democratic structures, nor do most countries feel it is in the general interest to compel parties to adopt such rules. (Germany and Finland are among the very few countries whose laws stipulate a certain degree of internal democracy in the conduct of party affairs.) Self-consciousness about organizational principles is perhaps most characteristic of parties that want to transform the political order: such parties often use their own charters to demonstrate how alternative models of political organization might work. In the past, this was most evident in the organization of Communist and Fascist parties. More recently, new Green parties have tried to practise new models of internal democracy to give credibility to their proclaimed aim of bringing about a new type of democracy in the wider polity.
66
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
Other parties are less concerned about trying to model themselves as ‘states within states’, in need of democratic constitutions that would befit a state. Instead, they explicitly or implicitly adopt organizations such as corporations and armies as their models. In adopting these models, they are asking to be judged not on the basis of the processes but by the outcomes they achieve. In parties that run according to this ethos, party leaders may resemble a board of directors who have to face occasional shareholder meetings, and fear shareholder rebellions if performance declines too drastically, but who are for the most part free to direct party activities as long as they deliver promised returns. These parties, too, may adopt democratizing reforms with an eye to electoral advantage, but their test of organizational quality remains electoral success. In short, because parties prepare their constitutions with an eye towards external as well as internal effects, a study of changes in formal structures may provide useful clues about what they assume will appeal to—or will alienate—public support. One way to think about the degree of democracy within parties is to examine the formal distribution of power, asking who selects and removes party leaders and who is involved in the selection of party candidates. Based on the answers to these two questions we can distinguish five basic models, roughly corresponding to traditional distinctions between rule by the one, the few, or the many: the leader-dominated party, the oligarchic party, the party of individual representation, the corporatist party, and the party built on the basis of democracy. Leader-dominated parties originate as, or become, the vehicle for a single dominant leader. In such parties, the leader may be self-selected (often, for instance, the party founder), or may be selected and re-selected by a party conference over which the current or outgoing party leader exercises strong control. Such parties may be loosely organized around a single leader (personalistic) or may be more formally organized (Leninist cadre/cell party). In either case, power is tightly held at the centre. Oligarchic leadership is a variation on this organizational style that is more common among long-established parties. Leaders in oligarchic parties are a self-selected group, generally consisting of elected officials, notables from the party's constituency, and party bureaucrats. Whatever the mixture of career paths leading to the top in such parties, the defining characteristic of the oligarchic party is its Fig. 4.2.Models of party governance
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
67
thinly veiled efforts to claim the mantle of democratic leadership selection. One good example of this oligarchic style was the British Conservative Party until the early 1960s. Of parties that can credibly claim to be governed by bottom-up structures, most rely primarily on representative procedures, though votes by the whole membership are becoming more common. Some representative parties operate on models which resemble what Held (1987: ch. 5) has described as competitive elitism: society is seen as fragmented, voters are viewed as poorly informed, and political leaders play prominent roles. Parties also may be founded on visions of society that appeal to individuals with diverse social identities, and may cater to such interest coalitions by offering ascriptive representation to members of communities which are considered important pillars of party identity (union members, women, ethnic minorities, and so forth). In more corporatist party structures, leaders of organized groups inside or even outside the party may have formal or traditional rights to participate in party decisions in order to ensure that sub-group interests will be represented within the party's legislative delegation (Poguntke 1998; 2002). Thus, parties may provide ‘linkage’ between government and citizens either through direct representation or through cooperation with collateral organizations. For the most part, the parties of the types described above have adopted outcome-oriented organizational strategies that are validated primarily by the parties’ electoral success, not by the extent to which they broaden the scope of individual participation. Under these models, citizens are presumed to have fixed interests and to choose among those who can best represent these interests, rather than to have a developmental need to participate in the political process. In contrast, a few parties adopt ‘basis democracy’ approaches to organization, affirming that one of the goals of party life is to expand the roles individual supporters can play within party debates. In such parties, the processes by which decisions are reached are carefully scrutinized for their conformity to the standards of democracy that the party promotes.
TRENDS IN PREVAILING MODELS OF PARTY DEMOCRACY The preceding discussion provides a framework for thinking about differences in models of party organization and leadership, and about differences in the organizational ideals parties espoused in their constitutions. When analysing change in the democratic practices within political parties, it is not enough to look solely at shifts in the extent to which parties provide opportunities for individual participation. Depending on their starting points, parties may democratize by moving from oligarchic, self-selected leadership to more representative structures. They also may change the nature of their representative structures, moving to or from group-based or individual representation. For instance, we may find that party constitutions are changing to include groups in different ways, or more groups, rather than becoming more broadly participatory. But these new forms of representation—even
68
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
indirectly through groups—may improve the quality of access among the rank and file. Thus, in this section we take a broad look at who gets to participate in party decision-making and how directly, and we ask whether parties’ recent organizational changes provide new access points for individual citizens who are seeking to influence the decisionmaking process.
How Inclusive is Legislative Candidate Selection? Almost a quarter century ago Austin Ranney (1981: 82–3) reported that the most common candidate selection method was by constituency party organizations with national or regional authorities holding an ultimate veto. Such quasirepresentative procedures provided the appearance of local control and broad participation without forcing party leaders to surrender total control over the process. Since then, a few very well publicized examples suggest that there may have been ‘democratization’ in party procedures for selecting both candidates and leaders. For instance, membership ballots to select candidates have been adopted by the Australian Democrats, New Zealand's Labour Party, and the Danish Centre Democrats. Likewise, parties as diverse as Irish Labour (Farrell 1994), British Liberal Democrats (Punnett 1992), and the Canadian Conservative Party (Carty and Blake 1999) have adopted membership postal ballots to elect party leaders. To move beyond the anecdotal, past cross-national research on the expansion of opportunities for participation within political parties finds some evidence of a shift in formal control. For instance, some research suggests that party leaders have granted members greater influence in candidate and leadership selection, although they retain veto rights (Scarrow, Webb, and Farrell 2000; see also Bille 2001). What is new is not that party leaders can overrule decisions made at lower levels but that the mass of members are included in the process, not just delegates or local party notables. One of the difficulties in looking for cross-national trends in this area is that candidate selection procedures are extremely complex, involving several sets of actors at various stages in the process from nomination to selection. Capturing the complexity of the process across a broad range of parties is virtually impossible, and more importantly, would likely obscure the patterns in which we are interested. In this study, we employ data on candidate selection that several other studies have assembled (Scarrow, Webb, and Farrell 2000; Katz and Mair 1992), and use it to investigate just the question of who enjoys formally guaranteed access to the candidate selection process. For purposes of this research, we classify parties according to the most inclusive body involved in the process. In other words, if party members are given a vote at any point in the process, for our purposes this party is categorized under the heading ‘membership vote’. The same logic is applied to other categories. If a regional party body plays a part in the selection process yet is subject to veto by the national leadership, the party is categorized according to the most inclusive decision-making level and placed under the
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
69
heading ‘regional delegates’. To illustrate the coding decisions, the Flemish Christian People's Party (CVP) is categorized under the heading ‘local party selectorate’ even though national leaders hold the right to veto locally made decisions. Similarly, the Finnish Greens are categorized as ‘membership can vote’ despite the ability of both the regional and national leadership to alter their decisions. The aim of this coding scheme is to highlight expanding access to the process, not to evaluate the internal power shifts that may prompt, or follow from, such changes. Table 4.2 presents evidence to assess the degree of inclusiveness of the party selectorate in legislative candidate selection. Of the eighteen nations for which we have data, fourteen have some parties that allow members a voice in candidate selection at some point in the process, and all nations have parties that give power to the local party organization or members in the selection process.3 At the party level, 45 per cent of the parties for which we have data allow members a vote in candidate selection, another 38 per cent allow the local party, and another 14 per cent allow regional delegates a voice in the process. Only in the French UDF and Dutch VVD does selection remain solely in the hands of the national party leadership. An asterisk in Table 4.2 denotes a move toward a more inclusive candidate selection process across recent decades. Of the ninety-five parties represented in the table, 41 per cent have moved toward a more inclusive process of candidate selection since 1960. For example, the Danish Socialist People's Party began in 1960 with a process in which constituency meetings open to all members selected the candidates. By 1976, a membership postal ballot became obligatory (Bille 2001). The patterns emerge largely along national lines. In Norway, Finland (since 1970), and Germany (since 1967) national laws have encouraged or required parties to adopt more open and democratic procedures. In other instances democratic reforms swept across party systems in a process of competition over votes, akin to Duverger's proposed party-level ‘contagion’ effect. For example, the entrance of Democrats ‘66 (D'66) into the Dutch party system and their dedication to transparent decision-making procedures set off a chain reaction in which other parties adopted similar rules. A similar process occurred across the Canadian parties (Carty and Blake 1999). At the same time, the party systems of both Italy and Norway maintained the status quo through this period. In sum, the movement toward allowing members access to the process of selecting legislative candidates occurs across most of these eighteen nations. These findings are consistent with Bille's (2001) conclusions regarding western European parties. Our evidence supports the contention that evidence of the democratization of candidate selection is more than anecdotal and represents a trend within a majority of these nations. Yet the trend within some nations is partially offset by the reluctance of parties in other countries to involve members. In Italy, Japan, Norway, and Switzerland parties limit participation in candidate selection to local, regional, and national bodies. Assertions about a democratizing trend are further tempered by recognition that members participating in the process do not necessarily wield the same amount of influence as other participants.
70
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
Table 4.2.Degree of inclusiveness of party selectorate in legislative candidate selection, 2000 Country
+
Australia (0) Austria (0) Belgium (2)
Oligarchical National leadership
Representative Regional delegates Lib, Nat FPÖ
UDF
RPR
Canada (5) Denmark (8) Finland (5) France (1) Germany (0) Ireland (2) Italy (0)
Japan (3) Netherlands (2) New Zealand (1) Norway (0) Sweden (0) Switzerland (7) United Kingdom (2) United States +
Workers', PD PSDI
Basis democracy Local party selec- Members can Non-members torate vote can vote Lab, Dem Gr, SPÖ ÖVP CVP, PSC, BSP, ECO, AGA, PSB, PRL* PVV* All parties* Progress SF,* RV,* V,* SD,* KF,*KRF* CD,* Liberals* All parties* PS* SPD, FDP, Greens, CDU FF, Labour, Gr*, FG* Greens, PDS, PSI, PR, PRI, PLI, PPI, AN, NL, FI, JCP,* JSP,* LDP*
VVD N
All parties*
Gr, D'66, PvdA,* CDA* Lab*
SV, DNA, V, C, H, KRF, FRP, Gr FP, CD, MP SAP, C, VP Con
Lab,* Lib Dem* All parties
Note: Table entries classified according to ‘most inclusive’ body involved in selection process. The abbreviations of party names are spelt out in the chapter Appendix. Number in parentheses indicates total number parties moving toward more inclusive procedures, 1960–2000. *Has moved towards more inclusive selection procedure since 1960s. Sources: Scarrow, Webb, Farrell (2000); Katz and Mair (1992); Bille (2001); McCarthy (2001); and personal communication with Michael Thies, UCLA; Ian McAllister, Australian National University; Marian Sawer, Australian National University.
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
71
Despite these caveats, we conclude that on the whole there has been a modest shift to more democratic candidate selection procedures since the 1960s.
How Inclusive is National Party Leadership Selection? In an era of ‘candidate centred campaigns’ in which the party leader embodies the party image (Wattenberg 1991; McAllister 1996) the selection of the party leader becomes another potential measure of intra-party democracy (see LeDuc 2001). Drawing again from the original data collected by Scarrow, Webb, and Farrell (2000), we compiled the formal rules regarding leadership selection for all of the parties for which we could find accessible information. Following the same coding procedures as those outlined in the previous section, we simplified complex selection procedures to classify parties according to the most inclusive body involved in the selection of the party's leadership. Table 4.3 presents this categorization for all of the parties in our set of nations for which we could find information. As was the case of rule for candidate selection, there is mixed evidence of change in the roles that party members play in the selection of party leaders. In the 1990s parties in a number of countries show a strong trend towards granting more procedural access to the rank and file, but a significant proportion continues to leave decisions in the hands of the party congress. There is a bigger shift among established parties if we consider the extent to which they have switched from oligarchic leadership selection to more representative procedures, empowering party congresses if not the individual members. Of the seventy-four parties for which data are available, 25 per cent allow members a vote in the process, and another 7 per cent allow members to nominate through their constituency organization. Yet this combined percentage is still outweighed by the 42 per cent of parties that leave leadership selection in the hands of the party congress. Only 19 per cent of these parties continue to limit leadership selection to the parliamentary party, and a mere 7 per cent to the national committee. An asterisk denotes a shift toward a more inclusive process since the 1960s. The movement toward a more inclusive selectorate in leadership selection is slower than in candidate selection. Of the seventy-four parties for which data are available, only 15 per cent have moved toward more democratic procedures in the selection of candidates since 1960. For example, the Japan's Liberal Democratic Party began with a leadership selection process controlled by the members of parliament (MPs). Even today, MP votes still weigh heavily in the process, but a party membership
72
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
Table 4.3.Degree of inclusiveness of national party leadership selectorate, 2000 Country
+
Oligarchical National commit- Parliamentary tee party
Australia (0) Austria (0)
Lab, Lib, Nat
Belgium (0) Canada (5) Denmark (0)
RV, KF, Prog
Finland (0) France (0)
FR, PCF
Germany (2) Gr Ireland (0) Italy (0) PDS, PPI Japan (1) Netherlands (0) New Zealand (0) Norway (0) Sweden (0) UK (3) US
+
FF, FG, PD VVD, CDA Lab V, KRF
Representative Party congress
Gr, SPÖ, ÖVP, FPÖ
SF, SD, CD, V, KRF SKDL, SDP, Center, KOK PS, RPR, UDF, FN FDP AN JSP PvdA SV, DNA, C, H, FRP SAP, FP, C, VP, MP
Members nominate through constituencyorganization
Basis Democracy Members can vote Dems
ECO, AGA, PSB, CVP, PSC, BSP VU, PRL All major parties*
CDU,* SPD* Labour LDP* D'66
All major parties* All major parties (Pres.)
Note: Table entries classified according to ‘most inclusive’ body involved in selection process. The abbreviations of party names are spelt out in the chapter Appendix. Number in parentheses indicates total number parties moving toward more inclusive procedures, 1960–2000. * Has moved towards more inclusive selection procedure since 1960s. Sources: Scarrow, Webb, Farrell (2000); Katz and Mair (1992); Bille (2001); McCarthy (2001); and personal communication with Michael Thies, UCLA; Ian McAllister, Australian National University; Marian Sawer, Australian National University.
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
73
ballot was adopted in 1979. In the most recent selection process, on 26 April 2001, grass-roots members were allowed to vote through an electoral college. Even this indirect involvement proved to be politically important: although the informal consensus among MPs before the membership vote was to select Hashimoto, MPs quickly switched their allegiance after Koizumi unexpectedly won among the membership (Rosenbluth and Thies 2002: 339). Canadian parties also have adopted processes that mirror American-style primaries in some aspects (LeDuc 2001). In 1985 the Parti Québecois adopted a party membership ballot to select their leader, sparking a trend at the provincial level. By 1998 the Progressive-Conservative party had followed suit by adopting a membership ballot to elect its national party leader.
Representation on Parties’ National Leadership Bodies In examining whether more points of access have opened up for participation within parties, it is important to assess not only whether parties are providing more scope for individual participation but also whether they are becoming more ‘corporatist’ in the way that rules grant privileges to some groups. Previous research indicates that the structures of representation within parties are indeed changing. As a consequence of the socio-economic transformations of the post-war era that have weakened the relationships between citizens and fixed social groupings, parties certainly operate in a different political environment. In many cases in established industrialized democracies, the relationship between parties and interest groups has declined, especially links between trade unions and socialist parties on the one hand and business groups and conservative parties on the other (Pierre and Widfeldt 1992; Selle 1997; Thomas 2001). This pattern of attenuating links is especially pronounced in northern Europe. Thomas Poguntke's (1998; 2000; 2002) research on western European parties confirms these findings. Poguntke finds that links to external ancillary organizations are waning, while links with internal interests groups are on the rise. ‘Internal interest groups’ are those created by the party with the intent of targeting new supporters. This suggests a game of musical chairs in which organizations such as trade unions lose their historical positions on the top committees, and these seats are filled by new, more fluid, often identity-based interests from within the party. These changes come as an increasing number of countries grant subsidies to political parties, reducing their reliance on donations from organized interests (Thomas 2001). The result of all this is that a corporatist organizational style is replaced by more individualist notions of representation. We are interested in determining how inclusive contemporary parties’ top party decision-making bodies are, namely, their national executive committees (NECs).4 Given that our focus is on expanding opportunities for participation, we examine whether traditionally under-represented segments have gained greater representation on these bodies. We examine two types of ‘internal interest groups’ that have received guaranteed representatives on NECs in the past—women's and youth organizations. Table 4.4 displays the national patterns in parties that guarantee delegates from these party sections on the party NECs in 1960 and 1990. Entries in the first column represent the percentage of parties with guaranteed representation for women's section delegates, presented as a ratio from 1960 and 1990. Entries in the second column represent the percentage of parties with guaranteed representation for youth section delegates, presented as a ratio 1960 to 1990. These figures show that since the 1960s parties are increasingly granting women and youth sections formal representation in the highest echelons. The proportion of
74
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
Table 4.4.Guaranteed delegates from party sections on parties’ national leadership bodies, 1960 : 1990 Country Austria Belgium Denmark Finland Germany Ireland Italy Netherlands Norway Sweden UK USA Average percentage % countries more inclusive 1960:1990 *
Women's section (%) 33 : 0 0 : 50* 11 : 0 50 : 25 0:0 0 : 20* 50 : 67* 33 : 75* 57 : 71* 100 : 100 100 : 100 0 : 50* 32 : 48 50
Youth section (%) 33 : 0 0 : 75* 33 : 37* 50 : 50 0:0 0 : 40* 50 : 67* 50 : 67* 85 : 85 100 : 100 50 : 100* 0:0 38 : 54 50
Note: Table entries represent the percentage of parties with guaranteed representation for delegates from this section, presented as a ratio 1960 : 1990. Increase in the number of parties with guaranteed representation from youth and/or women's sections. Source: Katz and Mair (1992).
parties with guaranteed spots for women's section delegates rose from roughly one-third to one-half over this time period. Similarly, the proportion guaranteeing youth section delegates rose from 38 per cent to 54 per cent. Although these data extend only to 1990, they yield some evidence about how parties are changing their fundamental organizational models. Taken together with other authors’ findings, cited above, the evidence depicts a trend toward declining links between parties and interest groups across advanced industrial democracies, together with a modest trend toward the rising power of ‘internal interest groups’. This may signal a shift from ‘corporatist’ structures of representation towards more individualist competition for voice in party decisions. If so, this lends further weight to the argument that changes signalling greater inclusiveness for citizens as individuals may nevertheless benefit party leaders because more participation may now be channelled to a greater extent through internal groups over which party leaders can exercise more control.
IMPLICATIONS FOR THE DEMOCRATIC PROCESS AND THE FUTURE VIABILITY OF PARTIES At the outset of this chapter, we alluded to the pervasiveness of the competitive elites model of democracy by referring to Schattschneider's well-known quote.
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
75
In this view, it is assumed that democracy arises from the competition between parties that aggregate issues and present voters with a package of policies. Yet recent trends call into question whether parties continue to furnish the glue that is supposed to hold together such a model. Cross-national research finds that the ideological positions of major, established parties are moving closer together (Budge, Robertson, and Hearl 1987; Klingemann, Hofferbert, and Budge 1994; Mair 1995; Knutsen 1998; Caul and Gray 2000). As parties’ policy profiles become less distinct, voters may perceive fewer clear-cut differences between their choices at the ballot box. Thus, the democratic competition once generated by elections may be weakening. As such, the processes by which party policies are reached may become all the more important in determining whether the parties that converge are able to carry their supporters along with them. The organizational changes we have assessed show that parties seem to be aware of these imperatives, and are willing to adjust their rules in order to conform to shifting political currents. Some party leaders have seemingly recognized the political salience of the nature of participation opportunities within parties, and used perceived popular pressure to justify rule changes that expand the set of intra-party decision-makers. Of course, backing for such changes frequently coincides with efforts to gain advantage in intra-party battles. Nevertheless, the undeniable result of such jockeying for electoral and intra-party support is that parties are increasingly opening up the candidate and leadership selection process to allow members a voice or to empower the local party organizations. Further, in an era of attenuating ties between parties and historical class-based interest groups, parties are expanding opportunities for participation for traditionally under-represented party activists. These formal rule changes suggest more fundamental changes in party culture as old mid-level elites and interest-group links are displaced. Maybe there is no increase in participation in parties or elections as a result of these organizational changes (cf. Whiteley and Seyd 2002), but there certainly is a change in the nature of the opportunities offered. The changes expand citizens’ access to the sites where important political decisions are made. They also expand the transparency of decision-making processes, placing important decisions such as the selection of candidates and leaders into the arena of intra-party voting. Although back-room deals for support certainly continue to play an important role in political success, the new rules nevertheless ensure that the final decision process will receive much greater scrutiny. The observed changes in the ways that parties operate do not necessarily alter the role of parties in the wider polity. Rules may change in the hope that this will help the parties build and maintain popular linkage, but there is no guarantee that the revised rules will indeed fulfil this purpose. Even so, the fact that they are adopted is a notable sign of a shift in democratic norms and strategies according to which parties’ electoral success is judged to be inseparable from the democratic quality of their rules.
76
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
APPENDIX: PARTY-LEVEL DATA, 1960–98 Country
Party
Australia
Labour Liberals National Democrats Green SPÖ (Socialist) ÖVP (Christian Democrats) FPÖ (Freedom Party) Ecolo (Francophone Green) Agalev (Flemish Green) CVP (Flemish Christian People's) PSC (Francophone Social Christian) BSP (Flemish Socialist) PSB (Francophone Socialist) VU (Flemish People's) VB (Flemish Bloc) PVV (Flemish Liberals) PRL (Francophone Liberals) Reform ND (New Democrats) Liberals Conservatives Social Credit Bloc Québecois Communist Socialist People's Left Socialist SD (Social Democratic) RV (Radical Venstre) CD (Centre Democrats) V (Liberals) KF (Conservatives) Progress KRF (Christian People's) Green
Austria
Belgium
Canada
Denmark
Finland
1960s Salience 0.3 4.4 1.4
1970s Salience 10.5 2.5 7.8
1980s Salience 0.6 0.7 0.5 7.8 4.5
1990s Salience 1.7 1.1 0.9 4 10.1 2.6
6.3
9.7
6.0
7.9
7.6
27
10.5
% change
Time span
2.3 −3.7
1961–98 1961–98 1961–98 1990–98 1986–95 1962–95
Level cand selection 2 4 4 2 2 2
Level leader NEC wom- NEC wom- NEC youth NEC youth selection an 1960s an 1990s 1960s 1990s 4 4 4 1 3 3 0 0 0 0
1.4 −3.3 −0.5
1.8
−4.2
1962–95
1
3
1
0
1
0
8.1
4.8
−22.2
1962–95
4
3
0
0
0
0
3.0
3
0.0
1981–95
2
2
0
0
0
0
3.1
3.1
0.0
1981–95
2
2
0
0
0
0
1
0
1
2.2
4.7
4.0
6.2
4.0
1961–95
3
1
0
2.2
2.1
2.5
7.3
5.1
1961–95
3
1
0
1.0
8.4
3.9
6.7
5.7
1961–95
3
1
0
1
0
1
1.0
8.4
1.2
7.7
6.7
1961–95
2
2
0
0
0
1
3.2
5.5
3.6
3.5
0.3
1961–95
4
2
0
0
0
1
2.9
2.9
3.2
0.3
1978–95
2.8
2.9
2.8
7.5
4.7
1961–95
2
0
1
0
1
2.8
5.6
3.0
3.1
0.3
1961–95
3
1
0
1
0
1
0.8
0.6
1.7
5.6 4.4
3.6
1993–97 1962–97
2 2
1 1
3.3 0.3
0.6 4.9
2.3 2.4
2.3 2.4
−1.0 2.1
1962–97 1962–97
2 2
1 1
2.8
0.9
−1.9
1962–74
0
0
0
0
3
0
0
0
0
0
0
1
1993–97
2
−5.1 1.6
1960–84 1960–98
2
−4.7
1968–84
6.2 3.7
5.3 5.0
1.1 7.3
9.2
3.5
4.5
0.9
3.0
2.8
0.3
−0.6
1960–98
2
3
0
0
1
1
1.1
3.4
1.8
3.9
2.8
1960–98
3
4
0
0
1
1
3.4
0.3
1.2
−2.2
1973–98
2
3
0
0
0
0
3.1 0.3
2.6 0.9
1.3 1.2
1.3 0.9
1960–98 1960–98
2 3
3 4
0
0
0
0
0.2 1.2
0.0 0.8
1 0
0.8 −1.2
1973–98 1971–98
3 2
4 3
0 0
0 0
0 0
0 1
17.8
15
−2.8
1983–95
2
0.0 0.3
5.3
77
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
France
Germany
Ireland
Italy
Japan
SKDL (People's Democratic) SDP (Social Democrat) Rural Centre Swedish People's KOK (Conservative) Green PCF (Communist) PS (Socialist) RPR (Gaullists) Conservatives UDF (Democratic Union) FN (National Front) Greens/Alliance 90 CDU (Christian Democrat) SPD (Social Democrat) FDP (Free Democrat) PDS (Democratic Socialism) Greens FF (Fianna Fail) FG (Fine Gael) Labour WP (Workers) PD (Progressive Democrat) Greens PDS (Communist) PSI (Socialist) PSDI (Social Democrats) PR (Radicals) PRI (Republicans) PLI (Liberals) PPI (Christian Democrats) AN (National Alliance) Christian Democratic Centre Northern League Forza Italia Communist JSP (Socialist) LDP (Liberal Democratic)
5.7
8.9
17.1
11.4
1962–87
2
3
0
0
0
0
0.0
5.2
2.7
0
0.0
1962–95
2
3
1
0
1
1
2.3 2.3 0.4
1.1 3.3 6.1
3.4 4.2 1.1
0.9 0 1
−1.4 −2.3 0.6
1966–95 1962–95 1962–95
2 2 2
3 3 3
1
1
1
1
8.5
7.5
8.5
0
−8.5
1962–95
2
3
0
0
0
0
5.8
5.3
5.5
3 4.2
−1.6
1993–97 1962–97
4.1
11.3
5.2
7
2.9
1962–97
2
1
4.3
4.9
2.9
2.5
−1.9
1962–88
4
1
4.5
4.5
5.4
0.9
1962–88
3.6
3.6
4.1
0.5
1978–97
5
1
2.6
4.4
1.8
1986–97
3
4.5
3.3
−1.2
1983–98
5
5 5
0.8
1.4
1.1
0.4
−0.4
1961–98
2
3
1
1
0
0
2.7
2.8
2.0
1.8
−0.9
1961–98
3
1
1
1
0
0
4.7
4.1
2.8
0.5
−4.2
1961–98
3
3
0
0
0
0
8.5
1990–98
0.8
0.9
3.9 0.4
5.8 2.5
1.9 1.7
1989–97 1961–97
2 3
4
0
0 0
0
0 1
2.2
6.5
1.0
5
2.8
1961–97
2
4
0
0
0
0
4.2
7.0
1.6 1.9
3.6
−0.6
1961–97 1981–89
3 4
1
0 0
1 0
0 0
1 0
0.7
4.1
3.4
1987–97
4
0
0
0
0
5
0
0 1
0
0 1
4.8
2.7
3.5 4.6
3.4 2.3
−0.1 −2.5
1987–96 1963–96
3 3
0.7
4.7
0.3
0.6
−0.1
1963–94
3
0
0
0
0
1.6
2.0
2.8
1.6
0.0
1963–92
4
1
1
1
1
7.7
14.0
18.3
10.6
1976–96
3
0.4
0.7
0.0
1
0.6
1963–92
3
0
0
0
0
0.5
1.0
0.3
5.3
4.8
1963–92
3
1
1
1
1
2.0
1.2
1.0
3.1
1.1
1963–96
3
5
0.4
0.3
1.3
2.3
1.9
1963–96
3
3
5.6
1996
3
1
1
1
1
0.3
1992–6
3
7.0 4.7
7.9 3.8
4.9 3.8
3 2.4 5
−4.6 0.3
1994–6 1960–96 1960–96
3 3 3
3
4.3
2.7
2.1
0
−4.3
1960–96
3
1
78
MIKI CAUL KITTILSON AND SUSAN E. SCARROW Clean Government New Party Netherlands Greens D'66 PvdA (Labour) VVD (Liberal) CDA CVP (Flemish Christian People's) ARP CHU New Zea- Labor land National Alliance Social Credit NZ First Norway SV (Left Socialist) DNA (Labor) V (Liberals) Centre H (Conservatives) KRF (Christian People's) FRP (Progress) Sweden Greens SAP (Social Democrats) FP (Liberals) Centre Christian Democrats VP (Communist) MP (Conservative) Switzerland Greens Social Democrats Independents Radical Democrats Christian Democrats Democrats Protestant People's People's UK Labour Liberal Democrats Liberals SDP (Social Democrat) Conservatives USA Democrats Republicans
10
4.1
2.5
3.9
−6.1
1967–93
11.6 6.4
10.0 11.0
4.9 4.2 2.9
0 6.9 5.1 4
2.0 −6.5 −2.4
1993–96 1989–98 1967–98 1963–98
2 2 2
1 3
0 1
0 1
0 0
0 0
3.3
3.5
1.3
2.8
−0.5
1963–98
5
4
0
1
0
0
3.5
2.8 9.2
3.0
2.1
−0.7 5.7
1977–98 1963–72
2
4
2.5 2.4 0.6
6.4 9.2 1.9
4.1
1.8
3.9 6.8 1.2
1963–72 1963–72 1960–96
1.3
0.7
0.7
−0.6
0
2.5
3.5
0.7 10
3.5
1960–96 1993–96 1960–90
3.1
1.9
2.0
5 4.3
1.2
1993–96 1961–97
3
3
0
0
1
1
2.7
3.0
3.4
3.2
0.5
1961–97
3
3
1
1
1
1
4.1 2.4 1.6
2.5 2.7 2.9
2.3 1.4 1.3
4.3 4.2 1
0.2 1.8 −0.6
1961–97 1961–97 1961–97
3 3 3
4 3 3
1 0 1
1 1 1
1 1 1
1 0 1
2.5
1.7
0.6
1.1
−1.4
1961–97
3
4
1
1
1
1
1.2
1.4
2.3
1.1
1973–97
3
3
0
0
0
1
1.8
5.1
0.0 1.9
5.8 2.3
5.8 0.5
1988–98 1960–98
3 2
3
1
1
1
1
4
7.5
2.2
2.8
−1.2
1960–98
3
3
1
1
1
1
1.1
5.0
0.3 4.5
3.2 2.1
2.2 −2.4
1960–98 1985–98
2 3
3
1
1
1
1
4.7
5.3
2.7
6.1
1.4
1960–98
2
3
1.1
6.2
1.5
0.3
−0.8
1960–98
3
3
1
1
1
1
5.3
8.4
10.3 6.3
9.8 10.5
−0.5 5.2
1979–95 1963–95
4 4
6.9
19.5
19.2
11.4
4.5
1963–95
4
2.8
2.6
5.6
5.8
3.0
1963–95
3
7.8
4.1
1.7
5.8
−2.0
1963–95
3
0.4 6.2
0.4 6.2
0.4 9
0.0 2.8
1971–95 1971–95
3 3
1.9 1.3
3.3 4.5
3.3 2.9
2.7 1.9 4.4
0.8 0.6
1963–95 1964–97 1992–97
3 2 2
1 1
1
1 1
0
1 1
4.4
7.4
5.5 5.5
1.1
1964–87 1983–97
0.1
1.9
1.8
0.5
0.4
1964–97
2
1
1
1
1
1
0.5 0.4
1.8 0.8
2.9 1.8
2.9 0.7
2.4 0.3
1961–97 1961–97
1 1
1 1
0 0
1 0
0 0
0 0
2
4
0
1
0 1
1
1
1 1
4
1 0
1 1
Notes: Entries represent decade average percentage election platform dedicated to ‘Favorable mentions of democracy as a method or goal in national and other organizations; involvement of all citizens in decision-making; as well as generalized support for the country's democracy’. For candidate and leadership selection rules we followed the advice of various country experts and the standards set by Scarrow, Webb, and Farrell (2000) and Katz and Mair (1992), updated with McCarthy (2001); Bille (2001), Carty and Blake (1999); personal communication with Michael Thies, UCLA and Steve Reed, Chuo University, on Japan; Ian McAllister, Australian National University on Australia. Coding for candidate selection (where selection initiated): 1 = National leadership; 2 = Regional delegates; 3 = Local party selectorate; 4 =
POLITICAL PARTIES AND THE RHETORIC AND REALITIES OF DEMOCRATIZATION
79
Members can vote; 5 = Non-members can vote. For leadership selection: 1 = National Committee; 2 = Parliamentary party; 3 = Party congress; 4 = Members can nominate through constituency org; 5 = Members can vote. Rules on composition of National Executive Committees from Katz and Mair (1992). Entries represent 1 for presence of party women's or youth organization, 0 for no such organization. Source: Comparative Manifestos Project (Budge et al. 2001).
80
MIKI CAUL KITTILSON AND SUSAN E. SCARROW
NOTES
1 2 3 4
We would like to acknowledge the help we received from David Farrell, Adam McCarthy, Thomas Poguntke, and Paul Webb in finding the data used in this chapter. We appreciate the thoughtful responses of members of the group at our various meetings, and are particularly grateful to Russell Dalton for his careful reading and helpful editorial suggestions. In an effort to discern whether minor party deviations were driving the trends in democratic themes (such as the Austrian Freedom Party), we empirically examined the impact of small parties on national patterns. These outliers had little leverage over the results. Other proposed types include the stratarchy (Eldersveld 1964), the bureaucratic-professional (Panebianco 1988), and the people's party (G. Smith 1989). In this and the following section we should keep in mind there may be a bias towards the parties with the most transparent procedures, as details of their rules are more widely available. Decision-making at the highest national level often occurs in formal committees made up of elected party officials. These processes are complicated, and often occur at different levels. Simply put, we are interested in the committee in which the most fundamental debates on the party's future occur. For the most part, smaller bodies are usually more powerful within the party than the unwieldy and large party congresses. Poguntke (1998) demonstrates that there are functionally equivalent executive committees across the parties of western Europe. Though the title of the highest-ranking body varies from party to party, we shall follow Katz and Mair (1992) and refer to this broad category as the set of ‘national executive committees’ (NECs).
5 Changing Party Access to Elections Shaun Bowler, Elisabeth Carter, and David M. Farrell Among the many objectives of political parties one of the most important is winning office, something that requires them to run the gauntlet of rules and regulations relating to the process of elections. This chapter examines the relationship between political parties and electoral laws, specifically looking at how the nature of electoral laws can affect the fate of parties in the electoral game, given that changes in the rules of the game can expand or restrict access to the electoral arena. Elections link citizens to the state, but electoral laws link them to the electoral arena and, hence, shape whether and how access occurs. We pay particular attention to electoral laws. While electoral systems—the formal processes by which votes are translated into seats—are clearly important, and are also dealt with in this chapter, the conduct of real-world elections which precedes the casting and counting of votes is shaped by access to resources such as money, television time, and the ballot itself. Access to these electoral resources is governed by law-based rules; and it is these much less well-studied and understood rules that occupy our principal attention. Other chapters in this volume examine how changes in democratic institutions, such as the frequency of elections or the potential of ‘direct democracy’, affect voters. This chapter examines how changes in democratic institutions can affect those actors at the heart of operationalizing representative democracy itself, namely, the political parties. The irony, of course, is that it is the parties which, when in power, can affect this relationship by changing the electoral laws to best suit their needs. It has been suggested that the established parties have been doing precisely this: restricting access for parties outside the established ‘cartel’ and providing themselves with additional financial and other resources (Katz and Mair 1995). We explore this possibility by examining trends in the development of electoral institutions in our sample of OECD countries over the past forty years. In the first section we examine electoral system trends. As suggested above, for our purposes the most interesting trends are those concerning electoral laws covering the period prior to polling day, and the bulk of the chapter will be devoted to these.
82
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
ELECTORAL SYSTEMS Elections are the defining characteristic of democratic practice. Not surprisingly, the main political arena for most citizens is the electoral arena, and within that arena political parties have historically been the major source of linkage between citizens and the state (Lawson 1980). Even though very few constitutions grant a formal role to political parties, parties remain fundamental building blocks around which electoral politics is built: so much so that some scholars stress that parties are among the central practical requirements of democracy. As Schattschneider (1942: 1) famously noted, ‘modern democracy is unthinkable save in terms of political parties’. Among other functions, parties help mobilize voters into taking part in the political process, recruit people into the political system, and articulate and aggregate the interests of voters. Recent scholarship notes that the centrality of parties to the electoral process may well be changing (Dalton and Wattenberg 2000), but for the most part citizen access to politics consists of working through parties and the system of electoral competition. As is well known, however, the laws governing the electoral process are not neutral (Grofman and Lijphart 1986). The outcomes of elections depend crucially on the ways in which the electoral system adds up the votes, as much political science research has demonstrated in a variety of ways (for an overview, see Farrell 2001). For the most part, the rules by which votes are counted and voter preferences translated into outcomes are well understood. The study of electoral systems represents one of the better-developed branches of political science and, by now, we can discuss with confidence at least some of the main effects of electoral systems. Electoral systems determine who the ‘winners’ and ‘losers’ are. Given that the political elite is likely to be keenly aware of these distributional consequences, any change in the status quo is likely to generate very hard bargaining. Parties know whether a given electoral system will favour small or large parties, and so it will be hard to reach agreement between parties since the divisions are likely to be hard to negotiate over or bargain away. The possibility of change is presumably even more remote under proportional systems where the large number of parties increases the number of actors required to agree to any change. Electoral systems are also sometimes entrenched in constitutions (as, for example, in Ireland) and therefore may be harder to change than simple statutes. In a sense, too, electoral systems should not change that much since the legitimacy of elections as a whole may depend on maintaining a degree of continuity in the rules. A major feature of elections—the main arena for participation in democratic politics—is likely to remain quite static over very long periods of time. This fact could lead to a particular kind of analysis of electoral rules: one in which nothing much happens. Many games, such as chess, baseball, or cricket, share this property: long periods where nothing actually happens but the players (and not a few fans: for example, Geras 2002) are feverishly thinking about it. That may be fine for cricket, but it tells us little in this context about how responsive
83
CHANGING PARTY ACCESS TO ELECTIONS
Table 5.1.Indicators of electoral system change Country
Date
Australiag
1960s 2000 1960s
Austria
2000
Belgiumg
1960s 2000
Britain Canada Denmark
1960s 2000 1960s 2000 1960s 2000
Finland France Germany
1960s 2000 1960s 2000 1960s
2000
Ireland Italy
1960s 2000 1960s
2000
m
Tiera
Formulab
District Number of Assembly magnitudec districtsd sized
H L N
AV AV d'Hondt LR-Droop d'Hondt
1 1 41.25 6.60 183
123 150 4 25 1
H
d'Hondt LR-Hare d'Hondt LR-Hare d'Hondt LR-Hare Plurality Plurality Plurality Plurality LR-Hare MSL LR-Hare MSL d'Hondt d'Hondt Maj-Plur Maj-Plur d'Hondt
20.33 4.26 23.48 7.04 15.00 7.50 1 1 1 1 175 7.30 175 7.32h 13.21 13.33 1 1 496.88
9 43 9 30 10 20 631.69 655 268 301 1 18.50 1 19 15.15 15 470.14 577 1
200 200 470.14 577 496.88k
L H
Plurality LR-Hare
1 656
247.75 1
656k
L H
Plurality STV STV LR-Hare
1 3.75 4.05 625.75
328 40.5 41 1
5% (N) or 5 3 const seats
152 166 625.75
L H
LR-Imp LR-Hare
19.55 155
32 1
— 17.2 — 14.9 1 const seat 2.0† and 300,000 votes
630
4% (N) for 4 higher tier alloc only
H L H L H L
H L H L
123 150 165
Legal thresholde
Effective threshold (%)f — 37.5* — 37.5* 1 const seat 8.5†
183
1 const seat 4 or 4% (N)
211.33
0.66 of a Hare quota 0.33 of a Hare quota — — — — 2% (N)
150 632.85 655 268 301 175 175
Special rulesj — — 12%* (D) — 5% (N) or 3 const seats
4.8† 2.8 37.5* 37.5* 37.5* 37.5* 2 2 5.4 5.2 37.5* 37.5* 5
84
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
Japan
1960s 2000p
Netherlands
1960s
2000 New Zea- 1960s land 2000r
Norway
1960s 2000
Sweden
1960s 2000
Switzerland 1960s 2000 United 1960s States 2000 † * a
b
c d e f
g h j
k m n p q r
L
Plurality SNTV d'Hondt Plurality d'Hondt
1 3.95 16.4 1 150
475 123 11 300 1
150
37.5* 6.3(D) 16.4 — 4.3 — 37.5* 0.67% (N) 0.67
d'Hondt Plurality
150 1
1 85
150 85
0.67% (N) 0.67 — 37.5*
H
SL
120
1
120t
L
Plurality MSL MSL MSL MSL MSL MSL
1 7.80 165 8.26v 8.27 349 11.07w
67s 19.56 1 19 28 1 28
5% (N) or 5 1 const. seat
152.44 165
— 4% (N)
8.9 4
231.67 349
8.4 4
d'Hondt d'Hondt Pluralityy
8.20 7.69 1
23.91 26 435
195.55 200 435
— 4% (N)x 4% (N) or 12% in 1 const — — —
8.5 8.6 37.5*
Pluralityz
1
435
435
—
37.5*
H L
H L H L
486n 480q
See Lijphart, (1994: 38–9). Approximation (see Lijphart, 1994: 182–3; Taagepera, 1998). Abbreviation for tier: N: national level; H: higher level; L: lower level; Abbreviations for formulae: AV: Alternative Vote; LR-Droop: largest remainder Droop formula; LR-Hare: largest remainder Hare formula; MSL: Modified SainteLagüe formula; Maj-Plur: Majority Plurality 2-ballot formula; STV: Single Transferable Vote formula; LR-Imp: largest remainder Imperiali formula; SNTV: Single Non-Transferable Vote formula; SL: Sainte-Lagüe formula. Average district magnitude. Calculated by dividing the number of seats in the legislature by the number of districts. All figures with decimal points are averages. N denotes a legal threshold imposed at the national level, D refers to a legal threshold imposed at the district level. The effective threshold is either determined by the legal threshold or is inferred from the district magnitude, whichever value is higher. If inferred from the district magnitude, the effective threshold is calculated by taking the mean of the upper threshold and the lower threshold: 75 / (M+1). See Taagepera, (1998); Lijphart (1999). Compulsory voting. In Denmark only 139 seats are distributed among the constituencies. To receive seats at the higher level Danish parties must either (1) have won at least one constituency seat, (2) have obtained at least as many votes as on average were cast per constituency in at least two of the three regions, or (3) have obtained at least 2% of all valid votes in the country as a whole. Not including surplus seats (Überhangmandate). From 1994. After 1964 re-apportionment. From 1996. In February 2000 a law was passed reducing the House of Representatives from 500 to 480. System introduced 1993, first used 1996.
CHANGING PARTY ACCESS TO ELECTIONS s
t v w x y z
85
This applies to the current House elected in the 1999 election. Minor boundary changes were made in 2001 following a national census: there are now 51 list seats (rather than 53), 62 general seats (as opposed to 61) and 7 Maori seats (rather than 6). Not including ‘top up seats’. In Norway only 157 seats are distributed among the constituencies. In Sweden only 310 seats are distributed among the constituencies. Swedish parties that have obtained seats at the lower level through the 12% rule only are excluded from representation at the higher tier. The State of Louisiana uses the two-ballot majority system. As of 1962 (the House elected in 1960 had members elected from 437 districts). Sources: Carter (2002); Gallagher, Laver, and Mair (1995); Inter-Parliamentary Union Parline Database 〈http://www.ipu.org/parline-e/ parlinesearch.asp>; LeDuc, and Norris (1996); Lijphart (1994; 1997a); Lijphart Election Archive 〈http://www.dodgson.ucsd.edu/lij/〉; Penadés (1997); Taagepera (1998).
86
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
electoral systems have been to the pressures for change. And pressures there have been, as demonstrated by a series of academic studies on electoral change (for example, Dalton and Wattenberg 2000). Of late, such prominent examples as New Zealand's move to mixed-member proportional from single member plurality and the adoption of mixed-member systems in Italy and Japan have generated considerable academic interest in documenting those changes (for example, Shugart and Wattenberg 2001). Occasionally, too, French electoral practice may change (it did most recently in the mid-1980s). All of these changes were dramatic affairs, much talked about at the time and since. In part, though, the attention given to them indicates just how rare and important such events are. By and large, fundamental reform of an electoral system remains as rare today as it has tended to be in the past (Nohlen 1984). Table 5.1 provides details for the countries examined in this volume, comparing the situation in the 1960s with that in 2000. Apart from the three major cases of electoral reform, the overall message from this table is one of stasis. The most that can be said is that some countries have tinkered with some of the details (for example, Austria's adoption of a third electoral tier; the change of electoral formula for the list element in Germany's election), and assembly sizes have slightly increased (in large part reflecting population growths). The overall effect of such changes—which can quickly be ascertained by looking at the effective threshold trends in Table 5.1—has been quite small. The message is clear and unsurprising. Apart from the three cases of electoral reform in Japan, Italy, and New Zealand, for the most part the electoral systems of the OECD countries have changed little over the past 40 years.
ELECTORAL LAWS Electoral systems are only one part of the edifice of electoral laws and electoral competition. Whether new participants—and for all practical purposes this means political parties—can compete successfully depends not just on the counting rules but also on a range of other rules, including those relating to ballot access, campaign finance, and access to media outlets. This set of electoral rules can be grouped under the rubric of a nation's electoral laws (in which electoral systems are a distinct part but one that relates specifically to polling day and thereafter). It can be expected that electoral laws will shape election outcomes just as much as electoral systems, and in the same manner if not the same degree: that is, electoral laws relating to how to get on to the ballot or access to resources for mounting a campaign can hobble both new challengers and/or small parties as much as, if not more than, the choice between the d'Hondt and the St Lagüe electoral systems. Access to the electoral arena, then, may be either denied altogether by, for example, circumscribing the ability of minor parties to get on the ballot, as in the US, or, if parties and candidates do gain a place on the ballot, they may be neutralized by
CHANGING PARTY ACCESS TO ELECTIONS
87
denying them access to television time, which constrains their campaigning efforts. Such electoral laws not only have consequences; they may also be easier to change than electoral systems, especially those that are entrenched by constitutional super-majorities. Furthermore, electoral laws (such as those on ballot access) may be less visible and much more complex than the counting rules, and so less susceptible to public charges of ‘rigging the rules’. And their distributional consequences may be more muted, so that they do not arouse too much opposition. The current set of parties may not be able to agree on new electoral systems but they are likely to find common ground in giving themselves large public subsidies or cheaper access to television, or in making it harder for new parties to enter the election race. As argued by Katz and Mair (1995), parties are likely to be happy to act as ‘cartels’ in the electoral market place, skewing rules and laws in their own favour. Electoral laws, then, are likely to be areas in which all current parties can agree to heighten the ramparts against new challengers and new parties. Even if the demands of reformers are quite muted, not only may we expect to see electoral laws change, we may expect to see them change in ways that privilege current parties by raising barriers to entry. The question is: do they? But at which electoral laws should we look? There has been relatively little systematic treatment of this issue within the electoral studies literature to date (though work is in progress, see Elkit and Reynolds 2000). Absent any established set of standard issues raised by electoral laws, we propose to look at three examples that, we believe, relate to critical aspects of elections and campaigning: ballot access, access to media, and campaign finance laws (relating to access to state subsidies). All three of these examples affect the ability of parties to run an effective campaign in immediate and obvious ways. Ballot access affects the ability to enter a race. Perhaps the clearest example of this is seen in the US, where ballot access laws vary, not just by office, but by state as well (Katz and Kolodny 1992). Such laws are often surprisingly difficult to discover, and thresholds for qualification—in terms of signatures of voters subject to geographical distributions, for example—are usually high. Little wonder, then, that third parties and third-party candidates in the US often have enormous difficulties even appearing on the ballot. America's two-party system may simply reflect the workings of Duverger's tendency (Cox 1997) and the presidential system, both of which put a premium on building large coalitions. However, the two main parties themselves seem at least anecdotally to be skilled at using ballot laws to keep out challengers. Electoral thresholds operate similarly to either help or hinder smaller parties in straightforward ways. Rules governing access to electronic media also may have consequences for the conduct and outcome of elections. While there are very many critics of campaigns on and by television as well as a surprising fondness for so called ‘retail politics’, the reality of the modern world means that television campaigns are unavoidable, even desirable (Farrell 1996; 2002; Norris 1999a). Electronic media make reaching
88
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
out to mass electorates easier and more manageable. Furthermore, given how much recreational time electorates spend watching television, using it as a major means of communications during campaigns does have its advantages: how else to tell so many voters what is going on? Again there are extreme examples. One hallmark of Russia's presidential campaigns, for example, has been government dominance of the electronic media. To the extent that television images, advertisements, and ‘spin’ matter (as sections of the voting studies literature suggest they might; for a recent review, see Farrell and Schmitt-Beck 2002), then having access to electronic media should also make a difference. Finally, we consider the role of state subsidies to political parties. No election campaign, even for president of a university graduate student association, is costless. Campaign spending helps in the fighting, and possibly the winning, of elections. True, campaign spending is often a proxy for something else (such as campaign effort or name recognition) since, other than in cases of outright bribery, the act of spending money in and of itself would seem to do little good at election time. Thus, having some subsidy or subvention helps those receiving it at election time, at least if they know how to spend it sensibly (whatever that may mean in terms of micro-level voting decisions). Again, there are extreme examples. Mexico's PRI, for example, could use their control of state largesse to help their election campaigns, not so much by pursuing pork barrel politics as by having a number of campaign workers more or less paid for by the state (rather like the surprising number of supposedly amateur Olympic athletes who are policemen or soldiers). Less dramatic is the case of those parties and candidates that do not receive money from taxpayers while others in the same political system do. Again, minor parties in US presidential elections provide an example of how this may work. Crudely speaking, minor parties have to gain 5 per cent of the vote before receiving public money. Since the ability to gain 5 per cent of the vote depends on running expensive media campaigns, the offer of public money essentially comes too late (assuming, of course, the minor party can get on the ballot in the first place). What is not surprising is that the Democratic and Republican parties receive millions of dollars in public money. The US, of course, is not unusual. Australian parties receive state funding subject to the similar requirement that they are already successful. To some extent this ‘the rich get richer’ aspect to public subsidy is unavoidable, for how else could states give public money? Public funds could hardly be given to candidates because they promised to do well at election time. Even so, access to state subsidy, and rules governing the same, affect the flow and amount of resources to political parties.
COMPETING TRENDS IN PARTY ACCESS So far we have identified three main areas of electoral law that shape party access to politics. What kinds of trends can we expect to see in these laws over the past generation? Two broad patterns seem possible a priori.
CHANGING PARTY ACCESS TO ELECTIONS
89
First, we may see a trend towards greater access to the political system. It is reasonable to suppose that voters and citizen groups who are frustrated by a current set of institutions might want to change them in order to open up the electoral process. The political system could thus be opened up to allow more participation by a wider range of candidates and parties. If there is a trend towards greater participation we could see ever-easier ballot access, wider access to public money (by smaller and newer parties) and also easier access to television for more parties and candidates. All of those trends would show an opening up of the political system. Conversely, we might see a trend towards a more closed electoral arena in line with the ‘cartel party’ argument. Opening up the political system would make it easier for competitors of existing parties to enter the contest. But established political parties are not likely to accede to such demands; in fact, quite the reverse. Parties are not just vehicles for voter preferences, they also have interests of their own. Almost by definition, the current parties are the successful ones under the existing rules, especially if they are among the larger parties. Hence, currently successful parties are not likely to be very receptive to proposals to open up the system to wider participation. To allow more participation essentially asks the current parties to hurt themselves in the name of an abstract principle. Since the kinds of rules we identify are away from the public eye, we might expect existing parties to use the rules to give themselves advantages: by restricting ballot access for newer and small parties and candidates, by making it harder for newer parties to appear on television, and perhaps by giving public money to the established parties rather than new ones. Given the clearer motivations of parties to act to protect themselves, it seems reasonable to expect to find more evidence of a trend towards greater restrictions on access to resources across the OECD than of a trend towards opening up electoral systems. Of course, these two trends are relatively extreme. In some ways they mark the end points of a continuum of electoral law change: increasing access on one end point, decreasing access on the other. We should, then, mention a third option, the ‘muddy’ middle position. It may be the case either that these electoral laws are just as resistant to change as electoral systems, or that different countries move in different ways and that there is no secular trend across the countries we consider here.
CHANGES IN THE ELECTORAL ENVIRONMENT OF THE OECD COUNTRIES, 1960s–2000s Appendix Tables 5.A1–A3 list in detail the changes that have taken place in the electoral institutional environment within which political parties compete across the OECD over the past forty years. Table 5.2 provides qualitative summaries of the main changes. These tables show that in a number of respects the laws concerning ballot access, media access, and state subventions have changed since
90
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
Table 5.2.Change in the electoral environments of OECD countries, 1960s–2000s: summary measures Australia Austria Belgium Britain Canada Denmark Finland France Germany Ireland Italy Japan Netherlands New Zealand Norway Sweden Switzerland United States
1. Ballot access Easier (−1) Harder (+1) No change (0) Easier (−1) Slightly easier (−0.5) No change (0) Slightly harder (+0.5) Easier (−1) No change (0) No change (0) Harder (+1) Insufficient Info Slightly harder (+0.5) Slightly harder (+0.5) No change (0) No change (0) Slightly harder (+0.5) Varies by state
2. Media access Easier (−1) No change (0) No change (0) No change (0) Slightly easier (−0.5) No change (0) No change (0) Slightly harder (+0.5) No change (0) Slightly easier (−0.5) Easier (−1) Insufficient Info Slightly harder (+0.5) Easier (−1) No change (0) No change (0) No change (0) No change (0)
3. State subsidy Easier (−2) Easier (−2) Easier (−2) No change (0) Easier (−2) Easier (−2) No change (0) Easier (−2) Easier (−2) Easier (−2) Easier (−2) Easier (−2) No change (0) No change (0) Easier (−2) Slightly easier (−1) Slightly easier (−1) No change (0)
Summary 1–3 −4 −1 −2 −1 −3 −2 +0.5 −2.5 −2 −2.5 −2 +1 −0.5 −2 −1 −0.5
Notes: This table summarizes trends in the 2000s versus 1960s, based on material provided in Tables 5.A1–A3. Positive scores indicate that the electoral environment is becoming more restrictive. Negative scores indicate that it is becoming more liberalized. Codes are weighted, with state subventions considered twice as important as ballot access or media access (see fn. 2).
the 1960s. Clearly, then, electoral laws are much more malleable than electoral systems. In laws that govern access to the ballot (Table 5.A1), change has occurred in just over half the countries under observation in this study. In the majority of these cases, the requirements have become stricter, meaning that it is more difficult today for new parties to gain access to the ballot than it was in the 1960s. In a number of countries the deposit that parties or candidates must pay to compete in the ballot has been increased. In the Netherlands, for instance, parties must now pay a deposit of Dfl. 25,000 (approx. $10,400) to participate in the ballot.1 Although this deposit allows parties to compete in all electoral sub-districts, it is still substantially more than the Dfl. 1,000 deposit required per sub-district in the 1960s ($280 in 1965, worth around $1,600 today). Elsewhere, parties or candidate must be supported by more signatures today than in the 1960s. In Austria, parties now require up to 500 petitioners as compared with 200 in the 1960s; in Finland candidates must now muster 100 signatures to form an electoral association as compared with 30 in the 1960s; in Norway 5,000 petitions are now needed to register a party, whilst in the 1960s only 3,000 were required; and in Switzerland
CHANGING PARTY ACCESS TO ELECTIONS
91
electoral lists must now be supported by between 100 and 400 signatures, as opposed to 50 in the 1960s. Contrary to this general trend, in a limited number of countries requirements for ballot access have been relaxed since the 1960s. In Britain, for example, a candidate was required to pay a deposit of £150 in the 1960s (equivalent to about £2,000 or $3,000 in today's prices), but since 1985 the deposit has been set at £500. In addition to the deposit being lower in terms of current prices, it is also easier for a candidate to be reimbursed her deposit today than in the 1960s. Since 1985 only 5 per cent of the constituency vote is needed for a deposit to be refunded, as compared with 12.5 per cent of the vote in the 1960s. The situation for new parties trying to access the ballot in Britain is thus more favourable today than it was in the 1960s. Accessing the ballot has also become easier in France, for similar reasons. The deposit a candidate must pay to stand for election is 1,000 Fr., the same amount as in the 1960s, during which time the purchasing power of the franc has fallen by roughly eight times. There has, on the whole, been less change to the laws regarding media access than to those governing ballot access (see Table 5.A2). Indeed, requirements for media access have remained the same in over half the countries under investigation in this study. In those cases where change to these laws has occurred, however, the general trend has been towards a relaxation of the requirements, making it easier for new parties to receive air time. In Australia, for example, minor or new parties are now allocated free electronic media time during election campaigns. In the 1960s not only were there no provisions for election party broadcasts, but other broadcast coverage of the election tended to be restricted to those parties that were already represented in parliament. Similarly, in Canada, although guidelines did exist for the allocation of media time to new parties, it was left up to the Canadian Broadcasting Corporation to decide whether, and in what way, air time should be allotted to new competitors. In the event, no new party was awarded any free air time in the 1960s. Today, by contrast, as long as they are registered, new parties receive a limited amount of free air time, and may also purchase additional slots. It has also become easier for new parties to gain access to the media in Ireland, where parliamentary representation is no longer a prerequisite for being granted air time, and in New Zealand, where any party that is registered or that presents a minimum number of candidates may receive some media coverage. In contrast to these countries, in France it has become slightly harder for new parties to gain access to the media. Air time during parliamentary campaign periods is no longer distributed equally among the parties represented in the legislature, as it was in the 1960s, but is allocated according to the share of the seats parties have in the Assemblée Nationale. New parties that have gained representation in parliament thus receive considerably less air time than they did in the 1960s. For those parties not represented in the parliament, a minimum amount of air time is still available on public channels, while paid political advertising has
92
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
been forbidden in France since the early 1990s. It has also become more difficult for new parties in the Netherlands to obtain media access. In the 1960s participation in the ballot in one electoral sub-district was sufficient to gain access to free broadcasting time; today parties must compete in all electoral sub-districts to gain media access. If the picture has been mixed with regard to media access, with some the rules being relaxed in some countries tightened in others, the picture is much clearer when we turn to the rules concerning state subventions (see Table 5.A3). These laws have been changed in over two-thirds of the countries under observation here and, in all cases, the requirements have been loosened. In Austria, for example, parties may now receive state funds in election years if they poll just 1 per cent of the vote. They are also entitled to public funds if they have representation in the Nationalrat. This contrasts with the 1960s, when only parties with parliamentary groups (a minimum of five MPs) could gain access to state subsidies. In Denmark, parties now benefit from a per-vote reimbursement as long as they have polled a minimum of 1,000 votes, whereas in the 1960s the only state subventions available were awarded to the parliamentary groups. Similarly, in Norway and in Sweden parties are now eligible for subsidies as long as they poll over 2.5 per cent of the vote; in the 1960s only parliamentary parties received subsidies. In other countries the situation is more favourable today for new parties simply because state subsidies have been introduced. In Australia, for example, no provisions existed in the 1960s for parties or candidates to receive public money, but today parties and candidates may benefit from having a share of their campaign costs reimbursed. Similarly, in Belgium parties now receive grants from the state to support both their central organization and their parliamentary group, whereas in the 1960s no such funds existed. On the whole, then, the above discussion suggests that electoral laws have become less restrictive over the past forty years. Even in those cases where individual regulations have become tighter (like those rules regarding ballot access as described above), the restrictions have tended to be outweighed by the loosening of other laws. In particular, the loosening of the regulations regarding media access and state subventions by far outweighs the greater restrictions on ballot access. The costs new parties incur from having to gather more signatures to participate in the ballot in Austria and Norway, for example, are easily offset by the benefits these parties derive from gaining access to state subsidies. Thus, one conclusion that can be reached is that, on balance, new parties face a less restrictive environment today than they did in the 1960s. The situation might not be so simple, however. A closer examination of the changes in the regulations governing state subsidies reveals that, although new parties in most of the countries under observation have gained access to state funds in the 1990s, in many countries subventions to large, established parties have, at the same time, been significantly increased since the 1960s. Therefore,
CHANGING PARTY ACCESS TO ELECTIONS
93
even though new parties may benefit from some state money, the increased financial resources of the larger parties continues to hinder the competitive potential of new parties. While legal changes in Austria, for example, have enabled parties to receive some state funds with as little as 1 per cent of the vote, other legislation has brought in huge bonuses for parties represented in the Nationalrat. Likewise, in France, although any party that presents candidates in seventyfive constituencies may draw on state money, those parties already represented in parliament are able to access funds reserved for them. The large, established parties also benefit disproportionately in Norway and in Sweden. Here the practice has been to increase subsidies to parliamentary parties by considerable amounts. Thus, although new parties may still access some state money, those parties already in parliament can draw on even more.
ASSESSING CHANGE IN ELECTORAL INSTITUTIONS OVER TIME It is much easier to document such developments than to establish their consequences, for several reasons. The developments outlined in the appendix tables show changes on a given measure of election law restrictiveness, but do not provide an index that allows us to rank-order countries. While we can see the broad direction of change in, say, Belgium and Switzerland, we cannot see whether Belgium has a more restrictive set of rules than Switzerland or vice versa. Second, if restrictions do have consequences, one of them is inherently unmeasurable, namely, the number of parties which did not come into existence. Hence we are not likely to find the same kinds of robust statistical relationships that characterize much of the voting systems literature. In the final column of Table 5.2 we provide some judgmental codings for each of the entries in Tables 5.A1–3. Positive scores indicate that the electoral institutions are becoming more restrictive, so that small parties are finding it harder to gain access to the ballot, to media, and to state subventions. Negative scores indicate that the electoral institutions are becoming more liberalized, and that small parties are finding it easier to gain access to the ballot, to the media, and to state subventions. The codes are weighted with state subventions considered twice as important as ballot access or media access.2 Whatever the obvious likely criticisms of the nature of our coding and the assumptions underlying it, Table 5.2 does, nonetheless, show an interesting trend towards easier access across the board, that is, across nations and across most of the measures we examine. For fourteen of the sixteen countries for which we have sufficient data to report overall summaries, the net overall trend is towards easier access. For two of the three measures for which we have evidence of change, the shifts have clearly been towards easier access. Within those general trends there are some specific patterns and exceptions that we can
94
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
note the following: • • •
The exceptional countries are Italy, the Netherlands, and Switzerland. What this suggests is that those nations that began this period with many parties saw a tightening up of access to politics. The topic area of exception is ballot access, where, for ten countries in which we saw some change, six made access harder while four made it easier. (For media access the respective scores were two/five, and for state subsidies zero/ thirteen.) The area that saw most change was state subsidies, which saw a massive expansion across all nations. The opening of access to campaign resources appears to disproportionately benefit those parties already within the political system.
Table 5.3.Assessing the effects of electoral law change on measures of party system change, 1960s–1990s
Change in ballot ac- Correlation cessd Sig. (1-tailed) (N) Change in media ac- Correlation cess Sig. (1-tailed) (N) Change in state sub- Correlation sidies Sig. (1-tailed) (N) Summary measure of Correlation change Sig. (1-tailed) (N) Change in dispropor- Correlation tionalitye Sig. (1-tailed) (N) a b
e
Change in the effective number of electoral partiesb .097
Change in the effective number of parliamentary partiesc .384
.115 (16) −.570**
.360 (16) −.006
.071 (16) −.142
.008 (17) −.300
.491 (17) −.415*
.294 (17) −.388
.113 (18) −.298
.044 (18) −.252
.056 (18) −.173
.131 (16) .214
.174 (16) .085
.261 (16) −.136
.197 (18)
.361 (20)
.284 (20)
Note: correlations are pearson r; * significant at the 0.05 level (1-tailed); Based on Strom (2000).
c d
Change in the proportion of independentsa .318
See Table 5.2 relating to items 1–3.
**
significant at the 0.01 level (1-tailed).
CHANGING PARTY ACCESS TO ELECTIONS
95
In an effort to provide a more systematic treatment of these trends, Table 5.3 shows the correlations between changes in the electoral environment and changes in party systems across the period of study. Specifically, we examine the relationship between changes in electoral systems (using the Gallagher disproportionality index as a surrogate measure) and electoral laws and changes in three party-related variables: the effective number of electoral parties, the effective number of parliamentary parties, and the number of independent legislators. We expect to see a generally negative relationship: that is, as laws loosen (and our measures grow more negative) we would expect to see this correlate with larger numbers of parties and greater numbers of independents in the legislature. For the most part, the evidence presented in Table 5.3 produces mixed results. Very few of the trends are statistically significant, and they are in varying directions, indicating ‘muddy trends’. That said, there are two statistically significant cases that indicate a negative relationship: changes in media access over the period correlate with changes in the proportion of independents, and changes in state subsidies are associated with changes in the effective number of electoral parties. Of course, these data do not allow us to establish any causal connection: we cannot tell whether the opening up the electoral arena caused, or was brought about by having, greater numbers of parties or more independent legislators. What we can say, however, is that the electoral laws that govern access to the main political arena—the electoral one—have undergone considerable changes over the past generation in terms of shaping access and that these secular trends seem at odds with stricter interpretations of the cartel party thesis.
CONCLUSION Two broad conclusions can be drawn from this discussion. First, we do see evidence of changes in the electoral laws we have identified. There is considerably more development and change in these laws than in the counting rules (electoral systems). As can be seen from Tables 5.A1–A3, the environment in which political parties compete has changed since the 1960s, as shown by trends in ballot access, media access, and state subventions. Second, the general evolution of these rules appears to have been towards a more liberal environment for all political parties, which on the face of it suggests an easier ride for small and/or new parties wishing to break into the system. But, as we have suggested, these developments can have another interpretation, which is that, if anything, they suggest an even easier ride for the established parties, flushing them with ever more resources and benefiting them disproportionately more than the smaller parties. This conclusion is suggestive of a cartel-type situation with a twist: nest-feathering and liberalizing of electoral laws to the benefit of all parties, but disproportionality more so for the established parties than for the new and/or small parties.
96
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
What, then, to make of these overall trends? For most of the countries under consideration we might summarize the trends as showing greater access in form but not in function. The parties have eased restrictions and allowed greater resources to flow to newer and smaller parties. The kinds of changes noted in Chapter 3 show various non-party reforms (direct election of mayors, direct democracy) that open up the electoral arena for different kinds of participation. Yet parties still remain central figures in the process and are not likely to introduce changes that mean they commit suicide. Nevertheless, the overall pattern of the reforms shows that to the extent that parties have introduced changes they have done so in a way that can be interpreted as allowing greater access. By and large, however, these reforms have done very little to increase the range and number of new parties in the system. A cynical view, then, would see the parties as having the best of both worlds: being able to claim that they have introduced changes in the name of greater democracy in the electoral arena without actually suffering increased competition or damaging their position in any real way. Does all this mean, then, that the kinds of laws we discuss above are essentially innocuous and have little impact on electoral competition? Here we can make the distinction between long- and short-term effects. The short-term effects of, say, changes to the laws on ballot access are likely to shape the competition between parties, but perhaps do not have longer-term impacts on the overall shape of party systems. But one interpretation of the exceptions that we find is that at least some party politicians thought these laws worth changing, as shown, for instance, by the Netherlands, with its highly fractionalized party systems in the 1960s, which made moves to tighten access to politics rather than loosen it, suggesting that at least in some places the parties themselves thought these reforms and changes would have consequences.
CHANGING PARTY ACCESS TO ELECTIONS
97
APPENDIX Table 5.A1.Requirements for ballot access Country Australia
Austria
Belgium Britain
Canada
Denmark
1960s Nominations by 6 electors per district. Deposit of A$100 (US$100 or £50 in 1966, worth US$550 today) reimbursed only if candidate obtained at least 20% of first preference votes Until 1970, 200 petitions per district.
2000a Nomination by 6 electors per district or by registered party.b Deposit of A$350 (US$180 or £125), reimbursed only if candidate obtains 4 % of first preference votes.
Candidates nominated by parties. 200–500 petitions per district required (according to population), or support of 3 outgoing MPs. Deposit of ATS 6,000 (US$430, £290) paid by each party in each constituency—not reimbursed. Support of 200–500 voters required As in 1960s, but since 1976 candiper district (depending on popula- dates may alternatively be supported by 3 outgoing MPs. tion). 10 signatures and £150 deposit per Since 1985, 10 signatures and £500 candidate (worth US$2,930 £2000, (US$735) deposit per candidate. today) refunded if candidate wins 12.5% of vote. Refunded if candidate secures 5% of vote. Nomination by at least 25 electors in Nomination by at least 100 electors, district. (50 in sparsely populated districts). Deposit of CAN$1,000 (US$640, Deposit of CAN$200 (worth US$940 £700 today) required, re- £435). Until 2000, half the deposit turned if candidate is elected or if was refunded when the candidate he/she receives at least half as many filed the auditor's report, and half was refunded if the candidate was votes as the winning candidate. elected or obtained at least 15% of the votes in constituency. Since 2000 deposit is refunded in full if reporting requirements are met regardless of number of votes won.c 1960–5: parliamentary parties do not As late 1960s. Independents need need petitions. 150–200 petitions from their district. Other parties need 10,000 signatures. From 1965: parties not in parliament need a number of signatures equal to 1/175 of total valid votes in last general election (approx. 20,000).
98
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
Finland
France
Germany
Ireland
Italy
Japan
Until 1969: candidates required backing of an electoral association in their constituency (30 voters needed to form an association). From 1969: candidates may alternatively be nominated by a registered political party (5,000 adherents eligible to vote required for registration). Parties not succeeding in getting any candidates elected in the two latest elections are struck off. Only registered parties may present candidates. Deposit of 1,000 FF (US$200 in 1965, worth US$1,100, £700 today) per candidate, returned if the candidate obtains 5% of the votes. Parties with 5 members in Bundestag or any Landtag are automatically eligible. Others must prove democratic credentials. Land lists must be signed by 1/1000 voters in respective Land to a maximum of 2,000. Individual candidatures for constituencies, or candidatures for constituencies submitted by party without 5 representatives require support of at least 200 electors in constituency. Candidates may nominate themselves or are nominated by a registered elector of the constituency. From 1963, parties must be registered. Only registered parties can have their names on the ballot paper.
As 1960s except since 1975, 100 voters necessary to form an electoral association (30 in Aland).
As 1960s. (Today 1,000 FF equals US$140, £95.) As 1960s.
The nomination process for candidates as 1960s.
Deposit now IR£300 (US$350, £240), refunded if candidate is elected or secures at least a quarter of the applicable electoral quota. Support of at least 500 electors (for Party symbol and name must be registered. At constituency level, lists candidates in single-member conof candidates require support of at stituencies) or backing by least 300 electors. Parties with one 1,500–4,000 signatures for party lists. seat in either house do not need signatures. Deposit required Deposit of ¥3 million (US$22,400, £16,090) required, reimbursed if candidate obtains at least one fifth of electoral quotient.
CHANGING PARTY ACCESS TO ELECTIONS
Netherlands
New Zealand
Norway
99
Since 1989 only 10 signatures re25 signatures and deposit of Dfl. quired to present a list in a sub1,000 (US$280 in 1965, worth US$1,600 £1,050 today) required to district. present a list in a sub-district, reimbursed if party wins three-quarters of electoral quotient. Parties already represented in parliament are exempt from this deposit. Deposit of Dfl. 25,000 (US$10,400, £6,800) required to present a list in one or more sub-districts, reimbursed if party wins three-quarters of electoral quotient. Parties already represented in parliament are exempt from both requirements. Parties must be fully recognized legal entities. Candidates require nomination by at Constituency candidates must be least two registered electors in the nominated by either at least two registered electors in district or as district. part of a bulk nomination submitted by a registered party. Deposit of NZ$300 (US$125, £90) Deposit of £10 required (worth, required per candidate, reimbursed if US$190 £130 today), forfeited if candidate received less than quarter candidate wins 5% of votes in the of the total number of votes received district and completes return of by the successful candidate in that election expenses. constituency. Independent candidates and members of unregistered parties may also contest constituency seats. Parties presenting lists must be registered. Deposit of NZ$1,000 (US$440, £300) required, reimbursed if party wins at least 0.5% of all party votes or wins a constituency seat. List of candidates submitted by 500 As 1960s except 5,000 signatures registered voters in the constituency required to register a party. or by a registered party. 3,000 signatures required to register a party. Continuous registration unless party fails to nominate any candidate in any constituency in two consecutive elections.
100 Sweden
Switzerland
United States a b c
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
No requirements to participate in ballot, although parties polling less than 1% of the vote must pay for ballot paper production. In cantons with 2 or more seats in the National Council, 50 signatures required per electoral list. No signatures required in single-seat cantons.
Requirements for ballot access determined at state level
As 1960s.
Until 1994, as 1960s. Since 1994, number of signatures per list varies by canton and depends on the number of seats the canton has in the National Council (which depends on population). In cantons with 2–10 seats (14 cantons) 100 signatures required; in cantons with 11–20 seats (5 cantons) 200 signatures required; in cantons with more than 20 seats (Zurich and Berne) 400 signatures required. No signatures required in the 5 single-seat cantons. Parties may present several lists in the same canton. As 1960s.
In some instances, the most recent data that could be gathered relates to the 1990s rather than 2000. Registration is not compulsory but unregistered parties do not qualify for rights and entitlements such as public funding. Registration is not mandatory but the benefits of registration are sizeable and include tax credits for contributors and refunds for election expenses. Parties become registered when they meet certain requirements and when they endorse at least 50 confirmed candidates. Failure to nominate 50 or more candidates results in de-registration. Sources: ACE Project; Elections Canada On-Line; 〈http://www.elections.ca/〉; Elections New Zealand 〈http://www.elections.org.nz/ elections/index/html〉; Hand, Georgel and Sasse (1979); Inter-Parliamentary Union Parline Database 〈http:www.ipu.org/parline-e/ parlinesearch.asp〉; Katz (1997); Katz and Mair (1992); Nozari (1991); Qualter (1970), Robertson (1997). Personal communications with colleagues listed in the acknowledgments.
CHANGING PARTY ACCESS TO ELECTIONS
101
Table 5.A2.Requirements for media access Country Australia
Austria
Belgium
2000* Free broadcasting time on basis of number of seats contested (parties must contest seats in majority of States and Territories to be granted free time) and demonstrated public support— established parties must have either one MP at previous election or have polled at least 5% of the vote in last election; new parties must demonstrate a minimum of 5% support across the nation in polls. Government and opposition parties are granted 60 minutes free time on radio and television with 30 minutes for the party election launch and six five-minute spots for policy announcements. Minor and new parties are allocated free time at the discretion broadcasting commission based on public support and of the number of seats contested. Minimum time allocated = 5 minutes. Paid political advertising permitted as long as the sponsor is identified. Stations must provide equal opportunities for advertising to all parties. No rules until 1966. From 1966, parliaAs 1960s except paid political advertising mentary representation required for media permitted since early 1980s. access. Shares according to party strength. Media time allotted to parties in and between election campaigns. Paid political advertising is prohibited. French-speaking media: no rules until 1964. French-speaking media: as 1960s. From 1964 ‘tribunes’ of eight minutes allocated proportionally according to the number of seats in Conseil Culturel (1 programme per year for a group with 2–5% of the seats, 2 programmes for groups with 5–10% of the 6 seats, 4 programmes for groups with 10–15% of the seats, programmes for those with 15–20% of the seats, 8 programmes for groups with 20–30%, 10 programmes for groups with 30–40%, 12 programmes for groups with 40% or more of the seats.) Dutch-speaking media: no rules until 1979. Dutch-speaking media: since 1979 groups represented by 10 members of the Nederlandse Culturraad can create an organization entitled to make programmes. Time allocated equally for 50% and proportionally for 50%. Since 1982 every Fraktionenen in the Vlaamse Raad can create a broadcasting organization.
1960s No free broadcasting time, but requirement for balance between parties with parliamentary representation if any political matter is broadcast at election time.
102 Britain
Canada
Denmark
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
Limited amount of free access to major As 1960s but since 1987 ratio is 1 : 1 : 1. parties according to mutually agreed definition of ‘balance’—decided on partly according to the parties’ vote share in the previous election. 1960–2 ratio 5 : 5 : 1 for Conservative: Labour: Liberal; 1963–73 ratio 3 : 3 : 1. Paid access to broadcast media forbidden. Applies to broadcasts during election campaigns and to inter-election broadcasts. At election time other parties may qualify for airtime if they field 50 or more candidates. Paid access to broadcast media forbidden. Until 1974, free airtime allotted to estab- Since 1974, free time is available during lished political parties during election cam- campaign periods to registered parties only. paigns. Time allotted according to number of parliamentary seats before dissolution and at dissolution, party's vote in previous election, and number of candidates nominated. For new parties, requirements were: to have policies on a wide range of national issues, have a recognized national leader, to have a nation-wide organization, to nominate candidates in at least three provinces and in at least one-quarters of the constituencies, and to demonstrate national significance. Air time to new parties was discretionary however, and no new party was awarded free air time in the 1960s—only the four established ones were. Requirements must be met to register and 50 candidates must be endorsed. Time allocated on a similar basis to 1960s, that is, according to parties’ results in the previous election, the number of seats held before dissolution and the number of candidates nominated. Paid political advertising is also permitted during election campaigns, for registered parties only. The total time available for purchase is 6.5 hours and time available for each party is calculated on similar basis as the free time allocated to each registered party. Each party is free to purchase as much of its allocated time as it wishes. Newly registered parties may purchase up to 6 minutes of time, up to an aggregate of 39 minutes per broadcaster. Paid political broadcasting is not permitted between elections. Equal shares to any party participating in As 1960s. ballot. Governing parties are not favoured in terms of time allocated, though they choose their television spots first. Paid political advertising forbidden.
CHANGING PARTY ACCESS TO ELECTIONS
Finland
France
Germany
Ireland
Italy
103
No official rules governing media access but Media access still only governed by informal informal agreements. rules. By and large, parties appear in broadcast media during election campaigns only. Election debates are organized in which all parliamentary parties take part, although on rare occasions other parties are invited. The leaders of the three main parties feature more prominently in these debates than other politicians. There are also short election spots in which parties present their policies. All parties have their spot, but parties not in parliament tend to be pooled together and receive less air time than larger parties. Equal rights for each party to appear on Paid political advertising is permitted during public channel during the election campaign. election campaigns (since 1990) but tends to Party interviews equally long, independent be rather limited. ‘Negative’ advertising is of size of party. prohibited. Between elections: 20 minutes of free time Since 1986 airtime during campaign period available every month for parties repreno longer distributed equally but according sented in parliament. to party strength. During election campaigns: parliamentary Paid political advertising forbidden since parties receive additional air time at no early 1990s. charge—approx. 3 hours distributed equally among parties prior to first ballot. Parties not represented in parliament given 7 minutes of free air time for first ballot and another 5 minutes for second ballot if they had nominated at least 75 candidates for first ballot. Free air time in campaign periods, allocated As 1960s, but paid political advertising in proportion to party's previous vote. permitted since 1980s on private channels. Minimum time allotted to new or previously These channels must offer equal opportuunsuccessful parties. nities to all parties, and are not allowed to discriminate against parties not in parliament. Smaller parties tend to get fewer spots than larger parties however. From 1965 parties with minimum of 7 TDs Election broadcasts now available to any entitled to election broadcasts. Other parties party or group fielding at least 7 candidates. entitled to short spots during news programmes. Coverage must to be ‘balanced’ and ‘objective’. Government may intervene to prevent any particular broadcast from going on air. Time allocated to each group is based on the group's votes in last election and number and geographical spread of its candidates. Only parliamentary parties have media Since 1995 equal opportunities for all parties access. and movements participating in the election to appear on broadcast media during elections. Paid political advertising permitted, but within limits. Law in progress as of 1999 proposing to prohibit any form of political advertising on national television channels during election campaigns, allowing it on local channels only.
104 Japan
Netherlands
New Zealand
Norway
Sweden
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
Provisions for free media time.
Candidates of parties that present at least 12 candidates are awarded equal free campaign broadcasting time. Every candidate is entitled to 4 television appearances (2 on public and 2 on private channels), and 2 radio broadcasts (1 public, 1 private). Each broadcast is limited to 5.5 minutes. Candidates receive less time if their party is running fewer than 12 candidates. Candidates are not permitted to buy broadcasting time but parties are, provided they do not mention the names of any candidates. Radio: during campaign times equal shares Since 1982 media access during the camallocated to any party presenting a list in at paign period is limited to parties that least one sub-district. During inter-campaign participate in all 19 sub-districts. Time is period equal shares to any party in parlia- distributed equally among these parties. ment. Television: until 1962 as for radio but only During inter-campaign periods, media acduring campaign periods. After 1962 as for cess is distributed equally, but only to parties radio including inter-campaign periods. in parliament. No legislative requirements until 1989, but Broadcasting Act of 1989 provides for broadcasting corporation provided parties allocation of free broadcasting time during with free air time for election broadcasting. the election period, and for the allocation of For 1963 election free time provided to funds to enable eligible parties to meet costs candidates only. Allocations made by Prime of election broadcasting. The state-owned Minister in consultation with the parties. broadcasters are obliged to provide time for Communist Party not given air time, Liberal the parties’ opening and closing addresses. party given limited air time. For the 1966 Political parties qualify for allocation of time and 1969 elections broadcasting corporation and/or money by being registered at least 3 determined total time available and its months before dissolution of parliament or allocation among the parties (rather than by having at least 5 constituency candidates candidates), but party leaders in effect had a at least 3 months before dissolution. veto. In 1966 the broadcasting corporation Allocation of time and money based on a proposed to grant air time to all parties but number of criteria including vote in previous the Labour leader vetoed this on grounds election, popular support for party in polls, that it would allow broadcasts to Commu- party membership figures, and number of nist party candidates. Time could not be MPs at time of dissolution. Broadcasters purchased. must provide identical terms for all parties and candidates. Paid political advertising forbidden. No official rules but national broadcasting As 1960s. corporation requires parties to have been represented in parliament in at least one of the last two parliamentary periods, to nominate candidates in majority of constituencies, and have a national organization if they are to have media access. Paid political advertising forbidden. No formal rules—decisions made by As 1960s. broadcasting companies. Free broadcasts during election campaigns—interviews and debates are the main form of broadcast. They have been generally restricted to parties in parliament (these parties have been given equal coverage) although exceptions have occurred on a very few occasions when smaller parties have been included. In other cases smaller parties have been given short spots and/or interviews, which, though smaller than those of the major parties, have been equal among themselves.
CHANGING PARTY ACCESS TO ELECTIONS
105
Paid political advertising is forbidden.
Switzerland
United States
*
Paid political advertising allowed on satellite channels since early 1980s. Paid political advertising is forbidden. Parties' portraits presented on television before elections. However, these spots are not produced by the parties, but by the media companies (public and private). The companies decide how much coverage they wish to accord to which parties; they provide equal coverage of the four largest parties. Smaller parties with representation in parliament do receive airtime, though this is more limited. Smaller parties with no representation in parliament are accorded coverage if they have had success in previous cantonal elections. Decisions over how much coverage to accord to a party are made according to the regional strength of the parties. There are also debates or discussions in which two or more parties are present. Paid political advertising is forbidden. No state sponsored access to media. No state sponsored access to media. Paid political advertising only. Broadcasters are required to provide equal Broadcasters must offer equal broadcasting opportunity (upon demand) for candidates opportunities to all candidates (upon deto utilize their facilities. mand) and must supply time at lowest rate charged to non-political advertisers. Advertisements must include indication of sponsor. No limits on media expenditure. Broadcasters may withhold access to their Since 1974 broadcasters can no longer facilities for certain electoral races if they choose to withhold access to their facilities wish from candidates for federal office.
In some instances, the most recent data that could be gathered relate to the 1990s rather than 2000. Sources: ACE Project 〈http://www.aceproject.org/〉; Australian Broadcasting Corporation (1999); Avril (1994); Borre (1988); Canadian Broadcasting Corporation; 〈http://cbc.radio-canada.ca/htmen/6%5F5/htm〉; Courtney (1978); Curtis (1988), Day (2000); Drysch (1993); Elections New Zealand 〈http://www.elections.org.nz/elections/index.html〉; Farrell and Webb (2000); Gamaleri (1995); Gunlicks (1988); Holtz-Bacha and Lee Kaid (1995); Katz and Mair (1992); LeDuc, Niemi, and Norris (1996); Marletti and Roncarolo (2000); Moring 1995); Nassmacher (1989); Pujas (2000); Ricolfi (1997); A. Smith (1981). Personal communications with colleagues listed in acknowledgements.
106
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
Table 5.A3.Requirements for state subventions Country Australia
Austria
Belgium
2000* Election funding introduced 1984. Amount of funding to parties is calculated by multiplying the number of first-preference votes won by the rate of payment applicable at the time. (Rate indexed to inflation—at 2001 federal election the rate was 179.026 cents per vote). Registered political parties are entitled to election funding where an endorsed candidate receives at least 4% of the total first preference votes. Independent candidates also entitled to election funding if they receive 4% of the vote. Unregistered parties are not entitled to election funding. Since 1995 election funding takes the form of a direct payment rather than a reimbursement. Since 1963 subsidies to parliamentary Frak- Subsidies to parliamentary Fraktionen as tionen. 5 MPs required to form a Fraktion. 1960s. Funds for public relations work as Base amount and supplementary amount 1960s. awarded according to number of seats held by party. High base amount, so small parties benefit more. From 1967, extra funds to parliamentary Grants to party academies/educational inFraktionen for public relations work. stitutions since 1972. Base amount and amount per seat. One institution per party, funded if party has at least 5 MPs. Press subsides since 1975. Subsidies to central party organizations since 1975: fixed sum for parties with at least 5 MPs; yearly funds for parties represented in parliament (base amount and supplementary amount per seat); funds for public relations for parliamentary parties. Campaign subsidies: (I) since 1975 parties polling 1% of the votes or more without gaining representation in parliament can be subsidized in election years. Subsidy based on share of the vote; (2) since 1989 parties in parliament receive campaign contribution in election years—takes the form of a per voter subsidy (20 ATS per voter, US$1.33, £0.91). No state subventions to political parties. Subsidy to parliamentary Fraktionen since 1971. 3 members required to form a Fraktion. Sum per Fraktion and sum per MP—awarded and adjusted for inflation yearly. Subsidies to central party organization since 1989. Parties with at least 1 MP receive a fixed amount and an amount per valid vote cast at the last election on its list. Sums paid yearly and adjusted for inflation. (1995 fixed amount =5 million BEF (US$113,700, £77,350) and the amount per vote =50 BEF (US$1.14, £0.77). 1960s No state subventions to political parties.
CHANGING PARTY ACCESS TO ELECTIONS
Britain
No state subvention to political parties.
Canada
Yearly allowance of CAN$4,000 (US$3,700 in 1969) to prime minister and leader of the opposition. From 1969 financial support for opposition parties in form of subsidy to parties’ research bureaux. In 1969 this was CAN$195,000 (US$180,500 £120,000) in total, and was allocated in accordance with the number of MPs in each party. To qualify parties had to have 12 members.
Denmark
1960–1964 no state subventions to political parties. From 1965: grants to parliamentary Fraktionen (1 MP required to form group) for administerial assistance. Sum based on number of seats held. From 1969 grants for ‘expert’ assistance paid to Fraktionen. Sum per month per seat plus sum for opposition parties in committees.
Finland
France
107
State subvention available to opposition parliamentary parties since 1975 to cover parliamentary costs. Takes the form of an annual payment from the party in government to the other parties in parliament. Amount payable is calculated according to number of seats held and vote share won in last election. (The sums for the House of Commons for 2001 −2002 =£11,012 (US$16,180) per year per seat won plus £21.99 (US$32) per 200 votes won, with no maximum limit.) To qualify a party must win a minimum of 2 seats or 1 seat and 150,000 votes. Reimbursement of campaign costs from 1974. For candidates: reimbursement of 50% of costs if they have won 15% of the vote in their constituency and have filled in election expenses returns. For parties: reimbursement of 22.5% of costs for registered parties provided they spend at least 10% of their spending limit. Reimbursements in election years only. Since 2000 registered parties that obtain 2% of the total valid votes cast in the election or 5% of votes cast in the constituencies where they presented candidates are entitled to reimbursement of 22.5% of their election costs. Grants awarded to parliamentary parties for research and administrative costs on basis of number of seats held. Subsidies to parliamentary party: since 1986 each Fraktion receives 15,000 DKK (US$ 1,840, £1.250) per month per seat plus 60,000 DKK (US$7,360, £5,010). Still 1 MP required to form a group. In addition, Fraktionen receive 90,000 DKK (US$11,050, £7,520) per member for administrative assistance. Subsidies to central party: 1987–1994: 5 DKK per vote (US$0.61, £0.42) if minimum of 1,000 votes received. Received annually. From 1995, amount raised to 19.50 DKK per vote (US$2.39, £1.63). As 1960s.
Since 1967 state subventions to parliamentary groups and to central party in proportion to number of MPs. 1 MP required to form group. Candidates with more than 5% of the vote Since 1988 candidates who win less than 5% on first ballot reimbursed costs of campaign of the vote are reimbursed for printing and letter, flyers and posters. mailing of programmes, posters and ballots only. Candidates winning more than 5% are reimbursed 50,000 FF (US$6,990, £4,760) as of mid-1990s. Also were reimbursed their deposit of 1,000 Since 1988 public funding to parties. DisFF (US$200 in 1964). This was lost for tributed in two equal parts: (1) funds candidates with less than 5%. allocated to parties that have presented candidates in at least 75 districts during last parliamentary session, distributed in proportion to vote won by candidates on first ballot, (2) funds allocated to parties on basis of their parliamentary representation.
108 Germany
Ireland
Italy
Japan
Netherlands
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
1959–1966: direct public subsidies to parties As late 1960s but also: represented in Bundestag (outlawed in 1966). 1967–8: election reimbursement of 2.50 DM (US$0.63 in 1967, worth US$3.30 today) per eligible voter distributed proportionally according to vote percentage of parties. Only parties winning 2.5% of second (list) votes or 10% of first votes eligible for reimbursement. From 1969: threshold lowered to 0.5% of Since 1983 federal election reimbursement second (list) votes. raised to 5.00 DM (US$2.35, £1.60) per eligible voter, distributed proportionately between parties according to vote won. From 1967 rule that only 50% of parties' Since 1979 independent candidates also income may come from state sources. entitled to reimbursement if they win 10% of vote in their constituency. Funds to parliamentary Fraktionen (15 MPs 1989–93 parties winning at least 2% of the required to form Fraktion until 1969; vote in the prior elections are eligible for thereafter 26 MPs required). Opposition very modest reimbursement of campaign bonus. Funds for educational activities since costs—declared unconstitutional in 1992. 1959. Grants to party foundations since 1962. Small sums payable to leaders of opposition Grants to leaders of opposition parties parties that contested previous election and extended to leaders of governmental parties. had at least 7 TDs. State funding of party organizations and campaign subsides introduced in late 1990s. Parties must win at least 2% of the vote in previous election to be entitled to funds. Expenditure limits apply. No state subventions to political parties until After 1993 referendum on party finance 1974. public funding for parties (for both their parliamentary groups and their campaigns) was abolished. A new law introduced a system based on reimbursement of campaign spending to candidates. Reimbursement is in proportion to vote won by candidate. Expenses reimbursed to parties with more than 4% of the vote or with at least 1 elected candidate and 3% of the vote. Ceiling of L80 million (US$40,500 or £26,400) per candidate, plus 100 L (US$0.05 or £0.03) per citizen in single seat constituency or L10 per citizen in multi-seat constituency. 1997–9: provision whereby taxpayers could allocate L4 lire of every L1,000 of their income tax to their chosen party. Since new law of 1999 parties are reimbursed L4,000 Lire (US$2.00, £1.30) for every voter they attract. Parties entitled to this if they poll 1% of the vote. No state subventions to political parties. Since 1994 subventions for parties with 5 or more MPs. Funds assigned annually according to number of seats held. Overall sum set at ¥30 billion (US$250 m, £155 m), which represents about half average yearly spending of politicians. Until 1964 no state subventions to political As 1960s but also: parties. From 1964 subvention to parliamentary groups (1 MP required to form group)—fixed amount plus amount that increases with number of seats held.
CHANGING PARTY ACCESS TO ELECTIONS
New Zealand Norway
Sweden
109
Subvention for research institutes since 1972 for parliamentary parties. Fixed amount per party plus amount per seat. Must be matched with private resources. Subvention for educational institutes since 1975 for parliamentary parties—fixed amount plus amount per seat. Private resources must make up at least 30% of funds. Subvention for parties’ youth organizations since 1976, dependent on size of parliamentary group. No state subventions to political parties. As 1960s. Secretarial assistance for parliamentary par- Secretarial assistance as 1960s. ties, according to number of MPs. Subventions to parliamentary groups from Subsidy to parliamentary groups: since 1983 1966: 1 MP needed to form group. 1966–9: the smallest parties receive increased sup17,500 NOK (US$2,450 in 1966, worth port. 1991: small parties received a basic US$13,400 today) per party plus 2,500 support sum of 432,136 NOK (US$66,585), NOK (US$350, worth US$1,900 today) per whereas the basic support for large party MP with minimum of 35,000 NOK groups was 216,279 NOK (US$33,325). (US$4,900, worth US$27,000 today) per Parties still receive support per MP (1991: party. this was 30,856 NOK, US$4,750), with extra support for small party groups since 1991. Press subsidy since 1969. Subsidy to central party organization: since 1970 parties are entitled to per vote subsidy. Sums adjusted annually (1991: 22.10 NOK, US$3.40). Parties must have presented candidates in at least half the constituencies and have won at least 2.5% of the votes to be eligible. Subsidies to youth and educational organizations since 1978. Subvention to party central organization: Subvention to party central organization: 1966–1970: per seat subsidy to parties with still per seat subsidy but since 1972 parties at least 2% of the vote and parliamentary with more than 4% of the vote but no seats representation (1966: 60,000 SEK per seat, receive a subsidy for every 0.1% above 2.5% US$11,600, worth US$63,000 today). Paid of the vote. Parties with seats but less than annually. 4% of the vote receive the per seat subsidy and are also awarded a subsidy for every 0.1% above 2.5% of the vote. Parties that won 4% in the previous election but not in the present election receive smaller subsidy, de-escalated in 3 subsequent years. Subvention to parliamentary Fraktionen: Subvention to parliamentary Fraktionen: as well as per seat subsidy paid for by the state (as 1960s), since 1972 party groups receive basic grant from the state, basic grant from parliament and another per seat subsidy from parliament. The two per seat subsidies favour the opposition party groups. Basic grants are the same for all parties. No rule as to how many MPs needed to Parliamentary groups of parties with 4% or form Fraktion. more in the previous election but less in the present election receive a portion of the basic grant from the state. The amount is de-escalated in 3 subsequent years.
110
Switzerland
United States *
SHAUN BOWLER, ELISABETH CARTER, AND DAVID M. FARRELL
Awarded on per seat basis with opposition Subsidies for youth and women's organizaparties granted bonus. 1966: government tions. parties received 3,000 SEK per seat (US$580, worth US$3150 today) while nongovernment parties received 4,500 SEK per seat (US$870, worth US$4,700 today). Press subsidies from 1969. No state subventions to political parties. Subsidies to parliamentary groups since 1972. Amounts based on number of MPs in each group. Minimum of 5 members per group but members do not have to come from same party. Each group receives an annual sum and a sum per member—in 2000 these were 90,000 CHF (US$56,750, £38,620), plus 16,500 CHF (US$10,400, £7,080) per member. No state subventions to political parties. As 1960s.
In some instances, the most recent data that could be gathered relates to the 1990s rather than 2000. Sources: von Arnim (1993); Avril (1994); Bille (1996, 1997); Ciaurro (1989); Courtney (1978); Curtis (1988); Drysch (1993); Elections Canada On-Line; 〈http://www.elections.ca/〉; Elections New Zealand 〈http://www.elections.org.nz/elections/index.html〉; Ewing (1987); Ferdinand (1998); Gentile and Kriesi (1998); Gidlund (1991, 1994); Gunlicks (1988); Inter-Parliamentary Union Parline Database 〈http://www.ipu.org/parline-e/parlinesearch.asp〉; Johnston and Pattie (1993); Katz and Mair (1992); Klee (1993); Koole (1989, 1994a); Landfried (1994); Linton (1994); Nassmacher (1989); Nozari (1993); Paltiel (1989); Pierre, Svåsand, and Widfeldt (2000); Pujas and Rhodes (1999); Rhodes (1997); Robertson (1997); Sear (2001); Svåsand (1991); Swyngedouw (1998). Personal communications with colleagues listed in acknowledgements.
CHANGING PARTY ACCESS TO ELECTIONS
111
NOTES We wish to acknowledge the help of the following colleagues who responded to our various queries: James Babb; Luciano Bardi; André Blais; Michael Brändle; Michael Bützer; Ken Carty; Paul Harris; Hans Hirter; Robin Kolodny; Romain Lachat; Andreas Ladner; Georg Lutz; Marguerite-Lise Menard; Wolfgang Müller; Shawn O'Brien; Tapio Raunio; Steven Reed; Martin Rhodes; Pascal Sciarini; Campbell Sharman; Jan Sundberg; Lars Svåsand. We are grateful to this research group for its feedback and guidance, and to Susan Scarrow for her detailed comments on previous drafts. The usual disclaimer applies. 1
All financial data used in this chapter precede the adoption of the euro by most of the OECD countries that are members of the European Union, and so we report national currency trends throughout.
2
Following Alexander (1989), we consider money to be one of the most important sources of political power and perceive a party's electoral success to be more dependent on its access to state subventions than on its access to the ballot or the media. In Table 5.2 we therefore attach a double weighting to state subventions.
This page intentionally left blank
II Change in Non-electoral Institutions
This page intentionally left blank
6 Towards More Open Democracies: The Expansion of Freedom of Information Laws Bruce E. Cain, Patrick Egan, and Sergio Fabbrini Calls for more direct and participatory democracy in the advanced industrialized countries have, in most cases, been accompanied by demands for a more informed citizenry as well. As noted in other chapters in this volume, citizens have successfully obtained reforms at the ballot box (such as the initiative and referendum) and administrative changes (such as open meetings laws and sunshine laws) that inject them more directly into the democratic decision-making process. Greater participation in policy-making by citizens and public interest groups requires more knowledge of government by these actors and therefore more transparency in government. Thus, in most nations democratic change has included rising expectations about openness and transparency in government. Even states that have had strong paternalistic and state secrecy traditions have been pressured in recent years into adopting laws that grant—on paper at least—more citizen access to information. In this chapter, we trace the expansion of freedom of information (FOI) laws and assess their effectiveness in holding democratic governments more accountable to their citizens. How much information do citizens need? By all accounts, information is critical to the expansion of meaningful citizen participation and influence in contemporary democracies (see, for example, Dahl 1971; 1998). But political scientists disagree on how much information citizens require in order for them to be competent voters and civic participants. Minimalists believe that citizens can rely almost entirely on easily digested information cues and short cuts in order to make political decisions without investing enormous time and effort into the tedious task of becoming informed (see, for example, Lupia and McCubbins 1998). Others believe in a more classical, deliberative model, in which citizens—or a representative subset of them—are heavily immersed in the nuances and details of public policy, for example, Fishkin 1991). But whether information is filtered through opinion leaders or directly disseminated to the general public, few, if any, contemporary voices argue against the proposition that democratic governments have the responsibility to make their documents available to the public as a general policy.
116
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
Nonetheless, getting information from the government is never a simple matter. Governments withhold information for many reasons, some insidious and some defensible. Information can be embarrassing to those in power, giving advantage to the opposition and political ammunition to critics. Controlling information, governments have learned, is an effective way to manage public opinion—and that is no small advantage in a democracy. On the other hand, secrecy sometimes serves the national interest or preserves essential commercial and individual privacy rights. It is important to recognize that there can be good, and maybe even compelling, reasons for governments to restrict the flow of information to citizens. But information restrictions can also have deleterious democratic consequences. Citizens are unable to participate or to choose properly when they are denied critical information about the government and its actions. Democratic accountability assumes that voters know what their agents are up to, and can reward or punish them accordingly. The core obstacle to this accountability is the asymmetry of information that exists between governments and their citizens. As principal-agent theory instructs, a government typically has more information about its activities than does its citizens, making it difficult to monitor. A further twist, particularly relevant in the Internet era, is that governments in the developed world have become vast repositories of information about their citizens due to their capacity to combine electronic data from multiple sources in sophisticated ways. So governments generally hold more information about citizens than citizens do about their governments. Historically, governments have varied in the degree of secrecy they have adopted. Some of this variation has its roots in cultural and historical circumstances. For instance, Britain's more hierarchical and paternalistic government culture, combined with its institutional doctrines of parliamentary sovereignty and ministerial responsibility, support a stronger secrecy tradition than is found in the US. A restrictive law like the British Official Secrets Act would seem wholly foreign in the context of America's First Amendment tradition and a populist political culture that is so suspicious of government that it divides power across multiple branches and layers. One recent trend has been towards expanding citizen access to government-held information through FOI laws, which have been enacted in sixteen out of the eighteen nations included in the scope of this volume's analysis. In this chapter, we track this expansion of FOI laws in advanced democracies and examine the patterns of usage by citizens and the press in a few selected cases. In the first section of this chapter, we briefly review the literature and theory of FOI laws and the purposes they serve. We then document the expansion of FOI laws across our eighteen-nation sample and speculate on why nations like Italy have apparently adopted FOI laws in form only. We then examine the frequency of FOI usage in three cases (the US, France, and the European Union, EU) and the connection between FOI laws and press coverage in two of them (the US and France).
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
117
Our major themes are as follows. First, the right of access to public documents is increasing, creating more possibilities for citizens than in the past. Second, the right to obtain information from the government is bounded by the rights of individual and corporate privacy and by important state interests such as national security. Third, these information laws are heavily utilized in some states but not in others, and not always in the ways envisaged by democratic theorists—causing us to wonder about the extent to which FOI laws change the traditional relationship between citizens and governments or merely replicate it. And fourth, when combined with an aggressive, independent, and investigative press tradition, FOI laws can be used effectively to hold governments, past and present, accountable for their actions.
POLITICAL SCIENCE AND GOVERNMENT INFORMATION Political scientific research and theory regarding citizens’ access to government information is somewhat scant. An important line of scholarly research—based largely on the experience of the United States—treats reforms such as FOI not as principled endorsements of democratic values but as a way for legislatures to enlist the help of interest groups in monitoring the bureaucracy. Such innovations allow legislators to forgo ‘police patrol’ monitoring (that is, intense, time-consuming scrutiny of government activity) for the ‘fire alarm’ method, which relies on interested parties to alert Congress when the government is acting against public preferences (McCubbins, Noll, and Weingast 1987). Given the limited amount of resources legislators have at their disposal, FOI laws can be in legislators’ interest by making agencies much more easy to monitor by citizens. FOI laws essentially allow legislators to share the costs of monitoring the bureaucracy with citizens. But added power to monitor comes at a cost: rather than the monitoring taking place only among legislators, it now can be performed by any citizen (or, indeed, as in the case of the US, any interested party). Arnold (1987: 281) notes that procedures like FOI … permit interested parties to watch and participate in agency decisionmaking and to appeal unfavorable decisions to the courts and Congress. In essence, such administrative rules and procedures force agencies to hear and consider the full range of policy preferences that Congress itself would hear if it had retained jurisdiction over these decisions. In other words, FOI laws introduce new players—citizens themselves—into the information game. Their entry into the game brings new preferences to bear directly on the agencies under scrutiny. Empirical studies of FOI laws tend to be case studies of one or more nations (for example, Robertson 1982; Marsh 1987) or outside the realm of political science, whether in law reviews (see US Department of Justice 2002) or in public policy studies
118
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
(for example, Doyle 2001). Perhaps most relevant to our analysis here is Bennett's (1997) study, which attempts to explain the diffusion of FOI laws throughout the world. He rejects the thesis that the expansion is caused simply by state growth or democratization, and instead contends that FOI laws have spread because advocates of information access in newly adopting nations have used already-adopting nations to legitimize their cause. Bennett (1997: 217) also argues that FOI laws exhibit substantial similarities, and that ‘… cross-national differences center on relatively tangential issues such as time limits, the permissible costs for providing the information, the definition of exemptions … and to the operation of the instruments of appeal and redress’. We will take issue with this statement shortly: our preliminary findings are that relatively similar laws on the books can have substantially different impacts in practice.
THE EXPANSION OF THE RIGHT TO INFORMATION The trend line with respect to freedom of information laws in OECD nations is quite clearly upward. As of August 2002, only two countries included in this volume's analysis—Germany and Switzerland—had not enacted national FOI laws, although the laws vary enormously in what they cover and how they are enforced. (FOI laws are currently under serious consideration in both Germany and Switzerland. And below the national level in Germany, several Länder—Berlin, Brandenberg, North Rhine-Westphalia, and Schleswig Holstein—have FOI laws; Banisar 2002a: 6.) The timing of FOI enactments is displayed in Fig. 6.1. A few notable patterns are clear. Sweden had the earliest FOI law, dating back to its Freedom of the Press Fig. 6.1.Enactment of national freedom of information laws
Note: Dates are earliest FOI law enacted by national legislature. Sources: Banisar (2002b); Bennett (1997).
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
119
Act in 1766. Now incorporated into the Swedish constitution, this law requires that ‘every citizen shall have free access to official documents’ (Banisar 2002b: 38). Given that Finland (1951), Denmark (1970), and Norway (1970) also were among the first nations to adopt FOI laws, there seems to be something of a Nordic tradition in this area. The US law (1966) was the product of an eleven-year investigation of unnecessary secrecy by a House of Representatives subcommittee headed by John E. Moss. Initiated in 1955 when the Republicans controlled the presidency and the Democrats the House, the move to open up government documents to inspection was a product of partisanship in a divided government situation and strong encouragement from the press corps, including former reporters who served on the subcommittee staff (Archibald 1993). Democrats strengthened the 1966 law in 1974 after overriding a veto by Republican President Gerald Ford. More countries added themselves to the FOI list in the late 1970s and early 1980s, including France and the British Commonwealth countries of Canada, New Zealand, and Australia. Not shown in Fig. 6.1 is that by the late 1980s and early 1990s these reforms were spreading to southern and central European nations, and then in the last wave, beginning in the mid-1990s, to Asia and the new eastern European democracies. FOI laws are on the books in twentysix of the current thirty OECD nations, and FOI laws now span the globe from Albania to Zimbabwe. There is some evidence of innovation diffusion through imitation. In several instances, countries explicitly modelled their FOI laws on the US example (for example, the Czech Republic), or the laws were reviewed by outside experts such as the American Bar Association and Swedish Ministry of Justice (for example Moldova). A very important point to bear in mind is that FOI laws are about the right, not really to information writ large, but to public records and documents specifically. Hence, citizens need to ask either for documents related to a specific topic or—in some nations—for a specific document. Beyond that basic feature, FOI laws can vary in important ways. Strong FOI laws—such as the ones in the US, Australia, France, Belgium, and Canada—contain several critical components. In a strong FOI law, the citizen's right to access to copies of the public documents is presumed, whereas in a weak one there is a presumed governmental right to determine whether to grant access to inquiring citizens. Second, the exceptions in strong FOI laws to access rights are limited to core concerns about state security and privacy rights rather than giving the government the right to exempt documents in any subject area it sees fit. Third, strong FOI laws provide for some third party to whom disputes over access can be referred. In France, it is the Commission d'access aux documents administrats. In Australia, adverse decisions are reviewed by an Administrative Appeals Tribunal and, if necessary, by the federal courts. Similarly, in the US, the Department of Justice Office of Information and Privacy reviews a department's decision to deny a request, and decisions can be further appealed to the federal courts.
120
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
In some OECD countries, the FOI law is either relatively weak or has not been implemented in any serious way. In Austria, for instance, the FOI law requires the government to answer questions about the content of documents, but does not permit citizens to have access to the documents themselves. The British enacted an FOI law in 2000 after debating the issue for over twenty years. It has been widely criticized by information-access advocates for exempting a broad class of information from exposure. For example, the act gives authorities discretion to deny requests when their reasonable opinion was that it would prejudice the ‘effective conduct of public affairs’ (Banisar 2002b: 41). Italy's 1990 law provides for general access to government documents, but requires a legitimate reason for the request. Moreover, our attempts to obtain data about public usage of the Italian law proved to be futile, and the Italian journalists we interviewed indicated that they never used it in their stories. The Italian case deserves a more detailed discussion because it is an interesting example of what we label as ‘reform by emulation’.
ITALY AND REFORM BY EMULATION In Italy, despite ten years of experience with a FOI law (the law was introduced in 1990), it was impossible to collect any data from the administrative offices about who used the law and why, or to find journalists who said that they relied on Italy's FOI law to research stories. Since they received so few FOI requests, the Italian administrators felt there was no need to collect data about them. What explains this situation? Several plausible arguments come to mind. First, the reputed inefficiency of the Italian administrative system may present formidable obstacles to citizens and journalists who wish to use the FOI law. In the face of likely administrative inefficiency, potential petitioners may simply give up in advance. This argument is certainly plausible, but cannot be the whole story. In fact, Italy has passed through several important reforms since the 1970s, partly in response to the growing impatience of educated Italian middle-class citizens with the traditional routines of public administration. Also, especially in the 1990s, the Italian administration felt the pressure to reform in order to deal with anti-centralistic sentiments in the Northern and Central regions and more competitively with the administrative structures of the other EU member states. In short, the Italian bureaucracy improved a great deal during the period that its FOI law was implemented. Another possible explanation is that Italian journalists do not use the FOI law, because the Italian media's operators have closely cultivated personal connections with politicians, public officials, interest groups representatives, and private entrepreneurs. Again, this argument is plausible but it is not entirely convincing. To be sure, roughly half a century of consensual democracy in Italy promoted deep and tortuous connections between politics, economy, and the media. Given the importance of party allegiances in the quest to find and retain a job and the capacity of
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
121
the political parties to colonize social life, it was inevitable that journalists, like other professionals, would move back and forth from politics to the (especially, public) economic sector or to the media. Given this web of acquaintances, of ‘friends of friends’, it was easier for a journalist to call directly ‘somebody who knows’ in order to get a desired piece of information than to fill out an impersonal form to submit to an unpredictable bureau. But, even if we grant that point, the Italian media system was radically transformed by the demise of the consensual politics system triggered by the ‘Clean Hands’ investigations. To be sure, the emergence of the Berlusconi media empire resurrected a tie between politics, the economy, and information, but it was the Berlusconi media people who colonized politics and the economy this time and not the other way around as in the consensual period. Still, a new generation of journalists has emerged in recent years without strong allegiances to political and economic groups. In theory, at least, they should use the FOI law, but have not. In addition to the first two explanations, there is a third: citizens and journalists do not make use of the 1990 Italian FOI law, because they lack the cultural predispositions to make effective use of the law. The 1990 law was not adopted out of political necessity. There was no citizens’ movement that demanded administrative transparency. Nor was there a press campaign to open administrative procedures and acts. In fact, we found no reference to FOI law either in the official documents of the Italian business association (Confindustria) or in the main editorials of its daily newspaper (Il Sole-24 Ore). It seems that the law was transposed from outside the Italian system rather than fostered by demand from within the system. ‘Outside’ in this case means the transnational network of administrative and political elites operating in the institutional context of the European Union. Italian administrative and political elites were gradually trained in administrative reform through this network. The lessons they learned were based on other countries’ experience. Italian elites attempted to translate those reforms into an Italian form. Many scholars believe that the EU is a system of network governance, where decisions bubble up through an informal process of shared information, consultation, and negotiation involving private and public actors. The Europeanization process favours common solutions for specific policy problems in the EU member states. However, a solution that answers a specific problem in some states may not be motivated by an analogous problem in other states. Europeanization and globalization create hospitable environments for the diffusion of common ideas and practices, but sometimes these solutions are adopted in national contexts that are culturally unprepared to accept them or even to acknowledge the problem that justified importing them. In sum, sometimes reforms might look like transnational solutions looking for national problems rather than the other way around. Thus, the Italian FOI law of the 1990 seems to a case of reform by emulation. Reforms by emulation are generally elite-driven in the sense that they are imposed from above or from the side as part of a more general strategy of modernization.
122
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
Reforms by emulation can be transposed into states that lack the appropriate supportive cultural and institutional contexts. In the case of Italy's FOI laws, citizens, journalists, and public officials have apparently not figured how to use the reform or envisaged the opportunities it might offer to them.
EXEMPTION POLICIES COMPARED A critical aspect of FOI comprehensiveness is the number and type of exemptions from public access allowed by law. The public's right to know has to be balanced against the public's right to be protected both individually and collectively. Therefore, FOI laws typically delineate subject areas that are exempt from the general disclosure requirements. Table 6.1 displays these exemption categories allowed by law for a sample of nine countries. Several patterns emerge quite clearly. First, there is one universal exemption: information related to defence secrets. Indisputably, it is reasonable for states to withhold information that might weaken their defence postures and bring harm to their citizens. A citizen's right to know is clearly not greater than the rights of fellow citizens to be safe from external threat, and this is acknowledged in the FOI exemption policies of all the sampled countries. A few other policy areas are almost universally exempted. One of these is the general category of commercial and individual privacy. With respect to the former, since a nation's welfare can hinge on trade negotiations, and secrecy can serve the national economic interests, it can be in the national interest to keep these matters from public inspection. Commercial privacy also means keeping company secrets from other companies. Another form of privacy is individual: the right of individuals to keep government information about them out of the hands of others. Individuals should be able to demand to know what information the government has about them, but they should also have the right to keep that information confidential with respect to others. The right to access to government documents, therefore, has to be limited to protect individual confidentiality. Another common exemption is for law-enforcement documents. Like defence matters, information about internal security or police operations could compromise the government's effectiveness in protecting the public. In some countries, this extends to court proceedings or internal agency documents as well. Among the less common exemptions, many seem to be derivative in the sense that they protect either the government's capacity for effectiveness or commercial and individual rights. For instance, US national law exempts documents related to the government's internal personnel rules, Britain and Australia exempt information obtained in confidence from any person. But some exemptions are quite idiosyncratic. Sweden exempts documents that are necessary to preserve animal or plant species, and the US exempts geological data concerning oil wells (at the
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
Table 6.1.Exemptions to FOI laws in G7,*Australia, New Zealand, and Sweden FOI exemptions National defence/foreign policy Trade secrets Invasion of personal privacy Law enforcement Internal govt deliberations Hurt economic interests of govt Obtained in confidence from other govt. Injurious to domestic govt relations Attorney-client privilege If info. will be published in future Court documents Legislative documents Endanger someone's safety Audits of other public authorities Obtained in confidence from any person Explicit exemption of certain depts. Exempted via later statute Communications with Crown Disclosure would be in contempt of court or parliament Necessary to preserve animal or plant species Internal personnel rules Supervision of financial institutions Geological data concerning wells Documents concerning administration of courts Electoral rolls Research whose disclosure would disadvantage govt agency Certain companies and securities documents Cabinet documents Disclosure would require substantial research Info. accessible by other means Disclosure would inhibit public business *
Australia •
Canada •
France •
Italy •
Japan •
New Zealand •
UK •
US •
Sweden •
• •
• •
• •
• •
• •
• •
• •
• •
• •
• •
•
• •
•
• •
• •
• •
• •
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
• •
•
•
• • •
•
• •
• •
•
•
•
• •
•
• •
•
•
•
• • • • • • •
• •
Excluding Germany, which has no national FOI law Sources: Banisar (2002b); Marsh (1987); authors' research.
• • •
123
124
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
insistence of the President at the time of enactment in 1966, Lyndon Johnson from Texas). Intuitively, one might think that, the weaker the FOI law, the more numerous the exemptions. This does not seem to be true, however. Italy, which has few exemptions, has a law that is rarely, if ever, used. France also has very few exemptions, but FOI activity appears to be quite heavy. Australia and the US have a sizeable number of exemptions, but also have strong FOI regimes. This point can be seen more clearly by looking more closely at both the usage patterns and the ways exemptions are handled in three cases: the US, France, and the EU.
Case 1: The US Freedom of Information Act The US Department of Justice provides very detailed records of requests under the US Freedom of Information Act (FOIA) to all federal departments and agencies. Data in the following tables were compiled from the Department's reports for federal fiscal year 2000 (US Department of Justice 2001). The agencies comprising the data in Table 6.2 include all cabinet-level departments and offices plus any agency that amasses more than 1,000 yearly FOIA requests. While these reports are titled ‘FOIA Annual Reports’, they actually combine FOIA requests (that is, asking for access to records concerning a third party, an organization, or a particular topic of interest) with records requested under the US Privacy Act of 1974 (PA) which grants access to records concerning oneself. Since we cannot distinguish PA from FOIA requests empirically, we have no choice but to report a combined figure. The volume of US Privacy Act and FOIA requests is quite staggering. The first column of Table 6.2 indicates the more than 2 million total requests federal agencies received in 2000. (All data we report for the US in this chapter are for 2000. Federal officials only recently began tracking the number of requests received, so we are unable to report how this total compares with past years.) These agencies accounted for over 99 per cent of the total requests that year. There is considerable variation in the number of requests by agency. The Department of Veterans Affairs, the Social Security Administration, and the Department of Justice lead the list, each with more than 200,000 requests a year. A key feature of these agencies is that they run programmes that provide services for and keep records on many citizens (for example, veterans’ health and benefits, immigration and naturalization, social security). Veterans, for instance, might need their health records for disability and health insurance purposes, and citizens doing genealogical searches depend upon Social Security Administration records. Given the high volume of requests, how many of these are denied, and does the amount vary from agency to agency? Answering these questions gives us a sense of the extent to which FOIA is a true instrument of transparency or simply window dressing. In addition, varying denial rates by agency provide a glimpse of the topic areas where information access is a contentious issue. Overall, the denial
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
Table 6.2.United States FOIA request, denial, and appeal statistics, 2000 Department or agency
FOIA requests received 1,239,844
Denials as % of total processed 2
Denials and non-disclosures % of appeals resulting in as % of total processed reversal of denial 3 50
Department of Veterans Affairs Social Security Administration Department of Justice Department of Agriculture Department of Defense Department of Health and Human Services Department of the Treasury Department of Labor Department of Transportation Environmental Protection Agency Equal Employment Opportunity Commission Consumer Product Safety Commission National Archives and Records Administration National Labor Relations Board Department of the Interior Central Intelligence Agency Department of State Small Business Administration Department of Energy Securities and Exchange Commission Office of Personnel Management Department of Commerce Federal Reserve System Peace Corps National Aeronautics and Space Administration United States Postal Service General Services Administration Federal Trade Commission Federal Emergency Management Agency Office of Management and Budget Office of the United States Trade Representative Office of National Drug Control Policy Total
262,450
3
4
23
235,042 140,239 97,266 61,971
15 3 14 17
30 4 39 27
11 42 22 34
46,854 22,469 19,750
17 62 14
36 83 42
21 31 18
15,906
5
30
31
15,729
79
84
32
13,138
6
11
22
9,187
5
9
24
5,526
17
29
69
5,161 3,820 3,611 3,126
19 46 30 15
42 62 61 24
17 13 58 22
2,935 2,875
14 23
34 66
30 23
2,716
7
14
31
2,035 1,882 1,793 1,740
20 11 1 34
35 26 2 56
29 12 0 39
1,574 1,545
29 12
50 25
7 25
1,320 230
38 26
81 51
37 29
225
11
35
18
70
27
71
75
25
15
45
n/a
2,222,054
6
11
18
Source: US Department of Justice (2001).
125
126
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
rate as a percentage of the total number of requests processed is fairly low (see the second column in Table 6.2). Only 6 per cent of all requests were denied, but there is quite a bit of variation by agency. As one might expect, agencies dealing with exempted areas such as defence, trade, and foreign intelligence have some of the highest denial rates—for example, the CIA (46 per cent), the Federal Trade Commission (38 per cent), and the National Aeronautics and Space Administration (34 per cent). The Equal Employment Opportunity Commission (EEOC) and the Department of Labor, which also have high denial rates, are harder to explain. Perhaps it is because they are agencies with low ratios of PA to FOIA requests, and that PA requests are almost always granted. Hence, having fewer PA requests raises the overall denial rate. In addition, both the EEOC and Labor are heavily involved in regulating and investigating legal disputes regarding employment. Parties under investigation by these agencies may be making requests in the hope of gaining information valuable to their cases. Why are requests denied? Agencies may invoke statutory exemptions in denying FOI requests. The reasons for denial invoked vary from agency to agency, and a close look at them sheds light on why the US government keeps certain information secret. Surprisingly, the protection of personal privacy—rather than defence or corporate secrets—is by far the most invoked exemption. Table 6.3 shows the frequency with which the exemptions set out in the US FOIA were used to deny requests in 2000 by the agencies listed in Table 6.2. In addition to the exemptions written into FOIA, agencies may also decline information requests for several other reasons, such as the non-existence of records, referral to another government department, or (in a very small number of cases) the requestor's inability or refusal to pay the cost of processing the request. When we put these non-disclosures together with the denials detailed above, the third column of Table 6.2 presents a more complete picture of the rate at which requests Table 6.3.Reasons for denials of FOIA requests, 2000 Reason Invasion of personal privacy Intra-governmental correspondence Would interfere with law enforcement proceedings Would disclose confidential law enforcement source Exempted by special statute Would disclose law enforcement techniques Related to internal personnel rules Trade secrets National defense secrets Other Source: US Department of Justice (2001).
% of all denials 53 14 7 6 5 5 5 3 2 1
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
127
are fulfilled (not simply the ones that are denied because they fit into an exemption category). Across all departments, 11 per cent of FOIA requests were not met, and in eight of the agencies more than half were not fulfilled. The new calculations, however, do not change our earlier conclusions about the types of agencies that are least likely to meet requests, namely, those that deal with privacy and security and those that have fewer PA requests. In the US, if a FOIA request is not fulfilled by an agency, requestors can file an administrative appeal with the Department of Justice's Office of Information and Privacy. There is a fairly wide variance in the proportion of appeals that result in the Department of Justice's reversal of an agency's denial of a FOIA request (the rightmost column in Table 6.2), ranging from a high of 75 per cent for the Office of the US Trade Representative to zero for the Peace Corps and Office of National Drug Control Policy. Again, there is no obvious pattern to this data. For example, compare two offices responsible for international affairs: the State Department is reversed quite frequently (58 per cent of the time) whereas the CIA is reversed about 18 per cent of the time. The US Postal Service (incidentally, the agency that receives the highest number of appeals) also has one of the lowest reversal rates (about 7 per cent). In all, denials of information under FOIA were either completely or partially reversed about 18 per cent of the time. A final important aspect of FOIA responsiveness is the timeliness of the responses. After all, if information that is embarrassing to the current administration is kept bottled up for many years, FOIA can hardly be said to be a strong instrument of transparency and accountability. Table 6.4 shows that in three critical departments—the State Department, the Department of Energy, and the CIA—substantial numbers of requests languish considerably before being granted a response. (Departments categorize FOIA requests as either ‘simple’ or ‘complex’. These were the only three departments that required more than 100 days to process a substantial number of either type of request in 2000.) The waiting period may be justified by the fact that all three handle information that is vital to US security interests. But it also raises suspicions as each of these departments has in the past engaged in activity that, in retrospect, has called their accountability into question. Table 6.4.Departments with notably long delays in responses to FOIA requests, 2000 Simple requests Number processed
Department or Agency Department of State 413 Department of En- 1,001 ergy Central Intelligence 1,075 Agency
Source: US Department of Justice (2001).
Complex requests Median days to proc- Number processed ess 37 2,637 133 375
Median days to process 694 531
7
176
3,415
128
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
Case 2: FOI in France France provides a second example of a strong and active FOI law. The French system operates a little differently from the US, which unfortunately makes it hard to compare usage data between the two nations. Whereas in the US agencies make the initial decisions about—and report data on—information requests, French agencies rely much more heavily on an intermediary for both: the Commission d'accès aux documents administratifs (CADA). CADA has two main functions: to provide non-binding advisory opinions (conseil) to government agencies regarding whether they should grant information requests; and to issue appeals rulings (avis) regarding denials of document requests. Decisions by CADA can be further appealed to an administrative tribunal. We attempted to get data comparable to the US on the number of requests and other relevant information from a sample of French governmental agencies, but were denied access. Instead, we rely on CADA avis and conseil data (CADA 2001), which is clearly an underestimate of total requests. CADA usage statistics are probably more comparable to the percentage of FOIA requests in the United States that are refused or not decided than to the total number of FOIA requests in the US. Annual usage of the French law has grown substantially if the annual number of CADA rulings—which increased from about 500 in 1980 to nearly 5,000 in 2000—is any indication. Appeals rulings by CADA outnumber its advisory opinions to government agencies by a ratio of about eight to one. We can get a sense of who is denying FOI requests through CADA appeals rulings data (see Table 6.5). The two largest sources of denials are the national government, and the district governments (the smallest territorial divisions in France). Table 6.5.CADA appeals rulings by level of government, 2000 Level of government that denied request National government Districta Local public agencyb State public agencyc Private agency providing public service Departmenta Regiona Other a
b c
Number
%
1,673 1,227 475 399 259
39.4 28.9 11.2 9.4 6.1
145 36 30
3.4 0.9 0.7
District, Department, Region: territorial and administrative divisions of France. Department is the level between District (the smallest division of the territory) and Region (the biggest division). Local public agencies are those managed locally, including the state-run hospital system. State public agencies are managed by the national government, and are usually linked to one ministry. Source: CADA (2001).
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
129
Fig. 6.2.Sources of appeals to CADA, 1986–2000
Source: CADA (2001). Most requests to CADA are by individuals (see Fig. 6.2). This illustrates that French requests are empowering individual citizens primarily. Another third of the requests comes from private organizations. Only a minuscule number of requests comes from public entities. Trend data since 1986, however, show that individual appeals have dropped a little from the very early years of the law's implementation. The top three categories of requests to CADA deal with civil service matters (for example, information about hiring contracts of non-permanent civil servants), urban affairs (for example, permits to build and general planning documents), and social service benefits. Again, we must bear in mind that these numbers reflect only the matters that have been appealed to CADA or about which French agencies have requested an advisory opinion, not the total population of requests. But we do see that, as in the US, information that relates to individual or commercial privacy can pose dilemmas for the government's FOI policy. In the US case, we saw that the rate of fulfilment of requests was very high but that denials were reversed by the Department of Justice only 18 per cent of the time. CADA acts more favourably in its appeals rulings, reversing agency denials of documents about half of the time (see Table 6.6). In about another third of the cases, the request was withdrawn or the document did not exist and had been lost. This is again higher than the non-fulfilment rates in the US (which were on average 11 per cent). Of requests denied for exempted reasons (only about 8 per cent of the total cases in the 2000 data), the invasion of personal privacy is by far the most frequently cited reason for denial in France (see Table 6.7). Security and
130
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
Table 6.6.Appeals rulings by CADA, 2000 Ruling Reversed Upheld: Document does not exist, or has been lost Improper request Denial per statutory exemption FOI law does not apply Request was granted after appeal filed, or request was withdrawn:
% cases 48.8 12.2 9.0 7.8 4.7 17.5
Source: CADA (2001).
Table 6.7.Reasons for denial per statutory exemption, CADA, 2000 Reason Invasion of personal privacy Preparatory document Industrial or trade secret Unachieved document Court document Public and individual security Tax or duty secret Secret protected by the law Internal government deliberation Previous advice from the Conseil d'Etat or the administrative courts National defense secret Foreign policy secret
% 52.1 24.7 7.4 4.8 4.4 2.3 1.6 1.1 0.7 0.5 0.2 0.2
Source: CADA (2001).
commercial privacy is less often given as a reason to deny requests in France than in the US. It also seems that there are more instances in France of procedural problems: people asking for documents that do not exist, requests that went to the wrong department, imprecise requests, and so forth. This may be a function of the relative newness of the law. Moreover, a significant proportion—8.5 per cent—of agencies receiving an advisory opinion from CADA indicating that a document should be released went on to deny those requests anyway. The rates of not following CADA's recommendations were even higher in the mid-1990s. In sum, while the French FOI law is clearly being used more frequently by French citizens over time (we assume that CADA rates are an unbiased indirect indicator of total FOI requests), there does seem to be a more significant administrative hurdle, leading to higher rates of non-fulfilment in France than in the US.
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
131
Case 3: The European Union and FOI Our third example is the FOI laws that apply to the EU, which provides a wider view of FOI norms across Europe. This is an interesting case because EU openness became an increasingly sensitive issue in the 1990s. Criticism of the EU's so-called democratic deficit, raised three decades ago by the English political scientist David Marquand (1979), has become commonplace. Even the direct election of the European Parliament since 1979 has not relieved these concerns. The interesting aspect of the EU's FOI is that it operates in a more confused system of accountability. The EU appears to be organized in a way that hinders the clear allocation of responsibility to public officials. Members of the Council of the Union (with ministerial membership from the fifteen member states) are primarily accountable to their national governments. The European Commission structures the Council's agenda and influences decisions through the tailoring of directives and regulations. But since the 1997 Amsterdam Treaty, the Commission's president is required to get a vote of approval from the European Parliament, and the European Parliament has the power of codecision with respect to the development of common market policies. But, even with these increased powers, the European Parliament is still a comparatively weak legislature. Aware of the democratic deficit concerns, the Commission has looked for new ways to be more accessible to groups, lobbies, and movements. The White Paper on Governance issued by the Commission in 2001 requires each administrative unit of the Commission to consult interested or affected groups before reaching a decision. Of course, a policy of consultation and negotiation that involves private actors and public institutions can still seem opaque and closed to the average EU citizen. More than in the US, the EU-negotiated democracy is constrained by a technocracy that retains critical information about public policy. Given the complexity of many EU issues, information requests have a special importance. We were able to obtain data regarding FOI requests for three years: 1994, 1995, and 1996. As in the US and French cases, we are interested in the frequency of requests, the reasons for denial, and the subject matter of the information. In addition, we pay particular attention to the geographical distribution of requests given the federal structure of the EU. In terms of the geographical distribution of the requests, there seems to be a pronounced north/south divide that reflects the balance of wealth, power, and population within the EU (Table 6.8). The countries that generate the most requests are predominantly northern European, such as Belgium (30 per cent—quite high, because of its location as the headquarters of a large number of international associations), the UK (18.4 per cent), and France (10 per cent). On the other hand, the countries that generate the fewest requests are southern European nations such as Greece (0.5 per cent), Portugal (0.9 per cent), Spain (4 per cent), and Ireland.
132
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
Table 6.8.Geographical origin of FOI requests to the EU, 1994–96 Nation Belgium United Kingdom France Germany Italy Netherlands Spain Denmark Ireland Portugal Greece Luxembourg European Economic Areaa New statesb Non-EU countries Total a b
Number of requests 298 182 100 91 54 43 39 30 16 9 5 5 16 23 78 989
% 30.1 18.4 10.1 9.2 5.5 4.3 4.0 3.0 1.6 0.9 0.5 0.5 1.6 2.4 7.9 100.0
Iceland, Liechtenstein, and Norway, which are non-EU states that have commercial agreements with the EU. Austria, Finland, and Sweden, all of which entered the EU in 1995. Source: Commission of the European Union (n.d.).
It appears that residents and entities within more affluent nations have a greater desire for EU information and more resources to file FOI requests. The relatively low usage figure for Italy (5.5 per cent) reaffirms our earlier conclusion about the low salience of FOI provisions within Italy. What is asked for? As shown in Table 6.9, the subject matter varies widely. The subjects in most demand are the environment (16 per cent), industry (11.7 per cent), fisheries (11.5 per cent) and competition (8.2 per cent). The prominence of environmental requests is quite striking, as EU environmental policy is relatively new. In the early years of the EU the environment was considered the exclusive realm of the member states. But a more unified market has in effect ‘Europeanized’ market externalities, and the EU government has become the locus of action on that front. In addition, the growing role of Green parties in large member state parliaments (for example, the German Bundestag) has strongly contributed to making environmental issues more salient in the EU parliament. Finally, the number of environmental requests may also attest to the aggressive, legalistic tactics of pro-environment groups. By comparison, there are fewer requests for documents related to agricultural policy, even though that is a major EU focus. We suspect that this is because agricultural policy is more settled and less conflicted than environmental policy. As seen in Table 6.10, the most common categories of requestors are academics, followed by lobbyists, lawyers, and industry. This is consistent with the pattern in
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
133
Table 6.9.Subjects of FOI requests to the EU, 1994–96 Subject of request Environment Industry Fisheries Competition External relations Agriculture Budget/financial control Internal market Social affairs Otherb Total a b c
Number of requests 170 121 118 84 59 56 46 42 23 309 1,028c
%a 16.4 11.7 11.5 8.2 5.7 5.4 4.5 4.1 2.2 30.1
Does not total to exactly 100 due to rounding. No subject in this category totalled more than 2 per cent of requests. Some requests were classified into more than one category. Source: Commission of the European Union (n.d.).
Table 6.10.Categories of applicants under EU FOI laws, 1994–96 Applicant category Academics Lobbyists Lawyers Industry Public authorities Private individuals Members of European Parliament Journalists Total
Number of requests 245 165 151 143 142 95 24 24 989
% 24.7 16.7 15.3 14.5 14.4 9.6 2.4 2.4 100.0
Source: Commission of the European Commission (n.d.).
the US that individuals—as opposed to groups and public entities—tend to make information requests. If we leave aside the academics, it seems clear that freedom of information requests are generated pre-eminently by economic and functional actors who are involved in the transactions of the EU. Very few journalists and Members of Parliament make requests, which raises questions about whether FOI plays a major role in maintaining traditional political accountability in this instance. With respect to the rate of refusal and reasons for denial, EU data requests are overwhelmingly granted (Table 6.11). Only 12 per cent are denied, a figure comparable to US statistics, and an almost equal number are simply inadmissible requests (that is, the document had been published already, there was no document, and so on). National security and the confidentiality of police operations are
134
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
Table 6.11.Disposition of FOI requests to EU, 1994–96 Admissible requests Positive replies Negative replies Reason citied for denial: – public interest – private interest – commercial and industrial secret – financial interest of the commission – confidentiality requested by the person or the member state supplying the information – confidentiality of the proceedings – documents of the commission whose work is confidential – case where several exceptions were invoked Inadmissible requests – documents already published – documents not issued by the Commission – imprecise request – non-existent documents *
No. 962 845 117
%* 91.3 87.8 12.2
38 0 9 5
32.6 0.0 7.7 4.3
10
8.5
30 8
25.6 6.8
17
14.5
92 51 25
8.7 55.4 27.8
4 12
4.3 13.0
Does not total to 100% due to rounding. Source: Commission of the European Union (n.d.).
not the factors for the EU that they are for sovereign nations such as France and US. On the other hand, refusing to release information for reasons of ‘public interest’ and ‘confidentiality of the proceedings’ does little to dispel the democratic deficit problem. The EU rules allow appeals of FOI denials to the Secretary General of the Commission with the expectation of a reply within a month. In the years for which we had data, there were fifty-nine such appeals. Of these, nearly 60 per cent were denied. In sum, FOI has the potential to play an important role in increasing transparency in a governmental entity that struggles with a widely perceived ‘democratic deficit’. Policy decisions are taken through a complex interplay of supranational, national, and regional institutions, and the implementation of those decisions resides with the member states. While the data indicate that FOI laws are not heavily utilized, a number of key economic and functional actors have made requests under the law.
POLITICAL ACCOUNTABILITY AND THE PRESS: THE US AND FRENCH CASES We have established several points so far. Over the past 25 years, FOI laws have become the norm in OECD nations. But, as we have seen, there appear to be active (for example, France and the US) and inactive
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
135
(for example, Italy) FOI regimes. Citizens, interest groups, and businesses put these laws to many uses, not all of which conform to the model of political accountability that inspires the passage of these laws. These kinds of FOI requests range from the relatively mundane (like health records for insurance purposes) to requests with potentially greater import, such as those related to the marketplace and individual privacy. What evidence is there that FOI increases political accountability? In theory, FOI allows the press and critics to have access to documents that reveal the government's policies and motivations more clearly. Armed with this evidence, we would expect a free and active press to produce more stories exposing hidden aspects of government policy and for critics to use this information in politically embarrassing ways. Does this process actually take place? To answer this, we again focused our attention on the US and French experiences. To explore the connection between FOI requests and the press in the US, we recorded the press mentions of FOIA in four major US newspapers between October 1999 and September 2000 (the same time period covered by our analysis of FOI requests above): the New York Times, the Los Angeles Times, USA Today, and the Washington Post. We then analysed the articles’ content to see how the requested information was used. FOIA stories appear frequently in the four newspapers—on average, about three a week. We categorized the papers’ FOIA stories into four distinct types (Fig. 6.3). We classified a total of eighty-one unique stories (we did not doublecount topics that appeared more than once or in more than one newspaper). The largest category (37 per cent of all stories) is made up of stories about the actions and decisions of the present government. This includes stories that reveal potentially embarrassing documents, such as a USAID report that is critical of US disaster relief efforts, or an Air Force memo on the environmental impact of Air Force training flights. Fig. 6.3.Subjects of FOIA stories in major US newspapers, Oct. 1999–Sept. 2000
136
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
Sometimes FOI requests reveal changes in department policy, as in a story about the Department of Justice seeking a record number of wire-taps. However, FOIA stories are not only about the current government. Many stories (25 per cent of our sample) deal with wrongdoing, embarrassing information, and administrative incompetence in the past. Good examples of this are the various stories on FBI monitoring of historical figures such as Murray Kempton, Wilt Chamberlain, the Black Panthers, John Lennon, and Martin Luther King which came to light in 2000 as the result of FOIA requests. These stories illustrated the excesses of government, which provides a lesson to less famous citizens. In addition, these stories remind us that democratic accountability works not only by affecting public opinion towards, and hence the electoral prospects of, existing politicians, but also by the possibility that present actions might be judged in the future in the light of better information. Political figures that care about their reputation in history need to consider the possibility that they will be judged harshly if the FOIA allows evidence of their malfeasance to be known later. Certainly, the tarnished historical reputation of FBI chief J. Edgar Hoover was additionally diminished by the revelation of the FBI's decisions—at once both laughable and chilling—to monitor the lives of celebrities who posed no danger to national security. We identified two other important types of FOIA stories. The first (20 per cent of our sample) are stories based on government information about corporate, individual, or group malfeasance. Yet another common type of FOIA story (19 per cent) focuses on administrative refusals to grant FOIA requests, or when the government yields only heavily redacted versions. Clearly, the strong investigative tradition of American reporting leads journalists in the national newspapers at least to utilize the FOI law to uncover valuable information for their stories. The fact that papers will print stories when the government refuses to hand over the requested documents suggests strongly that the onus is on the government to defend its secrecy rather than on the press to define its right to discover. Through the press, FOIA can be a powerful tool for democratic accountability in a variety of ways. Does this accountability mechanism operate in the same way in France? Replicating the US study with French data, we selected three national newspapers (Le Figaro, Le Monde, and Libération) and searched for stories on FOI and CADA over a ten-year period ending in 2002. This yielded 122 stories for an average of twelve stories a year, which is a much lower rate per publication than in the US. About a quarter of the stories dealt with the law in general terms, and especially changes made to the law in 2000. Another quarter dealt with the state and public agencies. The most popular type of story, as in the US, focused on the exposure of bad management or the abuse of public money. But there were also stories about planning issues, police and prison problems, the right to obtain personal records, and health and environmental matters. A good example of citizen use of FOI in France is a story that appeared in Le Monde on 17 January 2002 concerning what the Department of Health
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
137
and Social Affairs (DDASS) knew about a factory linked to asbestos in Aulnay-sous-Bois. Operating between 1938 and 1991, the factory's operations left a fine white dust all over the neighbourhood, but especially on a school located near the factory. A resident, whose sister and brother-in-law had asbestos in their lungs, petitioned for documents from the DDASS about the factory for use in a trial. CADA granted the request. Citizen groups have also used the law. An association called ‘Un centre ville pour tous’ requested technical and financial reports to support its opposition to the plans by which the Marseilles town council intended to renovate the working class areas of Panier and Belsunce. Again, its appeal to CADA was successful. Most of the requests covered in the French newspapers came from groups or individuals, but the press instigated a few of them. A magazine, Sciences et Avenir, requested the reports of the Examination Commission for the Hospital Budgets in France, which were supposed to assess the management and quality of both private and public hospital treatment. Even though it obtained a favourable advisory opinion from CADA, it was unable to get the data it wanted for two hospital groups it suspected of having overrated reputations. All in all, it appears that the FOI laws in the US and France operate in somewhat similar ways. Although the press corps is a strong proponent of freedom of information, it is not the heaviest user of the laws. This seems to be especially true in France. But the information that groups and individuals uncover does make its way into the public realm through stories that are reported about their causes. The information that FOI requests unearth reveals things voters and citizens need to know about their governments that would not otherwise see the light of day. Our analysis of press coverage also encourages us to expand our thinking about FOI's effects. Not only do current governments come under scrutiny due to FOI, but so do previous governments and private entities about which the government has records. FOI therefore can be a tool to hold multiple actors accountable in a democratic system. Furthermore, as governments become ever more comprehensive repositories of information about the societies they oversee, FOI laws serve what Doyle (2001) terms an ‘archive’ function: they allow us to glean meaning from records that were not originally compiled in order to increase public information.
CONCLUSION The enactment of most national FOI laws was complete in OECD countries by the end of the 1990s. These new laws reflect a substantial shift in the predominant view (among citizens and elites alike) of the state's role from trusted guardian to merely that of an agent of the people that requires careful monitoring by citizens. Since governments sometimes withheld information to cover up malfeasance or to protect themselves from political embarrassment, citizens had to have the right to access that information in order to hold the government accountable for its
138
BRUCE E. CAIN, PATRICK EGAN, AND SERGIO FABBRINI
actions. Gone is the faith in the state's paternalistic role to determine what citizens should know about government processes and policies. Politicians in many countries run and win electoral campaigns by pledging to reform the public sector, to promote a new public management, to make the government more accountable to its citizens. At the same time, citizens are challenging administrative discretion and secrecy. Moreover, the adoption of these laws may be one of the unintended effects of globalization. In the 1990s globalization reached its apex. Especially among OECD countries, globalization supported the formation of transnational networks of cooperation and information between public officials, professional lobbies, and economic and financial actors from different nations. These networks proved to be formidable arena for the diffusion of best practices, for learning about the likely performance of specific reforms, and for the supranational training of administrative elites. Europe has been at the forefront of this process of integration and cooperation. Because the EU struggles with the problem of a democratic deficit, it looked for best practices in the crucial field of administrative accountability. Increased transparency in its decision-making processes was one of them. Lacking a specific electoral legitimacy, the European Commission pushed for opening its procedures and information to a growing number of interest groups, epistemic communities, business representatives, and member states’ officials. This need became particularly acute in the EU. The deepening of the integration process transformed the EU into an important public policy regime. A growing number of policy decisions, affecting EU member states’ courses of action, are now taken at the Community level. Many of these authoritative decisions are the outcome of policy processes that exist outside the formal Community institutions, involving both public and private actors. The boundary between government and governance has been blurred. This process is happening in all the OECD countries, an effect of both the complexity of the problems they have to solve and the transnational influences operating upon them. As the Italian case demonstrates, some of the innovative ideas forged in this dense network have been transferred into national settings that were not culturally and politically ready for them. As we noted, this is a problem that sometimes arises with reforms by emulation. In sum, there has been a dramatic expansion of the formal right to information in OECD democracies. In some cases, such as Italy, it appears to be a formal right only and not a real opportunity. But in other cases, the US and France especially, citizens and groups make frequent use of this right to obtain documents that would otherwise be denied to them. Some of the information serves noble purposes, such as reinforcing political accountability and checking corruption, but some of it serves more mundane private or commercial ends. It would be hard, if not impossible, for citizens to hold governments accountable for their actions if governments controlled access to critical documents. Governments will naturally seek to suppress information that might be harmful to their electoral prospects. But suppressing that information prevents voters from punishing bad decisions: they
TOWARDS MORE OPEN DEMOCRACIES: THE EXPANSION OF FREEDOM OF INFORMATION LAWS
139
cannot react to what they do not know. Although much of the information that FOI laws cause to be released is not important to electoral decisions, some of the documents are, and they can be important determinants of elections and policy-making. Clearly, a critical variable that helps to transform transparency opportunities into real political accountability is the press culture in a particular country. A free and aggressive press that is willing to challenge the government with investigative reporting will seize FOI opportunities to get important information that exposes government incompetence, inconsistency, and malfeasance. But, absent that tradition, the creation of FOI will not create a serious press challenge. Similarly, groups and individuals have to be willing to fight for their right to information, often taking their requests through several stages of appeal. If the political culture does not support and encourage that sort of behaviour, citizen participation will give only the appearance of having expanded.
7 Trends in Decentralization Christopher Ansell and Jane Gingrich In this chapter, we analyse trends towards decentralization in eighteen OECD nations. In a companion chapter (Chapter 8), we analyse administrative reforms that seek to achieve more direct accountability over public organizations. To describe those reforms, we distinguished between neo-liberal and democratic accountability reform agendas. In this chapter, we apply this distinction to decentralization. Advocates of each reform agenda agree that decentralization can increase the accessibility, the accountability, and the transparency of the state to its citizens, but differ on the mechanisms by which it does so. The neo-liberal agenda sees the devolution of authority from a central (monopolistic) authority to multiple authorities as a necessary step in creating a competitive market. In the field of economics, Tiebout (1956) developed this argument into a market model for analysing the provision of government services. This model suggests that multiple competing jurisdictions can produce an efficient provision of public services. Tiebout argues that, while jurisdictions retain the ability to collect a certain level of taxes in order to provide public goods, the mobility of citizens imposes a market-like discipline on taxation. If taxes are too high, citizens can move to new jurisdictions; if they want more public goods, they can choose a community with a higher tax rate. The democratic accountability agenda aims to enhance citizens’ access to, and direct participation in, the process of governance. Decentralization produces easy and direct access to government, making it possible for citizens to participate in the forms of advocacy democracy described in Chapter 1. In addition, as Chapters 2 and 3 have demonstrated, there has been an expansion of representative democracy and direct democracy at the sub-national levels of government. Access is easiest when government is within practical reach of the citizenry. Furthermore, people participate because they feel a sense of belonging to a community, which creates a public spiritedness that transcends narrow self-interest. From this perspective, institutions are held accountable by public sentiment and the direct participation of concerned citizens, rather than by market competition. Public participation is both a privilege and an obligation of citizenship, but only a small and immediate community can
TRENDS IN DECENTRALIZATION
141
act in such a public-spirited and active manner. Institutions must be decentralized to the level at which a meaningful participation can be sustained. As with administrative reform, the neo-liberal and democratic accountability agendas can overlap a great deal in practice. The goals of efficiency and democracy are called upon to legitimize decentralization reforms; but they are accorded different degrees of emphasis in the two agendas. We argue that the relative prominence of efficiency or democratic goals depends on institutional and partisan factors. This chapter begins by outlining a general institutional framework for describing variations in these reforms across advanced industrial democracies. We conclude by discussing the implications of these findings for the democratic process.
INSTITUTIONAL FRAMEWORKS Much of the impetus for decentralization comes from the rise of new social movements beginning in the late 1960s and from the fiscal crises of the 1970s and 1980s (see Chapter 1). In addition, internationalization, European integration, economic transformation, technological change, and democratic pressures have presented many common challenges to OECD nations. Although decentralization appears to be a common response to these challenges, countries have by no means adopted a uniform strategy (Wright 1994). In this section, we draw on Lijphart's (1999) typology of democracies to analyse variations in the extent and character of decentralization reforms. We distinguish four dimensions of institutional variation: (1) the constitutional status of sub-national government; (2) the electoral rules that shape the formation of the political executive; (3) the administrative structure of the state; and (4) corporatist traditions. We argue that different institutional structures will shape the extent of decentralization and its democratic character. First, Lijphart distinguishes between federal and unitary states. In federal states lower levels of government have a constitutional guarantee of sovereignty over specific matters; moreover, these states are typically bicameral, with the upper legislative house directly representing the interests of the regions as corporate actors. In contrast, lower-level governments in unitary states derive their statutory frameworks from parliamentary legislation or an executive order rather than from constitutional authority, and lower-level governments are not directly represented in national legislatures. Lijphart interprets federalism as a counter-majoritarian institution, and his analysis represents a legislative perspective on the federal-unitary distinction. We expect unitary states to produce more comprehensive reform movements and to present more structural potential for decentralization. In unitary states, where spending by lower-level government ultimately remains the responsibility of the national government, we expect top-down decentralization reforms to be particularly concerned with efficiency. Where the impetus for decentralization comes from below (for example, from regional nationalists), the reforms will be oriented in a more democratic direction. In federal systems, fiscal problems may
142
CHRISTOPHER ANSELL AND JANE GINGRICH
be equally severe, but the responsibility for these problems is already distributed across different levels of government. Moreover, federal states already provide the institutional framework in which incremental changes in decentralization may be accommodated. Unitary states, in contrast, require relatively systematic reforms before such incremental tinkering can begin. Second, Lijphart distinguishes between majoritarian and consensual democracies. In majoritarian democracies, dominant parties alternate in power. In consensual democracies, by contrast, successive multi-party coalitions govern. Given internal party discipline, strong party leadership, and clear electoral mandates, we expect majoritarian democracies to produce more systematic and comprehensive decentralization reforms than consensual democracies. In majoritarian systems, voters will hold governing parties—of both the left and the right—directly responsible for fiscal affairs. Hence, we expect majoritarian parties of left and right to use decentralization as a mechanism for enhancing efficiency. In addition to a common concern about efficiency, however, we also expect decentralization reforms in majoritarian countries to bear a clearer programmatic stamp, the exact contents of which will depend upon the ideological orientation of the party in power. Majoritarian parties of the right will more wholeheartedly embrace the neo-liberal agenda, while those on the left will frame decentralization reforms in terms of enhanced democracy. By contrast, as a reflection of the negotiated character of coalition politics, we generally expect consensual democracies to produce more incremental and less programmatic decentralization.1 Third, the organization of the state's administrative structure has an impact on the course of decentralization reform. The ‘Napoleonic state,‘ for example, integrates the nation through a single territorial administrative structure.2 Subnational units (for example, ‘departments’) represent subdivisions of the national administration and are governed by a ‘prefect’ who is an agent of the central government. From the administrative point of view, the vertically integrated structure of the Napoleonic state makes it an archetype of a unitary state. We expect territorially based, vertically integrated administrative systems to produce greater demand for comprehensive, democraticallyoriented decentralization reforms. As in all unitary states, the responsibility of Napoleonic states for local government expenditures may produce top-down administrative reform strategies with efficiency-oriented goals. For such vertically-integrated administrative structures, ‘off-loading’ responsibilities onto lower-level governments is not an effective solution to the fiscal problems of the central state. Seen from below, the comprehensive, centralized administrative structure of the Napoleonic state is also likely to generate comprehensive demands for decentralization. These demands from below for decentralization reform are likely to be oriented toward democratizing the administrative state. By contrast, functionally organized administrative systems do not, as a by-product of their own organization, produce territorial communities that can demand greater self-government. Demands for greater democratic accountability are as likely to reinforce centralization as they are to produce decentralization.
143
TRENDS IN DECENTRALIZATION
Table 7.1.Expected reform paths of OECD countries, by institutional type Unitary Majoritarian
Consensual Napoleonic Belgium, France, Ireland, United Kingdom, New Zea- Italy, Japan land (pre-electoral reform) Comprehensive pro- Comprehensive degrammatic reforms; centralization reoverall bias toward forms; democratically efficiency, but com- oriented reforms when decentralization bined with democratic goals when left demands come from below. in power.
Federal Majoritarian
Corporatist Denmark, Finland, Australia, Canada, Netherlands, Norway, USA Sweden Reforms marked by democratic agenda, while central government remains concerned about efficiency; corporatism reproduced as inter-governmental bargaining.
Less likelihood of dramatic decentralization, overall bias toward efficiency, but democratically oriented when left party in power.
Consensual Austria, Germany, Switzerland Least dramatic and programmatic reform; incremental change likely to occur.
Finally, corporatist traditions are a fourth institutional feature likely to effect decentralization agendas, particularly in unitary states. For example, states with strong corporatist traditions, like the Scandinavian countries and the Netherlands, are often described as unitary decentralized states. Their party systems are highly proportional and relatively fragmented, and these states produce unity through corporatist institutions and bargaining. Corporatism represents a strong mobilization of society through associational membership. With the erosion of corporatism through national ‘peak’ associations, however, corporatist institutions may themselves undergo decentralization and be partially replaced by local-level institutions. Municipal governments may replace functional peak associations as new actors in national-level corporatist bargaining (bargaining with national governments over funding). In this context, decentralization reforms are likely to be concerned with maintaining political participation in collective decisionmaking. Consequently, we expect corporatist traditions to push decentralization reform toward the democratic accountability agenda. Table 7.1 summarizes our expectations about variations in decentralization reforms produced by different institutional configurations.
STATE DECENTRALIZATION We begin our empirical analysis with a broad overview of decentralization trends. Table 7.2 demonstrates that several advanced industrial democracies have introduced a new tier of elected government. The most dramatic reform since 1965 has without doubt occurred in Belgium, which transformed itself from a unitary into
144
CHRISTOPHER ANSELL AND JANE GINGRICH
Table 7.2.Creation of new levels of sub-national government Number of sub-national governments,
a b c d e f g
h i j k l
m
Australia Austria Belgium
1965 3 a 2 2
2002 3 2 4
Canada
2
b
2
Denmark Finland
2 2d
c
2 2
France
2
e
3
Germany
2
f
2
Ireland
1
g
2
h
Italy
2
3
Japan Netherlands New Zealand
1 2 2j
i
1 2 2
Norway
2
k
2
Sweden Switzerland United Kingdom
2 l 2 m 2
2 2 2
United States
2
2
Description of significant changes/creation of new levels of government n/a n/a Constitution amended to recognize regions and communities and establish political institutions in the cultural communities (1970), the regions (except Brussels) (1979), in Brussels (1988); Belgium officially becomes a federal state (1993) The North-West Territories divided to form a third territory, Nunavit, which allows for some degree of Inuit selfgovernment (1999) n/a New regional institutions (Regional Council) (1994) Decentralization Act (1982) established regions as public authorities and increased the administrative freedoms of the municipalities Addition of new Länder during the process of German unification (1990) Eight regional authorities established (1994) Regional authorities are established (1970) n/a n/a A substantial reorganization of regional authorities includes the establishment of elected councils for the regional authorities (1989) Directly elected county councils established (1976) n/a n/a Regional assemblies established in Scotland and Wales (1999). The two-tiered system of local government was replaced with a single tier in England (1992), Wales (1996), and Scotland (1996) n/a
Municipal and Länder. There are also administrative districts at the sub-national level. Municipal and provincial governments. Some provinces also have two tiers of sub-provincial government. Two-tiered system (municipalities and counties). Municipalities and provinces. There are also administrative districts at the sub-national level. Municipalities and departments. Local (municipal, county) and Länder. Ireland technically only has one level of sub-national administration, the county; however, there are sub-county bodies (boroughs, districts, and towns). Provincial and municipal governments. Five ‘special regions’ were established in 1948. Municipal governments (cities, towns, and villages). There are also prefectures in Japan that play a key role in sub-national administration. Regional and territorial authorities (district and city councils). Two-tiered system (municipalities and counties). Canton and municipal government. Several different types of municipalities exist, and some cantons have established other forms of administrative districts. Two-tiered system (county and district councils). Source: OECD (1997).
TRENDS IN DECENTRALIZATION
145
a federal state in 1993. Italy (1970) and France (1982) also produced significant reforms, creating distinct regional levels of government with elected representation and administrative autonomy within the framework of a unitary state. In 1999, the UK created regional parliaments in Wales and Scotland and Northern Ireland (although the latter is now suspended). We also see since 1965 a significant shift of functions to lower levels of government (OECD 1997), with education, social services, and regional planning as the most commonly transferred functions. Table 7.3 and 7.4 illustrate shifts in responsibility for policy and organizational/fiscal autonomy to lower levels of government. Additionally, in Japan, decentralization is on the agenda but has not yet materialized. The Japanese Decentralization Promotion Law of 1995 created a commission to study the problem. The only significant shift towards centralization occurred in the case of the UK under the Conservative governments of Thatcher and Major. A more selective transfer of functions from local or regional governments to central governments has occurred in a few cases: in Ireland, for example, environmental protection functions were transferred to a national agency.3 Table 7.3.Decentralization of responsibility to sub-national government OECD country Austria Belgium Canada Denmark Finland
France
Germany Italy Netherlands Sweden
United States Sources: OECD (1997); Council of Europe (1998).
Examples of devolution of responsibility to lower levels of government Responsibility for housing transferred to the Länder; lower levels of government granted participation in EU decision making (1992). Powers transferred to new regions and communities, constitution amended to recognize this (1980). Administration of criminal code transferred to the provinces. Ongoing transfer of responsibilities to the counties and municipalities: for example, social security, regional planning, health services, environmental quality, secondary schools. Responsibilities shifted to the municipalities: Planning and Building Act (1958), National Health Care Act (1972), Social Welfare Act (1982), Comprehensive School Act (1983). Regional planning and environmental policy transferred to the new regional institutions. ‘Deconcentration Charter’ increased role of sub-national administration in implementing central government policy (1992). Education, regional planning, local support transferred to the regions and departments. Law on cooperation between the Federation and the Länder increases the role of the Länder in EU decision making (1993). Manpower training, health care, agriculture, transport, environmental protection, and economic development are transferred to the regions. After a period of municipal amalgamation, a range of responsibilities transferred to the local level: for example, nursing homes (1988), services for the mentally disabled (1991). Transfer of responsibility to municipalities: social services (1982), health (1983), conservation and natural resources (1987), planning and building (1987), education (1991). Municipalities given increased responsibility for long-term medical care (1992). Responsibility for welfare transferred to the states (1996).
146
CHRISTOPHER ANSELL AND JANE GINGRICH
Table 7.4.Increases in sub-national organizational/fiscal autonomy OECD country Australia Belgium
Canada Denmark Finland
France Ireland Netherlands New Zealand
Norway
Sweden United States
Examples of increased autonomy of lower levels of government Northern Territory (1978) Australian Capital Territory (1989) granted self-government Funding Act assigned regions a portion of income tax, federal solidarity contributions, and non-tax revenue; communities also given sources of revenue (1989), and taxes on radio and television licensing become community fees (1993). The Constitution Act (1982) gives the provinces the power to amend their own constitutions. Move to block instead of tied funds for health, post-secondary education, and social services (1996–7). Consolidation of health grants into block grants (1985), local governments given more freedom in administrative reform (1989). Municipalities given power to collect tax on real estate (1993), and there is a move towards lump sum instead of earmarked transfer funds. New Municipal Act gives municipalities greater freedom to determine their own structure and agree on arrangements between themselves, but requires them to have a municipal board, manager, and auditors (1995). Sub-national governments granted the power to borrow where they please (1990). Domestic property tax abolished and replaced with Rate Support Grant (1978) giving local government some increased control over resources. Municipal Act revised, providing more differentiation among government. Rating Powers Act (1988) grants local government more power to raise taxes on property. Amendments to the Local Government Act (1989) require local government to produce long-term financing strategies and funding policy, but permit them to borrow on the open market. Move to block instead of tied grants: hospitals (1980), municipal health and social services (1984), General Purpose Grants Scheme replaces 50 earmarked grants (1986). Local Government Act (1992) contains some centralizing measures but allows local government increased freedom in the structure of council and committees and increased control over the structure of taxation. New Local Government Act extends the principles of free municipalities to all municipalities granting increased organizational autonomy (1992). Unfunded Mandates Reform Act prevents the federal government from mandating activity without accompanying funds (1995).
Sources: OECD (1997); Council of Europe (1998).
Various measures can be used to probe the real extent of this decentralization. One measure is the share of public employment by level of government. Among those nations for which OECD reports the share of public employment for 1985/90–2000, the central government's share has grown at a slower rate (or decreased at a faster rate) than that of sub-national government in Austria, Canada, Finland, France, Germany, and the US. It has grown at a faster rate in Ireland, the Netherlands, and New Zealand (see Appendix A). Among the nations for which OECD reports share of public employment only for 1985/90 and 1994, the central government's share has remained relatively constant in Australia, Italy, and the UK, and decreased in Denmark, France, and Sweden (OECD 1997). Fiscal resources are another measure of effective autonomy and decentralization. Between 1990 and 1997, subnational governments’ share of total public
TRENDS IN DECENTRALIZATION
147
expenditure has increased in Austria, Ireland, Italy, New Zealand, Norway, and the US. During the same years, it decreased in Australia, Belgium, Canada, Denmark, Finland, France, Germany, Netherlands, Sweden, Switzerland, and the UK (OECD 1997).4 A third measure of decentralization is the relative share of tax revenue received by the sub-national government. Between 1975 and 2000, sub-national governments’ share of total tax revenues has increased in Australia, Belgium, Canada, Denmark, France, Italy, and the Netherlands. In the same period, sub-national governments’ share of total tax revenues decreased in Austria, Finland, Germany, Ireland, Norway, New Zealand, Switzerland, United Kingdom, and the United States (Appendix B). An important indicator of the relative autonomy of sub-national government is its ability to raise its own revenue. The OECD reports the percentage of resources that derive from the sub-national government's ‘own receipts’ as a percentage of total resources in 1980 and 1994. The percentage of ‘own receipts’ increased in Australia, Belgium, Canada, Denmark, Finland, France, Germany, Ireland (slightly), Italy, Japan, Netherlands, Sweden, and the US, and decreased in Austria and Switzerland (OECD 1997). On the basis of these measures, the conclusion must be a cautious presumption of a trend toward decentralization. However, there is a good deal of conflicting evidence. The aggregate data may not capture the significant shifts that have occurred, as measures of fiscal decentralization are problematic in a number of ways.5 To gain a clearer picture of the extent and nature of reforms, we now turn to an analysis of the decentralization reforms in individual nations.
INDIVIDUAL COUNTRY ANALYSIS In this section, we examine the trajectories of decentralization and de-concentration reforms across the five different categories of states: Napoleonic, corporatist, unitary majoritarian, federal-consensual, and federal-majoritarian.
The Napoleonic States There has been significant decentralization across several of the Napoleonic states, including France, Italy, and Belgium, while decentralization in Japan has been limited. In France, the national state tinkered with forms of regional government throughout the 1960s and 1970s, but there were no major decentralization reforms until 1982. The 1982 reforms created a regional level of government composed of multiple departments and a new tier of elected government. This act also restricted multiple office-holding among sub-national officials and reduced the power of the prefect. The 1992 ‘De-concentration Charter’ gives localities greater responsibility to implement policy and strengthens the role of the prefect in coordinating policy. This wave of de-concentration was intended not only to solve some of the
148
CHRISTOPHER ANSELL AND JANE GINGRICH
coordination problems within the system but also to increase local democracy. While democratization has been one of the goals of reform, economic policy reform was also important. It is significant, though, that support for decentralization and de-concentration exists across party lines. Initial reforms were implemented under the Mitterrand presidency, but de-concentration continued under both Socialist and Gaullist prime ministers. The post-war Italian constitution enshrined the status of the regions, but they were not established until 1970 (with the exception of five ‘special regions’). The devolution of responsibility to the regions was delayed in part because of political antagonism between the major parties, but by the 1970s there were increasing political demands and need for reform. Following the creation of the ordinary regions, a number of responsibilities were transferred to the regions during 1975–7. The creation of the National Health Service further transferred wide management autonomy to compulsory communal associations. However, 1972 reforms of municipal government finances decreased the autonomy of municipalities over their revenue and expenditure, and the financing of sub-national government remains heavily dependent on the centre. While questions have been raised about the (uneven) capacity of regional governments (Hine 1993), the regions have evolved and the municipalities have retained important functions (this is less the case with the provinces). The Italian experience of decentralization illustrates an uneven, but nevertheless clear, path towards democratization, with the creation of a new elected regional tier and the gradual transfer of responsibilities to it. Decentralization and federalization in Belgium has been evolving since the 1960s, with the impetus for decentralization arising from cultural and economic demands. Political pressure for devolution began to mount in the late 1960s, and a set of constitutional reforms in 1970 established three cultural communities (French, Flemish, and German) and regions. These reforms served to distinguish both geographic regions and linguistic communities (Witte 1992). The most significant changes, though, occurred in 1988. As nationalism became increasingly prominent, the federal solution seemed more attractive to the major parties than the breakup of the country (O'Neill 1998). An agreement was reached between the five parliamentary parties that significantly decentralized functions to the regions and settled the contested status of Brussels. The 1993 constitutional revisions finally named Belgium a federal state, and the first elections to the regional parliaments were held in 1995. The Belgian experience illustrates dramatic decentralization in response to discontent from below, and this decentralization has taken on a democratic character with the advent of elected sub-national bodies. Japan is difficult to categorize within our framework; however, its administrative structure best approximates the Napoleonic model. Post-war Japan has tended towards centralization, with national legislation and central bureaucracies encroaching on various local matters and restricting the power of the municipalities and prefectures (Osamu and Wright 1998). This trend encountered some resistance, but it was not until the late 1980s and the 1990s that actual legislation
TRENDS IN DECENTRALIZATION
149
promoting decentralization was introduced. Pressures for decentralization arose from below as increasing economic differentiation, the rise of progressive mayors, the Freedom of Information Act, and local initiatives created important and innovative policies at the sub-national level (Takao 1998). However, despite the 1995 Decentralization Promotion Law, actual decentralization has been extremely limited. This analysis of Napoleonic states provides partial confirmation of our hypothesis. Certainly, in all of the cases reviewed some decentralization has occurred. Although partisan dynamics have played an important role in the process of decentralization (particularly in Italy), it has been upheld across partisan lines and reflects the consensual institutional structure of these states. In France, Italy, and Belgium, we see a combination of pressures for decentralization stemming from both above and below, with the desire to contain expenditure and respond to fiscal pressures combining with social demands for regional governance. The trend toward devolution is seen also in two Napoleonic states not reviewed in this volume, Spain and Greece, both of which have established in the past several decades elected intermediate levels of government.6 This trend is not unproblematic or unidirectional (as the case of Japan illustrates); however, there is a general trend towards increasing the democratic character and responsibilities of intermediate levels of sub-national authority across the Napoleonic states, thus providing greater access points for citizens and increasing the accountability of sub-national governments.
Corporatist States There are a number of broad trends in public sector reform across all four Scandinavian countries. With the expansion of the welfare state in the post-war era, the role of the central government dramatically increased. A series of reforms in all four Scandinavian states led to municipal amalgamation between the 1950s and the 1970s. These reforms reduced the number of municipalities and counties, but allowed the central government to shift responsibility for the implementation of welfare programmes to the lower levels. Measured as a percentage of Gross National Product (GNP), Denmark, Sweden, Finland, and Norway (in that order) lead Europe in decentralization (Local Government Denmark7). This process of amalgamation and increasing local responsibility was followed in the 1980s by extending local autonomy. All four countries began to move from conditional to block grants or towards increased local taxing power in the 1980s, thus giving local communities more control over their finances. The Scandinavian countries all experimented with ‘free communes’ in the 1980s, which amounted to granting pilot local governments more control over their organizational structure. This led to reforms in local government laws in all four countries in the 1990s, with greater autonomy extended to local governments. Though the general trend towards increasing decentralization has been partly fiscal or efficiency driven,
150
CHRISTOPHER ANSELL AND JANE GINGRICH
attempts to increase citizen participation and involvement have been important in all four countries (Rhodes 1999; Hansen 2001; Christensen and Lӕgreid 1999). Blom-Hansen (1999) argues that, despite high levels of local government spending, central governments in the Scandinavian countries retain much macroeconomic control, using a combination of carrots and sticks. In Sweden the approach to controlling local government spending has traditionally been consensual; however, in 1991 the central government imposed a freeze on local governments’ ability to raise taxes. In Denmark, the associations of municipalities and counties meet regularly with the central government to enact spending ‘agreements’; however, punitive measures were introduced in the mid-1980s to control local government spending (but were used only once in the 1980s). There has also been some centralization alongside the general decentralist trend, including central government activism to ensure comparable quality standards in service delivery across local governments, and central governments mandating responsibility in certain arenas. It is also important to note significant variation among the Scandinavian countries. For example, Sweden's turn towards market-like mechanisms in the provision of education and health care have not occurred in Norway, and reforms in Norway have been less structured and more incremental than those in Sweden (Christensen, Lӕgreid, and Wise 2002). The Dutch state has some elements of a Napoleonic administrative structure, but it is better classified with the Scandinavian countries as a decentralized unitary state with strong corporatist traditions. The decentralist trend in the Netherlands is, however, more recent and less pronounced than in Scandinavia. Indeed, the trend up until the 1980s was towards centralization rather than decentralization, and it was not until the 1980s that serious proposals were enacted, by both the centre-right and subsequent centre-left cabinets. The stated objectives for decentralization were fiscal reform and enhanced citizen access to government. The trend in the Netherlands has been towards coadministration and co-governance between the levels of government, and many of the changes experienced in the Netherlands have occurred outside formal changes in legislation. The corporatist countries also provide mixed results for our hypothesis. In all five countries change has been gradual and consensual. For example, the Scandinavian countries all experimented with ‘free communes’ and tested the results of increasing municipal autonomy before enshrining it in law. The bargained character of reform is also notable in some of the countries, with Finland and Norway both illustrating a consensual reform path marked by few partisan effects (Pollitt and Bouckaert 2000). However, Sweden and the Netherlands both display strong partisan influences in the reform path, with the centre-right governing coalitions in both countries introducing significant variations in support for neo-liberal principles. The reforms in the corporatist countries have generally taken on a democratic character, increasing the power and autonomy of elected local governments; however, Swedish reforms also contain neo-liberal elements.
TRENDS IN DECENTRALIZATION
151
Finally, there is a replication of corporatist bargaining between levels of government, with consensus-based bargaining across levels of government. This fits with our general argument that decentralization in corporatist countries may reproduce the corporatist style in intergovernmental relations.
Unitary Majoritarian Much has been written about the reform of local government in the UK, particularly during the Thatcher era. As elsewhere, the fiscal crisis of the mid-1970s led to a degree of reform and modernization at the local level; however, the election of the Thatcher government in 1979 led to a radical shift in the institutional environment in which local governments operated. Some sixty pieces of major legislation under Thatcher directly dealt with local government authority, many of which reduced the autonomy and authority of local government (Clarke and Stewart 1993). It is clear that many reforms did centralize authority or limit local autonomy. The financing of local authorities and control of their expenditures became increasingly centralized and a growing number of ‘quangos’ (quasi-autonomous non-governmental organizations) assumed many of the responsibilities formerly carried out by elected bodies. There is some question of the extent to which the reforms under Thatcher and Major decentralized power from local authorities to the people themselves, through measures like Citizens’ Charters and the privatization of council housing. However, reforms such as the outright abolition of the Greater London Council and the six metropolitan counties illustrate a clear centralist tendency. The Major government began to move towards increasing the role of sub-national authorities. However, major changes did not occur until the Blair government established regional assemblies in Wales and Scotland. The financial power of these assemblies is limited, but they do form a new tier of elected regional government. The Blair government has also passed legislation allowing local authorities to hold referendums for establishing an elected mayor. New Zealand has radically altered the structure of both its political and administrative systems over the last two decades, and has gone the furthest in introducing market-based mechanisms to its administrative system (Halligan 1997). The 1989 amendments to the Local Government Act introduced increased private contracting, led to the application of new managerial and accounting techniques at the local level, and drastically reduced the number of local authorities while reinterpreting their competencies. A new tier of elected regional authorities was introduced and territorial authorities received a broadened competence; thus some decentralization accompanied the massive structural reorganization (Wallis and Dollery 2001). Another set of decentralizing reforms granted power to Maori communities in 1989 (Jacobs 1997). Most of the reforms in New Zealand have been dominated by fiscal concerns, the desire for efficiency, and a market-based ideology, but participatory
152
CHRISTOPHER ANSELL AND JANE GINGRICH
elements were included and served an important legitimizing function (Cousins 1999). New Zealand has traditionally been, and remains, a highly centralized state; however, recent reforms do indicate limited amount of decentralization. Decentralization in Ireland has been piecemeal. Local government statutes remained largely unchanged from the preindependence period until 1978, when local authorities were given new powers to levy charges following the abolition of local property tax on domestic housing (1978) and agricultural property (1983). Fundamental change in Irish administration did not occur until the 1990s. Re-emergent political interest in devolution has accompanied recent budgetary debates, particularly in light of pressures from the European Union for the creation of regional authorities. Reforms passed in 1991 and implemented in 1994 created eight regional authorities. These institutions remain somewhat lightweight politically, and local autonomy in Ireland remains limited. Our expectations here are only partially confirmed. The Irish case illustrates that the trend towards decentralization remains far from universal, even in the unitary majoritarian countries where the potential for radical reform is greatest. While some of the most radical administrative reform in the OECD has occurred in New Zealand and the UK, this has primarily been focused on reorganization of the public sector as a whole, and decentralization played only a small role. The experience in the UK and New Zealand illustrates reform motivated primarily by concerns over efficiency rather than democratization, and confirms our expectation that majoritarian institutions allow for more radical reform. The experience of the UK fits into our broader argument that unitary-majoritarian systems reflect partisan programmes. But this argument is complicated by New Zealand's experience, where the original neo-liberal reforms were implemented by a Labour government.
Federal Consensual Public sector reform in Germany has been incremental. The initial structure of government laid out in the Basic Law was highly decentralized, but the period following the Second World War saw some centralization reflecting the centralizing of functions related to the rise of the welfare state. However, rather than strictly separating tasks at different levels of government, the cooperative federal structure tends to separate policy making (Federal) from policy implementation (Länder). Some modernization has occurred at the local level, with the amalgamation and consolidation of many smaller units of government. The late 1970s and the 1980s were a period of incremental reform, and Hendriks and Tops (1999) note the absence of New Public Management (NPM) reforms. However, this period also witnessed a rise in grass-roots politics, which resulted in territorial decentralization to committees or councils of the neighbourhoods, self-help activities, juries of citizens to provide advice, and institutionalized platforms for discussion
TRENDS IN DECENTRALIZATION
153
between local government and citizenry. The NPM philosophy increased in popularity at the local and national levels in the 1990s, particularly with the new fiscal pressures of German reunification. Reunification did not create significant structural reform, however, and the impact of reunification on the relative strength of Länder vis-à-vis the federal government remains contested. The German case illustrates a combination of democratic- and market-oriented reforms that have resulted in little decentralization. Reform in Switzerland and Austria has also been incremental. Switzerland is the archetype of Lijphart's consensual democracy, combining a consensual mode of decision-making with federalism. Swiss cantons have residual power, and are responsible for the implementation of most federal programmes (Kloti 2001). Indeed, since Switzerland was so decentralized at the outset of the post-war era, the trend has been toward greater centralization. On the other hand, while Austria is also a consensual democratic system, the power of the central government is particularly strong for a federal state. The federal government retains residual power and has relatively strong supervisory powers over the Länder. Austria also maintains strong corporatist institutions, creating an additional centralizing influence. Thus, in both Austria and Switzerland the last twenty-five years have witnessed an incremental centralization of power (with the growth of federal tasks) rather than decentralization. Both Austria and Switzerland have experimented with various administrative reforms, but neither has implemented major decentralist reforms (Kloti 2001). The Swiss, Austrian, and German cases confirm our expectations. There has been little change, and the change that has occurred takes the form of incremental modernization rather than comprehensive reform. Democratic legitimization of reforms differs across the three countries. Switzerland enjoys a tradition of direct democracy and local participation, while the Austrian system contains fewer access points for citizens. The trend in Germany and Austria has been towards more open government, but this has not been accomplished through decentralization. Thus, the incremental and consensual logic of reform dominates, and the existing federal structure moderates the need for decentralization.
Federal States: Majoritarian In Canada, provinces retain constitutional jurisdiction over most social policy areas, but the expansion of the welfare state from the 1950s to the 1970s increased federal involvement in these areas. Beginning in the 1980s, decreasing federal financing of social programmes and increasing provincial government deficits became a sign of federal withdrawal from areas of provincial and joint responsibility. In conjunction with reduced federal funding, however, there has been a shift from conditional to block-grant funding, giving the provinces more fiscal autonomy. Both federal and provincial governments have experimented with reforms aimed at increasing efficiency, with both decentralization and consolidation of
154
CHRISTOPHER ANSELL AND JANE GINGRICH
services as the outcome. Provincial governments, in particular, have looked to rationalize the delivery of health care services, often turning to a reduction in the number of delivery points while transferring power to regional health care boards. The Patriation of the Constitution in 1982 has been significant for relations between levels of government. The Quebec government's decision not to sign the Constitution added to the ensuing constitutional crisis, resulting in two failed constitutional amendments that would have further increased the powers of the provinces. This has also led to a debate over Aboriginal rights to self-government, and several Aboriginal tribes have negotiated self-government agreements with provincial governments. In 1999, a new northern territory, Nunavit, was created which allows for a degree of Inuit self-government. In sum, there have been several overt attempts at decentralization and deconcentration of government services; however, much of the increased autonomy at provincial and local levels has occurred because of the reduction of financial support from higher levels of government. Australian States have a great deal of policy, but not fiscal, autonomy, and there has been a gradual trend towards centralization over the last several decades (OECD 1997). Although there were demands in the 1990s for States to institutionalize a ‘cooperative federalism’, constitutional amendments proposing more cooperation have been rejected in referendums. Martin Painter (2001) argues that this lack of cooperation illustrates the susceptibility of the joint decision process to veto power, which can create constant gridlock. Although the Australian state has enacted less drastic reform than New Zealand, it has been a leader in implementing NPM reforms. While a recent rhetorical shift in central government towards the concept of community-based voluntarism may mark a move away from the NPM-based emphasis on efficiency-centred technocracy, Australian participatory policy remains wedded to the diffusion of citizens’ ‘charters’ and other such reforms more indicative of the neo-liberal reform agenda (Adams and Hess 2001). Thus, the degree of decentralization reform remains limited, and the nature of reform is more oriented towards efficiency than democracy. Much of the contemporary dialogue on decentralization and citizen participation in the United States turns on the apparent successes and failures of the so-called devolution revolution—a term coined by Richard Nathan (1996). Nathan argues there has been a secular trend away from centralization, beginning under former President Richard Nixon in the 1970s. This dovetails with the broad NPM axioms advanced by Osborne and Gaebler (1992), who emphasize the role of citizen as consumer and demand greater efficiency and flexibility at the State and local levels. These concerns were mirrored by the National Performance Review, established in 1992 under Vice-President Gore, whose first report in 1993 called for the decentralization of decision-making, bureaucratic simplification, and an increased customer-service orientation in sub-national government (Kellough 1998). This agenda increased in prominence when the Republican Party captured
TRENDS IN DECENTRALIZATION
155
Congress in 1994 with the Gingrich caucus demanding the conversion of federal entitlement payments into block grants, essentially turning welfare over to the States. The passage of the Unfunded Mandates Reform Act in 1995 forbade Congress to pass regulatory mandates without providing the necessary funding for States to conform to them. Yet the jury remains out as to whether the ‘devolution revolution’ is deserving of the name. Although federal mandates have been reduced, they have commonly been replaced by State mandates, and the ‘revolutionary’ content of devolutionary reforms like the Balanced Budget Act of 1997 may have been mitigated by policies of compromise (Tannenwald 1998). The level of decentralization in the federal majoritarian countries generally confirms our expectations. First, decentralization is less pronounced than in a number of the unitary states. Federal institutions leave the central government with both less capacity to enact wide scale decentralization and less need to do so as the system is already decentralized, and thus the level of decentralization is lower and less uniform than elsewhere. The partisan nature of reform is also more pronounced than in consensual systems, with centre-right governments in the US, Canada, and Australia being the biggest proponents of a neo-liberal decentralization agenda.
EDUCATIONAL DECENTRALIZATION Decentralization is a highly variegated phenomenon and difficult to grasp by looking only at the broad reforms. Therefore, in this section we focus our analytical lens on the specific domain of education. Educational decentralization is a useful focus because it reveals both the possibilities and the limits of the link between decentralization and democratic participation. Decentralization does not necessarily mean greater democratization of schools or even greater community control over schools. Decentralization can shift decision-making authority to teachers (professional control), to principals, or to parents, and can take the form of decentralization from central education ministries to local municipal governments or from local school districts to individual schools. In fact, Walberg et al. (2000) identify twenty-two meanings of ‘decentralization’ in education. Hence it is important not to presume that decentralization is equivalent to the democratization of school decision-making. One of the most important frameworks for educational decentralization is what has been called School-Based Management (SBM). Beginning in the 1980s, SBM has sought to decentralize decision-making to the school level, including important curricular, budgetary, and staffing decisions. SBM seeks to make schools ‘self-managing’, hence reducing the need for centralized bureaucratic administration. SBM does not necessarily lead to the democratization of schools, because greater school autonomy may imply greater decision-making autonomy for principals or
156
CHRISTOPHER ANSELL AND JANE GINGRICH
expanded authority for teachers. Many, however, see SBM as necessarily including a greater role for community participation in school-based decisions. Abu-Duhou (1999) argues that the core of SBM is ‘participatory decisionmaking at the school site’. SBM is a good illustration of our more general argument that neo-liberal and democratic accountability agendas are distinct strategies of administrative reform that often intertwine in practice. Both the market-oriented and the democratically oriented aspects of these reforms are intended to increase the direct accountability of schools to parents. Voucher systems, for example, exemplify the market-oriented approach to SBM. Voucher systems seek to empower parents as consumers of education by creating a quasi-market in the provision of education. Schools are given decision-making autonomy with respect to curriculum, personnel, and budget allocation and then encouraged to compete for student enrolments. Competition among schools is designed to encourage them to improve the quality of their services to consumers. In contrast, the school ‘site council’ exemplifies the democratically oriented SBM reforms. The site council approach treats parents as citizens of the school who need to be represented as stakeholders (along with teachers, the principal, and sometimes students) in a democratic community. In theory, the site council becomes the main organ of school governance. Whereas vouchers produce direct accountability to parents by granting parents the power of ‘exit’, the school site council aims to produce it by granting parents the power of ‘voice’. The unusually comprehensive SBM reforms initiated in the Australian State of Victoria between 1992 and 1996 illustrate the way in which market and democratic mechanisms intertwine in practice. The reform devolved administrative, financial, and personnel matters to schools, while at the same time centralizing certain accountability and standard-setting functions. A centre-right government, operating under financial crisis, sought to shift from a ‘managerial’ to a ‘contractarian’ style that fits our description of neo-liberal reform. At the same time, the Victorian reform also sought to empower parents to contribute to school governance through strengthened school councils (Pascoe and Pascoe 1998). The Charter School movement in the US provides another example of the intertwining of market and democratic mechanisms in practice. Abu-Duhou (1999: 52) argues ‘charter schools subscribe to and embody the democratic process of the common school; decentralization; a balance of autonomy and accountability via the charter; and a more market driven system’. Table 7.5 summarizes the major educational decentralization reforms in OECD countries in the last two decades. A 1995 OECD report analyses the distribution of decision-making authority across different levels of governance. The report notes that only a few countries (Germany, Switzerland) had not reallocated decision-making authority between 1975 and 1995 (OECD 1995: 10). However, this report finds variation in decentralizing reforms, with some transferring power from the national to the regional level (Belgium), from the intermediate levels to the extremes (England,
TRENDS IN DECENTRALIZATION
157
Table 7.5.Major OECD education decentralization reforms OECD country Austria
Australia Belgium
Canada Denmark
Finland
France
Germany Italy Japan New Zealand Netherlands
Sweden
Switzerland United Kingdom
United States
Reforms This highly centralized education system began taking steps toward decentralization in 1993 with devolution of authority over hours, curriculum content, and class organization to schools; school partnerships involving parents and teachers were created. Systemic devolution to schools in the State of Victoria, 1992–7; market-oriented but with increased participatory role for parents (Pascoe and Pascoe 1998). Responsibility for educational matters transferred to the communities (Flanders and Wallonia) and the central government no longer has a major role in educational matters (1989). Since 1991, local school councils in Flanders have been given increased decision-making power over matters such as personnel hiring (Amelsvoort 1995). School-Based Management experiment in Edmonton beginning in 1970s was extended to all Alberta beginning 1994. All provinces except British Columbia have taken steps to create school councils, with Yukon going the furthest. Decentralization of much decision-making authority from the national to the municipal level and from the municipal to the school level (1990). Authority and responsibility now shared by municipalities and school boards. 1994 reforms give greater authority to school boards and increase the involvement of parents in school governance. Reforms in 1983, 1985, and 1992 increased the powers of municipalities in teaching and school finance decisions. The 1985 law created a board of governors at the compulsory school level, which includes parents, teachers, staff, and pupils (in upper stages). Decentralization reforms (1983–5) gave local and regional authorities new responsibilities, with further responsibilities extended to primary schools (1989). Recent reforms (1999) increase the power of regions and schools over personnel (Cole 2001). Education is primarily a Länder responsibility. In a study of North-Rhine Westphalia, Amelsvoort (1995) finds little reform. Bill of Autonomy (1997) transferred more autonomy to schools in the areas of organization, management, and curriculum (Bottari 2000). Central government control over the superintendent of schools was abolished and there has been an attempt to make the curriculum more diverse. The system remains relatively centralized and standardized (Lee 2000). Systemic decentralization beginning in 1989; boards of trustees created to govern schools, with parent participation; regional authorities abolished (Gordon and Whitty 1997; Perris 1998). There is a centralized education policy with decentralized management and administration, and the trend is towards greater autonomy. A switch to block grant funding was designed to increase school autonomy. A 1982 Education Participation Act requires schools to create participation councils that include parents and teachers. 1992 amendments to the law increase the power of these councils and expand the role of staff (Karsten 1999). Increased decision-making power given to local authorities during the 1980s, with decentralization to committees appointed by each municipal and county council (1988). In 1991, school funding (block grants) and management were decentralized from the National Board of Education to municipalities. In 1992, the governing conservative coalition established a voucher system, which provided a grant (85% of cost of educating a student in public school) to be used in private schools. The Social Democrats continued the programme (with a reduction to 75% of costs) when they returned to power in 1994. More recently, municipalities were given the authority to decide the value of the voucher (Lidstrom 1999). Few major reforms. Some reforms in individual cantons. In England and Wales, the 1988 Education Act created an element of competition between schools and the 1993 Education Act created a framework for schools to ‘opt out’ of control by Local Education Authorities and receive money directly from Whitehall (Gordon and Whitty 1997). This creates an effective voucher system among public schools, where funding follows the student. At least twenty states have passed some variant of School-Based Management. A number of these states require the creation of school councils for purposes of shared decision-making, with many of them requiring parent participation. Voucher systems have been created in several cities (for instance, Cleveland).
158
CHRISTOPHER ANSELL AND JANE GINGRICH
Table 7.6.Percentage of decisions taken at each level of government in public lower secondary education, 1998
Austria Belgium (French community) Denmark Finland France Germany Ireland Italy Netherlands New Zealand Norway Sweden UK (England) UK (Scotland) United States
Central
State
35 6
18 10
Provincial-regional
Sub-regional
2
61
26 32 4 47 39 24 34 35 13 20 9
28
11 15
27
25
Local
School
22
25 26
43 64
31 36 29 37 53 33 73 66 9 66 62 40 29
16 3 3
2
55 22 18 51 69
Source: OECD (1998).
New Zealand, the United States) and from the national to local level (Nordic Countries). The OECD finds that, while France, Italy, Sweden, and the UK were all centralized systems in the mid-1970s, all of them have since undertaken decentralization reforms. Table 7.6 summarizes the OECD findings on the distribution of decision-making authority across levels of government, illustrating that in many OECD states the local and school levels play a significant role in educational decision making.
TRENDS IN DECENTRALIZATION
159
The character of these reforms follows the logic that we have described for decentralization reforms more generally. In general, countries with centralized education systems have had more scope for decentralizing their system. In general, unitary systems like New Zealand and Sweden have been able to undertake more systemic national reforms; federal systems in which education was controlled at the State level have had systematic reforms at the State level (Canada, Australia). Education systems controlled by local government (like the US) have had much more piecemeal reforms. The impact of state structure also interacts with the nature of electoral systems and party politics to shape the character of educational reforms. Abu-Duhou (1999: 21) argues that most experiments in decentralization have been imposed on school systems from outside the system, by politicians seeking to improve the efficiency and quality of schools. Majoritarian systems have generally undertaken more systematic reforms than ‘consensual’ democracies, particularly when the prime minister throws political weight behind those reforms (Perris 1998). For instance, in New Zealand in the 1980s the Labour Prime Minister, David Lange, took the education portfolio for himself. In contrast, two democracies that exemplify Lijphart's description of a consensual democracy—Switzerland and Germany—have produced little systematic reform in the distribution of authority. However, changing conditions in another exemplar of consensual democracy—the Netherlands—have encouraged decentralization. Karsten (1995: 305) notes that pillarization in the Netherlands encouraged centralization: ‘. . . centralization was also linked to the fact that the confessional parties did not want to grant the municipal councils a policy-making role in the area of education.’ Thus, as seen with decentralization reforms more generally, the institutional structure plays an important, but by no means determinate, role in shaping the extent of decentralization. In turning from this description of structural reforms to examine the character of democratic participation brought about by these reforms, we find a great deal of uncertainty and some important tensions. Participatory decisionmaking often challenges the professional authority of teachers, the administrative power of principals and superintendents, and the representative authority of school boards (Beck and Murphy 1999; Opfer and Denmark 2001). The case of Spain, although outside the parameters of this study, is instructive. In Spain, voting by parents in school council elections has declined since the school council idea was first introduced: 45 per cent of the parents voted in 1985–6; this figure dropped to 10 per cent in 1991–2 and to 6 per cent in 1996–7 (Hanson 2000: 43–4). In contrast, teachers have participated more actively and have sought to enhance their authority vis-à-vis parents (Hanson 2000: 43–4). This illustrates that decentralizing reforms that increase access of citizens to the decision-making process do not guarantee increased participation, but actual improvements in parental involvement require a strong commitment on the part of various groups to increase the role of parents (Beck and Murphy 1999).
160
CHRISTOPHER ANSELL AND JANE GINGRICH
CONCLUSION Decentralization is a complex and heterogeneous process. This chapter has argued that both the extent and the character of decentralization differ across the advanced industrial democracies. Some countries have created new tiers of sub-national government or transferred significant responsibility and autonomy to lower levels of government, while others have done little to change the structure of the state. Moreover, decentralization has a variety of aims. While the goal of decentralization may be to increase participation and access to the system, increasing efficiency and controlling costs is an equally important impetus to decentralization. The means used to achieve efficiency aims often differ from those used to create increased participation or democratization, illustrating different strategies of decentralization. Thus the means and ends of the decentralization process differ across countries. We have argued that four dimensions of the institutional structure (majoritarian/ consensus, unitary/federal, Napoleonic administration, and corporatist traditions) shape the extent and character of decentralization. Our analysis of the cases offers some, but by no means conclusive, evidence in favour of this analysis. Overall, more decentralization has occurred within unitary states than federal states, but the nature of decentralization within unitary states differs. Our differentiation of unitary states by electoral and administrative structures explains part of this variation. The trend towards decentralization is pronounced in the Napoleonic states. France, Belgium, and Italy have all created new levels of elected government, increasing local autonomy and democracy. Two of the three unitary majoritarian states we examined, New Zealand and the United Kingdom, have undergone radical reform and exhibit the influence of a neo-liberal agenda. However, decentralization has been limited in these cases. Finally, the corporatist states have all transferred significant responsibility and autonomy to lower levels of government, and the character of these reforms is marked by a democratic orientation. Clearly, the institutional structure plays an important, though not determining, role in structuring the extent and character of decentralization. Decentralizing reforms have a potential for democratizing the administrative structure of the state in terms of the multiple goals of access, accountability, and transparency that were discussed in Chapter 1. In creating a new tier of government closer to citizens, decentralization may allow for new points of access to new locally elected officials, to new local referendums, or to administrative processes. Decentralization can expand democratization in all three modes. Decentralizing responsibility and autonomy over decisions that affect local communities to local or regional government bodies may also enable citizens to hold local or regional decision-makers more accountable for their actions. While the actions of policy-makers in a national capital may seem remote and distant, the decisions in the state capital or the city council appear more immediate and intelligible to many citizens. Finally, bringing decision-making closer to citizens may make governance more open and visible, thereby enhancing transparency. Thus it is often the case that citizens feel more trustful of local officials than of national politicians.
161
TRENDS IN DECENTRALIZATION
But the full practical implications of decentralization for democracy are ambiguous. An increasing number of access points may decrease the efficacy of citizens’ actions, diffuse responsibility, decrease the lines of accountability, and create complexity that decreases the transparency of the process. Thus, decentralization has democratic potential but is not equivalent to democratization. To understand decentralization and its democratic potential, we must carefully examine the strategies of decentralization employed. This chapter has shown that certain advanced industrial democracies have gone further than others in realizing the democratic potential of decentralization.
Appendix A: APPENDIX A: TOTAL PUBLIC EMPLOYMENT BY LEVEL OF GOVERNMENT Country/level of 1985 government
1990
AUSTRIA Federal n/a 163,545 Municipalities n/a 99,445 State n/a 142,178 TOTAL 405,168 CANADA Federal 399,176 406,366 Municipalities 759,715 869,120 Regional 1,217,671 1,387,076 TOTAL 2,376,562 2,662,563 FINLAND Central 213,000 146,948 Municipalities 404,000 433,539 TOTAL 617,000 580,487 FRANCE Central 2,460,200 2,489,600 Municipalities n/a n/a Regional 1,208,200 1,326,400 GERMANY Federal 858,000 873, 000 Municipalities 1,177,000 1,287,000 Länder 1,839,000 1,878,000 TOTAL 3,874,000 4,038,000 IRELAND (structural change) Central 184,858 169,210 Regional 32,566 26,468 TOTAL 217,424 195,678 NETHERLANDS Central 521,933 512,228 Municipalities 198,513 177,712 Regional 43,157 40,800 TOTAL 763,603 730,740 NEW ZEALAND (full-time equivalents) Central n/a 189,200 Municipalities n/a 20,700 TOTAL n/a 209,900 UNITED STATES Federal 2,843,298 3,008,323 Municipalities 9,328,854 10,564,447 State 3,518,193 3,968,265 TOTAL 15,690,345 17,541,035
Source: OECD (2000c).
1997
1998
1999
2000
% change from 1990 to last year available
171,167 140,442 131,393 443,002
170,876 140,355 132,411 443,642
169,003 139,700 132,857 441,560
n/a n/a n/a
3 40 −7 9
337,713 891,483 1,315,126 2,544,322
330,981 891,560 1,314,617 2,537,158
330,003 893,709 1,312,806 2,536,519
336,603 890,915 1,320,619 2,548,137
−17 3 −5 −4
123,214 418,172 541,386
124,943 416,143 541,086
125,481 411,151 536,632
n/a n/a n/a
−15 −5 −8
2,488,300 1,110,500 1,220,500
2,270,100 n/a n/a
n/a n/a n/a
n/a n/a n/a
−9 — −8
526,400 1,683,300 2,401,900 4,611,700
516,000 1,648,500 2,363,100 4,527,600
510,200 1,609,700 2,313,700 4,433,600
501,700 1,585,600 2,276,800 4,364,100
−43 23 21 8
192,493 26,500 218,993
195,569 26,500 222,069
199,463 26,900 226,363
207,926 27,400 235,326
18 2 16
596,288 170,347 38,405 805,040
605,996 172,775 38,257 817,028
614,439 174,953 38,641 828,033
n/a n/a n/a n/a
20 −2 −5 13
189,560 19,600 209,160
190,660 20,150 210,810
190,350 20,200 210,550
186,610 18,695 205,305
−1 −10 −2
2,810,489 11,844,330 4,214,451 18,869,270
2,782,888 12,077,513 4,240,779 19,101,180
2,785,268 12,344,332 4,298,341 19,427,941
2,777,000 13,049,000 4,746,000 20,572,000
−8 24 20 17
162
CHRISTOPHER ANSELL AND JANE GINGRICH
Appendix B: APPENDIX B: SUB-NATIONAL REVENUE AS A PERCENT OF TOTAL REVENUE AND CENTRAL GOVERNMENT REVENUE Country
Sub-national as a 1975 % of: Australia Total 19.88 Central 24.82 Austria Total 22.98 Central 44.48 Belgium Total 4.62 Central 7.08 Canada Total 42.35 Central 88.95 Denmark Total 29.76 Central 43.44 Finland Total 23.62 Central 41.72 France Total 7.57 Central 14.77 Germany Total 31.29 Central 93.34 Ireland Total 7.27 Central 9.40 Italy Total 0.89 Central 1.68 Japan Total 25.56 Central 56.24 Netherlands Total 1.15 Central 1.96 Norway Total 22.37 Central 44.18 New Zealand Total 7.71 Central 8.35 Switzerland Total 43.76 Central 159.77 United Kingdom Total 11.05 Central 15.68 United States Total 34.16 Central 75.29
1985
1995
2000
18.55 22.78 23.83 48.70 4.81 7.69 45.37 110.26 28.39 41.53 22.39 40.16 8.69 18.40 30.81 97.37 2.26 2.76 2.32 3.72 6.03 59.55 2.39 4.61 17.65 29.58 6.48 6.93 38.80 132.86 10.25 14.80 32.73 77.73
22.47 28.99 20.89 42.46 27.83 77.27 47.04 120.61 31.33 48.14 21.87 45.06 10.42 24.80 29.01 92.29 2.36 2.84 5.40 8.60 24.23 61.27 2.69 4.96 19.65 33.84 5.26 5.55 35.80 132.50 3.88 5.13 33.01 78.81
n/a n/a 19.32 36.72 n/a n/a 44.22 106.51 32.80 52.76 21.17 38.42 9.37 22.08 29.84 97.58 1.80 2.09 11.20 18.59 25.62 68.50 3.00 5.32 16.22 26.00 6.12 6.52 34.04 106.39 3.93 5.07 — —
Note: Total, central, sub-national, and other revenue are measured in local currency. ‘Sub-national’ is an aggregate of all regional and municipal state spending and excludes separate revenue from social security funds and supranational. Source: OECD (2001i).
TRENDS IN DECENTRALIZATION
163
NOTES 1 2 3 4 5 6 7
We recognize that multi-party systems may also govern according to a majoritarian logic and, conversely, that majoritarian systems may behave according to a more coalitional logic under some circumstances. Lijphart does not explicitly discuss the Napoleonic state. Scholars often distinguish between ‘decentralization’ and ‘de-concentration’. Decentralization implies the relocation of authority or responsibilities from central to local or regional government, while de-concentration refers to a transfer of function from central government agencies to the field offices of those agencies. Sharpe (1988: 370) summarizes the relative decline in the central government's share in general government expenditure between 1950 and 1973. For example, examining the relative share of expenditure by sub-national government in order to uncover decentralization assumes government spending is zero-sum over time; that is, in order for increases in local government spending to translate into an increased proportion of total spending, total spending must not increase. Spain (1983, extended in 1992) has granted considerable political autonomy to particular regions and is sometimes described as a ‘quasi-federal’ or ‘regionalized’ state; Greece developed a structure of regional governance in 1994. 〈http://www.kl.dk/1/〉.
8 Reforming the Administrative State Christopher Ansell and Jane Gingrich Over the last three decades, the popular distrust of government institutions described elsewhere in this book has produced a wave of administrative reforms. This distrust represents a fear that bureaucratic agencies will substitute their own interests for those of a sovereign citizenry and a concern that the public sphere of the state has been coopted for private purposes. The remedies proposed are complex and sometimes contradictory. Sometimes the remedy is greater control over government institutions through heightened accountability, a change that can paradoxically lead to more rules and more bureaucracy. At other times, institutions are re-engineered to increase the responsiveness of administrative agencies to citizen demands. The common theme of such reforms is the desire to make the relationship between citizens and government more direct by streamlining or eliminating the layers and complexities of government. These reforms thus reflect the theme of accountability that runs through other analyses in this book. In this chapter, we argue that some of these administrative reforms represent an equivalent to demands for ‘direct’ or ‘participatory’ democracy. Demands for greater access to bureaucratic decision-making, more direct accountability over public agencies, and greater transparency in administrative procedures have led to an administrative revolution—with the well-worn caveat that, the more that things change, the more they may stay the same.1 There are two different reform agendas. A neoliberal agenda proposes to restore access, accountability, and transparency by imposing market (or market-like) discipline on administrative agencies. A democratic accountability agenda seeks to make administrative agencies accessible, accountable, and transparent by ensuring direct participation or representation of citizens in administrative affairs.2 The neo-liberal agenda sees producers or consumers as sovereign and sees the state as infringing on that sovereignty. The ‘state’ is seen as burdening producers with unnecessary and illegitimate controls in the interest of narrow minorities or state elites. State bureaucracies come in for special abuse as bloated and unresponsive monopolies. At best, these bureaucracies are inefficient. At worst, they are corrupt, extracting rents from legitimate producers. We associate this neo-liberal agenda
REFORMING THE ADMINISTRATIVE STATE
165
with deregulation, privatization, and other attempts to reorganize the state's monopoly over certain kinds of services and functions (such as education). The democratic accountability agenda sees the citizen as sovereign and views the state as usurping that sovereignty. The process of democratic governance should be brought ‘closer to the people’ so that this sovereignty can be directly expressed. As gatekeepers, political parties and legislatures distort the will of the people, keeping certain issues off the agenda. These representative institutions are beholden to private interests—typically producer groups—which have privileged access to politicians and who ‘capture’ public agencies. Instead, advocates of the democratic accountability agenda want agencies to be held directly accountable to citizens rather than indirectly accountable through a system of electoral representation. We also note two different tendencies within the democratic accountability agenda. A legal version sees citizens as rights-bearing actors in the political process and relies on a formal legal framework to guarantee democratic accountability and the observance of rights. It places value on procedural fairness, equity, and transparency. Citizens should be treated as juridically equal.3 A deliberative version, in contrast, is more substantive than procedural. It seeks to maximize the exercise of citizen voice in public affairs and is less concerned with formal guarantees of fairness or transparency; the quality of participation and voice matters more than representational equity. This deliberative approach is often communitarian: those most directly affected by a policy ought to have the greatest opportunity to exercise their voice. Despite these differences, and in contrast to the neo-liberal agenda, both legal and deliberative versions interpret citizenship as a democratic role. In the administrative revolution we describe, the rhetoric of reform is hardly so stark or as clearly demarcated as we describe above. In fact, we would suggest that the administrative revolution has occurred over the last three decades because of these overlapping agendas. While the two agendas may disagree about the ultimate goals of reforms, they share an antagonism towards complex, large-scale, and centralized state institutions. Efficiency reforms and reforms oriented toward democratization may overlap a great deal. They arise from a common sense of the unresponsiveness of highly bureaucratized public administration and the lack of accountability over administrative behaviour (Nikos 2001). Both derive from a lack of citizen trust in government. They can agree either on imposing greater controls on the state or on making institutions more responsive to the sovereign producers or citizens. At the same time, the distinction between neo-liberal and democratic accountability agendas provides a perspective on different aspects of administrative reform and the different trajectories of the reform process.
THE NEO-LIBERAL AGENDA: NEW PUBLIC MANAGEMENT The most important wave of reforms of the administrative apparatus of OECD member nations has been the set of reforms lumped under the category New
166
CHRISTOPHER ANSELL AND JANE GINGRICH
Public Management (NPM). NPM is an umbrella label for a broad range of reforms utilizing private sector models to bring efficiency and accountability to public sector bureaucracies. These reforms seek to apply the discipline of the market to government, and hence to return sovereignty to both producers (policy-makers) and consumers (the public). Producers have sovereignty in general when they control their own property rights and, specifically, when they have autonomy to make managerial and investment decisions. Consumers have sovereignty when producers are disciplined by market competition to respond to market demand and to produce goods efficiently. In the NPM model, sovereignty is returned to producers by ‘letting managers manage’. NPM grants agencies greater autonomy to manage their own affairs while providing them with market-like incentives to perform on the ‘output’ side. NPM restores sovereignty to consumers with a variety of techniques that seek to make agencies responsive to citizen ‘demand’. Kettl (1997: 447) writes: ‘At the core of let the managers manage is the customer service movement, which focuses managers on serving citizens instead of the needs of the bureaucracy …’ Hence, ‘letting the managers manage’ is complemented by ‘making the managers manage’— the imposition of market discipline on managers. In this section, we describe some of the specific NPM reforms that facilitate more direct access and accountability for consumers and greater administrative transparency. One of the central themes of NPM is making bureaucracies more ‘customer’ or ‘client’ centred. Citizens are conceived of as sovereign consumers of state services, and public bureaucracies are expected to operate efficiently and responsively to deliver these services. To be responsive to its customers requires ‘breaking through bureaucracy’ to become a ‘customer-driven agency’ (Barzelay 1992). As Pratchett (1999: 618) writes of Britain: The move towards consumer-oriented consultation was also a broader transformation of public sector management in the 1980s, which stressed private sector methods were superior and placed emphasis on understanding consumer needs. Against this background, most public organizations initiated customer-orientation programmes within their workforces while introducing complaints and suggestion schemes, customer surveys and other such methods for ‘getting closer to the consumer’. This customer-oriented perspective has been prominent in Anglo-American nations. In the US, for instance, the customer-oriented perspective was a core idea of the Clinton-Gore Administration's campaign to ‘reinvent government’ (Osborne and Gaebler 1992; Gore 1993). One specific reform strategy to make agencies more responsive to citizens is the citizen's charter or service charter, in which individual government agencies make public declarations of their standards for service and their goals for improving service. It is an implicit contract with consumers to which the agency is then presumably held accountable. As Clarke (2000: 162) writes, ‘The charters … embody a populist appeal to new or revitalized principles of public service …’. Rhodes (1999: 347–8)
167
REFORMING THE ADMINISTRATIVE STATE
summarizes the major elements of a citizen's charter: ‘published explicit standards; full and accurate information about running services; choice for the users of services; courteous and helpful service; effective remedies; and efficient and economic delivery of services.’ The Conservative government of John Major initiated the first charters in 1991 and variations on the British model then diffused to other OECD countries. As Nikos (2001: 6) observes: ‘Citizen's Charters have become a common policy option among administrative systems in the European Union.’ Table 8.1 identifies at least twelve OECD nations that have experimented with citizen's charters since the early 1990s.4 The scope of these reforms varies. In Britain, for example, about forty charters cover key national services and another 10,000 charters have been instituted at the local level.5 In Australia, all Commonwealth agencies and government business enterprises that engage with the public now have service charters. In Italy, the use of charters has been more restrictive, being used primarily in education, health care, and gas. Another institutional mechanism to facilitate the delivery of public services and regulatory information to citizens is the multi-service centre, single-window service, or one-stop shop. A good national level example is Service Canada, whose mandate is to respond directly to ‘Canadians’ demands for easy access to programs Table 8.1.Selected New Public Management reforms aimed at access, accountability, and transparency
Australia Austria Belgium Canada Denmark Finland France Germany Ireland Italy Japan Netherlands New Zealand Norway Sweden Switzerland United Kingdom United States
Citizen's charter
One-stop shops
+ (1997)
+ (1997)
+
+ (1999) +
+ + (1997) + (1992)
+ +
+ (1997) + (1994)
+ (1999) + (2000)
+
+ (1992)
+ (1998)
+ (1993)
+ (1991)
Satisfaction surveys
Procedural simplification
+ (1998) + (1998) + (1998)
+ (1998)
+
+ + (1983) + + (1998) +
+ (1999)
+ (1999)
+ (1999)
Sources: The information for Table 8.1 and for this section on NPM reforms was primarily drawn from country reports posted on the OECD's Public Management and Governance Programme (PUMA) websites, and in particular, from the following reports and surveys: Public Management: OECD Country Profiles, Issues and Developments in Public Management—Country Reports, Public Management in OECD Countries, and Government-Citizen Relations Country Profiles.
168
CHRISTOPHER ANSELL AND JANE GINGRICH
and services, less “red tape”, and service that is timely and courteous’. The Service operates 229 ‘in-person’ offices across Canada that provide access to over 1,000 government programmes and services. Service Canada claims to represent ‘the “new face” of the federal government—a new “citizen-centred” approach to government for a new millennium’.6 Table 8.1 identifies the countries examined within this project that have initiated one-stop shops at the national level (though one-stop shops may be organized at the local level). Again, the scope and timing of reform varies.7 The Netherlands began four integrated service-delivery pilot projects in 1992. Finland began implementing a one-stop shop concept in 1993 and now has 150 one-stop multi-service centres. Norway also began experimenting with onestop shops in 1993 and presented a plan in 1999 for an expansion of the system. Greve and Jespersen (1999: 151) report that local governments in Denmark were required in 1997 to make declarations of all their services so that citizens are ‘better able to judge the quality and amount of services they receive from local governments’. In 1997, Australia organized Centrelink—a one-stop shop to represent thirteen government bureaus. In 1998, Austria initiated the ‘government help’ project— an Internet portal to government services. The Flemish regional government in Belgium developed an agreement in 1999 to create an integrated office for municipal, provincial, and Flemish services. In 1999, the Irish government initiated the REACH programme to provide integrated service delivery. In Italy, a May 2000 ‘Action Plan’ outlined a plan to create one-stop shops for businesses (Battini 1999). These had been set up in fifty municipalities by 2001. In short, this reform has rapidly spread across many advanced industrial democracies over the past decade. Surveys of citizen satisfaction with public services are a third reform that seeks to make governments more responsive to citizen ‘demand’ (Table 8.1). National citizen surveys appear to be relatively recent experiments and we found less evidence of their broad diffusion. Ireland administered its first survey of public satisfaction with government services in 1997. Belgium, Canada, and Denmark conducted customer satisfaction surveys in 1998. The Danish Ministry of Finance has since administered a biennial survey of citizen satisfaction. Norway and the US conducted governmentwide customer satisfaction surveys in 1999, and a variety of US agencies conduct their own surveys. A reform closely connected with the proceeding reforms and justified as increasing transparency is procedural simplification—the streamlining and elimination of administrative procedures (Table 8.1).8 Germany began a programme of ‘de-bureaucratization’ and administrative simplification as early as 1983. But other campaigns to simplify administrative procedures began in the late 1990s. In Italy, the Simplification Bills of 1998 and 1999 initiated programmes to simplify administrative procedures. Belgium created an agency for administrative simplification in 1998. Norway developed a programme called Simplifying Norway to simplify government procedures in 1999. France has introduced a number of measures to encourage
REFORMING THE ADMINISTRATIVE STATE
169
administrative simplification, with a focus on business regulation. Ireland initiated regulatory streamlining as part of its Delivering Better Government programme. In Japan, the Cabinet decided on a policy to streamline administrative procedures; and Austria has identified administrative simplification as an important need.
DEMOCRATIC ACCOUNTABILITY REFORMS As many scholars and practitioners have noted, the NPM reforms tend to treat citizens in market-oriented terms as customers.9 In contrast, the democratic accountability agenda sees citizenship less in terms of a contractual and consumer relationship and more as a political status granting both rights and obligations to directly participate in the process of self-government. We distinguish between legal (procedural) and deliberative (substantive) versions of this agenda.
Legal Reforms: The Ombudsman One of the major reforms that seek to empower citizens to hold public agencies legally accountable is the ombudsman system.10 Sweden established the first office of the ombudsman in 1809. Today, there are ombudsman-type institutions in over ninety countries, with wide-ranging goals, organizational structures, and activities.11 Table 8.2 summarizes the founding dates of ombudsman systems for the set of nations examined in this project. The vast increase in the number of ombudsman-type institutions occurred primarily in the second half of the twentieth century, across both OECD and developing nations. While the office of the ombudsman traditionally functioned as a legislative check on the executive, it increasingly functions in practice as a mechanism for citizens to voice complaints and concerns about government administration. Indeed, the expansion of the ombudsman system was spurred in part by a desire of democratic governments to provide accountable administration to citizens (Gregory and Giddings 2000). The ombudsman institution plays an important and ever-increasing role in connecting citizens to government.12 A number of advanced industrial democracies have established ombudsman offices, though there is a great deal of variation in their character and structure. In Canada, ombudsman offices exist only at the provincial level, although some federal agencies have similar offices. In the US, no ombudsman exists at the federal level, but seven States and a number of US cities have embraced the system. Ombudsman offices in Italy exist only at the regional level. Australia and Belgium have ombudsman offices at both the State and federal levels. The German Petitions Committee operates in a similar manner to the ombudsman, but is a parliamentary committee composed of elected members of the legislative. However, two Länder do have ombudsman, as well as several German cities. The UK's Parliamentary Ombudsman is accompanied by a Health Service Commissioner as well as local ombudsman offices.
170
CHRISTOPHER ANSELL AND JANE GINGRICH
Table 8.2.Ombudsman offices in advanced industrial democracies Country Sweden Military ombudsman Finland Denmark Norway United Kingdom New Zealand Canada Alberta New Brunswick Quebec Manitoba Nova Scotia Saskatchewan Ontario Newfoundland British Columbia Yukon United Statesa Hawaii Nebraska Iowa New Jersey Alaska Florida Arizona a b
Year established 1809 1915 1919 1955 1962 1967 1969 1967 1967 1969 1970 1971 1973 1975 1975 1979 1996 1969 1971 1972 1974 1975 1980 1996
Country France Germany Petitions Committee Rhineland-Palatinate Schleswig-Holstein Italian Regions Tuscany Austria Australia Commonwealth South Australia Western Australia Victoria Queensland New South Wales Tasmania Northern Territory Capital Territory Cook Islands Netherlands Ireland European Union Belgiumb Flanders Wallonia
Year established 1973 1949 1974 1995 1979–89 1974 1976 1976 1972 1972 1973 1974 1975 1978 1978 1989 1990 1981 1981 1995 1995 1992 1995
Several major US cities have also established Ombudsman offices. Several major Belgian cities also have Ombudsman. Source: Gregory and Giddings (2000).
Roy Gregory and Philip Giddings (2000: 4–5) list the classic definition of the Ombudsman, outlined in the 1974 International Ombudsman Institute by-laws: (i) (ii) (iii) (iv) (v)
To investigate grievances of any person or body of persons concerning any decision or recommendation made, or any act done or omitted relating to a matter of administration, by any officer, employee or member or committee of members of any organization over which jurisdiction exists; to investigate complaints against government or semi-government departments and agencies; a responsibility to make recommendation resulting from investigation to organizations under jurisdiction; to discharge the role and functions as an officer of the legislature in a role which is independent of the organizations over which jurisdiction is held; to report to the legislature either directly or through a Minister on the results of its operations or on any specific matter resulting from an investigation.
REFORMING THE ADMINISTRATIVE STATE
171
Despite these common characteristics of ombudsman systems, there are significant variations across national contexts. The Swedish and Danish ombudsmen exemplify two different models. The Swedish ombudsman functions as a public prosecutor and acts as an alternative to the administrative courts, targeting both individual office holders and general administrative problems. In contrast, the Danish ombudsman functions as an investigator separate from the courts, and usually examines general administrative problems rather than individuals (Lane 2000). Both models increasingly represent citizens rather than parliament. The proliferation of ombudsman-type institutions parallels their rise as a genuine citizens' institution. Contemporary ombudsman offices seek to protect and represent citizens' legal rights and interests vis-à-vis administrative structures. Unlike the more deliberative institutions we will describe below, the ombudsman generally does not actively solicit citizen participation and deliberation, but rather provides an access point for citizens wishing to protect their legal or formal rights. Variations in national ombudsman systems create different capacities to serve citizens and respond to citizen concerns. Most national ombudsmen accept complaints directly from the general public. Britain, and until recently France, employed an ‘MP filter’ system, where complainants must be directed through members of parliament and are then passed along to the parliamentary ombudsman. Visibility also varies significantly. Only 36 per cent of British citizens report familiarity with the ombudsman system, compared to 93 per cent of Dutch citizens (Gregory and Giddings 2000; Hertogh 2000). Users of the ombudsman system are also not always fully representative of the general population. Table 8.3 summarizes the number of complaints received by national ombudsman systems, comparing statistics from the late 1970s to the end of the 1990s. In each of the nations, the growth in the use of the office is substantial. The sources of these complaints, of course, vary across nations. In France, for instance, an increasing number of complaints are received from non-citizens and concern the speed of processing immigration applications (Médiateur de la République 2001). The powers and jurisdiction of ombudsmen also vary across context (Table 8.4). Many national ombudsmen have the power to initiate investigations without receiving complaints; others, such as the British and Belgian ombudsmen, do not. Several national ombudsmen possess power over municipal administration. The French ombudsman has such power, and operates through a decentralized system with a number of local ombudsman offices. In 1997, the Danish ombudsman also received jurisdiction over local ombudsman offices. Ombudsmen's jurisdiction has also increased with the expansion of freedom of information and human rights legislation, as ombudsmen often have remit over investigating access to information cases. Ombudsmen often have wide powers of investigation, and are usually allowed access to all government documents and buildings. The ombudsman in Australia has the power to compel testimony from witnesses, as does the Swedish ombudsman. Swedish, Canadian, and Danish ombudsmen, among others, also have the power to inspect prisons and hospitals, and to receive confidential communication
172
CHRISTOPHER ANSELL AND JANE GINGRICH
Table 8.3.Trends in ombudsman complaint loads: number of cases Australia Denmark Finland France Germany Netherlands New Zealand Norway Sweden United Kingdome a
a
b
c
d
e
f
1978–1979 5,000b 1,751 1,739 4,316 (1979–80) 1,635 1,644 (1979–80) 2,990d 1,25f
Most recent year available 14,867 (2001–2) 3,390(2000–1) 2,442 (2001–2) 51,189 (1999–2000)c 20,666 (2000) 8,242 (2000) 5,358 (2001–2) 2,209 (2001–2) 4,798 (2000–1) 2,139 (2001–2)
Note: Unless specified, these data do not include ‘own initiative’ cases or cases referred to other departments, and include written, oral, and electronic contact. Includes only oral complaints to the Commonwealth Ombudsman (including the Defense Force Ombudsman, Australian Federal Police and Australian Capital Territory), and does not include complaints to the separate state ombudsmen. The Commonwealth Ombudsman was established in 1977, and the volume of complaints increased rapidly in the first several years, with 8,535 oral complaints received in 1979–80. The French Médiateur de la République now has a central office as well as regional offices. The 1999–2000 data refer to the combined total of the offices, with 6,836 complaints submitted to the central office and 44,353 to the departmental offices. These data refer to the number of complaints completed (this number is similar to the number of complaints received in the year) and is supplied in order to ensure data comparability. Since the UK Parliamentary Ombudsman receives complaints only from Members of Parliament (MPs), the data refer to the number of cases referred to the Ombudsman from MPs rather than the number of complaints received from the public. The data refer only to the Parliamentary Ombudsman, and does not include the regional bodies or the Health Service Commissioner. The year 1978–9 experienced a spike in the number of complaints, with 901 received in 1977–8 and 758 in 1979–80. Sources: Caiden (1983); Médiateur de la République (2001); Norway, Parliamentary Ombudsman (2001); New Zealand, Office of the Ombudsman (2001); Australia, Commonwealth Ombudsman (2001); Sweden, Parliamentary Ombudsman (2001); Denmark, Folketingets Ombudsman (2001); Finland, Parliamentary Ombudsman of Finland (2001); United Kingdom, Parliamentary Ombudsman (2001); Netherlands, National Ombudsman of the Netherlands (2001).
from inmates. In other countries, the powers of investigation are more limited, with certain parts of the administration excluded from investigation. Finally, powers of compliance also vary across contexts. Generally, ombudsman systems have limited powers to enforce compliance; and most national ombudsmen attempt to use persuasion and reporting as mechanisms of accomplishing change. Some ombudsmen have greater powers. The British ombudsman can award financial compensation to complainants, and the Australian ombudsman can recommend compensation. In Sweden, the ombudsman can issue official reprimands. Alongside the proliferation of the institution of the ombudsman, there has also been a diversification of ombudsman offices. Many countries are entertaining suggestions
173
REFORMING THE ADMINISTRATIVE STATE
Table 8.4.Powers of the ombudsman, 1999 Report to a parlia- Allowed to initiate Guaranteed by mentary committee disciplinary action constitutional arrangements Australia and states Tasmania
Austria • Belgium Canadian Provinces Alberta, New Alberta, New Brunswick, Quebec Brunswick, Nova Scotia Denmark Finland France German Länder Greece Ireland Italian regions
Netherlands New Zealand Norway Portugal Spain and regions Sweden Switzerland subnational UK US states
• • Rhineland-Palatinate
• • •
Tuscany
Lombardy, EmiliaRomagna, Friuli Venetia
National and Amsterdam and Rotterdam • • • •
Amsterdam, Rotterdam
Nova Scotia
British Columbia, Nova Scotia, Saskatchewan
• •
Alberta, British Columbia, New Brunswick, Nova Scotia, Quebec, Saskatchewan • • Rhineland-Palatinate
•
• Emilia-Romagna, Friuli Venetia, Lombardy, Piedmont, Tuscany Haarlem
• • • • National and Andalucia • • • • Basel County Zur- Basel County Zur- Basel County, Zurich ich ich • Local AdministraLocal Admin., tion Northern Ireland, Scotland, Wales Iowa Alaska Ohio Nebraska
Source: Kempf and Mille (1999).
• • • Andalucia
•
Limited to cases of Have the right to maladministration initiate their own proceedings Northern Territory, Commonwealth, South Australia, Northern Territory, Western Australia Queensland, South Australia, Tasmania, Victoria •
• Emilia-Romagna, Tuscany National, Haarlem and Rotterdam • National and Andalucia • Basel County, Zurich
Alaska, Hawaii, Iowa, Nebraska, Ohio
174
CHRISTOPHER ANSELL AND JANE GINGRICH
for establishing special-purpose ombudsmen, such as children's ombudsmen or department-specific ombudsmen. Usage of ombudsman services has also increased, with most countries reporting increases in complaints (see Table 8.3). Indeed, this increase often overburdens the ombudsman offices, as financial resources available for ombudsman are often set in advance and limited by parliament (France is an exception, where funds are assigned after the event). Diversification and fiscal pressure pose challenges to national ombudsman systems, but they also illustrate the increasing popularity and usage of the system. The number, usage, and increasing orientation towards defending citizens' rights illustrates the increasing importance of the ombudsman as an instrument of democratic accountability across the advanced industrial democracies.
Legal Populism: Administrative Procedures and Public Consultation Administrative procedure laws are another mechanism by which public agencies are held directly accountable to citizens. Administrative procedure laws require administrative agencies to follow a particular set of procedures in issuing new regulations. In many cases, such laws require a specific process of public consultation, which may take the form of creating advisory committees, the holding of public hearings, or ‘notice and comment’ procedures.13 The first administrative law—the US Administrative Procedure Act—was passed in 1946. As Table 8.5 indicates, many OECD nations now have formal administrative procedure laws. While the diffusion of administrative procedure laws began in the 1970s, Australia, Italy, the Netherlands, New Zealand, Sweden, and the UK have all enacted new or revised laws or policies since 1998. In assessing the experience of different OECD countries, we find a trend towards the formalization of public consultation. De Vries (2000) and Beierle and Long (1999) argue that formalization of consultation may in fact reduce effective informal participation. By contrast, Franklin (2001) argues that the 1993 Government Performance and Results Act, which instructs US federal agencies to consult stakeholders in preparing their strategic plans, did have a modestly positive impact in opening up the strategic planning process. Without being able to fully assess these claims, we note that in some cases formalization does appear to supplement or supplant strong informal systems of consultation. For example, although public consultation had been ‘widespread in practice’, Australia adopted a formal administrative procedures law at the national level (OECD 1996b: 14). In Ireland, public consultation prior to rulemaking was a ‘well-anchored custom’. Still, the OECD regulatory reform report on Ireland (2001c) observes that ‘Since the advent of the Public Service Management Act of 1997 and the Freedom of Information Act 1997, the flow of information between departments and the public has been considerably enhanced’ (OECD 2001c: 48). In the UK, there is still no legal basis for public consultation, but in November 2000 the Blair government issued a Code of Practice on Written Consultation (OECD 2001e: 5). This new
REFORMING THE ADMINISTRATIVE STATE
175
Table 8.5.Laws on administrative procedures Country Australia Austria Belgium Canada Denmark
Date of law 2000 (1977, 1975) n/a n/a n/a 1985
Finland France
1982 1979
Germany Ireland Italy Japan Netherlands
1976 1990 1999 (1990) 1994 1998
New Zealand
2001 (1969)
Norway Sweden
1970 1986 1998
Switzerland United Kingdom
n/a 2000
United States
1946
Title of law Administrative Reform Act
Law on Administrative Procedures (Law no. 571) Administrative Procedure Act Law on Justification for Administrative Acts Act on Administrative Procedures Administrative Procedure Law Administrative Procedure Act General Administrative Procedure Law Act Administrative Amendment Act (Administration Act) Public Administration Act Administrative Procedure Act 1986 Government Public Administration Bill Code of Practice on Written Consultation Administrative Procedure Act
Sources: OECD (2001a: 241–7; 2002a: 175–82).
emphasis on consultation ‘builds on the practice of establishing non-statutory consultative or advisory groups, which draw together expert witnesses and representatives from the public and voluntary sectors to advise the Government on specific issues’ (OECD 2001e: 5). In a few OECD countries, public consultation is strong but remains informal. In Denmark, for example, public consultation is not specified by law but by tradition and internal government policy. Informal consultation is common (OECD 2000a: 48–9). Formal administrative procedure laws may be most important where public consultation is firmly established by neither law nor tradition. Although the OECD (2002a: 152–3) describes Japan as having a strong tradition of informal consultation, this consultation is not necessarily with the public at large. In a review of recent Japanese administrative reforms, a former official in charge of administrative reform observes: ‘One can cite a number of praiseworthy characteristics of the Japanese bureaucracy, but this list would not include transparency—about which it has never shown much enthusiasm. The Administrative Procedures Act has already changed bureaucratic behavior to some degree, and we can hope for further improvement under the new Freedom of Information Act’ (Masajima 1999: 227). Italy may be another case where an administrative procedure law might have a positive impact on public consultation. Italian administrative procedures are ‘scattered in a series of laws and decrees’ and public consultation ‘is not mandatory unless required by an explicit rule’ (OECD
176
CHRISTOPHER ANSELL AND JANE GINGRICH
2001d: 153–4). Notice and comment procedures are rarely used before approving regulations. Notice and comment procedures and advisory committees appear to be among the most common techniques of public consultation. A Finnish response to an OECD survey observes that ‘[Our] two main approaches to public consultation are the “committee institution”, which includes a wide variety of consultative and advisory committees, and circulation of proposals for comment’ (OECD 2001g: 13). Some evidence suggests that notice and comment procedures are becoming more widely used. Although the Dutch have been disappointed with low levels of participation, notice and comment consultation is increasingly used in the Netherlands. In 1999, a Japanese cabinet decision introduced a systematic notice and comment procedure (OECD 2001a: 38). In Canada, a formal cabinet policy established a notice and comment procedure in 1986 (OECD 2002c). By contrast, several countries have recently limited their use of advisory committees. The use of advisory committees in Denmark, the Netherlands, and Finland has either declined or become more restrictive (OECD 1999; OECD 2000a: 48–9; OECD 2001g: 13). The reason may be that advisory committees are sometimes criticized as being limited to experts or as over-representing special interests.
DELIBERATIVE ACCOUNTABILITY: TRENDS TOWARDS COLLABORATIVE GOVERNANCE If one trend is towards the increasing formalization of administrative procedures in the name of increasing transparency, two other trends can be observed. First, some evidence exists of a trend to make the administrative process more interactive and deliberative. Second, attempts are being made to render participation in administrative decision-making more inclusive. We examine these two trends in turn. Traditional modes of public consultation are often criticized as ineffective in producing meaningful public participation. King, Felty, and Susel (1998) argue that public hearings are not an effective means of public consultation, being too orchestrated and formal to generate real deliberation.14 The ‘rule and comment’ procedure now enshrined in many administrative procedure laws is also criticized as producing only limited dialogue. For example, Palerm (2000) emphasizes the lack of two-way communication inherent in the Environmental Impact Assessment ‘notice and comment’ procedures. In contrast, forms of ‘collaborative governance’ create more deliberative forums in which citizens may interact more intensively with government agencies (Fung and Wright 2001; Hunold 2001). King, Felty, and Susel (1998) suggest that more interactive participation at the municipal level and in federal agencies like the Environmental Protection Agency indicate that more collaborative consultation is possible. Collaborative governance can take many forms. In the regulatory arena, collaborative governance can take the form of ‘regulatory negotiation’ or ‘reg-neg.’
REFORMING THE ADMINISTRATIVE STATE
177
Regulatory negotiation brings stakeholders together earlier in the rule-making process to produce more intensive dialogue. In an analysis of the possibilities for reg-neg in the US, Seidenfeld (2000) finds that reg-neg does succeed, within limits, in encouraging adversarial stakeholders to resolve differences earlier in the regulatory process. In Australia, the technique of ‘sequential consultation’, which has been developed to assist in the preparation of major policy statements related to health issues, also seeks to overcome the limits of traditional rule and comment regulation. As described by Leroux, Hirtle, and Fortin (1998: 455), The sequential approach involves, first, notifying the general public of the Council's intention to draft/revise notices, guidelines, or recommendations on a topic, and next encouraging the public to make comments on the issue. The comments must then be taken into account while drafting the preliminary version of a notice, guideline or recommendation, and this version will be submitted to the public to allow it to react and make further comments. Many other forms of collaborative governance can be observed. For example, ‘collaborative learning’ processes have been used in the US and Finland to encourage mutual learning among stakeholders that can, in turn, lead to more systemic solutions to complex issues (Blatner et al. 2001; Saarikoski 2000). Fung and Wright (2001) use the term ‘empowered participatory governance’ to describe a similar type of collaborative governance at the local level in Brazil, India, and the US. Fung describes reforms in the Chicago police and school systems that sought to allow citizens to ‘participate continuously and directly in the micro-governance of two important institutions of urban life’ (2001: 79). Institutional reforms to increase parent participation in educational decision-making are, in fact, quite widespread, as we describe in Chapter 7. The Scandinavian countries and the Netherlands appear to us particularly active in devising new strategies of collaborative governance. In the Netherlands, collaborative government is referred to as ‘interactive policy making’ (Schedler and Glastra 2001). Hendriks and Tops (1999), for example, describe Dutch cities as shifting from a New Public Management focus on efficiency in the 1980s to an attempt to promote citizen participation in the 1990s.15 Klijn and Koppenjan (2000) observe that Dutch ministries and local government have been experimenting with ‘interactive governance’ and examine some of the difficulties of the process in the city of Rotterdam. The Scandinavian countries seem particularly concerned to encourage political participation. Finland, for example, launched a Participation Project in 1997 under the Department of Interior in order to encourage more participation in communal affairs (OECD 2001g). Sweden has also encouraged a number of proposals to strengthen political participation; Norway is experimenting with ‘joint consultation’ in which cabinet ministers visit public and private institutions to engage in informal (and not pre-scripted) discussions (OECD 2001b). ‘User boards have been established in Denmark to provide input to social service agencies (OECD 2001h). Rhodes (1999: 349) notes that ‘boards’ are a traditional method in Denmark of incorporating input and advice from experts and interest groups. However, ‘The 1980s and 1990s saw a new variety;
178
CHRISTOPHER ANSELL AND JANE GINGRICH
user boards. They are used to integrate service users, managers and employees in education, child care, and care of the elderly’.
DELIBERATIVE ACCOUNTABILITY: THE BROADENING OF PARTICIPATION If one trend in public consultation is to increase the ‘interactiveness’ of decision-making, a second trend is to be more inclusive. Consultative strategies, such as advisory councils, have traditionally been used for soliciting expert opinion but less often as a forum for public involvement. Thomas McGarity (2002) argues that the advisory committee model of public participation is very close to what he calls the ‘exclusionary model’. ‘Only credentialed experts’, he writes, ‘are invited to participate on the advisory committees, and the experts are not necessarily chosen to reflect different scientific perspectives.’ Although the US Federal Advisory Committee Act demands that committees be ‘balanced’, this has typically been interpreted as requiring the selection of experts representing different interest groups.16 One attempt to broaden the traditional ‘expert’ basis of advisory committees is to create ‘citizen advisory committees’. John Applegate (1998) claims that the rise of citizen advisory boards is a response to the failures of ‘review and comment’ regulatory procedures required under the Administrative Procedure Act. In a study of the use of citizen advisory committees in Great Lakes water quality planning (where they have been mandated), Knaap, Matier, and Olshansky (1998) conclude that these groups did improve input into the planning process, but also failed to fully represent affected constituencies. In another study of the same groups, Beierle and Konisky (2001) arrive at the ‘guardedly optimistic’ conclusion that stakeholder involvement can lead to an improved representation of public values, less conflictual relationships among stakeholders, and greater capacity for implementation. However, they also find that the advisory committees do not always greatly broaden the participation of the public (stakeholders being drawn from those groups and organizations already actively participating) and are relatively unrepresentative of socially disadvantaged groups. Judith Petts (2001) evaluates the use of community (citizens') advisory committees (CACs) in waste management in several British counties. She finds that they were representative of a wide range of community interests and were successful in encouraging both effective dialogue and learning. In a study of the use of advisory groups in a local planning exercise in the UK, Reeves (1995) comes to relatively positive assessment of the advisory group as a means of public consultation. Teske (2000) finds more mixed results in an innovative US programme designed to aid local communities to develop independent scientific expertise to deal with local waste management. Chandler (2000) expresses a negative view of these attempted reforms and argues that they confuse participation with democracy.
REFORMING THE ADMINISTRATIVE STATE
179
Many experiments in collaborative governance have a local focus because participation is more easily mobilized where potential outcomes have direct, immediate, and tangible impacts on local communities. Public participation and attention become more problematic as scale increases and the impacts are less tangible and specific (Wolfe, Kerchner, and Wilbanks 2001).
EXPERIMENTS IN DELIBERATION In the last several decades, public administrators across several OECD countries have begun experimenting with novel modes of engaging citizens in decision-making. While the trend towards using policy tools that actively solicit citizen participation remains relatively limited in comparison with traditional methods of linking government and citizens, the fact that a number of national and local governments across the OECD are using similar innovations warrants attention. The recent growth of interest among both academics and practitioners in citizens' juries, citizens' panels, consensus conferences, and other methods of citizen engagement illustrates new currents in thinking about democracy and democratic practices. Scholars of these new participatory initiatives have linked these practical innovations to contemporary theories of deliberative democracy. Proponents of deliberative democracy present a model of democratic participation that differs from both traditional representative democracy and modern pluralism, stressing not only increased citizen participation in decision-making but the value of the process of participation itself (Smith and Wales 2000).17 The aim is to go beyond stakeholders and organized interests groups and involve citizens in the process of decision-making, not simply by recording their preferences but by engaging them in the activity of politics itself. Peter Dienel, founder of the ‘planning cell’ (Planungszelle), a German participatory tool, claims that citizens should be involved in the decisions that affect them, but, recognizing time and information constraints, he proposes a participatory division of labour—something he argues is accomplished by initiatives that bring groups of ‘normal’ citizens into the process of decision-making (Dienel and Renn 1995). This perspective stresses that citizen participation does not just legitimize policy decisions but actually leads to better decision-making and enhances citizens' experience of politics (Lenaghan 1999). While the question of whether citizens' juries, forums, or consensus conferences fulfil the ideal of deliberative democracy is a matter of debate, these new policy tools do attempt to bring ordinary citizens into the process of decision-making.
Citizens' Juries The term ‘citizens' jury’ was coined by Ned Crosby of the Jefferson Center in Minneapolis in the 1970s. Crosby branded the name with the United States government in the early 1990s in order to prevent misuse. Crosby's model of a citizens'
180
CHRISTOPHER ANSELL AND JANE GINGRICH
jury is similar to Dienel's concept of a planning cell. Various adaptations of citizens' juries have been used across Europe, the United States, Australia, and New Zealand, and have been strongly championed by the Institute of Public Policy Research in London (Coote and Mattinson 1997). The basic structure of a citizens' jury involves bringing together between twelve and twenty-five randomly selected citizens to consider a ‘charge’ or a specific set of policy questions over the course of between two and four days. The citizens are generally selected so as to be representative of the demographic make-up of their community, although some juries attempt to select citizens who represent the range of different views and interests in the community. Jurors are paid for their time, making the citizens' jury process a relatively expensive forum for citizen participation. The deliberation roughly follows the juridical model, involving different sets of expert witnesses that the citizen panel may cross-examine. The discussion is usually facilitated by an independent moderator, who may have a greater or lesser role in guiding discussion depending on the structure of the jury. Jurors are given a chance to deliberate on the issues, and then submit a report that is delivered to the organization commissioning the jury. The jury's decision is not binding, but often the commissioning government agency must issue a follow-up report and respond to citizen suggestions. Proponents argue the jury system is advantageous because it provides a forum for a group of representative citizens to conduct informed deliberation of important issues, and offers a genuine dialogue between citizens and policy-makers (Crosby 1999; Dienel and Renn 1995). However, some critics have challenged the value of the process. In evaluating whether the citizens' jury model embodies the ideals of deliberative democracy, Smith and Wales (2000) argue that the process substitutes a statistically representative panel for the ideal of inclusive participation, running the risk of essentializing members of the juries as representatives of their social and economic groups. With respect to a range of new participatory initiatives, Chandler (2000) identifies the risk of institutionalizing passive citizen participation rather than encouraging active and autonomous citizens to join together and approach government on their own terms. The fact that the decisions of the jury are not binding leads sceptics to argue that the process is little more than a public relations exercise. In contrast, others express the concern that, because citizen jurors are not accountable to the public for their decisions, jury results suffer from an accountability deficit (Pickard 1998). Citizens' juries are clearly not a replacement for representative democracy or other channels of citizen participation, but experimentation with them by various local and national governments does illustrate an attempt to engage citizens in a more extensive process of deliberation and decision-making.
Consensus Conferences Consensus conferences are similar to citizens' juries, but were developed specifically for the purpose of citizen participation on scientific and technical issues.
REFORMING THE ADMINISTRATIVE STATE
181
The consensus conference model was developed by the Danish Board of Technology (DBT), an independent organization established by the Danish parliament in the mid-1980s. The format has been actively promoted by the Loka Institute, a non-profit institution dedicated to research into social and environmental issues.18 The rationale of the consensus conference is that citizens, not just experts, should be involved in questions of science and technology; and consensus conferences focus on these issues from the perspective of public and social responsibility (Andersen and Jaeger 1999). The consensus conference organizes a citizen panel of about fourteen people, recruited through local media campaigns and then selected by organizers to form a representative or diverse sample. The organizers form an advisory panel that conducts preparatory meetings with the citizen and expert panels, organizing the structure of the conference and providing citizen panellists with background information. During the conference, the experts answer questions developed by the citizen panel; and at the conclusion of the conference the citizen panellists present their final report and recommendations. Like citizens' juries, the events are highly publicized and the final reports are distributed to politicians and the media (Sclove 1996). Consensus conferences are open to the same criticisms as citizens' juries. As with citizens' juries, the extent to which these processes affect actual policy-making is unclear. Even more than citizens' juries, consensus conferences are open to criticism on the selection of the citizens' panel. Citizens must apply to be members of the panel, leading to concerns of self-selection and a lack of representativeness among panellists (Andersen and Jaeger 1999). Again, like citizens' juries, they are not a replacement for other channels of political decision-making but provide new ways of involving the public in high-profile scientific debates.
Other New Initiatives for Citizen Participation A number of local and national governments across the OECD have developed other tools to increase citizen participation. Local councils in Britain and Denmark have used visioning and community planning exercises that bring citizens together in public forums to discuss priorities or broad visions for their community (Pratchett 1999). British local governments have also experimented with standing citizen panels, which use a reference group of 100–200 citizens who meet periodically to discuss and vote on council policies (Stewart 1996). Local and national governments in Finland, Norway, and Germany have experimented with youth forums, which involve young people in organized and ongoing discussions about the issues that affect them.19 National and local governments in Australia and Britain have used citizens' panels, which are groups of citizens regularly contacted by postal questionnaires on a range of policy issues (Pratchett 1999).20 Other techniques such as citizen advisory councils, focus groups, public round tables, and neighbourhood councils have also been established across a number of countries (see Table 8.6).
182
CHRISTOPHER ANSELL AND JANE GINGRICH
Table 8.6.New participatory initiatives across the advanced industrial democracies Citizens' juries (CJ) Australia
A CJ was conducted in New South Wales (NSW) on Container Deposit Legislation in January–February 2001, in conjunction with a televote (type of deliberative poll). The NSW Minister for Environment commissioned the Institute of Sustainable Futures to conduct the CJ.a Several other CJs have also been
conducted by sub-national governments.
Austria
Citizen (or Consensus) conferences (CC) Australia's first CC on Gene Technology in the Food Chain was held in 1999. The conference was sponsored by a number of governmental agencies and was conducted by Science Museum. A lay panel of fourteen citizens deliberated over the course of several days on a range of issues relating to genetically modified organisms (GMOs).b In 1997 a CC on Tropospheric Ozone was held in Baden. The project was sponsored by three Länder, and was not carried
Other new participatory initiatives Brisbane City Council established a Community Reference Panel in 1998 which involves over 6,000 residents in surveys and participation experimented with different participatory exercises.c Other local governments have programmes.
An advisory council on waste incineration was held in the city of Linz.
out by an independent organization. The lay panel was composed of young people (age 18–26), who deliberated over three days. d
Belgium
Canada
Security and Society Contract in the City of Mons established in 1993, addresses quality of life issues in the city and provides a number of public participation forums (including open house, public meetings and hearings, community outreach, surveys, focus groups, citizen advisory groups, resource centre, mediation, and joint project teams.e Some CJs have been conducted, The City of Hamilton, Region of National Forum on Health was but these have been primarily established in 1994. The Forum Hamilton-Wentworth with sponsored by non-governmental McMaster University hosted a brought together citizens and organizations (although govern- Waste Management CC in 2000. stakeholders across the country. mental officials have been inIt used a number of tools to This involved eleven citizens volved). Examples include the who deliberated over three days.f engage citizens, such as study Canadian Broadcasting Corpo- In 1999, the University of Calgary hosted a circles, discussion groups, stakeration's CJ on the federal budget, consensus conference on food biotechnology, holder and citizen conferences, and the University of Calgary's involving citizens from across western Canada and surveys (OECD 2001a: (this project was sponsored by the Social Science 85–103). Other examples of CJ on xenotransplantation. and Humanities Research Council and the initiatives involving citizens inAlberta Agricultural Initiative, results were clude the Round Table on the forwarded to seven government agencies).g Environment and Economy, and the Rural Dialogue Process.
REFORMING THE ADMINISTRATIVE STATE
Denmark
The Danish Board of Technology (DBT) has sponsored over fifteen CCs since 1989. The DBT is an independent agency sponsored by the Danish Parliament, and the results of CCs are passed along to decision-makers. Most recently (1999), the DBT held a CC on genetically modified food, and the problem of noise (2000).h
Finland
France
In 1998, the French Ministry of Health used a number of CJs composed of between ten and fifteen people to discuss healthrelated issues. This was part of larger consultation process (‘État Généraux de la Santé, EGC’) examining the state of health care in France (other regional boards on hospital politics and user groups were also used— and over 200,000 people were involved in the various consultation and participation schemes).k
A Citizens’ Conference CC on genetically modified organisms was held 20–1 June 1998. The initiative was initially suggested by Prime Minister Jospin, and was implemented by the Office Parlementaire d'Evaluation des Choix Scientifiques et Technologiques (OPECST), which organized an independent steering committee. The CC included a lay panel of fifteen members, who discussed a number of issues relating to GMOs.l
183 Danish local governments have experimented with scenario workshops, which are also organized by the DBT (Anderson and Jaeger 1999). Other ex-
amples of local initiatives include that activities of the town of Stevns, Sjaeland, which organized a future vision workshop involving fifty citizens in the process of long-term town planning.i
In 1997 the federal government in conjunction with the Association of Finnish Local and Regional Authorities launched the Participation Project, which is aimed at increasing participation in local government. Over fifty local authorities participated with over seventy projects launched between 1997 and 1999, ranging from establishing a youth council in Helsinki to the ‘democratic dance night’ in Järvenpää.j
184 Germany
Japan
Netherlands
CHRISTOPHER ANSELL AND JANE GINGRICH
A CC on genetic testing was held 23–6 November 2001 in Dresden. This was sponsored by Germany's Federal Ministry for Education and Research and the Donors’ Association for the Advancement of Science in Germany and was conducted by Deutsches Hygiene-Museum Dresden. The conference was Germany's first, and was modelled on the Danish consensus 117– 40). The usage of m planning cells has increased conference. ‘Planning cells’ have been used in Germany since 1972–3, and since then dozens of communities have used the technique. In 1982 the German Ministry of Research and Technology conducted twenty-four planning cells with twenty-five participants each—the most comprehensive use of this technique to date (Dienel and Renn 1995:
in recent years (Geißel and Oels 2000).
There have been several CCs in Japan. Most recently the Society for Techno-Innovation of Agriculture, Forestry and Fisheries, commissioned by the Ministry of Agriculture, Forestry and Fisheries, held a conference on genetically modified crops. The conference involved eighteen citizen panellists, and was held over several meetings from September to November 2000. Other CCs in Japan have examined gene therapy and the high information society. o The Rathenau Institute, an independent body, has conducted several CCs. After a series of public workshops and preparation undertaken by the Rathenau Institute, in 1995 a CC on Predictive Genetic Technology was organized by Platform for Science and Ethics, in collaboration with the Health Ministry and Health Council. Fifteen lay citizens were involved in the three-day process, and the discussion was open to the public (Aarts, Hofman, and Smink 2001).
Examples of local initiatives include the city of Munich, which established a children and teenager's forum. Beginning in 1989 and continuing to present, the city uses a permanent working group on children and teenagers forums that establishes a number of participatory activities that involve children in local politics. Municipal law also specifies the group be consulted, and administrators must respond to the outcomes of children/ teenager forums.n
Several local governments in the Netherlands have experimented with new schemes of public involvement. Starting in 1991, the city of Deventer developed a neighbourhood-based participatory approach, which created citizen councils within five neighbourhoods, and used techniques such as public meetings and hearings, forums, joint-project teams, and surveys to bring citizens into local governance.p The city of Ahern used visioning exercises in 1996, incorporating citizens into public hearings, workshops, forums, and neighbourhood discussions on the future of the city. q
REFORMING THE ADMINISTRATIVE STATE
New Zealand
Norway
Switzerland
185
The city of Wellington conducted a CJ in 1995, and considered whether the city of Wellington should sell its shares in Capital Power. The city contracted a marketing company to conduct the jury, which involved fourteen randomly selected citizens.
New Zealand has conducted a number of ‘Talking Technology’ conferences, which are similar to the Danish CC. The first such conference was held in 1996, on plant biotechnology, and a second panel was held in 1999. This was organized by the New Zealand Consumer's Institute, and results were publicized and forwarded to government.r A CC titled ‘Fast Salmon and Technoburgers’ was held in 1996. The conference was conducted by the Biotechnology Advisory Board, an independent advisory body appointed by the government and the National Committees for Research Ethics. The conference involved sixteen lay citizens in a discussion about genetic modification and food.s Several CJs, following the Ger- There have been several ‘Pubman ‘planning cell’ model, have liforums’ in Switzerland, which been organized in Switzerland. are similar to CCs. In 1999 a In 1992, the Swiss Building publiforum on genetic technolDepartment commissioned the ogy and food was held, organSwiss Federal Institute of Tech- ized by the Swiss Science and nology to organize a planning Technology Council's Technolcell on the construction of a ogy Assessment Programme. landfill. This deviated slightly SSTC and CTA work with the from the German model in that, Federal Council, advising it on rather than selecting citizens issues of science and technology. randomly, the various political The 1999 publiforum involved communities involved nomina- seventeen lay citizens and took ted citizens (Dienel and Renn place over several weekends.t Several other publiforums (electricity and 1995: 117–40). society, human stem cells, transplantation and medicine) have also been conducted.
The Norwegian government has established a youth forum, bringing young people into an organized structure of consultation (OECD (2001a)).
186 United Kingdom
United States
a
b
c
d
e
f g h i
j
k
CHRISTOPHER ANSELL AND JANE GINGRICH
CJs have been used by a number of local authorities across the country. Examples include juries in Portsmouth (1997), Walsall (1997), and Nottingham (1999), which considered issues on health care and rationing (Lenaghan 1999). A jury in Ely
There have been two CCs in the UK. The first, in 1994, looked at plant biotechnology. This was organized by Science Museum and funded by the Agricultural and Food Research Council (now the Biotechnology and Biological Sciences Research (1997) which examined the Council.u The second conference, held in
issue of creating a Fens wetland area. These juries were organized by the Institute for Public Policy Research and the King's Fund, and commissioned by local authorities or a consortium of local authorities.
The Jefferson Center has conducted a number of CJs in Minnesota and across the US; however, many of these were sponsored by organizations independent of government. The first CJ in the US with governmental sponsors was conducted in 1984 by the Center for New Democratic Processes (now Jefferson Center) on water quality in Minnesota. Since then, a number of CJs have been held with and without government sponsors, including two (independent) national juries (Crosby 1999: 401–18)
1999 on radioactive waste, was organized by the UK Centre for Economic and Environ-
mental Development (CEED, an independent
The British government has established a national citizens’ panel, the People's Panel, which involves 5,000 randomly selected citizens from across the country. It uses surveys, workshops and CJs to canvass their views on a number of politically relevant topics, and the information is available to all government agencies (OECD 2001e).
charitable organization) funded by Office of Science of Technology, National Environment Research Council (NERC), and Nirex and industry.v
The Loka Institute conducted a consensus conference on Telecommunications and the Future of Democracy with the Massachusetts Foundation for the Humanities, the Education for Public Inquiry and International Citizenship Program at Tufts University, MIT's Technology Review magazine, and University of Massachusetts Extension. This was an independent citizen's panel, which lacked government sponsorship (Sclove 1997).
The Institute for Sustainable Futures has an online description of the container deposit legislation process, available at 〈http://www.isf. uts.edu.au/CDL_Review/index.html〉. Hendriks (2002) examines the relationship between interest groups and the citizens’ jury in New South Wales. Australian Museum. ‘Australia's First Consensus Conference, On Gene Technology in the Food Chain 〈http://www.austmus.gov.au/ consensus/〉. Brisbane City Council. ‘Community Reference Panel.’ 〈http://ycys.brisbane.qld.gov.au/about_ycys/community_reference_panel. shtml〉. Helge Torgersen, ‘EUROPTA: The Ozone Consensus Conference in Austria’ 〈http://www.tekno.dk/subpage.php3?article=797&language=uk&category=11&toppic=kategori11〉. Toolkit Citizens Participation in Local Government ‘Belgium, Mons: The ‘Security and Society Contract’〈http://www.toolkitparticipation. com〉. 〈 http://www.eng.mcmaster.ca/stpp/consensus2000/index.html〉. 〈 http://www.acs.ucalgary.ca/~pubconf/index.html〉. Danish Board of Technology 〈http://www.tekno.dk/〉; Anderson and Jaeger (1999). Toolkit Citizens Participation in Local Government. ‘Denmark, Stevns, Sjaeland, Future Vision in Stevns’ 〈http://www. toolkitparticipation.com〉. Tiala, Toni, Ministry of the Interior, Department for Municipal Affairs ‘Participation Project In Finland.’ Presented at Euro-PA Convention, Rimini, Italy. 12/4/2002; 〈http://www.intermin.fi/intermin/images.nsf/files/AABF90A3CEA6C433C2256BA500386D39/$file/europa.pdf〉; the ‘Participation Project’ . OECD (2001f). French government EGC 〈http://www.sante.gouv.fr/egs/1-demarche/index.htm〉.
REFORMING THE ADMINISTRATIVE STATE l m n
o
p
q
r s
t
u
v
187
OECD (2001f).See also the Loka Institute 〈http://www.loka.org/pages/Frenchgenefood.htm〉. Citizens' Conference on Genetic Testing 〈http://www.buergerkonferenz.de〉. Toolkit Citizens Participation in Local Government. ‘Germany, Munich. Development of a Children's and Teenagers Forum' 〈http:www. toolkitparticipation.com〉. Hirakawa, Hideyuki. ‘Provisional Report on the GM crops Consensus Conference in Japan.’ 28 February 2001. 〈http://www.cs.kyoto-wa. ac.jp/~hirakawa/GMO/cc-report.lisbon.html〉 Toolkit Citizens Participation in Local Government. ‘Netherlands, Deventer. The Deventer Neighborhood Approach: A Neighborhoodbased Participatory Decisionmaking programme’ 〈http://www.toolkitparticipation.com〉. Toolkit Citizens Participation in Local Government. ‘Netherlands, Arnhern. Hotel Arnhern 2015’ 〈http://www.toolkitparticipation. com〉 Biotechnology Resources Index 〈http://www.agresearch.co.nz/scied/search/biotech/resources_biotech.htm〉. OECD (2001a: 223–37). ‘Fast Salmon and Technoburgers’ 〈http://www.etikkom.no/NENT/fast.htm#I%20GENE%20MODIFICATION〉. Technology Assessment Program. ‘Report on Publiforum on Genetic Technology’ 1999 〈http://www.ta-swiss.ch/www-remain/ reports_archive/publications/1999/ta_p_1_99_e.pdf〉. Science Museum. UK National Consensus Conference on Plant Biotechnology 1994 〈http://134.225.167.114/NCBE/GMFOOD/ ccreport.html〉. CEED ‘Radioactive Waste Management UK National Consensus Conference: Citizen Panel Report’ 1999 〈http://www.ukceed.org/ consensus_conference/contents.htm〉.
188
CHRISTOPHER ANSELL AND JANE GINGRICH
The Extent of New Participatory Tools The OECD (2001a), relying on responses to an OECD survey distributed to member states, argues that the use of tools of active participation remains limited, particularly in comparison with ‘passive’ measures such as information provision and consultation through surveys. The OECD points to several national governments that have experimented with new participatory measures, such as Denmark's consensus conferences or France's use of citizens' juries in its general review of the health system. However, the OECD survey examines only policy tools used at the national level, and many of innovative measures are being used by local and other sub-national governments. The use of such procedures is piecemeal, and with few exceptions it depends primarily on the resources and initiatives of individual local governments. Several national governments have attempted to create more experimentation among local government. In Britain, the Department of Transport, the Environment and the Regions set new guidelines and rules for local governance, setting out a new ‘Best Value’ framework that redefines the responsibilities of local government and encourages (and requires) participatory measures (Chandler 2000). The Finnish national government has also encouraged experimentation with participatory initiatives at the local level, launching the ‘Participation Project’ that works in partnership with the Association of Finnish Regional and Local Authorities and local communities to develop new mechanisms for participation. However, most local participatory initiatives appear driven by local circumstances rather than by national governments. Moreover, many participatory initiatives are initiated and promoted by non-governmental organizations. For example, the Jefferson Center has conducted a number of citizen's panels funded by independent sponsors, and several universities and think tanks have sponsored consensus conferences independently of government (see Table 8.6). Thus, innovations and use of participatory measures has often developed outside of government. Even in Britain, where citizens' juries have received much academic and political interest, under 5 per cent of respondents in a Department of Transport, Environment and the Regions 1998 Survey reported using them (Lowdnes, Pratchett, and Stoker 2001a). However, this survey does find significant use of deliberative measures, with nearly 50 per cent of respondents using focus groups and over 40 per cent reporting the use of community planning exercises; moreover, the use of such measures began to accelerate in the mid-1990s. Similar results are found in Germany. In a survey on local initiatives relating to Local Agenda 21 (United Nations Conference on the Environment and Development project), 24 per cent of the local authorities responding had used future vision workshops in 1999, compared with less than 5 per cent in 1996 (Geißel and Oels 2000).
CONCLUSION A common thread through the great diversity of reforms described in this chapter is the desire to make the governing relationship between citizens and administrative
REFORMING THE ADMINISTRATIVE STATE
189
decision-making more direct. We discerned two different but overlapping reform agendas: a neo-liberal agenda that seeks to hold public agencies more directly accountable to citizens as consumers of public services, and a democratic accountability agenda that seeks to enhance the direct participation or representation of citizens as bearers of rights and obligations in administrative decision-making. We further distinguished procedural and substantive versions of the democratic accountability agenda. We then examined the implementation of specific administrative reforms that exemplify these agendas. Citizen's charters, one-stop shops, consumer surveys, and procedural simplification were analysed as characteristic reforms of the neo-liberal agenda. We found that these NPM reforms, particularly citizen's charters and one-stop shops, spread across many of our OECD countries in the 1990s. Ombudsman systems and administrative procedure laws represent the procedural/legal version of the democratic accountability agenda. Ombudsman systems have also spread throughout nearly all the eighteen countries we analysed, beginning in the 1960s. The spread of administrative procedure laws started at an earlier date, and a number of countries had adopted these laws by the end of the 1990s. With respect to the substantive/deliberative version of the democratic accountability agenda, we described new forms of collaborative governance (for example, regulatory negotiation, collaborative learning) and some fairly recent experiments with citizen deliberation (for example, citizens' juries, consensus conferences). While examples of these reforms can be found earlier, these were mostly experiments of the 1990s. Although we have distinguished these agendas theoretically, our larger finding is that the reforms are often intertwined in practice, just as Chapter 1 described how past reforms of representative and direct democracy tended to occur in parallel. The citizen's charter idea, for example, may owe its rapid and widespread diffusion to its appeals to the virtues of efficiency and democracy. After conducting this research, we no longer find the conventional association between the neo-liberal agenda and Anglo-American nations so neat. On the one hand, NPM reforms were much more widely diffused than we initially expected. On the other hand, Anglo-American countries were also active in experimenting with the democratic accountability agenda. Finally, although we were initially sceptical about the significance of new participatory tools like citizens' juries and consensus conferences, our research suggests that these still experimental techniques may represent an emerging trend in OECD nations.
NOTES 1 2 3
On the broad outlines of this administrative revolution, see Kettl (1997). For similar categorizations of administrative reform agendas, see Peters (2001). This is the ‘procedural republic’ described by Sandel (1996).
190 4 5 6 7
8 9 10 11 12 13
14
15 16
CHRISTOPHER ANSELL AND JANE GINGRICH
For more details on citizen's charters, see Toonen and Raadschelders (1997). United Kingdom Cabinet Office, Charters Unit. 〈http://www.servicefirst. gov.uk/index/list.htm〉. Service Canada 〈http://www.servicecanada.gc.ca/〉. The OECD's International Regulation Database includes responses from a 1997–9 survey that asked: ‘Are there single contact points used for getting information on licences and notifications (one-stop shops)?’ Of the eighteen countries in our analysis, Australia, Austria, Canada, Italy, and the UK reported affirmatively. See also the survey of ‘One-Stop Government’ in Austria, Belgium, Denmark, England and Wales, Finland, France, Germany, Ireland, Italy, Netherlands, and Spain, edited by Hagen and Kubicek (2000). OECD reports that twenty-six of twenty-eight OECD countries report taking measures to streamline administrative procedures (OECD 2002a: 58–9). Rhodes (1999: 349) puts it succinctly: ‘It is important to distinguish between clients, users, customers, and citizens. Clients depend on benefits. Customers shop in markets. Users have formal rights. Citizens have a broad participation role covering both the community and individual.’ The term ‘ombudsman’ is not universally used. For example, the French ombudsman is called the ‘Médiator de la République’. International Ombudsman Institute 〈http://www.law.ualberta.ca/centres/ioi/brochure.htm〉. De Vries (2000) describes the ombudsman system as an example of the bureaucratization of participation and argues that it may detract from the tendency of local political elites to take account of citizen perspectives. Various kinds of environmental and regulatory impact analysis require consultation with the public. Environmental impact assessment (EIA), in particular, is an area that has often led to innovation in the public consultation process. The basic process can be seen as a variant of the more general ‘rule and comment’ procedure widely used in regulatory rule-making. In a series of focus groups with UK citizens to discover their perspectives on local government consultation, Lowndes, Pratchett, and Stoker (2001b: 452) found that citizens declined to participate because they felt that nothing would actually come of their participation. Surveys of local government in Britain found that about a third of the respondents identified a positive impact of public participation on final decisions. Twenty per cent reported that participation simply confirmed decisions and another 20 per cent suggested that participation had little impact. However, 13 per cent suggested that participation was becoming more important (Lowndes, Pratchett, and Stoker 2001a: 222). German cities, they argue, undertook the opposite trajectory: from a greater focus on citizen participation in the 1980s to an increasing attraction to New Public Management in the 1990s. In the US, a Federal Advisory Committee Act (FACA) was passed in 1972. Its purpose, according to Beierle and Long (1999) was to make the advisory system more open. They argue, however, that FACA had an unintended ‘chilling effect’ on collaboration with the public, brought about by its procedural limitations on creating advisory committees. FACA has also discouraged agencies from creating advisory committees because of the litigation that can result from the legal ambiguity over what constitutes
REFORMING THE ADMINISTRATIVE STATE
17 18 19 20
191
an advisory committee. FACA is a good example of the way the bureaucratization of participation can actually reduce participation. In her examination of a citizens' jury in the New South Wales, Australia, Hendriks (2002) notes that these participatory procedures are actually in conflict with some of the central tenets of pluralism, and she documents interest-group antagonism towards the involvement of citizens through the citizens' jury process. 〈http://www.loka.org〉. On the Finnish case, see OECD (2001a). The web site of an organization of national local government associations 〈http://www.toolkitparticipation.com〉 provides examples from Finland and Germany. For a critique of the citizen panel concept, see Rippe and Schaber (1999).
9 Participation, Representative Democracy, and the Courts Rachel A. Cichowski and Alec Stone Sweet This project has focused attention on the various ways in which citizens participate in government beyond (1) electing their representatives and (2) joining or supporting interest groups that lobby elected officials on matters of public policy. No one denies that legislatures remain central arenas of policy-making and thus of interest group activity. Our collective purpose, however, is to explore the emergence of new arenas for political participation and their impact on the theory and practice of modern democratic governance. This chapter concerns the relationship between representative democracy and judicial authority. In some respects, our topic is hardly a new one, being at the very core of more than two centuries of constitutional and state theory in the US (Bickel 1962; Black 1960; Bobbit 1982) and in Europe (Carre de Malberg 1922; Dyson 1980; Troper 1994). In recent decades, the authority of courts to review the acts of public authorities—including those of legislatures, executives, and administrative agencies—has deepened, sustaining important debates about the democratic legitimacy of judicial power. Scholars, lawyers, and the judges themselves routinely produce elaborate justifications for judicial review, arguments that are just as regularly countered by opponents in the academy, at the bar, or on the bench. We will respond to these more normative issues, in light of our findings, in the concluding section of this chapter. Our larger and more immediate aim, however, is to address a set of questions adapted from the discussion in the introductory chapter of this volume. First, to what extent can citizens activate judicial institutions to pursue more diffuse public policy interests, and has such use grown over time? Second, what are the main factors that explain cross-national variation in such use? In any given polity, of course, a complex set of factors and circumstances help to determine whether and how the courts will be used to make or obstruct policy. To make matters more complicated, within the same polity the precise mix of factors will vary across time and policy domain. How citizens’ groups are organized, the types of resources they command, and the extent to which any given policy
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
193
domain is entrenched or in flux will interact with the specific policy-making dispositions of judges, legislators, and administrators or agencies operating at various levels of government. If, when moving from one situation to another, one observes only different patterns—as through a turning kaleidoscope—one is unlikely to achieve the broad-gauge comparability to which this project aspires. We have therefore chosen to focus on three institutional variables: (1) separation of powers, those constitutional rules that establish the authority of, and organize relationships between, governmental organs; (2) the nature and scope of judicially enforceable rights that individuals may plead in courts against public acts; and (3) rules governing legal standing to sue state authorities. These are relatively stable, systemic factors that condition how the more specific mix of variables operate in any given situation. If one finds transformation in these basic structural factors, then one can be more confident that the politics organized by these institutions are also changing. Our goal is to make a start in explaining the variance in the use of courts as sites of policy innovation, across time and political system. To assess change over time, we use two basic indicators: (1) case load, especially levels of publicinterest suits initiated by private actors and groups; and (2) levels of judicial discretion, defined as the capacity of the courts to make law and to shape the policy-making activity of other official actors, such as legislatures and executives. Our principal findings can be summarized as follows. Over the past half-century, separation of powers doctrines, whose purpose was to restrict judicial discretion, have weakened, in part because the domain of rights has steadily expanded. During this same period, legislators have dealt with the increasing complexities of government by delegating some of their authority to administrative agencies and by bringing citizens into rule-making and enforcement processes by liberalizing standing rules. Outcomes vary, but these factors have tended to interact, over time, in ways that reinforce a creeping judicialization of policy-making (Shapiro and Stone Sweet 2002). These dynamics depend upon the development of a particular symbiosis between judicial law-making and decisions to litigate. We discuss the microfoundations of these dynamics at this point, since they are not discussed further in the chapter. One straightforward way of analysing the sources and consequences of change in our three variables is by specifying levels of judicial discretion as the dependent variable of the study. Judicial discretion is the power of judges to choose between two or more (legally justifiable) policy alternatives (Hart 1994; MacCormick 1978).1 Bodies of legal norms and principles do not apply themselves directly to resolve legal disputes. Instead, the judge uses her discretion to classify fact contexts, interpret norms, and adapt rules to situations. Rule interpretation and application entail choice; choice entails law-making; and
194
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
the law that judges make will shape the behaviour of those they govern, if (1) judges justify their decisions with reasons, and (2) potential litigants regard these reasons as possessed of some authoritative, prospective value.2 When combined with some minimally robust conception of precedent, the exercise of judicial discretion will gradually institutionalize as a stable set of discursive practices that enable the continuous adaptation of rules to situations. In any case, we have good reason to expect that, if judicial discretion expands, creative law-making will ensue; and, as law-making proceeds, judges will use precedent both to help justify their powers and to organize markets for litigation, on a sectorial basis (Stone Sweet 1999). We now turn to litigants. Without a case load relevant to policy, a court's law-making powers are a moot point. The questions of whether and which individuals and groups actually choose to activate the courts to influence, pre-empt, or block policy that would otherwise be made in representative bodies depend upon a host of factors. Given the expense of litigation, private actors will pursue their interests through adjudication only to the extent that they expect the returns to litigation to exceed costs. This calculus is conditioned by perceptions of the relative costliness and likelihood of achieving the same policy goals through other means, such as lobbying the legislature. Further, a potential litigant has an interest in taking seriously what judges have done in the past, in order to make informed guesses about what judges will do in the future. It follows that some kinds of judicial rulings (precedents, law-making) will stimulate litigation, and others will deter it. And it follows that those who litigate more frequently in any given domain will invest more heavily in mastering the intricacies of relevant case law and in charting the evolution of legal doctrine over time. To the extent that they do, the centrality of courts, and the modes of governance they propagate, is strengthened. In the rest of this chapter, we examine the expansion of public-interest litigation and judicial discretion in three stages, corresponding to each of our variables: change in separation of powers doctrines, expansion of rights claims, and liberalization of standing rules. Following the template laid down by the editors, the analysis covers countries from three geographical areas: North America (United States and Canada), Europe, and Asia-Pacific (New Zealand, Japan, and Australia). We seek to identify general trends, explain variation, and evaluate the role of litigating on policy processes and outcomes. We note in advance three important limitations of our research. First, although the field of comparative judicial politics has grown substantially in recent years, systematic data of the kind one would wish to have for this project have not been collected. We make use of existing data, and have compiled new data-sets, but the problem persists. Second, on some of the cases (for example, Norway) included in this project, we found no scholarly literature relevant to our research questions. Third, given space limitations, we do not address variation across policy sector. Instead, in the section on standing to sue, we focus on one domain—environmental protection—across time and nation-state. Although we do not generalize from this sector to others, we
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
195
think that the environmental case is illustrative of certain general trends, as well as the fluidity and complexity of contemporary judicial politics.
SEPARATION OF POWERS AND JUDICIAL AUTHORITY Separation of powers doctrines are embedded in the constitutional law; they are at least partly enforceable by supreme and constitutional courts, and they underpin normative theories of the state. Such doctrines reflect, but also condition, how political legitimacy is to be understood and assessed. Read descriptively as rule systems, they purport to model the relationship between structures and functions in any complex system of governance. Read prescriptively as sets of idealized expectations about how structure and function ought to be related, they provide standards for evaluating the appropriateness of governmental action and official conduct. Traditional separation of powers doctrines are in deep crisis. Conceived as a set of prescriptions, they appear increasingly obsolete and incoherent. Conceived as descriptions, they obscure more than they clarify what is actually going on in the world, undermined, notably, by the politics of rights adjudication. We nonetheless discuss traditional separation of powers doctrines here in order to provide a common baseline against which changes in different systems might be measured. In general, the breakdown of relatively strict separation of powers norms reduces the centrality or primacy of legislatures as sites of rule innovation and enhances the role of judges. Cross-national differences between how courts actually operate, as sites of lobbying and policy innovation, are gradually being reduced. Nonetheless, we note the existence of three relatively distinct ‘families’ of constitutional separation-of-powers arrangements in the countries under consideration. We limit the discussion to the ways in which separation of powers doctrines are designed to minimize the legislative impact of courts. In the British, or Commonwealth, model, the constitutional relationship between the legislature and the courts is formally organized by the doctrine of parliamentary sovereignty. Judicial review of statute is prohibited; and substantive limitations on the legislative powers of parliament, such as rights provisions of a supra-legislative character, may not be presumed. Courts engage in the review of administrative acts, but administrative judicial review can always be justified as the preservation of the prerogatives of parliament, ensuring that the will of the legislature is not subverted by administration. The common law, an explicitly judge-made law, governs in areas not pre-empted by statute. As discussed below, the development of (treaty-based) European law has undermined parliamentary sovereignty, conferring on UK judges the power, for example, to review the validity of parliamentary statute with EU law. In Canada and Australia, judges are also called upon to umpire disputes between the federal and the federated governments, and to
196
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
interpret federal arrangements more generally. Much more important, Canadian judges now exercise substantive review of statute through rights jurisdiction, and Australian judges are poised to do more of such review. The European Continental model views judicial power as a necessary evil (Merryman 1985; Stone 1992: ch. 1). Lawmakers need no help from judges in making the law, which would be a ‘confusion of powers’, but do require judges to help enforce it. Separation of powers doctrines explicitly distinguish the ‘political function’ (legislating and administering) from the ‘judicial function’ (enforcing statutes to resolve legal disputes). Constitutions commonly express these doctrines in three linked ways: (1) they prohibit judicial review of legislation; (2) they deny courts jurisdiction over fundamental rights; and (3) they formally subjugate judicial to legislative authority. As in the United Kingdom, these doctrines assert the supremacy of statute within the legal order. But the Continental model goes much further, formally excluding the discretionary law-making power of judges. Instead, judges are conceived as ‘slaves of the codes’, the codes being the various bodies of legislative provisions which, taken together, purport to govern every possible situation in a more or less comprehensive manner. While the denial of judicial discretion has always been exaggerated, the development of constitutional rights and of European law has rendered traditional separation of powers doctrines obsolete, although they have not been replaced. In the US and now Canada, by contrast, separation of powers means ‘checks and balances’ wherein the judicial branch of government is explicitly granted co-equal status to that of the legislative and executive branches. Judicial review of federal statutes was implied in the US Constitution, at least as the US Constitution was interpreted by Marshall in Marbury v. Madison (1803).3 Further, the US Constitution contains a bill of rights and provides for federalism, both of which generate commitment dilemmas that a hierarchically organized legal system, under a Supreme Court, can help to resolve. Nonetheless, judicial review is limited by the requirement of ‘case or controversy’, which for our purposes is defined as litigation in which one of the disputing parties alleges to have been directly injured by the unlawful enforcement of a public act. Among other things, the case or controversy rule forbids advisory opinions and insists that only a ‘concrete’ injury can lead the courts to consider the ‘abstract’ policy issues raised by a given dispute. The case or controversy rule, and the various doctrines of non-justiciability that have been derived from it (for example, mootness), are meant to protect the other branches from judicial incursion into their work. As discussed further below, during the past half-century the requirement has steadily eroded, and doctrines of non-justiciability, including standing doctrines, are famously incoherent. The orthodox view has long been that the American experience is exceptional and that, the more any legal system comes to resemble the American system, the more courts will matter to the kinds of politics that an undergraduate political science text might take as noteworthy. The Commonwealth model (that is, common law adjudication) typically follows behind; and the Continental systems are usually
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
197
dismissed as being of little or no interest to the political scientist. There are several problems with conceptualizing the world in this way. First, there has been a fundamental change in the system of constitutional justice in several European nations that expands citizen access beyond the European Continental model. We discuss these developments in the next section. Second, some European countries, such as Norway, Sweden, and Switzerland, do not fit this tripartite typology, but blend rules and practices associated with the American and Continental systems. Japan is another example. The Japanese legal system is a civil law system topped by an American-style Supreme Court; all judges possess judicial review authority but few actually exercise it. Third, even where a given ranking might hold across some time scheme, since the Second World War virtually all legal systems have developed in ways that increase the power of judges to control policy outcomes. More specifically, one finds relatively higher levels of judicial discretion to the extent that (1) legislative sovereignty has been rejected or eroded, (2) judges have been given jurisdiction over constitutional rights, and (3) non-justiciability doctrines have weakened. We now turn from these traditional models to the politics of rights, that is, to how rights jurisdiction has served to undermine separation of powers, expand judicial power to impinge upon or control policy outcomes, and to create new opportunities for citizens to use the courts in the service of their own policy goals.
RIGHTS POLITICS The more judges are constitutionally required to protect rights in an effective manner (the pan-European situation today), or the more judges consider effective rights protection to be their constitutional duty (the American situation), the less likely judicial power will conform to, or be contained by, traditional separation of powers doctrines. Put differently, jurisdiction over constitutional rights radically expands the capacity of judges to participate in policy processes. There is no obvious way to separate rights protection from the making of public policy, since rights, by definition, comprise substantive limitations on policy. Further, the judicial route to policy reform can be maximally efficacious, since judicial law-making grounded in an interpretation of a constitutional right is immune from legislative override; such rulings can be changed only through a subsequent judicial decision or by constitutional amendment. Thus, other things equal, policy outcomes produced through adjudication are often ‘stickier’ than those produced by legislators, and this can attract public interest groups on both sides of controversial political questions. We now explore the extent to which citizens can bring rights to bear on the making of public policy, and how this has changed over time.
United States and Canada In the US and Canada, jurisdiction over constitutional rights gives judges opportunities to intervene in virtually any area of domestic public policy. After the
198
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
1940s in particular, American judges began to relax standing requirements, thereby expanding their own discretion and the attractiveness of courts to public-interest litigation. In some domains, litigants who have not been directly injured by a statute can nonetheless plead the interests of third parties—those who might be injured by the enforcement of a law in the future. In practice, this has meant that interest groups routinely use litigation for broad public policy purposes. Furthermore, judges may review the constitutionality of a statute before it has been enforced if petitioners can show that a vague or over-broad statute is likely to produce a ‘chilling’ or ‘deterrent’ effect on the scope or enjoyment of a right. Where substantial rights interests are at stake, American courts view the standard techniques of adjudication (for instance, a concrete ‘case’ between two parties with an opposed interest in the settlement of that ‘controversy’) to be inappropriate. It is not atypical for several dozen interest groups to file suit against a state seeking to reduce access or provision of abortions; the same is true for major free speech cases. In these areas, litigation typically leads to the ‘pre-enforcement’, or ‘abstract’, constitutional review of the statute: provisions found to be unconstitutional ‘on their face’—that is without reference to how they have in fact been applied against a specific person in a specific situation—are voided (Shapiro and Stone Sweet 2002: ch. 6). In virtually all cases, balancing is the preferred means of resolving the dispute and fixing (constitutional) policy, and balancing is a relatively pure form of policy-making. Figure 9.1 depicts the rise in rights cases heard by the US Supreme Court for the terms between 1933 and 1990.4 The Court's attention to civil liberties and rights has grown in fits and starts with a steady increase following the 1960 period. Rights cases constituted only 9 per cent of the Court's case load in the 1933 term, whereas they comprised almost 65 per cent of the docket by the 1971 term. While liberal judicial interpretation of the Bill of Rights sustained this litigation, organizations committed to promoting civil rights and liberties successfully activated and pressured the courts in that direction (Epp 1998). The dramatic increase in rights claims after the 1960s is linked to the creation and development of regional and national women's groups5 as well as growing cooperation on litigation strategies between more established groups such as the American Civil Liberties Union (ACLU), the National Association for the Advancement of Colored People (NAACP), and the American Jewish Congress. Today there exists a dense rightsoriented litigation network that provides legal support and financial resources for a wide range of causes. In Canada, the ‘rights revolution’ is a more recent phenomenon. Canadian judges, whether because of a less individualistic conception of the law or a lingering deference to parliament, have been relatively cautious when confronted with new rights claims (Hartz 1969; Horowitz 1968; Lipset 1990; Epp 1998). Even when a strong legal basis for rights litigation was created (the 1960 Bill of Rights), courts were notoriously modest when it came to enforcing rights against legislative action (Epp 1998: 159). Nonetheless, rights consciousness emerged in Canadian
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
199
Fig. 9.1.Civil rights claims as a percentage of the US Supreme Court's case load, by term 1933–90
Note: Data points represent the annual civil rights claims as percentage of cases heard by the Court. Source: Pacelle (1991). society during the 1970s and, with the adoption of the 1982 Constitution Act and the Charter of Rights and Freedoms, rights litigation experienced explosive growth. Prior to 1982, Canadian government, including the judiciary, remained formally subject to the oversight of the Privy Council of the House of Lords, the highest court in the United Kingdom. The Constitution Act patriated the Canadian Constitution (previously a statute of the British parliament), marking a final step in the country's independence. Most important for our purposes, the Act departed significantly from its Westminster roots, limiting parliamentary sovereignty by way of the Charter and expressly conferring judicial review powers on the courts in a wide area of substantive rights: from fundamental freedoms similar to the US First Amendment rights to special language rights.6 Further, the Charter not only radically expanded the scope of judicial discretion but provided virtually unlimited basis for bringing rights litigation (see Bogart 1994; Hiebert 2002; Russell 1992). In particular, section 24 of the Charter enables individuals to seek judicial remedies for infringement of these rights and freedoms and empowers Canadian courts to apply a ‘just and appropriate’ resolution. In practical reality, the situation is now quite similar to the US, as courts throughout Canada are given considerable power to dismantle legislation on the basis of protecting a wide scope of individual rights. This stands in stark contrast to the pre-Charter period, in which judicial review was largely restricted to federalism cases. Figure 9.2 charts the growth of individual rights litigation before the Canadian Supreme Court between 1960 and 1990.7 Rights cases comprised nearly 60 per cent
200
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
Fig. 9.2.Civil rights claims as a percentage of the Canadian Supreme Court's case load, 1960–90
Note: Data points represent the annual civil rights claims as a percentage of cases heard by the court. Source: Epp (1998). of the Court's case load in 1990, compared with less than 13 per cent before 1975. Today, Canadian citizens are increasingly utilizing judicial institutions to pursue diffuse public interests, a form of participation that was extremely limited before the Charter came into effect.
Europe and the New Constitutionalism The institutional foundations of judicial power in Europe have been transformed by the development of rights jurisdiction, which has fatally weakened parliamentary sovereignty while expanding the nature and scope of judicial discretion. We discuss this transformation here as the product of three separate processes: the constitutionalization of the national legal order, the constitutionalization of the European Union (EU), and the domestication of the European Convention on Human Rights (the Convention). Each of these outcomes was heavily conditioned by the experience of fascist dictatorship, the Holocaust, and the devastation of the Second World War, which showed that legislative sovereignty was not enough to guarantee the viability of democracy and public liberty.
Rights and the ‘Constitutionalization’ of the National Legal Order As Table 9.1 shows, there exists considerable cross-national variation in types of review powers possessed by such courts in Europe. Austria, France, Germany, and Italy (as well as Portugal, Spain, and now virtually every new democracy in central and eastern Europe) possess specialized constitutional courts charged with
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
201
Table 9.1.Review powers held by courts, by country Enumerated rights Austria Belgium Denmark Finland France Germany Ireland Italy Netherlands Norway Sweden Switzerland United Kingdom
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No
Judicial or constitutional review of rights claims Yes Yes Yes Yes Yes Yes Yes Yes No No Yes No No
defending constitutional rights against infringement by all public officials, including legislators. Belgium, Denmark, Finland, and Ireland have supreme courts with jurisdiction over rights, or a mix of supreme courts and a constitutional court. Although most countries now have some form of rights review, Switzerland, the UK, the Netherlands, and Norway do not, although there is a movement to bring such review to the Netherlands (ten Kate and van Koppen 1995). Sweden has a system of judicial review, but it is virtually never used (see Holmstrom 1995). The most important changes have taken place in the zone of the civil law (Austria, France, Germany, Italy), with the establishment of new constitutional courts with extensive rights jurisdiction, or powers to police federalism, or strong regionalism (Austria, Belgium, Germany, Italy). Constitutional rights are more extensive, broader in scope, and more social (collectivist and positive) than the classically liberal rights announced in the American bill of rights (see Stone Sweet 2000: chs 2, 4). Not surprisingly, these new systems have developed inexorably to encourage more publicinterest policy-driven rights litigation. There are three basic ways in which constitutional courts have been instrumental in enhancing the law-making discretion of judiciaries. The first is through ‘concrete review’ of public acts, including statute. The Austrian, German, and Italian constitutional courts possess this type of review authority. Concrete review is initiated by the judiciary (the ordinary or non-constitutional courts) in the course of litigation in the courts, most often when one of the litigants claims her rights have been violated by government. Every presiding judge is authorized to refer constitutional questions—is a given law, legal rule, judicial decision, or administrative act constitutional?—to the constitutional court. Referrals suspend proceedings pending a review by the constitutional court. Once rendered, the constitutional court's judgment is sent back to the referring judge, who then decides the case
202
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
with the help of the ruling. Constitutional decisions are binding on the judiciary; and the judge's failure to send a referral or to properly apply the constitutional court's decision is reviewable in Germany and Austria through an individual complaint procedure (see below).8 Ordinary judges are not permitted, at least in theory, to determine the constitutionality of public acts on their own, but they are usually permitted— or obliged—to interpret the codes as if they conformed to rights provisions. When ordinary judges engage in such ‘principled interpretation’, they in effect rewrite the codes in light of rights. Second, after judicial remedies have been exhausted, German (Singer 1982) and Austrian citizens have the right to petition the constitutional court directly in what are called constitutional complaints (Verfassungsbeschwerde). The German court now receives more than 4,000 such appeals annually, the vast majority of which are summarily rejected. Of the more than 200 complaints a year that actually result in a decision, a modest but rising percentage deals with matters more public than purely private. To take just one example, the famous 1993 decision of the German Federal Constitutional Court on the Maastricht Treaty (in which the Court asserted the power to review the ultra vires nature of EU acts) was reached pursuant to an individual complaint asserting that the treaty on European Union had weakened German democracy. Third, interactions between constitutional courts and the judiciary have gradually but profoundly transformed the role and function of the courts, a phenomenon known as ‘the constitutionalization of the law’ (Stone Sweet 2000: 114–25). Constitutionalization liberates the ordinary judges from the supremacy of statute by adding a higher rung, a constitutional rights level, to the hierarchy of law that judges must enforce. Outcomes are partly the logical, normative consequence of the direct applicability of rights provisions in the legal order, and partly the product of complex dialogues between constitutional judges and the judiciary. There is space here to mention only the most important outcomes of this process and to discuss variation among cases. Cross-national differences in the scope, pace, and intensity of constitutionalization is closely tied to the existence, or non-existence, of particular modes of review. In Germany, where concrete review and individual complaint procedures coexist, extensive constitutionalization has proceeded rapidly. Indeed, well over 90 per cent of all individual complaints allege that a judicial decision has failed to take into account constitutional rights or the constitutional court's decisionmaking. For the constitutional court to decide on the merits of such claims, it must delve deeply into the workings of the presiding judge, and it has the power to impose its own preferred outcome on any recalcitrant judge. Constitutional complaints thus have the effect of forcing the ordinary judges to take rights claims seriously. In Italy, the absence of an individual complaint mechanism has reduced the capacity of the constitutional court to control judicial outcomes. In consequence, constitutionalization has been more gradual and piecemeal, and the high civil court, Cassazione, has retained substantial interpretive autonomy. In practice,
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
203
this has meant that the Italian judiciary is capable of interpreting the codes in light of rights on their own, without a prior reference to the constitutional court. Indeed, in the area of labour law the ordinary judges tend to be more ‘socialist’ on rights than the constitutional court, and labour unions have used the courts to great benefit. In France, where promulgated statute retain their formal, sovereign character, and no formal links between ordinary and constitutional jurisdictions exist, a primitive form of constitutionalization can nevertheless be observed.
The European Court and ‘Constitutionalization’ of the EU The death of parliamentary sovereignty and the rise of rights politics in Europe cannot be understood without attention to the constitutionalization of the European Union: the process by which the Treaty of Rome has evolved from a set of legal arrangements binding upon sovereign states into a quasi-federal, vertically integrated legal regime in which individual rights are juridically guaranteed. The process has been driven by interactions between the European Court of Justice (ECJ), national judges, and private litigants, pursuant to the ECJ's announcement of the doctrines of direct effect and supremacy in the 1960s. The doctrine of direct effect holds that, under certain conditions, Community law confers upon individuals rights that all national officials, including judges, must uphold. The doctrine of supremacy holds that, in the event of a conflict between Community law and national law, the former must always take primacy over the latter. These moves integrated national and supranational legal systems, establishing a decentralized enforcement mechanism for EC law.9 The mechanism depends entirely upon the initiative of private actors, through Article 234 of the Rome Treaty, which creates a kind of concrete review reference procedure for the EU. The doctrine of direct effect empowers individuals and companies to sue member-state governments or other public authorities for either not conforming to obligations contained in the treaties or regulations, or for not properly transposing EU legislation into national law. The doctrine of supremacy prohibits public authorities from relying on national law to justify their failure to comply with EC law, and requires national judges to resolve conflicts between national and EC law in favour of the latter. Article 234 was designed to promote the consistent application of EC law in the member states. The member states did not mean to provide a mechanism by which individual litigants could sue their own member states, nor did they mean to confer on national judges the power of judicial review of national legislation. Both of these outcomes, however, inhere in supremacy and direct effect, and related doctrines. Table 9.2 tracks Article 234 references on an annual basis since the first reference in 1961 by country. This measure is the best indicator now available of the degree to which litigants have claimed rights issuing from EU law in national courts (although it is far lower than the number of cases in which EU law is material that courts actually process). Levels of references were very low during the 1960s, and then began to pick up in 1970 as EU rules entered into effect and as
204
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
Table 9.2.Total number of individual rights claims pursuant to Article 234, EU Treaty, by country and policy domain, 1961–97 Country Austria Belgium Denmark Finland France Germany Ireland Italy Netherlands Sweden UK TOTAL Policy Agriculture Free movement of goods Social security Taxation Competition Establishment Social provisions App. of laws Free movement of workers External Environment Transportation Commercial policy TOTAL
1961–9
1970–9
1980–9
1990–7
Total
0 10 0 0 7 31 0 3 22 0 0 73
0 78 6 0 86 275 6 84 108 0 20 663
0 147 25 0 282 346 15 124 185 0 85 1,209
37 155 38 9 194 405 13 294 148 11 139 1,443
37 390 69 9 569 1,057 34 505 463 11 244
13 14
270 165
309 307
236 314
828 800
26 14 8 1 0 1 1
111 33 46 23 9 13 28
130 98 66 56 52 72 62
176 249 153 163 171 130 86
443 394 273 243 232 216 177
1 0 0 0
21 1 15 11
36 20 18 21
43 57 42 36
101 78 75 68
79
746
1,247
1,856
Notes: Entries are the total number of Article 234 references. All time periods include ten years except the 1990s, which contains data for only eight years. The data are not adjusted for varying dates of entry into the European Union. Accession dates are as follows: Austria (1995), Belgium (original member), Denmark (1973), Finland (1995), France (original member), Germany (original member), Ireland (1973), Italy (original member), Netherlands (original member), Sweden (1995), UK (1973). Source: Data compiled by the authors from Stone Sweet and Brunell (1999).
national judges accepted the doctrines of supremacy and direct effect. References doubled by 1980, then levelled off again until 1985, after which they shot up once more (post-Single European Act in 1985; see Fligstein and Stone Sweet 2002). The data are also picking up the effects of the ECJ's tendency to interpret member-state obligations under the treaty as substantive rights for individuals. To take two examples (Stone Sweet and Caporaso 1998), although the treaty forbids states
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
205
from engaging in trade protectionism and requires member states to ensure that men and women be paid equally for the same work, the ECJ converted these provisions into individual rights provisions: traders have rights that they enforce against national governments in national courts (Poiares-Maduro 1998), and the equal pay for equal work rule has been interpreted broadly to catch many forms of workplace discrimination based on sex (Cichowski 2001). As important, over the past three decades the ECJ has incorporated into the Rome treaty a kind of charter of rights which is based on the European Convention of Human Rights and what the Court calls the shared ‘constitutional traditions’ of the member states (Weiler and Lockhart 1995a, b). Working to catch up with the Court, the member states are presently engaged in drafting a formal, written charter of rights for the EU. Table 9.2 also shows the extent to which EU law has expanded in density and scope. Early on, reference activity was dominated by producer interests (free movement of goods and agriculture products) which claimed over 50 per cent of the references in the 1970s. This percentage falls steadily thereafter, to 27 per cent in the 1990s. At the same time, we see an important expansion of reference activity to public policy domains such as environmental protection, taxation, commercial policy, and the free movement of workers. Strikingly, in the 1990s nearly one in twelve references concern social provisions, the majority of which are claims to gender equality. Clearly, as the density EU law expands the legal system becomes a vehicle not simply for economic interests but also for more diffuse ‘public interests’. Further, citizens and non-governmental organizations—including groups concerned with environmental protection and consumer rights—are increasingly utilizing this legal avenue to pressure for policy change.10
Domestication of the European Convention on Human Rights The expansion of a pan-European rights politics has also been sustained by gradual absorption of the European Convention on Human Rights into national legal orders. In 1950, twelve states signed the Convention (there are now forty-four members), creating not only a European charter of rights but a set of procedures and organizations for protecting those rights. The procedures are complex, but the treaty provides for judicial review of alleged rights abuses by the European Court of Human Rights (ECHR) after legal remedies available in the national system have been exhausted. It bears emphasis that legal systems that would not otherwise possess rights review through domestic constitutional provisions can do so through domestication. Table 9.3 charts the annual number of individual claims made directly to the Convention's organizations (both the Commission prior to 1998 and directly to the Court thereafter) between 1961 and 2000. The data show a steady increase in rights claims, with an upward spike beginning in the 1980s. The latter jump is explained by two factors. Under Article 25 of the Convention, member states could opt to allow direct claims by nationals to the Convention's organizations (instead of being filtered through other processes). By the mid-1980s, every country had done so.11 The attractiveness of the system will be further enhanced by a 1998 decision to reform the regime's institutions: member states abolished the
206
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
Table 9.3.Total Number of rights claims brought before the European Court of Human Rights, by country, 1961–97 Country Austria Belgium Denmark Finland France Germany Ireland Italy Netherlands Norway Sweden UK TOTAL
1961–9 3 3 0 0 0 1 3 0 0 0 0 0 10
1970–9 4 5 1 0 0 3 0 0 2 0 2 5 22
1980–9 17 21 4 0 6 20 3 17 10 0 13 42 153
1990–7 48 20 11 6 77 10 4 130 30 3 22 53 414
Total 72 49 16 6 83 34 10 147 42 3 37 100
Source: Data complied by the authors from Council of Europe (1996; 1997) and Gomien (1995). The authors would like to acknowledge the invaluable research assistance of Tamir Moustafa in collecting these data.
European Commission on Human Rights (formerly the main access point for claims), and strengthened and enlarged the European Court of Human Rights. All contracting states are now obliged to permit direct access to the ECHR. The most regularly invoked provisions of the Convention concern rights to due process and fair trial. Between 1960 and 1995, the European Court of Human Rights issued 552 judgments, of which 332 involved such claims (60 per cent of the total). These rulings, which are binding, have provoked legislative changes and eroded national parliamentary autonomy over delicate human rights issues (such as treatment of radical or terrorist groups). The data also show considerable cross-national variation. The Convention alters the opportunity structures for rights claims at the national level, but differentially across legal regimes. Most important, the Convention places the decision-making of national judges and administrators of the criminal justice system in the shadow of European human rights law. National judges may use the Convention to enhance their power vis-à-vis the legislature and the executive12 to the extent that expanded rights jurisdiction gives powers to the courts more generally. As important, lawyers, citing an ever-expanding case law, routinely threaten to appeal national decisions to the ECHR, and these threats can provoke changes in national practices, both formal and informal. Considered cross-nationally, the effectiveness of the Convention is partly dependent on how it has been incorporated into national law. In some countries, including the Netherlands, Austria, Belgium, France, and Switzerland, the Convention has quasi-constitutional status: that is, it is both directly effective and
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
207
takes primacy over conflicting statute. The domestication of the Convention creates a judicially enforceable bill of rights for those countries that would not otherwise have one, such as France, the Netherlands, and Sweden. In others, such as Germany and the United Kingdom,13 incorporation has taken the form of a statute rather than a quasiconstitutional act, which limits the ability of individuals to use the Convention to question the legality of a public act or practice. Where national courts cannot or do not enforce the Convention on their own, individuals are all the more likely to go to the ECHR to force them to do so.14 The fact that individual claims from the United Kingdom make up a large portion of the ECHR's case load may be partially attributable to the lack of opportunities to make the claims within the domestic legal system of the UK.15
Australia, New Zealand, and Japan As in Canada, the development of rights politics in Australia finds its historical roots in the American and British experiences. Looking to the US, the framers of the Australian Constitution (1900) created an independent supreme court, the High Court, and gave it significant judicial review powers, primarily in the area of federalism (Galligan 1987: 62). From Great Britain, they adopted the tradition of parliamentary sovereignty and therefore did not include a bill of rights (Kennett 1994: 582). For most of its history, the High Court's primary concern was with patrolling the boundaries of federal and state powers, not with constitutional rights review. In the 1990s, the dogmatics of parliamentary sovereignty began to weaken as consensus grew among judges, scholars, and even some politicians that strong prime ministerial democracy had displaced parliament as the centre of gravity of political life, thereby jeopardizing individual rights and freedoms (Brennan 1991; Toohey 1993). At the same time, judges began to assert new powers to review governmental acts, despite the absence of a charter of rights (Toohey 1993). Between 1990 and 1993, for example, the High Court handed down a series of rulings that in effect revised the Constitution with respect to rights.16 Relying on both its own precedents as well as those of certain British courts, the High Court extended the traditional doctrines of judicial review, which apply ‘not only to rights against the state, but also to claims upon the state’ (Bailey 1995: 1). The Court is now in position to develop the rights implied by the Constitution as a new ‘common law of the Constitution’ (Bailey 1995: 1). This doctrinal shift, in effect, opened the door to new forms of rights-based challenges to the government. There is virtually no available literature on judicial politics in New Zealand (but see Hirschl 2000), despite some important recent developments. In 1990, however, the parliament adopted the Bill of Rights Act, which lists rights and freedoms, requires the courts to interpret statutes in light of rights provisions, and elicits the advice of the Attorney General whenever he or she has reason to think that a bill being debated in parliament would violate the Act. Judicial review of statutes, however, remains prohibited.
208
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
To this point, our analysis of rights has been dominated by the Western experience. The Japanese legal system provides an important corrective to the view that rights naturally empower courts and private individuals to the advantage of public-interest litigation. The word ‘right’ is in fact quite new to the Japanese legal system. Following the adoption of the French Civil Code in the 1870s, the word kenri was created to denote this concept (Port 1996: 560). Of course, the concept of individual obligations to others implies the concept of an entitlement to be treated in particular ways, and duties and entitlements have a lengthy historical tradition in Japan. Still, the Japanese make use of rights differently from their Western counterparts (Port 1996; Feldman 2000). Most Westerners would agree that rights create a legal interest that can be enforced against the will of another person or the government. In the Japanese culture of dispute resolution, the public assertion of rights is rarely used in one-on-one arguments. Often, when the Japanese feel angry or aggrieved, they are much more likely to walk away, change the subject, or adopt an extraordinary politeness (Feldman 2000: 4–5). The appeal to a legal right, in everyday life or before a judge, is therefore far more indirect or embedded in a thick web of circumstance. Institutional factors lead to comparatively lower rates of rights-oriented public-interest litigation. In the 1870s the Japanese legal system adopted the traditional Continental model: legislative sovereignty and no judicial review. Today, despite an American-style Supreme Court created after the Second World War, the courts continue to be relatively more deferential to the legislature on questions of policy (O'Brien and Ohkoshi 2001). The role of the courts in actions brought by citizens against the government is particularly limited. While Japanese courts can suspend administrative actions, the Prime Minister can overturn such rulings; and there is no clear doctrine that empowers the courts to force an administrative agency to take positive action (Port 1996: 120).17 Furthermore, the Japanese legislature has institutionalized systems of mediation and arbitration, partly to discourage individuals from asserting rights before judges (Feldman 2000: 34). Nevertheless, rights litigation is found in disputes where the parties have lost interest in continuing a harmonious relationship and there is little hope for informal agreement. Figure 9.3 tracks the number of cases filed with the Supreme Court between 1962 and 1991. The case load of the Court has actually decreased over time, which makes Japan a significant outlier in this study. The system has not worked to encourage such litigation or to counter cultural proclivities to avoid litigation. Japanese citizens have successfully used courts to achieve policy reform, as witnessed by the environmental protection movements of the 1960s and 1970s, a story taken up in the next section.
STANDING TO SUE Rules governing standing to sue affect public-interest litigation in obvious ways. For our purposes, the relevant rules are those that enable interest groups to attack
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
209
Fig. 9.3.Total number of cases before the Japanese Supreme Court, 1960–91
Note: Data points represent the annual number of cases heard by the court. Source: O'Brien (1996). the legality of a governmental action in court. Traditionally, legal systems barred interest groups from taking court action on the grounds that they lacked, by definition, the kind of standing over which courts possessed jurisdiction. Most important, a plaintiff had to be able to show that he or she had been personally injured by state wrongdoing or negligence. Extending standing to sue to groups constitutes a relaxation of such ‘case or controversy’ requirements, and turns the courts into sites for public-interest lobby activity. This form of lobbying, of course, differs somewhat from that which is directed at legislatures and executive agencies in that it proceeds as litigation, with all the trappings of judicial process. Standing rules are fixed through legislation and judicial rulings (precedent); they therefore vary sector by sector. Scholars have not done a good job cataloguing such rules: comprehensive data, even within any one country in our study, do not exist. In this section, we focus on environmental protection sector as an example, and in particular on the standing of non-governmental organizations (NGOs) in that policy domain. Environmental protection regimes exemplify a more general move toward delegated governance (Shapiro 1988; 2002). The standard line is that the delegation of governmental authority to an executive agency is functional for legislative bodies in so far as delegation reduces the overall transaction costs associated with governing: legislative bargaining and commitment; monitoring and enforcing agreements; and developing rational policies in the face of technical complexity, incomplete information, and powerful incentives for rent-seeking (see Thatcher and Stone Sweet 2002). In North America and Europe today, environmental protection legislation is typically administered and enforced by specialized agencies, and their work is reviewable by the courts. Legislators may provide standing to
210
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
citizens’ groups to sue the agency or other public authorities, thus creating a decentralized means of monitoring the performance of the agency and of enforcing legislative priorities.
United States and Canada Although the United States and Canada share a common British legal lineage for standing rules (classic case or controversy requirements), the two systems have diverged widely over the past four decades.18 While the US once led the movement in Western legal systems towards opening the courts to public-interest litigation, it is now the one country in our study that has experienced the opposite trend in recent years. Table 9.4 displays the main statutes and case law relating to standing for citizen litigation adopted between the 1970s and 1990s in both the United States and Canada. All of these moved in the direction of liberalization, until the 1990s in the US. In the 1970s, the US Supreme Court's decision in Sierra Club v. Morton apparently consolidated judicial acceptance of broad standing to sue for citizens groups committed to protecting the environment.19 The court accepted ‘aesthetic’ harms to the environment as a qualified ‘injury-in-fact’ to allow standing for the Sierra Club, a nationally recognized environmental NGO. This doctrine was further Table 9.4.Important statutes and case law relating to standing rules for citizen environmental litigation in the United States and Canada, 1970s–1990s
United States
(restrictive) (restrictive) (restrictive) Canada
Statutes and case law relating to standing Clean Water Act Sierra Club v. Morton United States v. SCRAP Japan Whaling Association v. American Cetacean Society Lujan v. Defenders of Wildlife Endangered Species Act Steel Co. v. Citizens for Better Environment Friends of the Earth Inc. v. Laidlaw Env. Services Thorson v. Attorney General Ontario Environmental Protection Act Minister of Justice v. Borowski Charter of Rights and Freedoms Fisheries Act Minister of the Environment v. Finlay Canadian Environmental Assessment Act Act Respecting Environmental Rights in Ontario
Sources: Owens (2001); Carlson (1998); Handley (2002); Corriveau (1995).
Year 1972 1972 1973 1986 1992 1994 1998 2000 1975 1980 1981 1982 1985 1986 1992 1993
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
211
elaborated in United States v. SCRAP.20 In that case, a group of George Washington University law students challenged a federal rate increase for scrap metal transporters, arguing that this would discourage the transport and reuse of metal, leaving behind litter that would cause aesthetic harm to the group's environment. The Court took the case. In the 1980s, the Court heard only one major standing case in the environmental area (Japan Whaling Association), but did not reduce access to the courts for citizens and groups.21 This trend came to a halt in the 1990s. In Lujan (1992), the Court rejected a suit brought by an NGO that sought to bring certain animal species found abroad under the enforcement provisions of the Endangered Species Act.22 The Court denied standing on the grounds that the group had not shown that it had suffered a legally recognizable injury capable of judicial redress within the range of interests protected by the statute (Handley 2002).23 The ruling stands as a far cry from the liberal interpretations of ‘injury-in-fact’ adopted in the previous decades. The Court appears to have adopted a consistent posture toward standing, returning to a more traditional orientation.24 The change of heart can largely be attributed to the current conservative make-up of the Court (Owens 2001; Pierce 1999), whose priorities appear to be (1) reversing ‘judicial activism’ and (2) better husbanding the scarce resources of the federal courts in the face of frivolous, or purely ‘political’, lawsuits.25 Canada, however, continues to liberalize its requirements for standing. In the 1970s and 1980s, Canadian courts opened their doors wider to citizens by allowing constitutional challenges against statutes on the grounds of a ‘generalized grievance’, a departure from earlier doctrine requiring a ‘special injury’ (Owens 2001: 343). The court made this argument first in the Thorson decision (1975) and reaffirmed this doctrine later in the Borowski (1981) and Finlay decisions (1986).26 Canadian provinces, for their part, have adopted statutes that include provisions for citizen enforcement and litigation (Schofield and Thompson 1994). The Yukon Territory's Environmental Act is an example. The measure dismantled the common law special-injury requirement for standing, allowing citizens and groups greater ease in bringing cases against both private parties and government agencies for violations of the Act. Similarly, Ontario has adopted both an Environmental Protection Act and the Fisheries Act, which offer the same permissive standing requirements (Schofield and Thompson 1994). The 1990s witnessed the extension of this access. At the federal level, the Canadian Environmental Assessment Act (1992) allows citizens and NGOs greater participation in the planning stage of clean-up projects, as well as standing to activate judicial review of projects with potentially harmful environmental effects.27 Similarly, the Act Respecting Environmental Rights in Ontario (1993) provides citizens with a set of environmental rights, including the ability to initiate an enquiry with the Environmental Commissioner regarding potential polluters and the right to prosecute the implicated polluter (Corriveau 1995). The Canadian pattern is gradually taking hold in Europe.
212
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
Europe Most European countries have moved toward a more permissive approach to standing. Table 9.5 lists the main statutes and judicial rulings relevant to environmental litigation in the 1960–90 period. Prior to the 1980s, British citizens interested in bringing suits against the government in the area were required to show that they had ‘suffered a special injury apart from the general populace’, thus, making NGO public-interest litigation especially difficult (Owens 2001).28 Beginning with the Fleet Street Casuals case in 1982, the British House of Lords came to interpret this doctrine more flexibly, encouraging those bringing valid complaints, and filtering out the ‘busybodies, cranks and other mischief-makers’.29 In the 1990s, two rulings, pursuant to suits brought by Greenpeace and the World Development Movement, were particularly important for NGOs. The courts took the cases on the grounds that the concerns of these internationally recognized organizations represented the public interest, and not just the work of a meddling ‘busybody’.30 The doctrine now centres on determining whether a lawsuit is frivolous or the expression of a judicially defensible public interest. Table 9.5.Important statutes and case law relating to standing rules for citizen environmental litigation in European countries, 1960s–1990s Belgium Denmark France Germany Ireland Italy Netherlands
Norway Switzerland United Kingdom
Statutes and case law relating to standing Right of Action of Environmental Associations Act Environmental Protection Act Law No.76–629 Nature Protection Decree No. 85–400 Environmental Organizations Federal Nature Conservation Act Local Government Act Air Pollution Act Law No. 349 Ministry of Environment Nuclear Plant Borsele case De Nieuwwe Meer case Environmental Protection Act General Administrative Law Act Administration Act Civil Procedure Act Federal Nature & Heritage Conservation Act Federal Environmental Protection Act Supreme Court Act Fleet Street Casuals Regina v. Inspectorate of Pollution ex Greenpeace Regina v. Secretary of State for Foreign and Commonwealth Affairs ex The World Development Movement
Sources: Führ et al. (1995); Owens (2001); Führ & Roller (1991); Betlem (1994).
Year 1993 1982 1976 1985 1976 1976 1987 1986 1975 1986 1993 1994 n/a n/a 1966 1983 1981 1982 1993
1994
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
213
Italy began allowing citizens to help with the enforcement of certain environmental protection laws as early as the 1960s. However, the courts initially interpreted these provisions narrowly, recognizing standing only for those who had suffered a ‘special injury’. In reaction, parliament adopted a statute that conferred standing on groups with recognized ‘national reputations’, a law that has subsequently been followed by the courts (Parker 1995). More recently, Law No. 349, adopted in 1986, created a comprehensive environmental protection programme and explicitly granted standing to ‘nationally-certified environmental organizations’ irrespective of whether they could show ‘special injury’. This has resulted in a long series of NGO-sponsored suits against government agencies for failure to implement and enforce environmental laws (Cichowski 1998). One finds the same trend in other European countries. The Netherlands, for example, now has a well-established tradition of broad standing to sue, with administrative courts upholding this doctrine as early as 1975. The Dutch Supreme Court expanded these rights in the 1980s and 1990s (Betlem 1994).31 Generally, Dutch courts recognize standing for well-established public interest organizations, and require only that the groups’ goals must be related to the agency's actions that they are challenging (a rule contained in the General Administrative Law Act of 1994 as well). In Ireland, almost any citizen may bring a suit against certain classes of decisions taken by government agencies in the environmental sector (Scannell 1995). Many other European countries have adopted new legislative reforms to permit NGOs access to both administrative and civil procedures to enforce compliance with environmental laws, including France (1976), the Netherlands (1979), Denmark (1982), Italy (1986), and Belgium (1993).32 The German case is a more nuanced one. There are virtually no German federal statutes that explicitly allow citizen suits (Greve 1989). Further German federal law prohibits private parties filing lawsuits on behalf of a ‘public interest’ and goes so far as preventing environmental NGOs from suing on behalf of members (Handley 2002: 126). Yet NGOs are given significant access to other avenues in the political process to pressure for enforcement of environmental protection laws.33 Law-making and compliance decisions in the area are subject to unrelenting lobby activity in the direction of enhanced protection (Greve 1989). Given that environmental law is primarily implemented at the state level, it is worth noting that standing to sue in Länder court is far more permissive. By the 1990s, eleven out of the sixteen Länder possessed statutes granting a right to legal action for accredited associations in certain areas of environmental law.34 That said, Länder courts have adopted somewhat narrow interpretations of these standing rights (Führ et al. 1995: 86). Finally, we note changes in standing rules at the EU level. Until recently, environmental interest groups had enjoyed little success in direct actions before the ECJ, forcing them into the national courts. Indeed, environmental organizations, such as Greenpeace and World Wildlife Fund, have systematically been turned away by the Court for their lack of standing.35 However, in May 2002 the Court of
214
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
First Instance of the ECJ overruled its previous restrictive position and,36 in particular, the requirement that each plaintiff show that he or she had been directly injured by a Community action, and that the specific situation could be differentiated from that of all other individuals. The Court held that it is now no longer relevant how many persons are negatively affected by a measure as long as such affects can be shown to have been directly caused by it. This liberalizing of the standing criteria bodes well for citizen litigation in the future.
Australia and Japan Historically, Australian standing-to-sue rules closely followed the traditional British model, making public interest litigation extremely difficult.37 Table 9.6 lists the main judicial rulings and statutes related to standing in the environmental area in the 1960–2000 period. In its 1980 decision in Australian Conservation Foundation,38 the High Court of Australia denied standing to a conservation organization challenging a government agency for failure to follow an environmental impact assessment procedure. The Court, holding to classic doctrine, ruled that the NGO did not have standing to sue in so far as it had not shown that it had suffered ‘special injury’. The Australian state legislatures, as well as various State-level courts, have nonetheless begun to adopt a more permissive posture. Victoria and New South Wales, for example, adopted environmental legislation that explicitly enables environmental groups to challenge agency actions.39 Similar to trends found in Europe, Australian courts increasingly grant standing to environmental organizations that possess a nationally recognized reputation and a long history of commitment to environmental protection.40 Further, Australian courts recognize a Table 9.6Important statutes and case law relating to standing rules for citizen environmental litigation in Australia and Japan, 1960s–2000
Australia
Japan
Statutes and case law relating to standing Australia Conservation Foundation (ACF) v. Commonwealth of Australia New South Wales Env. Planning and Assess. Act ACF v. Minister for Resources Ogle v. Strickland Truth about Motorways v. Macquarie Fundamental Act for Environmental Pollution Prevention Kumamoto Minamata Disease case Niigata Minamata Disease case Toyama Itai-Itai Disease case Yokkaichi Asthma case Fundamental Act for the Environment
Sources: Kawashima (1995); and Owens (2001).
Year 1980 1979 1989 1987 2000 1967 1973 1971 1971 1972 1993
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
215
‘vocational nexus’, or a special interest that can be shared by a group of individuals in a similar vocation, as sufficient to meet the ‘special injury’ standing requirement.41 Australian business has not welcomed this shift, and has begun to wage its own campaign in the courts. In the recent Truth about Motorways case, business leaders challenged a provision of the Trade Practice Act of 1974 which enabled citizen suits, on the grounds that conferring standing on ‘any person’ was unconstitutional.42 The High Court disagreed, thereby shunning the road taken by the US Supreme Court in the 1990s (Owens 2001). The trend towards adopting permissive standing rules is less evident in Japan. Table 9.6 lists four rulings, known as the Big Four, that paved the way for public-interest litigation beginning in the 1970s. The cases, brought by citizens severely injured as a result of unregulated industrial pollution, were important for three reasons. First, they led the courts to liberalize the standard of proof for negligence in pollution actions, thereby shifting the burden of proof to polluters and making it easier for judges to provide remedies (Goodman 2001: 776). Second, the rulings changed public attitudes towards the relationship between industry responsibility and individual rights. Traditionally, a head of a polluting firm that had damaged a community would assume responsibility, apologize, and resign, and thus absolve his company. The Big Four decisions exhaustively documented the firms’ negligence, publicized their legal wrongdoings, and established enforceable standards of corporate responsibility for the future. The transformative impact on traditional views of rights and litigation has been, by all accounts, extraordinary. Litigating for justice could be labelled an ‘appropriate vindication of rights’ rather than merely selfish or greedy (McKean 1981: 34). Third, these decisions undermined traditional patterns of citizen participation in public policy processes. On the one hand, litigation had gained a new place in politics as a reform tool; and a community's right to protection from pollution could be pleaded before a judge. On the other hand, these changes provoked a counter-reaction from the government, which saw in the availability of new judicial remedies a threat to its control over policy. Its solution was to create an administrative complaint procedure to draw off potential litigation (Goodman 2001: 777). In fact, environmental suits all but disappeared in the 1980s, and public-interest litigation in other areas of public policy, such as sex discrimination, was reduced in the same way.43 In the 1990s, the courts and the legislature began to move towards more permissiveness, if timidly. An important 1989 judicial decision allowed plaintiffs to plead the provisions of multiple statutes rather than simply the one that governed the allegedly negligent behavior (Kawashima 1995: 265). Further, the adoption of the Fundamental Act for Environment in 1993 suggested positive avenues of reform for public interest litigation. Historically, a group or individual claiming a more diffuse public interest (called a ‘reflex interest’ under Japanese law) would be denied standing by Japanese courts, making litigation by environmental groups difficult (Kawashima 1995). Joining individual interests through a
216
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
class action suit was also prohibited, similar to the requirements of European civil law systems. However, the Environmental Act requires government to take diffuse interests into account in its management of the environment, and these interests can be pleaded in court.
CONCLUSIONS This chapter shows that judicial policy discretion has expanded across the globe. How judges have used their discretion, the choices they have made, helps determine how attractive courts, as arenas of policy innovation, will be to interest groups or citizens pursuing more or less diffuse public interests. Judges can choose to widen or narrow gateways to the courts, in light of justiciability doctrines, their own policy goals, or concerns for managing burgeoning case loads. As we have seen, most notably in Australia, Canada, and the US, standing rules are relatively elastic: judges can expand or contract them. Even where such requirements are governed by statute, the courts determine, case by case, whether individuals or groups possess a justiciable interest under the law. With a few exceptions (the US and to some degree Austria and Canada), traditional separation of powers doctrines prohibited constitutional judicial review and denied judges rights jurisdiction. In Europe particularly, legislators, jealous of their sovereignty, refused to share their authority to govern with judges. The tragic inter-war experience of dictatorship and genocide, and the massive American presence in Europe and Japan after the Second World War, destroyed the normative appeal of parliamentary supremacy and placed rights review on the constitutional agenda. Today one observes that, in successive waves of democraticization (Germany, Italy, and Japan after the Second World War, southern Europe in the 1970s, central and eastern Europe in the 1990s), all new constitutions contain judicially enforceable rights review of legislation and other public acts. In the US, too, notions of judicial deference to legislatures have declined, not least because majority rule has at times failed to protect fundamental rights sufficiently (the rise of formal apartheid in the South after Reconstruction, Japanese internment camps during the Second World War, and McCarthyism thereafter). The same suspicion of majority rule has infected Canada and, more gradually, Australia and the UK. Judges are now increasingly required or expected to engage in judicial review, both to monitor compliance with statutory prescriptions and to protect the rights of individuals in their relations to government. Generally, judges have used these powers in ways that have undermined the monopoly or centrality of the ‘political’ branches of government as arenas for policy-making. Inevitably, certain classical themes of democratic theory are raised by the fact that the least representative branch of government has gained so much new law-making authority. How should we assess, normatively, the huge expansion in judicial discretion and the concomitant growth in the social demand for such discretion? There is no right
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
217
answer to the question, of course. However, a variety of defences of judicial power has emerged, and these deserve to be considered in light of this volume's overall concerns. We conclude by briefly mentioning three. First, to extend the citizenry's access to the courts is to increase public participation in the making, monitoring, and enforcement of laws. In the vast majority of countries in our survey (and in the EU), the formal legislative process is a closed, even secret, affair, dominated by the executive, not the legislature (the great exception is the US). As the purview of government has expanded so has the demand for more transparency and accountability, which judicial review can provide. Second, the fact that judicial power is delegated power requires us to think differently about the underlying bases and purposes of democracy from the perspective provided by, say, models of democracy that celebrate majority rule. In domains where legislators—that is, political parties—face acute commitment problems or situations of great technical complexity and uncertainty, delegating authority to non-representative bodies, including courts, can help them achieve better policy outputs than if they tried to make law on their own. Better policy—or output legitimacy—substitutes for representativeness as the criterion for assessing government action. As the administrative state has grown—that is, as legislators delegate law-making discretion to executive agencies or other non-majoritarian institutions—so has the domain of the judge. Courts are deployed, almost naturally, to review and control the use of agency discretion in the name of the legislature. The legitimacy of the courts in such a system is thereby indirectly tied to that of elected officials. Third, contemporary global constitutionalism means rights constitutionalism. Is it acceptable, or legitimate, for any modern democracy to go without judicially enforceable, constitutionally entrenched rights? The UK and a few Commonwealth countries notwithstanding, the question has been definitively answered: ‘no’. The salient issues now are these: is a polity better-off or worse-off without such rights?; do courts make better rights-oriented policy than the ‘political branches’; and to what extent do legislatures and executives become better policy-makers as a result of being placed in the shadow of rights adjudication?
NOTES 1 2
At a minimum, we assume that judges do not behave fraudulently; at most, we expect judges to work to produce defensible decisions. We will not deal with the social logics of precedent and law-making here beyond making the following points. Courts have a powerful interest in propagating and curating doctrinal structures, both to justify law-making and to structure their own environments. Judges portray precedent as an inherently legal constraint on their own discretion, although this constraint evolves endogenously, through law-making. In its idealized form, governance through precedent proceeds incrementally, through formal exercises in normative deliberation and analogic reasoning. A court that does not actually or always proceed in this way nonetheless typically pretends to the rest of us that
218
3 4 5 6 7 8 9 10 11 12
13
14
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
it has. Judges package decisions in ways that make them appear to be self-evident, redundant, deductive extensions of pre-existing law. In this way, precedent camouflages law-making while enabling it. See Shapiro and Stone Sweet (2002: ch. 2). Marybury v. Madison, 1 Cranch 137 (1803). These cases involve individual rights and liberties as protected by the US Bill of Rights and Equal Protection provisions of the US Constitution. See Epp (1998: 26). The Legal Defense and Education Fund of the National Organization for Women, the Women's Legal Defense Fund, and the Women's Rights Project of the ACLU played important roles in supporting women's rights litigation. These rights include ‘fundamental freedoms‘, ‘mobility rights’, ‘legal rights’, ‘equality rights’, ‘official languages of Canada’, and ‘minority language educational rights’. Similar to the US, these data include individual rights and liberties claims. This is litigation based on the Charter of Rights and Freedoms (1982) and the Bill of Rights (1960). See Epp (1998: 172). Individual complaint procedure is also present in the Swiss legal system. However, this system does not possess a constitutional court but instead a supreme court that mixes functions of both the European and the American judicial review models. Our discussion focuses on the systems possessing a constitutional court. In this chapter we consistently utilize the term European Community (EC) law to refer to the body of law adopted prior to the institutional reforms of 1992, since when this international organization has become known as the European Union. In the area of gender equality and the environment, this litigation is the product of considerable mobilization of national and transnational NGOs, from the European Women's Lobby to Greenpeace (see Cichowski 1998; 2001; 2002). Prior to this some countries that had ratified the Convention did not allow direct individual claims. Examples with date of acceptance of art. 25 include, France (1981), Turkey (1987), Spain (1981), and Greece (1985), to name just a few. In France, the Conseil d'Etat, after years of maintaining a reserved attitude towards the Convention, gave priority to an international treaty over a municipal law that was enacted after the treaty had been made applicable (No. 108243, 20.10.1989 Nicolo. Full text of the decision and the conclusions of the Commissaire du Gouvernement, RUDH 3 (1989), pp. 262.). In Germany, where the Convention has the status only of statutory, not constitutional, law, the Federal Constitutional Court has recently upgraded this rank. A recent decision by the Constitutional Court held that the German Constitution must be interpreted in light of the European Convention (Decision 26 March 1987. BverfGE 74, 358. This was later confirmed by Decision of 29.5.1990, EuGRZ 1990, 329). In Germany, Article 59.2 of the Basic Law assigned the Convention the status of federal law (Zusimmungsgesetz). After years of debate, the European Convention was finally incorporated into the British legal system through the Human Rights Act of 1998. This made the UK the oldest signatory state to incorporate. Much of this delay centred on the issue of the status Convention rights would have, as the repercussions of constitutional status, empowering both individuals and national courts vis-à-vis parliament, were not welcomed (see Loveland 1999). The German Federal Constitutional Court has continued to state clearly in its case law that a constitutional complaint may (indirectly) invoke the Convention along with the individual's fundamental right to equality under Article 3.1 of the German Basic Law to argue for an arbitrary misapplication or non-application of the Convention rights
PARTICIPATION, REPRESENTATIVE DEMOCRACY, AND THE COURTS
15 16 17
18
19 20 21 22 23 24 25 26 27
28
219
Decision of 17.5.1983, BverfGE 64, 135 (157); 13.1.1981, VverfGE 74, 102 (128). Essentially, Article 3.1 (‘All men shall be equal before the law’) is interpreted as a fundamental right protecting against unreasonable (arbitrary) distinctions. The Federal Constitutional Court has brought arbitrary misapplication of law by courts under this provision. See Polakiewicz (1996). The United Kingdom presents a similar dynamic, if more limited in terms of an individual's ability to alter public policy through litigation. The Human Rights Act does not give supra-legislative status to Convention rights (and thus allegedly maintains parliamentary sovereignty), yet section 3 of the Act does state clearly that national courts are required to address these rights: ‘So far as it is possible to do so, primary legislation and subordinate legislation must be read and given effect to in a way which is compatible with Convention rights.’ Between 1960 and 1995, 81 out of 552 ECHR judgments involved the United Kingdom. The number of judgments was second only to Italy (123). See Bailey (1995) for a complete list of the cases. For further discussion of both the political and legal ramifications of this case law, see Patapan (2000) and Kennett (1994). In any event, courts are often reluctant to do so. In the Kurokawa v. Chiba Prefecture Election case (1972), for example, the Supreme Court found an electoral system to be unconstitutional, yet refrained from invalidating that election. The decision was subsequently ignored by the government, and both the image and power of the Supreme Court were weakened. As former British colonies, both countries adopted England's liberal standing doctrine in their original founding constitutions or charters. These rules permitted all citizens, even those without a specific injury, to challenge government action in court (see Owens 2001 for an overview). That said, the standing doctrine in the United Kingdom did turn in a more restrictive direction following the Boyce decision in 1903 (Boyce v. Paddington Borough Council, 1 Ch. 109 (1903) App. Cas. 1 (H.L.)). Sierra Club v. Morton, 405 U.S. 727, 734 (1972). U.S. v. Students Challenging Regulatory Agency Procedures (SCRAP), 412 U.S. 669, 687–88 (1973). Japan Whaling Association v. American Cetacean Society, 478 U.S. 221 (1986). Lujan v. Defenders of Wildlife, 504 U.S. at 559 (1992). See Davis and Pierce (1994) for a discussion of standing law. See the following cases: Steel Co. v. Citizens for a Better Environment, 523 U.S. 83 (1998) and Friends of the Earth Inc. v. Laidlaw Environmental Services, 528 U.S. 167 (2000). See argumentation in Friends of the Earth, Inc. v. Laidlaw, 120 U.S. 693 at 710 (2000). See also O'Connor (1986) and Scalia (1983). Thorson v. Attorney General [1975] 1 S.C.R. 138 (Can.) (1974); Minister of Justice v. Borowski [1981] S.C.R. 575; Minister of Finance v. Finlay [1986] 2 S.C.R. 607. The CEAA's predecessor, the Environmental Assessment Review Process (in the 1985 Environmental Act), was the foundation for numerous successful NGO-sponsored cases, including Canadian Wildlife Federation v. Canada [1989] 3 C.F. 309 and Friends of the Olman River Society v. Canada [1992] 2 W.W.R., 193. See Hunt (1992) and Corriveau (1995) for a list of similar cases. See Boyce v. Paddington Borough Council, 1 ch. 109 (1903) for an elaboration of this standing doctrine.
220
RACHEL A. CICHOWSKI AND ALEC STONE SWEET
29 See International Revenue Commissioners v. National Federation of Self-Employed and Small Business Limited [1982] App. Cas. 617 at 653. 30 Regina v. Inspectorate of Pollution and Another ex parte Greenpeace Ltd., (No. 2), 4 All E.R. 329 (Eng. C.A. 1993); Regina v. Secretary of State for Foreign and Commonwealth Affairs ex parte The World Development Movement, Ltd. 1 W.L.R. 386 (Q.B. 1995) at 393. 31 See cases Nuclear Plant Borsele KB of 8 September 1975, Nederlandse Jurisprudentie/Administratiefrechte Lijke Beslissingen 1975, 275 vdV; De Nieuwe Meer (1986) HR 27 June 1986 Nederlandse Jurisprudentie 1987, 743. 32 See Ormond (1995) for an overview. 33 The Federal Nature Conservation Act (1976) has helped create these openings. 34 The following lists the Länder possessing at least limited standing statutes for environmental associations and date of adoption in chronological order: Bremen (1979), Hesse (1980), Hamburg (1981), Berlin (1983), amended in 1990, Saarland (1987), Lower Saxony (1993), Sachsen-Anhalt and Brandenburg after 1990. Notably absent from this list are three Länder with the greatest industrial sectors and thus the highest levels of potential environmental pollution: Bavaria, Baden-Württemberg, North Rhine-Westphalia. 35 Greenpeace v. Commission, Case C-321/95 ECR 1998: 1651; Greenpeace v. Commission, C-461/93 ECR 1995: 2205; An Taisce/WWF v. Commission, Case 585/93, ECR 1994: 733. 36 Jégo-Quéré and Co. v. Commission, Case T-177/01 ECR 2002: 34. 37 See n. 25 and associated text. In particular, Australian courts draw from the British ‘special injury’ requirement elaborated in the Boyce v. Paddington Borough decision. 38 Australian Conservation Foundation, Inc. v. Australia (1980) 146 C.L.R. 39 New South Wales Environmental Planning and Assessment Act of 1979. See Preston (1991) and Murchinson (1995). 40 See Australian Conservation Foundation v. Minister for Resources (1989) 19 A.L.D. 70. 41 See Olge v. Strickland (1987) 77 A.L.R. 41. 42 Truth About Motorways Pty Ltd. v. Macquarie Infrastructure Inv. Mgmt. Ltd. (2000) 11 Austl. 43 The government's response to sex discrimination lawsuits in the 1960s and 1970s was similar, with a transfer of power away from the courts to the bureaucracy following the adoption of the Equal Employment Opportunity Act in 1985 (Upham 1987).
III The Consequences of Political Reform
This page intentionally left blank
10 A Second Transformation of Democracy? Mark E. Warren In their earliest incarnations, democratic institutions were small-scale arrangements, limited to locales and city-states by their defining mechanism, decision-making by the assembled people. By the early modern period, most of the existing direct democracies of northern Europe and Italy had been eclipsed by larger political and economic units. Even when these democracies were not overthrown, their power was undermined through incorporation into larger political units. The first major transformation of democracy thus involved the development of institutions of representation—a second form of democracy that enabled it to spread to larger political units and incorporate large numbers of people. These institutions included national parliamentary government, elected representatives, and popularly elected local governments operating within the encompassing elected national governments (Dahl 1998: 17). The institutions of representative democracy spread gradually at first. Since the Second World War and particularly in the last two decades, however, the number of countries using the institutions of representative democracy has increased dramatically. Representative democracies now, for the first time in history, encompass a substantial majority of the world's population (Freedom House 2000: 2). While recent attention has focused on this important expansion of representative democracy, the familiar forms of representative democracy have themselves been undergoing transformation. This volume documents two trends in the OECD countries, that is, the wealthy, consolidated liberal democracies. The first trend involves a deepening of representative democracy over the last several decades. These changes include, for example, the participation of more groups and individuals within political parties, and an increasing number of elective offices. The second trend is of much more recent origin, and potentially represents a striking change in what democracy looks like and how it operates. These changes include the rise of social movements, dramatic increases in the numbers and activities of associations and interest groups, new forms of direct action, increasing use of referendums, devolution and deconcentration of decision-making and governance, stakeholder representation within bureaucracies, a growing use of the courts to press citizen interests, new experiments in collaborative governance and deliberative policy-making, more vigorous public debates about policies, increased
224
MARK E. WARREN
public monitoring of government and corporate activities, new political uses of communication technologies, and small groups aggregated into networks that are now often global in scale. The question I pose in this chapter is: do these changes so alter or complement the institutions of representative democracy that we are entitled to speak of a second transformation of democracy leading to a third historical form? Are the qualitative changes now occurring in democracy of such magnitude and importance that they compare to the transformation of local assembly democracy into representative democracy? This question is both empirical and normative. The empirical dimension involves documenting changes of the sort just mentioned, including the institutional changes documented in this volume. The normative dimension involves an assessment of whether these changes parallel the first transformation with respect to conserving and extending democracy. In the best of cases—and these were not typical—early republican forms of democracy were preserved, but within the context of new forms. Similarly today, representative democracy faces, if not obsolescence, at least diminished importance in the face of globalization, the de-centring of the nation-state, complexity, and functional differentiation (Habermas 2001: ch. 4; McGrew 1997; Held 1995; Beck 1997). The notion of a ‘second transformation’ would suggest that emerging forms of governance function to preserve electoral-representative democracy, but that democracies now function in ways that represent a qualitative break with the representative model. It is not possible, of course, to provide definitive support for this speculation. By their very nature, such changes are complex and incremental; they could be seen to count for something as global as a ‘transformation’ only in retrospect. What I offer here is a highly speculative and often purely stipulative sketch of what the notion of a second transformation would involve. I ask: (1) what kinds of phenomena would a second transformation encompass, should one be occurring? And (2) how might we judge these developments from a democratic perspective?
WHAT WOULD A SECOND TRANSFORMATION ENCOMPASS? In order to frame the first question, we should ask: What kinds of social relations might potentially be organized in democratic ways? By ‘democracy’ I shall mean something very generic so as not to limit by definition the conceptual terrain of possibilities. The long-established norm of democracy is that political systems should maximize rule by and for the people. Within the tradition of liberal democracy, ‘the people’ are understood as individuals, each of whom ought to benefit from collective self-rule. From this norm is derived that of equal inclusion: every individual potentially affected by a decision should have an equal opportunity to affect the decision. ‘Democracy’ in this generic sense could in principle extend to all social relations, institutionalized or not, that enable individuals to influence the collective decisions that affect them.
A SECOND TRANSFORMATION OF DEMOCRACY?
225
This conception of democracy casts a net over a domain of social relations considerably wider than the governmental institutions considered in this volume. There are important reasons to do so, however, owing to the structural pluralization of sites of collective decision and action in today's democracies. Because representative democratic institutions have evolved within liberal constitutional forms, they have functioned not only to guide and legitimize the actions of governments but also to enable collective decisions outside of the domain of government. These constitutional forms unleash not just forces of the market but also those of society, producing a growth in associational activity as well as public spheres within which public wills take shape (Preuss 1995; Habermas 1996; Cohen and Arato 1992). States themselves have become internally differentiated in ways that pace the differentiation of their environments, with hundreds of specialized agencies and thousands of departments. These broad, structurally induced patterns of pluralized collective capacity have generated patterns of political conflict that have exceeded the capacities of representative democracy simply because, as the powers of collective action have become de-centred, so have the patterns of political conflict (Offe 1996; Beck 1997). Globalization reinforces these structural developments by undermining the powers of territorially based nation-states (Habermas 2001; McGrew 1997). These structural developments have been paralleled by individual and cultural transformations. While there are many variations and exceptions, individuals in modern societies have become more reflexive, that is, more conscious of lifechoices and more likely to engage in self-monitoring (Giddens 1991; Beck 1997; Habermas 1990). In differentiated societies, individuals participate in multiple domains and institutions, each of which provides perspectives on the others owing to their differing and often competing norms and expectations. Combined with a dramatic increase in communications, these forces tend to produce individuals who, because they have the responsibility for their biographies, are also more likely to pay attention to those venues of collective action that seem most salient to lifecrafting. These venues may not be governmental in nature; indeed, there is evidence that in the OECD countries the celebrated apathy of citizens may be something quite different: those with educational and economic resources have the capacities to determine their lives, and maintain the involvements that enable them to do so. Government is more important for those with fewer resources while less relevant for those who have opportunities and can take good government for granted (Van Deth 2000; Mansbridge 1997). At the same time, there is evidence that individuals in the OECD countries are, on average, more interested in controlling their futures, more attentive to and critical of government performance, less deferential to authority, and more likely to participate in the domains most salient to their life-chances (Dalton 1999; 2003; Inglehart 1997; 1999; Norris 1999; 2002). It might seem that these developments would produce a crisis of democracy of the kind Crozier, Huntington, and Watanuki (1975) famously predicted on the basis of their demand overload thesis. These predictions have fared poorly,
226
MARK E. WARREN
however, in large part because Crozier and his colleagues made two mistaken assumptions. The first was that the continuing erosion of authoritative institutions in society (which they correctly predicted) would cause new demands to flow into the formal institutions of representative democracy. The second was that the institutional forms of conflict mediation would remain the same. Neither assumption has played well: citizens have been remarkably resourceful in inventing new collective agents while transforming the nature of authority within, for example, church, family, and workplace. And, largely as a result of newly demanding and sceptical citizens, governments have been creative in experimenting with new forms of governance (Dalton 1999; Norris 2002). One way to frame these developments—and thus to frame the question as to the territory a second transformation of democracy might encompass—is to map the hypothetical possibilities by distinguishing the most visible lines that differentiate political space. Table 10.1 represents these distinctions. In one dimension, we can Table 10.1.Venues of democracy Domain focus
Institutionalized Centred
State Representative democracy Parties, interest groups, corporatist representation Direct democracy Referenda, town meetings Non-elected institutions (agencies, courts)
Society
Economy
Devolved, reflexive
Non-institutionalized Political protest, demonstration, public discourse and influence, networks Political protest, demonstration, public discourse and influence, networks Political protest, demonstration, public discourse and influence, networks
Public hearings, deliber- Devolved and de-conative stake-holder rule- centrated administration, citizen use of making courts enabled by legal standing Devolution of public func- Political protest, demontions to associations stration, public discourse and influence, networks Democracy within associations Direct action Collective bargaining, stakeholder representation, workplace democracy
Note: Examples in boldface indicate topics discussed in this volume.
A SECOND TRANSFORMATION OF DEMOCRACY?
227
think of politics (and thus possible sites of democracy) as being located primarily within the state, within society, or within the economy. The significance of this distinction is that it captures the fact that developed societies make collective decisions and organize collective actions in three basic ways: through administrative power organized through the state, through norms expressed and organized through the associations of civil society, and through money organized by markets (Parsons 1971; Habermas 1987). As societies differentiate along these lines, so we should expect politics to differentiate as well. This same dimension should reflect the fact that democratic states combine electoral institutions such as legislatures with non-electoral institutions such as executive agencies and courts, leading to distinctive patterns of politics within each. Likewise, with the increase in referendums and other forms of direct decision-making, we should distinguish between representative and direct democracy. In a second dimension, we can distinguish between institutionalized politics, such as occurs through political parties, and non-institutionalized politics of the sort exemplified by new social movements. The significance of this distinction is that the success of democracy depends both on the existence of protected space for non-institutionalized forms of politics, such as protest and public debate, and on institutions that are sensitive enough to the ‘output’ of noninstitutionalized politics to formulate these outputs into collective agendas (Habermas 1996: ch. 8).1 In order to account for the de-centring impact of liberal-democratic governance on decision-making, we should also distinguish between ‘centred’ and ‘devolved’ and ‘reflexive’ forms of institutions (Teubner 1983). Centred institutions receive input, process it, issue authoritative decisions, and then organize collective actions. Representative institutions combined with executive agencies operate in this manner. Devolved or reflexive institutions develop the conditions under which parties to a conflict can work out their differences and monitor solutions. An example would be laws establishing collective bargaining, since they serve to equalize the powers of labour relative to management and define fair processes without deciding or imposing outcomes. Combining these dimensions in Table 10.1 gives us a rough map of the potential spaces of democracy in societies like those of the OECD countries, from direct action in the economy and discourse and demonstration in public spheres to the more traditional politics of political parties. The terrain covered by this volume falls into the cells that represent institutionalized, state-centred politics, represented by the examples in boldface type. Although the volume does not track non-institutionalized politics, changes in institutions respond to changes in non-institutionalized domains, since these generate pressures for change. In addition, the impact of institutional changes on democracy can be reciprocal, so that non-institutionalized domains are politicized by changing patterns of institutional inclusion. Rights and their judicial enforcements, for example, create and protect the non-institutional dimensions of democracy, such as public spheres and associations. Likewise, freedom-of-information (FOI) and sunshine laws, freedom of
228
MARK E. WARREN
speech and association, as well as public education combine to enable and sustain public spheres, which in turn influence representation. I do not intend this map of political space as an argument that a second transformation is occurring; it is far too stipulative for that. It does provide, however, an image that suggests an emerging political system that is pluralized, multi-venued, and de-centred. The image also suggests quite clearly that the representative model of democracy lacks the conceptual tools necessary to judge the contributions of these emerging political spaces to democracy.
HOW CAN WE ASSESS A SECOND TRANSFORMATION? Let us assume, however, that such a map represents emerging domains of democratic possibilities. If this is the case, the challenges of assessing a second transformation are significant. Conceiving the normative criteria of democracy under the representative model of democracy is simple by comparison, encapsulated, for example, in Dahl's excellent and well-known list of criteria (1998: 37–8). Democracy requires: • • • • •
equal and effective participation in making views known; voting equality; equal opportunity to gain enlightened understanding; equal chances to control the agenda; and inclusion of all adults.
The meaning of these criteria is clear when there is one target, namely, a bounded political unit controlled by a state which is divided into territorial constituencies—even if, as Dahl rightly emphasizes, the challenges for achieving them are considerable (Dahl 1989). But if much contemporary politics occurs outside of, beside, underneath, and over statecentred electoral institutions, the meanings of democracy in these spaces are unclear, and the apparent difficulties are of another magnitude altogether (Warren 2002b). To take just one example: a democratic virtue of voting for government representatives is that it imposes relatively few burdens on individuals. Democracy under this description is relatively easy to conceive and even achieve. But, as the venues of politics multiply along with opportunities for participation, so do demands upon individuals’ time, attentiveness, knowledge, and skills. Such expansions may produce not only trade-offs among political venues but also outcome biases in favour of individuals who possess more participatory resources (see Chapter 11). It would be at least paradoxical if the expansion of opportunities for participation resulted in less political equality. In the remainder of this chapter I suggest some general norms that would allow such possibilities to be formulated. Although it may be unclear what becomes of voting equality if non-governmental organizations become more influential in
A SECOND TRANSFORMATION OF DEMOCRACY?
229
global politics, for example, we can avoid prejudging such developments as undemocratic if we posit more general norms of collective self-government, and then ask what they might mean within emerging venues of politics. My method here is stipulative: I shall suggest standards for assessing the many developments that might potentially constitute a second transformation of democracy. The approach is not a substitute for argument, nor is it meant to deny that each and every assessment is contestable. But in the space available here it does provide a way of sketching a territory that has not yet been surveyed systematically by democratic theory.2 The strategy involves disaggregating the meanings of democracy according to their institutional and organizational forms. A particular trend—say, towards more referendums or increasing numbers of class action suits—can then be assessed in two complementary ways: • •
Does the trend enhance democracy according the criteria appropriate to the particular kind or dimension it represents? Does the trend detract from other dimensions of democracy? That is, what are its effects within the encompassing ecology of institutions and social relations? What are the trade-offs?
The first question involves disaggregating dimensions of democracy and providing criteria for each dimension. The second question, which involves re-aggregating these dimensions, is difficult to treat theoretically because it involves so many possibilities and contingencies. I therefore focus mostly on developing the first question, following the strategy summarized in Figure 10.1 and returning to the important but contingent question of trade-offs only in the last section. Figure 10.1 lists the most abstract and encompassing norms of democracy on the left, disaggregates the kinds and dimensions of democracy, and then relates these to institutions and venues. The conceptual strategy summarized in the figure makes it possible to relate the meanings of a variety of criteria of democracy, such as access, transparency, and accountability (the criteria used in this volume), to their justifications as means of democratic inclusion. Likewise, while it is difficult to generalize about contingent trade-offs, this strategy makes it possible not only to identify tradeoffs between norms that are instrumental to democracy on a case by case basis—accountability and efficiency, for example—but also to assess their relative contributions to more general norms of democracy.
INCLUSION At the highest level of generalization, democracy is the form of collective self-rule that includes each individual within the collectivity as an equal participant in collective decision-making. Democratic inclusion, or what I shall refer to simply as ‘inclusion’, means that every individual potentially affected by a collective
230
Fig. 10.1.Normative dimensions of democracy
MARK E. WARREN
A SECOND TRANSFORMATION OF DEMOCRACY?
231
decision should have an equal opportunity to affect the decision (Young 2000: 23; Habermas 1996: 107; Dahl 1998: 37–8; Held 1996: 324). For inclusion to be effective, individuals need to be empowered or enfranchised to participate in collective decisions. Individuals must also have the conditions and capacities to make autonomous judgements—that is, decisions should be free from coercion, threats, or blackmail, and based on a good understanding of the interests of self and others. These are general norms of inclusion because they apply to all dimensions of democracy. General norm 1: Franchise: Units of political decision making—their boundaries of inclusion and exclusion—should match the potential effects of collective decisions. That is, franchise—including voice and vote—should be appropriate to the scope of collective decisions. This criterion is expressed in debates and struggles over who should count as an enfranchised citizen of a given territory, since this is one way of deciding who is entitled to participate in collective decisions (Dahl 1989). A somewhat more complex version of this issue is exemplified in boundary disputes between levels of government in federal states. Those who impose injuries on others often aim to exclude these others by defining the issue as belonging to a level of government that is either favourable to their interests or ineffective. Firms that exploit natural resources, for example, often seek to enlist democratic norms by arguing for ‘local control’ over resource decisions. In so doing they seek to exclude those outside the locale who have interests in the future consequences and cross-border impacts of resource extraction—a violation of the franchise norm. European integration has put this criterion on the table under the norm of subsidiarity: that is, the norm that the levels at which decisions are taken should match the scope of potential effects. In the terms used in this volume, individuals should have access to the collective decision-making institutions and processes that affect them. General norm 2: Autonomy: Individuals should be autonomous in their persons and judgments. Inclusion makes sense as a norm only if individuals have the protections and capacities necessary to participate in making collective decisions, a norm increasingly referred to as autonomy (Held 1996: ch. 9). Individuals should be autonomous in their persons, meaning that they should be protected from arbitrary coercion, threats, or blackmail. Such protections are institutionalized in the form of rights to physical security as well as to basic welfare sufficient to protect against economic desperation. In addition, individuals should be autonomous in their judgements: they should know what they prefer for themselves or their collectivities; they should be able to provide reasons to others for the positions they hold, and know how their judgements connect to collective decisions. Conditions for autonomy of judgement are institutionalized through rights of speech, association, and petition, openness in decision-making, and access to information (Habermas 1996: chs. 3, 8; Cohen and Arato 1992: ch. 9). This volume summarizes some of these institutional conditions under the norm of transparency.
232
MARK E. WARREN
These two norms, however, would seem to set impossibly high standards for democracy, especially under the conditions of large scale, globalization, and complexity that define today's societies and which necessarily limit individuals’ participation in those collective decisions that affect them. We should ask, therefore, another kind of question: is active inclusion in collective decisions always necessary for democracy? The answer is that active inclusion is most important for democracy when decisions are ‘political’. By ‘political’ I mean social relations defined by a need for collective action under circumstances in which (1) there is disagreement about what to do, and (2) one or more of the parties has the power to force the issue (Warren 1999a). Active inclusion in decision-making— democracy—matters most under ‘political’ conditions since members of a political collectivity cannot (and should not) assume that decisionmakers will take their interests and values into account. Much of what goes on in democratic politics, of course, is contest over the boundaries of what should or should not be treated as a political question. Those who benefit from a status quo at the expense of others will have an interest in keeping issues from becoming politicized, while those who perceive harms will try to bring an issue into political domains. But there are vast areas of collective decision-making that are not political in this sense because they are based on convergent interests and identities. Indeed, the vast majority of acts of legislatures—which are specifically designed for political conflict—pass unanimously and without contest. In many areas of collective decision—say, airline safety or public health—interests may mostly converge around safety and health, enabling divisions of decisionmaking labour based on the competences of aerospace engineers or medical experts. This is not to say that such areas are not potentially political but rather that they often operate on the basis of a consensus about what should be done, within institutions that can be structured to minimize potential conflicts of interest. Under these circumstances, relations of warranted trust and authority enable passive inclusion of those whose interests are affected. For passive inclusion to be consistent with democracy, however, the assumption of convergent interests must be testable: trusted authorities should be answerable for their decisions even if they do not answer for every decision they make. Domains of passive inclusion should be structured with oversight, access, and the possibility of transparency just in case the activities of authorities depart from the interests entrusted to them. Ideally, in a democracy, institutions and organizations would be designed to revert to democracy as needed, on an issue-by-issue basis. Then most decisions can be made by authorities entrusted with serving convergent interests without harm to the democratic norm of inclusion (Warren 1996; 1999b). Thus, if democratic theory is to be relevant to complex societies it must include a norm that enables a political division of labour between authorities and participants as follows (Fig. 10.1, column 2): Political arena norm:Political arenas, defined by conflicts of interests, norms, and identities, require active democratic inclusion; nonpolitical arenas do not.
A SECOND TRANSFORMATION OF DEMOCRACY?
233
Individuals are passively included in non-political arenas as part of the consensus that defines these arenas as non-political. However, the consensus must be testable by democratic means, and the institutions within the domain should revert to actively inclusive decision-making if the consensus is suspect or fails.
DIMENSIONS OF DEMOCRACY The norms of democratic inclusion, franchise, autonomy, and political arena apply to democracy generally. Their meanings, however, depend upon the many ways they can be expressed, organized, and institutionalized in today's democracies. Column 3 of Fig. 10.1 provides some general meanings to which I shall refer as ‘dimensions of democracy’. Column 3a distinguishes the means of active inclusion (participation)—voting and persuasion—from its scope, which has to do with the domains over which democratic means of decision-making are effective. In what follows, I suggest some judgements about what would count as ‘more’ or ‘less’ democracy in each dimension, all other things being equal—that is, not taking into account trade-offs among dimensions. These judgements should be read as illustrative rather than argumentative in the sense that I intend them only to indicate the normative terrain that any comprehensive account of a second transformation should traverse.
MEANS OF PARTICIPATION Voting Voting is a power mechanism consisting in the distribution of the right to decide across some population of individuals or their representatives. Within democracies, voting comes in two basic forms, direct and representative (Fig. 10.1, column 3c). Direct voting is one form of direct democracy, and means that citizens make policy directly by voting. Direct voting defines some institutions such as referendums, and is central to the political life of other institutions, such as town meetings, party conventions, unions, and many kinds of associations. Voting for representatives means, of course, that voters elect individuals who make decisions on their behalf.
Direct voting How do we know when direct voting is democratic and when more of it means more democracy? The general criteria above apply, of course: if direct voting is within an exclusive political unit—that is, the political unit excludes people who are affected by the decision—it does not contribute to democracy, and may very well detract from it. In rights language, if the right to vote is restricted, the results
234
MARK E. WARREN
are undemocratic, as was certainly the case in slaveholder and apartheid democracies. Thus: direct voting is more democratic the more voting includes those affected by the issue. Two other conditions should be present if direct voting is to count as democratic. First, the agenda, whether at a town meeting, union meeting, or in a referendum, should itself be broadly inclusive of the concerns of the political unit. There is no a priori way to know when this condition has been achieved. But it is possible to gauge the processes that precede the vote in order to determine whether the agendas and questions are broadly inclusive. In the case of face-toface forms of democracy, this can be determined by the degree to which the participants are able to affect the agenda, place new questions on the agenda, and debate the terms in which they have been formulated. Referendums are more difficult to judge since they often lack institutionalized deliberative forums through which questions can be developed and challenged. Sometimes referendums are held on topics that have been well debated and the questions are clear. In other cases, however, political activists or elites place vague or manipulative initiatives on ballots that have had little public exposure, often as a way of avoiding arenas (such as legislatures) within which the agenda might be more fully debated. In such cases, referendums actually decrease democracy, because voting is not combined with an inclusive process of agenda formation. Thus: direct voting is more democratic the more inclusive the process of agenda formation of all affected by the issue. The second condition is closely related, and follows from the general norm that individuals should be autonomous in their judgements. Autonomy cannot be directly measured or observed. But we can infer autonomy from the processes that precede a vote: has the issue had broad public input? Has it been debated in the public domain? Have voters had a chance to form opinions? It is likely that the more an issue is debated and the more familiar voters are with an issue, the less subject they are to manipulation by political entrepreneurs. This is an especially important consideration with referendums, which political entrepreneurs sometimes seem to prefer because they can bypass deliberative venues and ‘go directly to the people’. Thus: direct voting is more democratic the more inclusive the deliberative processes preceding the vote.
Voting for representatives When democracy is representative, inclusion is achieved by holding representatives accountable to voters. But whether more elections and electing—one of the trends documented in the volume—translates into more democracy depends upon the extent to which the numerous conditions of accountability are fulfilled. First and most basically, in order to hold representatives accountable, individuals must be included within the constituencies affected by the representatives’ decision. While there remain disenfranchised individuals such as guest workers within territorially based districts, more subtle versions of the same problem arise
A SECOND TRANSFORMATION OF DEMOCRACY?
235
in other ways. As issues become less territorial in nature, representatives elected by district or region increasingly hold powers that affect those who cannot vote for them and cannot hold them accountable. Problems of franchise can be exacerbated within federal systems where territorial entities are represented rather than the individuals within the entities, as is the case with the US Senate. Such systems of representation magnify the franchise of some individuals and diminish the franchise of others. In both cases the distribution of the franchise violates the equality of individuals stipulated by the democratic norm of inclusion. Thus, voting for representatives is more democratic the more the franchise reflects equal inclusion of those affected by the representative's decisions. Second, even if an electoral system enfranchises individuals equally, the design of the electoral system can increase or decrease voters’ abilities to use the vote to enforce accountability. Winner-take-all or ‘first past the post’ (FPTP) systems are based on single-member electoral districts. These systems are majoritarian and provide incentives for representatives to represent only those who will provide a majority in the next election. These systems are exclusive of minorities because they provide no means through which minorities can enforce accountability. In contrast, systems based on multi-member districts and designed for proportional representation (PR) allot representation by the proportion of the vote received by each party that reaches a threshold percentage of the total vote. Typically, PR systems are designed to increase inclusion (Lijphart 1999). Thus: voting for representatives is more democratic the more the electoral system is designed for proportional representation. Third, to the degree that political parties mediate the principal-agent relationship between individuals and representatives, a secondary principal-agent relationship is embedded in the individual-party relationship. While in principle all of the above criteria apply to this relationship as well, we can compress these into the proposition that more internal party democracy is likely to secure the individual-party relationship. Voting for representatives is more democratic the more parties practise internal party democracy. Fourth, even with the above conditions fulfilled, elections must be contested for voters to enforce accountability, since it is the possibility of removal from office that presumably motivates representatives to represent the interests and values of voters. Voting for representatives in uncontested elections or plebiscites is not an indication of democracy, because the vote contributes nothing to the enforcement of accountability. Thus: voting for representatives is more democratic the more elections are contested. Fifth, for the vote to enforce accountability, electoral and legislative systems must be sheltered from non-democratic forces, that is, forces such as money and coercion that override the force of the vote. Competitive elections, for example, are not sufficient to guard against plutocracy if all candidates in an election find they must raise large amounts of money in order to mount competitive campaigns. While lines of accountability and influence are difficult to observe directly, it is
236
MARK E. WARREN
possible to ask about the degree to which institutional checks exist against the influence of money, such as effective campaign finance laws, enforceable codes of professional ethics, and independent oversight of elections and election financing. Thus: voting for representatives is more democratic the more non-democratic influences on representatives, such as money, are blocked. Sixth, voters can hold representatives accountable only if they can know what the representatives stand for, both before elections and during the representatives’ terms. In the terms of this volume, transparency is a condition of accountability. Political parties can serve this function by disciplining candidates to run on a platform, and then to vote consistently with the platform during their term. In addition, the legislative process itself should be open, both in its deliberations and in its votes, and the available information should be widely disseminated by the media and interested publics. Under the scenario of a second transformation, however, we should expect parties to become weaker rather than stronger owing to the de-centring of politics. But we should also expect more openness in legislative processes and more dissemination of information by the media and concerned publics owing to increasingly politicized societies. These criteria should be considered together since greater openness in and mobilization around legislative processes may compensate for weaker parties. Thus: voting for representatives is more democratic the more political parties are able to enforce platforms, both with respect to their candidates and within legislative processes; voting for representatives is more democratic when legislative processes are more open, both in deliberation and in voting; voting for representatives is more democratic when the media and interested publics function to disseminate information about legislative processes.
Persuasion Several decades ago, perhaps, policies were viewed by most people as legitimate by virtue of having been produced through the institutions of representative democracy. Today, when citizens are much less trusting of political elites and more sceptical of their claims, legitimacy is not so easily gained. Even when political elites win fair elections or are duly appointed, their proposals and policies need to stand on their own, justified not only to constituents but to a broad variety of stakeholders who now often have capacities for organization that provide credible threats for obstruction. Part of a second transformation would be the emergence of publics formed around multiple policy areas and reflected in the politics of networks, protests, public forums, hearings, and mass media. We should thus expect governments as well as civil society associations to focus increasingly on communication, not simply because media savvy is ever more important to political aims but also in the more substantive sense that decisions, policies, and actions must increasingly be justified to relevant publics (Beck 1997). This volume provides evidence of such developments, including a dramatic trend toward more openness and transparency in the
A SECOND TRANSFORMATION OF DEMOCRACY?
237
government, increasing uses of public hearings, new experiments in deliberative policy-making especially in local government and administrative law, collaborative policy-making, and decentralization of administrative structures. From the perspective of democratic theory, these trends are important not only for reasons of accountability but also because they enable and promote persuasion as a means of doing politics. Voting is not sufficient for democratic decision-making; it merely provides an incentive for people to talk about what they should do by excluding force, personal authority, and other non-democratic means of decision-making. Once talk becomes the key medium of decision-making, however, it becomes a kind of influence in its own right, the influence of persuasion. Through persuasion, opinions are cultivated, reasons developed, and justifications offered, so that in one way or another voting as an exercise of power also expresses an act of judgement. Moreover, the accountability of elites for their decisions ultimately depends upon persuasion because it is through persuasion that individuals come to understand the reasons for decisions and thus their potential acceptability. While there are many venues of persuasive influence, I shall simply distinguish between direct and representative forms because somewhat different norms apply. By direct participation, I mean that individuals are directly involved in persuasive influence, whether in face-to-face deliberations, letter writing, Internet chat rooms, or demonstrations. Representative participation occurs when an individual is not seeking to influence others but is part of an audience that follows the arguments of others who are directly involved in persuasive influence, such as advocacy groups, media opinion leaders, legislators, and the like.
Direct persuasion How could we determine whether increasing opportunities for direct communication count as increases in democracy? Unlike voting, we cannot infer equal inclusion directly from institutional designs, complex as even this may be. The powers of persuasion are not distributed equally, and it is hard to imagine how they might be. A particularly persuasive individual can exercise a disproportionate influence, while those who are less eloquent, less informed, or less connected to the concerns of an audience may have no persuasive influence at all. Within the dimension of persuasion, the best we can imagine is that ‘democracy’ refers to the equal opportunity for individuals to exercise persuasive influence, should they so choose. A democratic distribution of choice cannot, however, be directly observed. Rather, it can only be inferred from conditions that would enable the full range of voices salient to an issue to be recognized and heard. The first set of conditions has to do with the autonomy of participants. Participants in directly persuasive forums should be protected from vulnerabilities in ways that enable them to use the force of argument to try to get their way. Clearly, freedoms of speech and association are key. But also important are economic vulnerabilities, which should be reduced by welfare rights to the point that participants are not supplicants to those who control their livelihoods. Few persuasive
238
MARK E. WARREN
situations are as well balanced as those, say, engineered into legislatures. But often security, political, and certain welfare rights can translate into a rough equivalence of obstruction potential in ways that can bring dominant parties to the table. Thus: direct persuasion is more democratic the more rights and protections effectively limit the vulnerabilities of participants. Even when protected, individuals will lack the capacities to participate in persuasion without the information necessary to connect their interests and values to decisions. Institutions that support access to information come in a variety of forms, including FOI acts, sunshine laws, and active and vigorous media; mandated record-keeping by governments, firms, and other powerful organizations; and a court system that enforces rights of access to information. Governments can also support direct persuasion—for example, by enabling advocacy research and providing easy access to government records. In addition, a society that includes a multitude of public interest groups collecting, organizing, and disseminating information will support direct persuasion. Thus: direct persuasion is more democratic the more participants have access to information relevant to the issue. Like many forms of direct democracy, participation in direct persuasion suffers from limitations of scope, and is therefore most likely to be found in local arenas and in policy areas that are localized in their impacts—education, neighbourhood and urban design and development, policing, and the like. Nonetheless, often policies with broader implications, such as industrial policies affecting the environment, are negotiated within small-scale venues. Because of inherent limits of scale, direct inclusion in such cases can easily and almost naturally (in the name of ‘local democracy’) exclude others who are affected—those downwind of a power plant, for example. These are franchise exclusions. In the case of inclusion in persuasive influence, however, we usually speak of inclusion and exclusion as a matter of recognition of those potentially affected by decisions, and enabling their voice. Thus: direct persuasion is more democratic the more those affected by decisions are recognized and attain voice in the process. The extent of recognition and voice is difficult to measure because it is hard to know who is silent or unrecognized. Their extent may be indirectly indicated, however, by the relative density of associations devoted to groups typically excluded from decision-making, or the existence of political processes that are proactive in reaching those potentially affected by decisions.
Representative persuasion Although opportunities for direct access to persuasive influence increase dramatically in a second transformation of democracy, the imperatives of scale and complexity dramatically limit individuals’ abilities to influence all of the issues that affect them. Nonetheless, in principle individuals can be represented by others who provide their voice within public debates. And when individuals are attentive to debates they may be influenced in ways that increase their capacities for autonomous judgements. What I am calling ‘representative persuasion’ occurs
A SECOND TRANSFORMATION OF DEMOCRACY?
239
when individuals identify with others who provide their voice and follow attempts at persuasion by others. These representative voices and debates may be found in advocacy groups that go public with their causes, debates in the mass media, civic journalism, political candidates presenting and defending their positions, legislators involved in debate, deliberative polling, and debates staged by public interest groups for the purpose of political education. In order to capture the subtleties of a second transformation in today's large-scale, complex societies, we need to be able to describe this kind of attentiveness and support as a form of participation. For representative persuasion to count as democratic, however, individuals need to be presented with a wide array of voices and positions as well as a wide choice of deliberative representatives. Thus: representative persuasion is more democratic the more positions are presented and the more voices heard on a given issue. Of course, it often happens that representative persuasion undermines the autonomy of participants, as when speech and images play on subconscious fears, anxieties, and preconceptions. So perhaps we should say that representative persuasion, like political representations, should be accountable in some way to the audience. This sense of accountability, however, is unlike that embedded in the principal-agent model of voting, which presumes that voters know what they want and why they want it. In the case of persuasive influence, we must make a psychological distinction: ideally, influences would be conscious in ways that allow arguments and positions to be assessed and defended. This is not to say that subconscious influences—feelings, intuitions, and even prejudices—are not important sources of political issues. Psychologically, it is likely that most important new issues have their sources in ‘awakenings’ that have subconscious sources. But in the domain of politics, unlike art and literature, feelings and intuitions are often not widely shared but rather indicate latent injustices, harms, or exclusions. So while preconscious sentiments may have an important evocative force, in a democracy they should also be articulated as arguments of the sort that people can agree with, reject, or modify. We can speak of the accountability of those involved in representative persuasion in precisely this sense: they should formulate opinions and positions in such a way that a cognitive, language-based response is possible. Thus: representative persuasion is more democratic the more positions are formed in ways that enable debate about the positions. Democratic theorists often refer to this norm as ‘reasonableness’ (Young 2000: 24–5), a norm that is virtually impossible to measure. We do, however, recognize the limiting cases of unreasonableness. Some kinds of religious rhetoric define communities of believers according to truths that have only faith behind them—a strategy that stops deliberation in its tracks. Derision of opponents, character assassination, as well as images, phrases, dogmas, and slogans that are repeated until they become mantras likewise allow no response. But recognizing these common rhetorical strategies is a different matter than providing criteria of
240
MARK E. WARREN
assessment, nor, pace Rawls (1993), could institutions be designed to prohibit unreasonable speech. So, for both epistemological and political reasons, it is better to focus on process criteria. In particular, what makes racist or religious rhetoric dangerous to democracy are political conditions that enable responses to be repressed. Thus: representative persuasion is more democratic the more open the society and culture, and the more interests, values, and voices are represented. Democracies, especially under conditions of a second transformation, generally lack the capacities for repression, in part because rights of speech and association usually protect responses, in part because the responses have many means for dissemination, and in part because the pluralistic, post-traditional cultures of the OECD countries now tend towards discursive openness rather than closure. The greater danger resides in resource distributions that are so unequal that an interest can saturate public debate with a single message, crowding out alternative voices. Against this possibility, openness is enhanced by the presence of advocacy groups that go public, since these activities keep public sphere deliberations alive, as does the presence of independent, professional, and pluralistic media. From an institutional perspective, democratic representative persuasion is supported by measures that promote openness, such as FOI acts and sunshine laws, and a representative pluralism of opinion within government, especially within the legislative branch.
SCOPE OF PARTICIPATION The final two dimensions of democratic inclusion involve scope (see Figure 10.1). With regard to the first, the thesis of a second transformation of democracy suggests that democracy is increasingly a part of domains within government, economy, and society where it did not formerly exist. Thus, the expansion of democracy should count as one way we should think about the question of more democracy (Dryzek 1996). Second, and perhaps less obviously, because democratic decisions are meaningless unless they can be put into effect, the scope of democracy can be expanded or reduced by the presence or absence of collective agents with capacities to organize collective decisions.
Domain Expansion The expansion of democracy into new domains of society follows from the complexities and structural differentiations that increasingly limit the capacities of governments to impose decisions, as well as from subjective transformations as individuals increasingly value control over the contingencies that structure their lives. In addition, we might expect that these same developments would generalize a certain ethos of democracy, in keeping with Tocqueville's and Dewey's observations that democracy is not just a form of government but also a kind of society. Indeed, over the last several decades it has seemed increasingly appropriate to
A SECOND TRANSFORMATION OF DEMOCRACY?
241
speak of democracy in the workplace, family, church, associations, social movements, and networks. Because this volume focuses on changes in institutions of government, it does not discuss those dimensions of scope that fall into society and economy. Nonetheless, domain expansions within society and economy are not independent on changes in the institutions of government, and so we should find changes within government that reflect these broader changes.
Administrative law and policy The most direct and visible domain changes are those occurring within the administrative branches of government, mostly through the development and implementation of legislative mandates and guidelines, but also in local governance. In the US, for example, school governance through parent-teacher committees are becoming commonplace, as are democratic experiments in community policing, urban planning, and environmental policies. Executive agencies are increasingly mandated to develop administrative rules through processes that include stakeholders, using public hearings, deliberative processes, and stakeholder representation. Moreover, the new serviceoriented approaches of government agencies are interestingly ambiguous. The trend toward treating citizens as ‘consumers’ of government services is not, in itself, democratic, since consumers do not engage in collective decisionmaking but rather make their choices as individuals (see Chapter 7). Nonetheless, because market analogies have limits in the area of government, the effects of a ‘consumer’ approach to citizens may have democratic implications that would not develop within market contexts. Mandates to treat citizens as consumers are, really, ways to increase accountability to those who depend upon government services, thus altering dependence relationships in potentially democratic ways. New strategies of administration can also result in indirect expansions in the scope of democracy, often the result of processes of conflict resolution or policing of laws and regulations that involve the concerned parties—what Teubner (1983) has referred to as ‘reflexive law’. In such cases, governments devolve political processes to non-governmental venues, and then oversee the processes of conflict resolution, as in collective bargaining, mandated worker-safety committees, and government requirements for stakeholder representation on corporation boards. These count as government-sponsored domain expansions, since conflict-resolution is shifted in democratic directions. In the area of administrative domain expansion there are no generalizable domain-specific criteria for what counts as ‘more democracy’. However, what I have referred to as a general ‘political arena norm’ is especially important, since expansions of democratic procedures to previously ‘non-political’ arenas are often, in fact, driven by political conflicts that have been suppressed or badly managed by bureaucrats and technocrats. Thus: expansions of democratic procedures into new domains lead to more democracy if these domains are characterized by conflicts of interests and values.
242
MARK E. WARREN
In addition, the general norms of franchise and autonomy apply as developed above: domain expansions of democratic procedures count as increases in democracy when all potentially affected individuals are included in the processes, and when individuals are autonomous in their persons and judgements.
Courts and expansions of rights Domain expansions also result from new political rights and forms of standing. As documented in this volume, the OECD countries have been undergoing ‘rights revolutions’, from the incremental reinterpretations of the Bill of Rights in the US to the adoption of new charters of rights in Canada and the EU (see Chapter 9). These developments include, on average, increased access to the courts. Many see this trend as detracting from democracy by virtue of judicial displacement of the powers of the ‘democratic branches of government’. Such views presuppose not only that the electoral model of democracy is working well but also that it exhausts the meaning of ‘democracy’. Under conditions of a second transformation, however, neither assumption holds true. Moreover, this view draws on a mistaken but widespread view that democracy and individual rights are competing goods when, in fact, gains in individual rights have historically been one of the most important ways in which democracy has expanded (Habermas 1996; Young 2000; Kymlicka 1995). On the one hand, rights are often the objects of democratic struggles, especially rights of political expression, the right to vote, the right to associate, rights to personal security, rights to due process, rights to welfare security, and rights to education. On the other hand, as rights are achieved, they tend to expand democracy because they mitigate the power relations that can be leveraged against democracy.3 Rights of individuals are relational empowerments which imply duties of forbearance of and equal treatment by other power holders—the police, government agencies, firms, and other individuals—while also requiring governments to deploy resources to guarantee forbearance and equal treatment. Thus: court decisions that equalize and expand individual rights tend to expand democracy. Some of the ways in which this generalization holds true are these. First, in so far as courts enable political rights such at those of speech and association, they also enable zones of political activity outside of government. These include the more traditional public spheres and political associations, as well as demonstrations, protests, and new forms of direct action. Direct action is increasingly important in democratizing global politics, since there are no institutionalized forms of democracy at the global level. More generally, where rights-based rule of law is strong and knowable, individuals can act openly and with confidence that they will not fall foul of unknown or arbitrary government actions. This in turn helps to secure those associational habits and capacities that are the precursors to ‘making democracy work’ (Putnam 1993; Warren 2001). Second, the courts are the ultimate arbiters of political inclusions and exclusions. Thus, if courts are more accessible, individuals or classes of individuals who have
A SECOND TRANSFORMATION OF DEMOCRACY?
243
interests in political decisions can appeal for inclusion. As noted in this volume, while this has been a long-standing element of judicial politics in the US, EU-level courts have become increasingly involved in expanding access, a contribution to democracy that overlaps with franchise. Third, increased standing for class-action suits can expand an element of democracy into new domains, as when classes of individuals are able to demand accountability from corporations for the consequences of their economic activities. Closely related, courts can become the vehicles of enforcement of administrative rules through the participation of citizens. The Clean Water Act of 1970 in the US, for example, gives citizens the standing to sue private entities for violation of the Act, a device that gives citizens a means to participate in creating and maintaining public goods. Fourth, rights can change power relations in democratic directions within society and the economy. Equalizing women's rights to own property, for example, democratizes the family by mitigating a wife's economic dependence on her husband. Likewise, rights-based laws against sexual harassment in the workplace remove one kind of power a boss can deploy over a worker, again tending toward more democracy in the workplace.
Capacities of Collective Agents The second dimension of scope has to do with whether there exist collective agents with the capacities to act on democratic decisions in ways that are relatively effective, efficient, and worthy of public trust. If organizations—not simply governmental organizations but also firms, churches, and charities—are clumsy, slow, unresponsive, inefficient, or corrupt, then democratic decisions will have limited impact. This counts as a problem of scope in the negative sense: lack of effectiveness and efficiency in, say, government agencies or schools or associations causes people to lose confidence in their abilities to get collective things done. Lacking such confidence, individuals will often prefer to seek individual self-sufficiency rather than collective solutions to collective problems, which in turn further erodes the scope of democracy. This vicious circle is best exemplified by the Italian case where, with its pervasive government corruption, citizens have had little control over government, expect little of government, and as a result have had few capacities to effect change (della Porta and Vanucci 1999; Putnam 1993). But other OECD countries have suffered from this cycle of disaffection, sometimes caused by corruption, sometimes waste and inefficiency, sometimes unresponsiveness (Pharr and Putnam 2000; Orren 1997). One of the consequences was a political climate that enabled elites to privatize many public functions, beginning in the early 1980s in the US and Britain. As documented in this volume, however, over the last two decades governments have sought to reinvent themselves though decentralization, new forms of oversight and accounting, an increased focus on consumer-friendly government, and
244
MARK E. WARREN
public–private partnerships with associations. While these experiments are quite varied, at least two generalizations are possible. First, collective agencies are most effective and efficient when the level of organization is appropriate to the problem (subsidiarity) and when the nature of the organization captures the skills and knowledge of participants without sacrificing accountability. One model, successfully implemented in Chicago public schools and policing, echoes the approach of reflexive law by devolving organization and administration to small units so as to maximize participation and gain from ‘local knowledge’ and flexibility of means, while centralizing oversight of processes and results (Fung 2003). Such models are not without problems, but in many areas of governance these democratic approaches promise to be more effective and efficient than the bureaucratic and market alternatives (Dryzek 1990). Thus: subsidiarity combined with centralized accountability enhances democracy by enabling more effective and efficient collective agents. In addition, since the 1970s we have seen increasing attention to corruption, as well as to remedies, including increased transparency, oversight of financial matters, increased uses of inspector generals and other oversight mechanisms (Rose-Ackerman 1999). Corruption scandals continue to emerge, but these are probably not the result of a broad trend towards increasingly corrupt governments but rather of rising public standards, increasing functional pressures for efficiency, more availability of information, and a greater number of groups paying attention to these matters (Thompson 1995). Thus: more transparency, institutionalized oversight, audits of financial records, and associational monitoring increase democracy by limiting corruption of collective agents and increasing public trust.
TRADE-OFFS AMONG DIMENSIONS OF DEMOCRACY The OECD democracies today are ecologies of complementary and competing institutions and associations. The contribution of each part to the overall functioning of democracy will depend upon its role within the broader ecology (Warren 2001: ch. 7). It is therefore not likely that the many norms that define ‘more democracy’ in each dimension would be simultaneously achievable, and it is certain that there will be trade-offs among at least some of them. The large number of possible trade-offs, however, renders impossible all but the most general observations. At the highest level of abstraction, the general criteria of franchise, autonomy, and scope can trade-off against one another (Dryzek 1996). Expanding the franchise in the form of referendums, for example, can reduce autonomy simply because the amount of time required to attend to political issues is limited, and individuals may be less able to make informed judgements the more ballot measures they must vote upon. Expanding the scope of direct democracy can also infringe on autonomy, again owing to insufficient time
A SECOND TRANSFORMATION OF DEMOCRACY?
245
for individuals to attend to all the areas in which they are entitled to participate. Under such circumstances, often only a few intensely interested individuals participate—individuals who function as political elites but who are not under the constraints of accountability that an elected representative would be (Warren 1996). Under these circumstances, more direct democracy can actually lead to less democracy overall. To take another example, at least since Carl Schmitt's (1988) critique of parliamentary democracy in 1929, it has been clear that accountability of representatives, secured by party discipline, can trade-off against representative deliberation, which will be constrained by party discipline. And there are well-known trade-offs between inclusion and the effectiveness and capacity of collective agents. As more individuals and voices are included, the ability of an institution to make decisions can be impaired. Again, sometimes the institutional means to attain a goal undermine the goal itself. Anti-corruption measures, usually in the form of oversight, can often add to cumbersome qualities of agencies, reducing their efficiency (Rose-Ackerman 1999). These are examples of trade-offs that might exist in principle. In practice, most trade-offs are contingent: they happen under some circumstances but not under others. The costs of oversight to stem corruption, for example, are lower within institutions with strong professional ethics and codes, and higher within institutions without such a culture. Inclusion can increase the capacities of collective agents—if the issues admit of consensus and the outcomes are legitimate. Alternatively, inclusion can reduce the capacities—if the issues are so tough that they lead to gridlock. Courts can easily adjudicate rights-based claims if rights and entitlements are broadly understood and legitimate, simply because violations will be more exceptional. But under circumstances where rights struggle against traditional injustices, courts may become so overloaded with cases that their effectiveness is diminished. In other cases, potential trade-offs can be mitigated by institutional innovations. In one of the oldest examples, public education systems were expanded in the late nineteenth century in order to reduce the trade-off between the expanding franchise and the autonomy of individual judgements. To take another example, the expansion of referendums can reduce the quality of individual judgements by spreading citizens’ sporadic attention even more thinly. But the lagged referendum with an agenda-setting vote followed by a binding vote, say, six months later would give time for public debate to do its work, enabling individuals to prioritize the salience of the referendum, to benefit from the advice of groups that represent their views, and to learn enough about the issue to make autonomous choices (Barber 1984: 281–9). Finally, limitations on democracy in one dimension may be mitigated by ecological context. Complex federal polities like the US, for example, contain many thousands of sub-polities. Even if each is internally a democracy, they commonly violate franchise by externalizing costs on to those who are excluded, such as
246
MARK E. WARREN
members of neighbouring municipalities. But if those who are excluded are organized in ways that allow them to resist the costs, or to impose their own costs, the patterns of exclusion could lead to a pattern of bargaining among political units, which could amount to an expanded franchise as the concerns of each unit are registered with one another (Young 2000).
CONCLUSION: IS A SECOND TRANSFORMATION UNDER WAY? Do all of these changes add up to something as dramatic as a second transformation of democracy? The empirical part of this claim rests on the observation that politics in the OECD countries is, in fact, being done today in ways that are dramatically different from several decades ago. The concept of a second transformation suggests that we should expect only moderate increases in democracy in political venues closely tied to the representative-electoral system, while we should expect more dramatic increases in those venues that supplement, complement, and compete with the representative system. We should expect somewhat weaker political parties, as they are less able to control political inputs. We should see an increasing number of domains where collective problems and actions are deliberated and negotiated rather than imposed. We should find an increasing number of boundary disputes, as the question of ‘what is political?‘ becomes more contentious. We should continue to expect noisy public spheres, with increasing amounts of public information, and increasing demands for transparency in government. We should see changes in administrative strategies fuelled by an increasing number of political conflicts that spill over from representative systems, within which they are inadequately or sometimes illegitimately resolved. Judicial systems, which not only must respond to inadequacies in representative outputs but become the front line for disputes fuelled by the rights revolution, will continue to develop as key political venues. The normative claim is that these developments are transforming democracy a second time to produce a third form, one that includes ways of doing politics that, in aggregate, preserve, extend, and deepen democracy. Although I have not offered any overall assessment as to whether this is the case, I am cautiously optimistic (Warren 2002b). The multiplicity of forms and sites of democracy makes assessing democracy a messy business. It is not likely, for example, that a general assessment would find democracy advancing in every dimension as it undergoes transformation. My more modest aim here has been to frame the question of assessment by indicating the many dimensions in which democracy is developing. It is possible in principle to develop multi-dimensional assessments that connect the basic and still radical meaning of democracy as collective self-government to the highly complex forms emerging within the OECD countries.
A SECOND TRANSFORMATION OF DEMOCRACY?
247
APPENDIX: SUMMARY OF NORMATIVE CRITERIA OF DEMOCRACY Basic Principle Inclusion: Democracy is the form of collective self-rule that includes each individual as an equal participant in collective decision-making. Democratic inclusion means that every individual potentially affected by a collective decision should have an equal opportunity to affect the decision.
General Implementation Norms General norm 1: Franchise: Units of political decision making—their boundaries of inclusion and exclusion—should match the potential effects of collective decisions. That is, franchise—including voice and vote—should be appropriate to the scope of collective decisions. General norm 2: Autonomy: Individuals should be autonomous in their persons and judgements.
Norm Dening the Scope of Democracy Political arena norm: Political arenas, defined by conflicts of interests, norms, and identities, require active democratic inclusion; non-political arenas do not. Individuals are passively included in non-political arenas as part of the consensus that defines these arenas as non-political. However, the consensus must be testable by democratic means, and the institutions within the domain should revert to actively inclusive decision-making if the consensus is suspect or fails.
Norms of Modes of Participation Direct voting is more democratic: the more voting includes those affected by the issue; the more inclusive the process of agenda formation of all affected by the issue; the more inclusive the deliberative processes preceding the vote. Voting for representatives is more democratic: the more the franchise reflects equal inclusion of those affected by the representative's decisions; the more the electoral system is designed for proportional representation; the more parties practise internal party democracy; the more elections are contested; the more non-democratic influences on representatives, such as money, are blocked; the more political parties are able to enforce platforms, both with respect to their candidates and within legislative processes;
248
MARK E. WARREN
when legislative processes are more open, both in deliberation and in voting; when the media and interested publics function to disseminate information about legislative processes. Direct persuasion is more democratic: the more rights and protections effectively limit the vulnerabilities of participants; the more participants have access to information relevant to the issue; the more those affected by decisions are recognized and attain voice in the process. Representative persuasion is more democratic: the more positions are presented and the more voices heard on a given issue; the more positions are formed in ways that enable debate about the positions; the more open the society and culture, and the more interests, values, and voices are represented.
Norms of Scope of Participation Administrative law and policy: Domain expansions of democratic procedures into new domains lead to more democracy if these domains are characterized by conflicts of interests and values. Domain expansions of democratic procedures count as increases in democracy when all potentially affected individuals are included in the processes and when individuals are autonomous in their persons and judgements. Courts and expansions of rights: Courts decisions that equalize and expand individual rights tend to expand democracy. Capacities of collective agents: Subsidiarity combined with centralized accountability enhances democracy by enabling more effective and efficient collective agents. More transparency, institutionalized oversight, audits of financial records, and associational monitoring increases democracy by limiting corruption of collective agents and increasing public trust.
NOTES 1 2
By ‘non-institutionalized politics’, I do not mean that institutions do not exist within these domains but rather that the institutions are not specialized for politics—collective decision-making and action under conditions of conflict. This is not to say that there are not important beginnings. From the spatial/structural side, two attempts to conduct systematic ‘democratic audits’ stand out: the Democratic Audit of Sweden (Micheletti 1998) and the Democratic Audit of the UK (Weir and Beetham 1998; Beetham 2001). From the perspective of normative democratic theory, Dryzek (1996) and Young (2000) go the furthest in providing criteria of assessment appropriate to a second transformation.
A SECOND TRANSFORMATION OF DEMOCRACY?
3
249
Although group rights sometimes support democracy, especially when they benefit disadvantaged minorities, it is difficult to generalize owing to the tendencies for group rights to empower group elites, often against group members (Young 2000: ch. 4; Kymlicka 1995).
11 Democratic Publics and Democratic Institutions Russell J. Dalton, Bruce E. Cain, and Susan E. Scarrow There is a tendency among some political analysts to look upon the politics of the 1950s and early 1960s with nostalgia—as a halcyon period when democracy had triumphed in the Second World War, economic growth was improving living standards, and government satisfied the needs and expectations of its citizens (Almond and Verba 1963; Stokes 1962). Today, democracies have also triumphed in the cold war, living standards are far better than a generation ago, and there are claims that democracy represents the end of history—but, in contrast to earlier times, public concerns about the internal challenges and crises of democracy now abound (Starke 1993; Nye, Zelikow, and King 1997; Norris 1999; Pharr and Putnam 2000; Hibbing and Theiss-Morse 2002).1 Indeed, citizens in contemporary democracies are more critical of the way that democracy functions (Dalton 2003). Where once the democratic norm led people to be allegiant and passive, citizens today are more sceptical of politicians, parties, and political institutions, as well as more assertive in their political participation. Although electoral participation is decreasing, engagement in new forms of democratic participation is increasing (Dalton 2002: chs. 3–4; Putnam 2002; Verba, Schlozman, and Brady 1995: 72). The share of the public that engages in direct forms of action, ranging from signing petitions to participating in protests, has grown over the last several decades (Inglehart 1997: ch. 10). Citizen groups, public interest organizations, and NGOs have proliferated, and these groups claim to represent issues that democratic governments have ignored. Thus, contemporary democracies generally face popular pressures to grant more access, increase the transparency of government, and become more accountable for their actions. In addition, as Mark Warren (Chapter 10) eloquently describes, the nature of advanced industrial societies is changing. These societies are more complex in their social and economic structures. This complexity and specialization leads to the fragmentation of political interests and identities. As interests proliferate and gain greater political voice, it becomes more challenging to reach a collective
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
251
solution to policy questions (Dahl 1982). At the same time, the units within these systems become more and more interdependent. This increases the need for coordination among diverse interests and actors. These twin forces of complexity and interdependence challenge traditional forms of democratic decision-making, which also stimulates new demands for political reforms that adapt democracy to a new age (Bobbio 1987; Zolo 1992). Our project began with an awareness of these pressures for contemporary democracies to reform and adapt. Often these calls are identified with the Jeffersonian principle that the cure for democracy's ills is more democracy: hence a call for the expansion of the democratic process in response to these changing social and political conditions. Thus, we sought evidence of the significant changes in the institutions and processes of democracy that were occurring in response to these reformist demands. For instance, public interest groups are now contentious new political actors in contemporary democracies: we asked whether democratic processes have responded by granting these groups new political standing or new access into the political process. Similarly, the protest politics of the late twentieth century provided a new form of political action and expression; but, unless protests (or their activists) move inside the institutions in some way, the impact on public policy is indirect and uncertain. Interdependence and a more dynamic political environment place a premium on access to information, so that actors can learn what others are doing and respond. So one of our initial questions was whether democracies have reacted to these demands by making governance more accessible and transparent to its participants. Once we assessed how political processes are changing, our attention broadened to the question of how these changes are affecting the nature of democracy: are we seeing a new democratic transformation to rival the populist transformation of the early twentieth century (Dahl 1989; Barber 1984; Warren 2001)? Democratic theorizing has become a growth industry, again. Democratic theorists, however, often disagree on the elements and overall structure of these new forms. At best, there is a Gramscian consensus that the old era of democratic politics is passing, but also an uncertainty about what form the new democratic reality will take. The hope is that the expansion of democracy will improve the democratic process, following the Jeffersonian dictum. However, some scholars—including within our project—worry that the wrong reforms might actually damage important elements of the democratic process. These questions fuelled our project. Each of the scholars assembled here assembled and analysed the evidence of democratic institutional reform in his or her field of specialization. Together, we have painted a picture of how contemporary democracies are changing, and in the preceding chapters we have begun to sketch the implications of these reforms for the functioning of the democratic process. This conclusion offers a broader view of the mural we have painted. The results give an image of how democracies are evolving, and how the democratic process is being transformed as we enter the twenty-first century.
252
DALTON, CAIN, AND SCARROW
THREE FORMS OF DEMOCRACY Democracy provides the potential for citizens to gain access to the political process in multiple ways (Verba and Nie 1972; Verba, Nie, and Kim 1978). Chapter 1 outlined three broad forms of democratic participation and group access to the governing process: representative democracy, direct democracy, and advocacy democracy. This section summarizes the evidence on institutional change for each of these forms. Much of the history of democratic theory and practice focuses upon the processes of representative democracy. Scholars ranging from Schumpeter to Dahl, for example, often treat competitive elections as the primary measure of democracy. Some analysts assess the health of the democratic process by the percentage of the public that turns out at election time. Elections give citizens their greatest single method of influencing the course of government, through the selection of political elites. The connection of the public through parties and elections to the institutions of governance is an essential element of the democratic process—some have said the defining element. The contributors to this volume first considered whether the contemporary wave of democratic reform has affected the processes of representative democracy. In Chapter 2, Dalton and Gray documented a growth in the number of electoral opportunities in these nations since 1960. Where once the typical democratic citizen (outside the United States and Switzerland) could cast less than a handful of votes during a four- or five-year electoral cycle, now the number of offices determined by elections has grown in most democracies. Today, Europeans select a parliament for the European Union, regionalization has increased the number of elected sub-national governments, and in several nations other elective offices have been created at regional or local levels. This expansion of the amount of electing has reached unprecedented levels in the United States. For instance, in the November 2002 elections, a citizen in Houston was asked to cast a choice on more than 100 items on the ballot! Change is qualitative as well as quantitative; citizens are participating in new types of decision-making, often in ways that give them more scope to cast a direct verdict on policies or individual politicians. As documented in Chapter 3, in most nations the public has more opportunities to offer verdicts on specific policies in referendum votes, a trend that is particularly clear at the local level. There are increasing opportunities for citizens to directly vote for the executive leader at the local level, instead of merely casting a vote for the parties who designate leaders. These changes suggest that citizens now have a larger and more direct role in important political decisions, though in a way that builds upon, rather than supersedes, representative institutions. In the democracies we study, representative processes still dominate, and within these processes political parties remain the primary actors, shaping political competition and structuring the avenues for individual political advancement. Yet even
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
253
the parties are affected by some of these trends towards greater inclusiveness and openness. In Chapter 4, Kittilson and Scarrow showed how many parties are changing their internal rules to give individual members a greater role in selecting party leaders and party candidates. Whether or not these changes will result in real power shifts within the parties, they are apparently motivated by parties’ desire to appear more in touch with members and other supporters. These changes also make parties appear more in step with prevailing democratic norms that value representative and transparent processes over murky and oligarchic alternatives, such as the classic ‘smoke-filled rooms’ of politics. In making these changes, parties recognize and thereby reinforce these norms. These reforms have strengthened the basis of representative democracy as a pillar of the democratic process. Together, these findings lead to the conclusion that the significance of representative democracy as a method of access and influence has probably held constant or even increased slightly during the later half of the twentieth century. It is true that turnout in elections has fallen by about 10 per cent across the advanced industrial democracies over the past four decades (Wattenberg 2002; Blais 2000; Franklin 2003), and this partially signifies a decrease in political access (or use of this means of access). But, at the same time, the amount of electing is up by an equal or greater amount (Chapter 2). Because of the expansion in electoral choices, citizens are travelling to the polls more often and making more electoral choices. An increased number of parties and an opening of electoral processes are additional examples of expanding democratic access. Moreover, reforms to develop internal democracy within political parties help to make political parties more accountable to their supporters and the decisions of party elites more transparent. In summary, much as we noted in the introductory chapter about the American populist reform era of the early twentieth century, the current wave of democratic change begins by reforming the institutions of representative democracy. A second set of reforms expands direct democracy. Instead of relying on the mediated influence of representative democracy, initiatives and referendums allow citizens to decide government policy directly. Moreover, actors outside of the government can control the framing of issues and even the timing of the policy debate, further empowering the citizens and groups that use this mode of action. Direct democracy was rarely used by citizens and public interest groups in the 1950s and early 1960s, but its usage is expanding. In Chapter 3, Scarrow found that the number of national referendums in our set of nations roughly tripled between the 1960s and the 1990s; referendum usage also expanded at regional and local levels of government. Similarly, over three-quarters of the increase in the amount of electing that Dalton and Gray (Chapter 2) document is attributable to increased use of referendums at national, state, and local levels. Referendum usage has grown dramatically in Italy, dealing with a host of policy issues. The expansion of the European Union has been closely linked to referendums on new treaty terms in several European nations. And in the United States and
254
DALTON, CAIN, AND SCARROW
Switzerland the use of direct democracy has grown substantially over the past several decades. Do these changes represent a fundamental expansion in political access and the means of making government accountable? On the one hand, the political impact of each referendum is more limited than an election to decide the national legislature, since only a single policy is decided in each referendum. On the other hand, the increasing use of referendums influences political discourse and principles of political legitimacy beyond the policy at stake in any single referendum. In Britain, for instance, the introduction of the first referendum in 1975—on European Community membership—reshaped democratic theory and practice. No longer was parliamentary sovereignty absolute; instead, the concept of popular sovereignty was legitimized. Subsequent devolution decisions required additional referendums, and today contentious issues, such as adoption of the euro, are considered issues that ‘the public should decide’. Thus, even though the use of direct democracy in Britain (and other nations) remains limited in terms of the frequency of usage and the scope of issues decided in this manner, the expansion of this mode of access represents a significant institutional change for contemporary democracies. A third area of institutional reform involves what we label ‘advocacy democracy’. In this form, citizens or public groups directly interact with government and even directly participate in the deliberation process, even if the actual decisions remain in the hands of government elites. One might consider this to be traditional lobbying activity, except that it does not involve traditional interest groups or standard channels of informal interest group persuasion. Rather, advocacy democracy empowers individual citizens or citizen groups to participate in advisory hearings, attend open government meetings (government in the sunshine), consult an ombudsman to redress a grievance, demand information from government agencies, and challenge government actions through the courts. The evidence of change in the use of advocacy democracy is less direct and less easily quantifiable than other aspects of institutional change; but the overall expansion is undeniable. Government was often conducted in the proverbial smoke-filled back room, where neither citizens nor other political actors knew what was discussed or what was decided. For instance, in 1960 most committee sessions of the US Congress were closed to the public, and a veil of secrecy covered much of the governing process. There have been dramatic increases in the transparency of government over the past several decades. Today, nearly 90 per cent of hearings in the US Congress are open to the public, and often they are televised. In 1960, only two nations had Freedom of Information (FOI) laws; Cain, Eagan, and Fabbrini (Chapter 6) demonstrated the spread of these provisions to all the advanced industrial democracies. Similarly, Ansell and Gingrich (Chapter 8) described the expansion of the office of government ombudsman, and formal provisions that give individuals and citizen groups access to administrative hearings and other policy forums.
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
255
In Chapter 7, Ansell and Gingrich demonstrated how the formation of new regional governments and the devolution of power to sub-national governments create new channels of popular access. This movement of governing closer to the people expands the potential for direct citizen involvement in policy formation and administration. Few individual citizens travel to Washington, London, or Tokyo to interact directly with national government officials; but many more travel to their city hall or other local agencies. Indeed, a variety of sources document increasing citizen participation in local community groups, self-help groups, neighbourhood associations, and other activities dealing with political matters (Verba, Schlozman, and Brady 1995: 72; Dalton 2002: ch. 3). Finally, Cichowski and Stone-Sweet (Chapter 9) described how the transformation of judicial processes complements these other institutional reforms. The expansion of rights legislation and adjudication provides new routes of political access for individuals and citizen groups; and use of them has dramatically increased in those nations where empirical evidence is available. In addition, changes in the concept of legal standing give new opportunities for citizen groups to challenge government legislation. For instance, environmental groups were once unable to sue to protect the environment unless they could demonstrate a direct loss from the disputed action. In many countries legal and administrative reforms of legal standing rules now allow such citizen groups to represent the ‘public interest’ in such cases. This pattern is repeated in other policy areas and with other citizen interest groups, creating a new route of access to politics. One could imagine these three forms of democratic politics as Venn diagrams with partially overlapping circles. At the start of our time frame in 1960, the predominant means of citizen access to the democratic process would be signified by a large circle for traditional forms of representative democracy: campaign participation and voting. Perhaps threequarters or more of all citizen access was channelled through the processes of representative democracy. Rather than shrinking over time, we maintain that the scale of this access had held constant or slightly increased by 2000 because of the growth of voting choices and reforms of the processes of representative democracy. By comparison, direct democracy would be signified by a much smaller circle in 1960, which has grown substantially over time. In imprecise terms, we might expect the size of this Venn circle to have doubled or tripled over this period. Direct democratic channels still represent a small part of the democratic process, but their share is much larger than a generation ago. Finally, advocacy activities were also quite modest in 1960s, but grew dramatically over time. Especially in European democracies where direct citizen action was initially quite rare, the expansion of public interest groups, Buergerinitiativen, and other citizen groups substantially expanded the repertoire of political action. Administrative reforms, decentralization, the judicialization of politics, and other factors created new opportunities for access and influence. Moreover, we have focused on institutionalized political access; the growth of unconventional
256
DALTON, CAIN, AND SCARROW
political action—protests, demonstrations, and similar activities—has also been substantial over this time span.2 The potential for access and actual use of new channels of advocacy democracy has increased several times over in most advanced industrial democracies during the later half of the twentieth century. In summary, institutional reform has expanded citizen access and participation in the political process, in all three styles of democratic action. Citizens and public interest groups today are more engaged, through different channels of access, than a generation ago. Political change on all three dimensions is transforming the relationship between citizens and their democratic state.
THE CHANGING RELATIONSHIP BETWEEN CITIZENS AND THE DEMOCRATIC STATE Beyond the simple conclusion that democracies are expanding access, how have these institutional reforms actually changed the democratic process in these nations? Robert Dahl's writings (1956; 1971; 1989; 1998) provide a benchmark for defining the essential elements of democracy. But like many other democratic theorists, Dahl tends to equate democracy with the institutions and processes of representative democracy. In On Democracy, for example, Dahl (1998) discusses democracy in terms of elections and the mass franchise, paying much less attention to other forms of citizen influence that may actually represent important and in some cases more influential methods of citizen influence over political elites.3 Similarly, Joseph Schumpeter (1943: 260) offered a minimalist definition of democracy: ‘The democratic method is that institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people's vote.’ In this tradition, democracy is often analysed solely in terms of representative democracy. Our broad examination of democratic change stems from a conviction that citizens do not necessarily just vote and then serve in chains until the next election, as Rousseau claimed in his critique of representative democracy. Recognizing the uses and limits of these other forms of citizen participation is essential to understanding democracy in its entirety. Representative democracy is a base, and citizens also work through other channels of access to influence policy-makers between elections. These other channels of democratic politics are largely missing from many treatises on democratic theory.4 As with Warren's discussion in Chapter 10, we draw upon Dahl's On Democracy to define the essential criteria for a democratic process. We want to build upon Warren's discussion and broaden Dahl's framework to compare the implications of a changing mix of representative democracy, direct democracy, and advocacy democracy. Such comparisons can determine the implications of a changing repertoire of political action for democratic publics. Dahl (1998: 37–8) suggested
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
257
that democracy be judged by five criteria: • • • • •
Inclusion. All, or most, adult permanent residents should have full rights of citizenship implied by the following criteria; Political equality. When decisions about policy are made, every member must have an equal and effective opportunity to participate, and participants must be given equal weight; Enlightened understanding. Within reasonable limits, each member must have equal and effective opportunities for learning about the relevant alternative policies and their likely consequences; Control of the agenda. The members must have the exclusive opportunity to decide how and what matters are to be placed on the agenda; Effective participation. Before a policy is adopted, all the members must have equal and effective opportunities for making their views known to other members.
We offer a complementary discussion to Warren's chapter, considering how the processes of institutional change described in this volume are related to these criteria. Warren focused on the criterion of inclusion, and we extend this discussion to consider how new forms of democracy fulfil other democratic criteria. The first column of Table 11.1 lists Dahl's five democratic criteria. The second column summarizes the prevailing view on how well the electoral process of representative democracy fulfils these criteria. For example, inclusion is a fundamental element of democracy (see also Chapter 10). Advanced industrial democracies addressed the inclusion criterion through the expansion of the franchise to all adult citizens—a process that required a long series of reforms but is now virtually assured in these nations.5 Success in meeting this goal is indicated by presenting ‘universal suffrage’ in boldface. The political equality criterion also posed an initial challenge to many supposedly democratic states because they offered multiple voting for some citizens or excluded other potential voters (de facto or de jure). Nearly all advanced industrial democracies now meet the equality criterion for elections based on the principle of one person-one vote, which is also presented in boldface. At the same time, some problems of equality beyond formal access remain. For example, contemporary debates about campaign financing and registration are linked to the equality criterion, and full equality in political practice is probably unattainable. Summaries of these remaining equality problems are noted in shaded areas. Overall, however, the principle of equality is now a consensual value when applied to elections and the process of representative democracy in these nations. It may at first appear that the expansion of the electoral marketplace discussed in Chapter 2 simply extends these characteristics to new elections. But increasing the number of times that voters go to the polls and increasing the number of items on the ballot tend to depress turnout—which works against the equality criterion.
258
DALTON, CAIN, AND SCARROW
Table 11.1.Dahl's democratic principles 1. Democratic criteria Inclusion
Political equality
2. Representative democracy Universal suffrage provides inclusion
One person/one vote with high turnout maximizes equality Problems of low turn-out, and inequality due to campaign finance issues, etc Enlightened understanding Problems of information access, voter decision processes
Control of the agenda
Effective participation
3. Direct democracy
4. Advocacy democracy
Universal suffrage provides inclusion
Equal citizen access
One person/one vote maximizes equality Problems of lower turn-out
Problems of access to nonelectoral arenas Equal opportunity Problems of very unequal use
Problems of greater information Increased public access and higher decision-making de- to information mands placed on citizens Problems of even greater information and decision-making demands Citizen initiation proCitizens and groups vides control of agenda control of locus and focus of activity Problems of control of campaign Problems of influence by special debate, selecting candidates, etc. interest groups Control through respon- Direct policy impact en- Direct access avoids sible parties sures effective participa- mediated participation tion Principal-agent problems: fair elections, responsible party government, etc
Notes: Criteria that are well-addressed are presented in bold, criteria that are at issue are presented in italics.
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
259
When local elections, or even European Union elections, draw less than half the electorate to the polls, one must question whether the gap between equality of access and equality of usage has become so large that it undermines the basic principle of political equality. If only a quarter of the American public actually votes for a local mayor or a local school board, this erodes the legitimacy of the electoral process. Moreover, low turnout may distort the representativeness of elections. Second-order elections tend to mobilize a different electorate, which may be more ideological than the public at large; and more second-order elections would mean more distortions within the electoral process. The tension between democratic theory and democratic practice becomes even more obvious when we turn to the criterion of enlightened understanding. The political behaviour literature has long debated whether the average voter possesses the necessary information and cognitive skills to make ‘enlightened’ political choices (Converse 1964; Delli Carpini and Keeter 1996; Popkin 1991; Dalton 2002). While we are fairly sanguine about the necessary access to information and the voters’ abilities to make rational choices when it comes to high-visibility elections, such as presidential and national parliamentary elections, the expansion of the amount of voting gives a new meaning to this debate. How does a resident of Houston make enlightened choices on the dozens of judges that appeared on the November 2002 ballot as well as other local offices? In such second-order and third-order elections, the heuristics that voters can use in high profile first-order elections may be insufficient or even lacking altogether, as in non-partisan races for low-visibility offices. Thus, the expansion of the electoral marketplace may empower the public, but it raises questions about the ability of voters to exercise these new electoral opportunities. Another problematic criterion of representative democracy is the control of the political agenda. If elites can structure the agenda-setting process, this can minimize the ability of the public to control the course of government. This is especially problematic when the opportunities for citizen input are limited, and elites dominate political discourse as well as the institutions of governance. The recent reforms of representative democracy partially address this criterion. Increasing the number of elected offices gives citizens more input and presumably more venues to raise relevant issues. Moreover, recent political finance reforms that aim to equalize campaign access and party support also facilitate greater openness in political deliberations. In short, the reforms discussed in the first section of this book generally work to strengthen this democratic criterion. Finally, a crucial issue for representative democracy is the effectiveness of participation. Do citizens get what they vote for? Often this principal-agent problem is solved through the mechanism of party government; voters select a party and the party ensures the compliance of individual MPs and the translation of electoral mandates into policy outcomes (Klingemann et al. 1994). It is difficult to deduce how the electoral reforms described in this volume affect this democratic criterion. On the one hand, more openness and choice in elections
260
DALTON, CAIN, AND SCARROW
should enable the public to express their political preferences more extensively and in more policy arenas. If voters can elect the mayor, for instance, the connection of voters to elites is more direct than if this were an appointed office. On the other hand, as the number of office-holders and levels of government proliferates, it may become more difficult for voters to assign responsibility for policy outcomes to multiple elites where divided party and political control become more common. We share Warren's concern (Chapter 10) that fragmented decision-making, divided government, and the simple multiplicity of elected officials may diminish the political responsiveness of each actor. In short, recent institutional reforms increase the potential for participation in the processes of representative democracy, but we are uncertain how this systematically affects the effectiveness of participation. Democratic theorists have focused on how voting in party/candidate elections fulfils these five democratic criteria, but we can ask the same questions of direct democracy (column 3 in Table 11.1). Since most referendums and initiatives are based on universal suffrage and one person/one vote, they operate in much the same way as representative elections in terms of inclusion and political equality. The notable exceptions are special elections and bond elections that define the voters as a subset of the total electorate; for instance, allowing only property holders to vote on a special tax assessment on property. As Mark Warren notes in Chapter 10, such elections and other participatory forms that violate the inclusive norm pose fundamental problems from the standpoint of democratic theory.6 We would argue, however, that when one moves beyond the institutions of representative democracy in first-order universal-franchise elections, the democratic criterion of inclusion becomes more complicated than the simple assessment of equal access. Equality of opportunity does not mean equality of participation when the threshold for becoming engaged is higher, and the level of participation falls far lower, than in national vote levels. Turnout in direct democracy elections is often lower than for comparable elections for public officials. Thus, the expansion of direct democracy potentially raises questions about political equality in low turnout elections. For instance, when Proposition 98 appeared on the 1996 general election ballot in California, barely half of the voting-age public turned out in this election and only 51 per cent of these voted for the Proposition. But, as a consequence of this election, California's constitution was altered to mandate that a specific part of the state budget be directed to primary and secondary education. This has reshaped state spending and public financing in California (and, some would argue, has not necessarily benefited the programmes the referendum was supposedly designed to help). Such votes raise questions about the fairness of elections in which a minority of the total eligible electorate makes crucial decisions that affect the public welfare. Referendums and initiatives also place even greater demands for information and understanding upon the voters. Many of the heuristics that voters can use in party elections or candidate elections are less relevant to referendums. Moreover,
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
261
the issues themselves are often complex and beyond what a typical citizen could understand. For instance, did the average Italian voter have enough information to make enlightened choices on the 1997 referendums dealing with television ownership rules, television broadcasting policy, the hours that stores could remain open, the commercial activities that municipalities could pursue, labour union reform proposals, the regulations for administrative elections, and residency rules for Mafia members? Arthur Lupia (1994; also Lupia and McCubbins 1998) presented provocative evidence that voters can rely on group heuristics and other cues to make informed decisions on referendums. But obviously the proliferation of policy choices and especially the introduction of lower-salience local referendums raise questions about the viability of such cue-taking models (LeDuc 2002). Thus, the expanded use of referendums generates questions about whether voters can fulfil the criterion of enlightened understanding in making their decisions. The strength of direct democracy primarily involves Dahl's last two criteria. Referendums and initiatives shift the locus of agenda setting from elites towards the public and interest groups. Indeed, issues that elites do not want to address can be brought into the political arena through processes of direct democracy: tax reform or term limits in the United States, abortion reform in Italy, and the terms of EU membership in Europe. Even when referendums fail to reach the ballot or fail to win a majority, the mechanism of direct democracy can force elites to be more sensitive to public interests (Bowler and Donovan 2001; Gerber 1999). Direct democracy can strengthen the public's ability to shape the agenda of politics. Furthermore, by definition direct democracy should provide for more effective participation than exists with methods of representative democracy. Direct democracy is unmediated, and so it ensures that participation is effective. Voters make policy choices with their ballot: to enact a new law, to repeal an existing law, or to reform the constitution. Even in instances where the mechanisms of direct democracy require some elite response in passing the law or re-voting in a later election, the direct link to policy action is clearer and more immediate than through channels of representative democracy. Direct democracy thus seems to fulfil the agenda-setting and effective-participation criteria of democracy, but there are questions in these areas as well (also see Warren's discussion in Chapter 10). Elizabeth Gerber (1999) and David Broder (2000) raise the important question whether special interest groups find it easier to manipulate processes of direct democracy to their own advantage than to work through representative processes. The discretion to place a policy initiative on the ballot can be appealing because it gives special interests a chance to make an unmediated appeal to voters when they cannot achieve what they want through the legislature. In addition, direct democracy decisions can be drafted to the special interests’ liking and are less susceptible to the bargaining and checks and balances that occur within the normal legislative process. They are often also more immune from revision by future generations than are normal legislative decisions, given that a legislature may find it politically difficult, and in some
262
DALTON, CAIN, AND SCARROW
cases constitutionally impossible, to repeal or revise policies adopted by a direct popular vote. Some recent referendums in California illustrate this style of direct democracy: wealthy backers pay a consulting firm to collect signatures to get a proposal on the ballot, and then bankroll a campaign to support their desired legislation. This is not grass-roots democracy at work, but the representation of special interests by other means. In summary, the expansion of direct democracy can potentially complement traditional forms of representative democracy. On several criteria, it represents a positive expansion of the democratic process by allowing citizens and public interest groups new avenues of political access, and new ways to shape the political agenda. But direct democracy also raises new questions about the equality of influence, if not access, and the public's ability to make fair and reasoned policy judgements. Perhaps the most important factor is not whether direct democracy is expanding but how it is expanding: are there ways to achieve the increase in access and influence without also increasing the problems of inequality and accountability? We return to this question later in this chapter. The final column in Table 11.1 considers the new forms of advocacy democracy that are examined in this volume (Chapters 6–9). These new forms of action provide citizens and public groups with valuable and politically significant new access to politics, but it is also clear that this access is very unevenly used. Nearly all can vote, and most do. But very few citizens can (or do) file a lawsuit, make requests under a Freedom of Information Act, attend an Environmental Impact Review hearing, or attend local planning meetings. Thus, advocacy democracy raises potential problems of democratic inclusion and equality. In addition, participation in these more demanding forms of access also places a larger conceptual burden on the citizen who must understand the workings of the policy process and complex issues, and not just show up on election day to mark an X on the ballot. There is no institutional or formal equivalent to one-person-one-vote for advocacy democracy. Thus advocacy democracy raises the question of how to address the criteria of inclusion, political equality, and enlightened understanding. Equality of access is not sufficient if equality of usage is grossly lacking, particularly if usage is highly biased by the skill and resource variables that predict such participation (for example, Verba, Schlozman, and Brady 1995). The extent of this inequality is seen in the following example from a European Election survey. When Europeans were asked whether they voted in the election immediately preceding the survey, social status differences in participation were very slight (Fig. 11.1). A full 73 per cent of the less educated said they had voted in the previous European Parliament election (even though it is a second-order election), and an identical percentage of the better-educated claimed to vote (Tau-b =.03). Education differences in campaign activity are somewhat greater, but still modest in overall terms (Tau-b =.11). A much larger inequality gap emerges for modes of participation that come closer to direct or advocate forms of democracy. For instance, only 13 per cent
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
263
Fig. 11.1.Inequality in participation by education group
Note: Results combine the twelve nations weighted to represent the total EU population. Source: Eurobarometer, 31 (March–April 1989) and 31A (June–July 1989). of the less educated said they had participated in a citizen action group; but nearly three times as many of the bettereducated had participated (Tau-b =.20). Similarly, quite large inequality gaps exist for signing a petition (Tau-b =.21) or participating in a lawful demonstration (Tau-b =.21). Like the old European proverb of beggars sleeping under a bridge, the law treats everyone equally when it comes to opportunity, but it is in the use of opportunities that real inequality exists. On the criterion of enlightened understanding, advocacy democracy has a mixed result. On the one hand, advocacy democracy can enhance citizen understanding and facilitate expanded inclusion in several ways. Citizens and public interest groups can increase the amount of information that the citizen has about government activities, especially through FOI laws and participating in government policy-making and administrative hearings. And with the assistance of media that disseminate this information, the public can better influence political outcomes. By ensuring that information is given to citizens in a timely fashion, advocacy democracy allows citizens to make informed judgements that make governments more accountable. And by eliminating the filtering that governments would otherwise apply, it ensures that citizens get a more accurate picture of the considerations that influence policy decisions, with fewer coverups and self-serving distortions. On the other hand, advocacy democracy makes greater cognitive and
264
DALTON, CAIN, AND SCARROW
resource demands on participants, and thus may similarly suffer from the inequalities of participation noted above. It requires much more of the citizen to participate in a public hearing or petition a government administrator than to simply cast a vote at election time. The most insightful evidence comes from Mansbridge's (1980) study of collective decision-making in New England town hall meetings. She finds that many participants are unprepared or overwhelmed by the deliberative and decision-making processes. Democratic practice falls short of the idyllic images of democratic theory. Advocacy democracy fares better on the remaining two democratic criteria. Advocacy democracy gives citizens greater control of the political agenda, in part by increasing their opportunity to press political interests outside of the institutionalized time and format constraints of fixed election cycles. Using the tools of advocacy democracy, participants often can choose when and where to challenge a government directive or pressure policy-makers; this is a strength of advocacy democracy. Similarly, even though these forms of action are often attempts at persuasion where elites still make the final policy decisions, they do nevertheless provide direct and unmediated access to government. Property owners might participate in a local planning hearing, a public interest group might petition government for information on past policies, and dissatisfied citizens may attend a school board session. Such direct and unmediated participation brings citizens into the decision-making process rather than having them rely solely on agents. In the end, unmediated access might not be as effective as the efforts of a skilled representative speaking for one's interests, but greater direct involvement in the democratic process should improve the accountability and transparency of the democratic process. Thus the strength of advocacy democracy is denoted by the boldface entries in these last two cells of the table. In summary, advocacy democracy increases the potential for citizen access in important ways. It gives citizens and public interest groups new influence over the agenda-setting process, and it can give the public unmediated participation in the policy process. These represent important extensions of democratic rights. At the same time, however, advocacy democracy may exacerbate political inequality because of the inequality of usage. As new access points are created through advisory panels, consultative hearings, and other institutional reforms, some citizens are empowered to become more involved in the democratic process. But other citizens who do not possess the skills or resources to compete in these new domains are left behind. A good illustration of this problem is seen in the realm of environmental policy. This is an area where citizen and public interest groups have gained new rights and new access to the democratic process, as has been described by several chapters of this study. This is seen as a genuine expansion of the democratic process. But this democratic potential is disproportionately used by citizens who are already participating in politics and who possess the skills and resources to engage in these new forms of action. Left behind are the poor and minorities. Thus
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
265
environmentalism is an example of democratic empowerment of the citizenry, but it also is an arena in which we may see democratic inequality increasing and producing problems of environmental justice. In summary, Chapter 10 discussed some of the trade-offs that exist between different forms of democratic action. Our consideration of other democratic criteria leads to an even stronger realization that each democratic form potentially has both advantages and limitations. Moreover, the different modes of democratic participation have other systematic implications for the functioning of democracy. For instance, representative democracy and advocacy democracy entail a shift in the primary political actors: from parties toward interest groups and social movements (see below). This affects both the nature of political access and the types of interests that may gain access. Moreover, the expansion of interest group liberalism through advocacy democracy creates new questions of representation.7 Advocacy democracy may thus respond to the strength of the claimants rather than to the strength of their claims. Even the basis of political expertise changes. While advocacy democracy values know-how and expertise in the citizenry, it devalues those same characteristics among policy-makers. Other questions involve the effectiveness and efficiency of government under different democratic modes. No form of democratic action is ideal; each contains advantages and limitations.8 As action repertoires shift from a predominant reliance on processes of representative democracy towards a mix including greater use of direct and advocacy democracy, this creates the need to find a new balance among democratic forms. As yet, democracies have not fully recognized this potential problem and thus have not attempted to find institutional or structural methods to address them. Our findings might contribute to the realization that democratic reforms represent opportunities, but also challenges in ensuring that promoting one democratic criterion does not weaken the democratic process in terms of other criteria.
COMPLEX DEMOCRACY AND THE STATE The expansion of citizen opportunities to acquire more information about, observe, and participate in a state's activities has important implications not only for representation but also potentially for the way the state itself operates and performs. From the citizen's point of view, direct democracy empowers citizens and groups to influence directly government policy and even the polity's constitutional structure. In addition, advocacy democracy provides more opportunities for citizens to learn about and influence legislative and administrative decisions. The articulation of their interests and preferences is no longer exclusively or even primarily channelled through the party system and the electoral process. New paths have been added on top of the old forms of representation, allowing citizens more effectively to make their voices heard between elections. From a
266
DALTON, CAIN, AND SCARROW
strictly governmental perspective, the expansion of citizen opportunities alters the administrative structure in very important ways, and might have significant long-term effects on systemic performance and policy legitimacy. In this section we consider how the expansion of direct democracy and advocacy democracy may affect the structure and workings of the democratic state.
Expanded Voting and Direct Democracy The expansion of voting opportunities and especially of direct democracy has profound implications for democratic processes in advanced democracies. Limited elections and decision-making through elected representatives were the hallmarks of most Western democracies. Frequent elections, it was thought, would throw representatives into a state of constant electioneering, depriving representatives of the time and independence to make sound judgements. In countries where both legislative chambers are elected bodies, delegates to the upper chamber are often elected for longer terms as a mechanism for insulating them from volatile public opinion. Countries like Italy were said to have too many elections and to suffer from too much political instability. Public opinion was too unstable and ill-informed to be the basis of day-to-day government decision-making. Representatives could be held in check by periodic fixed elections or contests called by the government within a set period of time. There was no need to consult compulsively with the people in the interim. Similarly, representative government was thought to be more stable than direct democracy. With the exception of Switzerland, advanced industrial democracies until recently resisted having frequent national referendums. Just as too many elections deprived states of stability and representatives of independence from an ill-informed electorate, initiatives and referendums were thought to strip away the protective judgement of politicians. At the worst, direct democracy could be used by populist politicians as plebiscites on personal popularity. Even in the best of circumstances, legislation produced by popular vote was seen as necessarily simplified and inflexible, and it is difficult to coordinate referendum choices with policy choices made in other arenas. This could produce more policy reversals, inconsistencies, and mistakes. As we have noted several times, while the old electoral structures have not been abandoned, they have been transformed in important ways. The democratic feedback of national contests has been supplemented with results from new types of local elections, EU elections, referendums, and the like. Combined with more widespread use of polling by governments to monitor public opinion as well as with political sensitivity to more widespread mediasponsored opinion polling, the effect is to strengthen the public opinion signal between electoral cycles. This is particularly true of direct democracy mechanisms, which also raise an institutional question: do referendums and initiatives supplement representative government or do they somehow subvert it? Supplementation implies the continued
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
267
importance of representative government, with referendums, recalls, and initiatives functioning as safety valves when representative government gets stuck and either cannot or will not make a needed decision. So far we have suggested that this is the case: that these mechanisms of direct democracy are overlaid on traditional representative forms. But in contrast to the safety-valve model is the more radical idea that direct democracy can replace indirect democracy. In this view, representation is an inefficient and potentially distorting mechanism for aggregating preferences. New technology, rising citizen sophistication, and increased expectations of public consultations can lead to direct democracy substituting for indirect democracy. Instead of the presumption that decisions should be made in representative policy arenas first, this view argues that policies can be taken up in either arena as it suits the strategies of policy advocates. This option to choose arenas leads to fundamental questions about the circumstances under which we are likely to see a divergence of interests between the people and their elected representatives. Frankly, although there is mounting scholarly interest in the diffusion of referendums in advanced industrial democracies (Butler and Ranney 1994a; Gallagher and Uleri 1996; Bowler, Donovan, and Tolbert 1998), the literature is still divided on the impact of direct democracy on the institutions and processes of representative democracy. Certainly the nature of democratic politics changes with the expansion of referendums and initiatives, but in what ways and with what consequences is still an open research question. Our findings suggest that an answer to this question should be a priority for further research. The change from indirect to direct government will also produce changes in how interest groups, parties, and other intermediary organizations operate. When representative institutions predominate, interest groups must deal with parties and elected officials. But in direct democracy, intermediaries can become policy-makers, bypassing the link with elected officials (Bowler and Donovan 2001; Gerber 1999). The majority party or government per se no longer tightly controls the policy agenda. And the interest groups that flourished under indirect democracy may give way to new policy entrepreneurs who can navigate the waters of direct democracy more adroitly. This yields a new type of representative democracy, one in which unelected interest groups gain status as policy-makers, no longer confined to trying to influence elected representatives but able to directly influence one type of legislative process.
Advocacy Democracy As previous chapters have documented, the expansion of advocacy democracy can also implicitly undermine some traditional assumptions about state structure. For instance, the classic Weberian government is a hierarchical bureaucracy that implements policy through a chain of command rising to a centralized peak. Centralization is a critical feature of modern state-building, overcoming the
268
DALTON, CAIN, AND SCARROW
cooperation problems that arose in feudal structures. The purest version of the modern state is founded on an undisputed monopoly of national security and revenue collection by the central government. Federal arrangements deviate from the central government ideal type by reserving some powers and authorities for local jurisdictions. This can be done in different ways, creating different kinds of federal arrangements. In its most decentralized forms, subnational jurisdictions retain autonomous powers to tax and spend in specified policy areas, thereby eroding a critical feature of the classic modern state structure. Therefore, the expansion of federal arrangements documented by Ansell and Gingrich in Chapter 8 implies an important re-conceptualization of postmodern state structures. A second aspect of the Weberian state is the concept of the neutral bureaucrat. Politics, in this conception, comes before administration. Parties and voters elect representatives who determine the policy that the state must administer. This neat theoretical demarcation between administration and policy has always been suspect, particularly in regulatory realms and corporatist arrangements that allow for interests to be consulted about policy between elections (Aberbach, Putnam, and Rockman 1981). The creation of citizen charters and juries, the adoption of administrative procedure acts, and expanded citizen opportunities to interact with administrators as they make decisions move the policy-making process forward in the funnel of decision-making. Citizens can influence policy-making through the party system and electoral process as before, but they can also intercede to influence elected officials before they make their decisions; and even after the decisions are made they can attempt to influence the administrators as they implement the policies. Depending upon the amount of discretion granted to the administrator, policy can potentially be remade and even reversed at the point of implementation if enough citizen pressure is applied successfully. To the degree that administrators take public demands into consideration when making decisions, administration gets fused with representation, and the bureaucrat's neutrality is sacrificed in the name of making policy democratically. Third, advocacy democracy undermines the norm of state paternalism as expressed in Burkean views of representation. Strong central states often presume to act on behalf of the national good and in the citizen's best interests. In this parent-child analogy, the state must sometimes act as the trustee of the public's interests. One aspect of this role is the presumption that the state can have secrets. There might be information the state must possess in order to protect citizens from external threats or to preserve marketplace integrity. The state is trusted with this information because citizens understand that it is ultimately in their interests to do so. The FOI laws described in Chapter 6 by Cain, Egan and Fabbrini challenge state paternalism in the sense that citizens can demand that governments reveal information that might otherwise be kept secret. Rather than trusting states to use secrecy for the common good, citizens think of transparency as a way of keeping governments honest.
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
269
Another challenge to state paternalism comes from administrative reforms that conceive of citizens as consumers. In the classic central state model, the bureaucrat implements the law by formulating specific rules and regulations that govern citizen interactions. A good process is one that goes by the books, treating all citizens impartially and evenhandedly. Many of the administrative reforms outlined in Chapter 8 conceive of the citizen as a consumer who must be satisfied. Following the rules is no longer enough, and may even be counterproductive to the ultimate goal of pleasing the customer. The bureaucrat is not presumed to know what is in the citizens’ best interests, and must be responsive to their demands and concerned about the outcomes that result from administrative decisions. The state no longer presumes to know best. It determines what is best on the basis of citizen input and participation in ongoing decisionmaking. The traditional central state also seeks to implement the law faithfully. The goal of advocacy democracy is to satisfy the consumers/citizens. The state adopts the norms of the private sector, with outcomes placed before procedures. Innovations from the private sector are imported into the public sector. No one presumes that corporations have paternalistic responsibilities. The potato chip company does not dictate the flavour of the potato chip to the consumer (at least, not for long). Rather, the features of the product are determined by marketplace behaviour. Some of the administrative reforms arrive at private-sector behaviour by creating a culture of public-sector competition, but more common are reforms that rely on ‘voice’ rather than ‘exit’, to use Hirschman's terms (1970). Democratization provides input and feedback in lieu of market signals. Lastly, Cichowski and Stone Sweet (Chapter 9) document the increase of legal venues and rights that can be used by citizens to check the state's administrative powers. They describe the proliferation of individual rights claims under EU Article 234, the evolution of judicial review, for example, and the expansion of citizen access to the courts across the OECD nations. These legal reforms provide yet another avenue for citizens to alter state policy. If they fail to achieve influence through the traditional party system and expanded electoral opportunities, and their attempts at advocacy through participation in administrative hearings fall short, there is one last arena in which to test the policy: the judiciary. Advocacy can continue in the court system, and citizens can prevail in the end even if they lose at all the preceding stages.
SYSTEMIC PERFORMANCE AND POLICY LEGITIMACY To this point, we have primarily discussed the impact of democratic institutional reforms from the perspective of individual citizens and their relationship to the state. But these changes in the structure of democratic politics also affect systemic performance. Indeed, an important aspect of the debate about democratization and the empowerment of the public is the potentially negative impact that democratic
270
DALTON, CAIN, AND SCARROW
changes might have on governance (Crozier, Huntington, and Watanuki 1975; Huntington 1981). This section considers the potential effect of these institutional changes on two related elements of the governing process: legitimacy and government efficacy.
Direct Democracy, Expanded Elections, and Government Performance Expanded electoral opportunities and direct democracy have a clear impact on perceptions of government legitimacy. Legitimacy is inherently a subjective conception involving the belief that decisions are properly made by the government and accepted by the people. Perceptions of normal and appropriate democratic processes have changed over time. It was legitimate in the early stages of democracy to restrict the franchise to a small, propertied segment of the population, or for elected officials to have a more Burkean approach to representation. It was perhaps inevitable that these early restrictions and the concepts of legitimacy associated with them would fall away, and that democratic expectations would broaden. Implicit within the logic of democracy is a norm of greater inclusion and more equality. Indeed, voters in advanced democracies seem to equate more elections with more opportunities to hold representatives accountable: that is, to punish elected officials by voting against them when they pursue policies that the voters do not like, and to reward them with continued office when things go well. More elections hence mean more chances to hold official feet to the electoral fire. Even if voters do not avail themselves of the opportunity to participate in elections as often as in the past, they now have more opportunities to express their discontent when they need to. The threat of electoral consequences can operate on the minds of officials even when voters have stayed home in the past. Similarly, direct democracy eliminates the potential principal-agent problems associated with representation and substitutes unmediated decision-making by the electorate. That assumes of course that the mediators have not manipulated the referendum and initiative process. As we mentioned before, there is some debate about this. Even so, it is fair to say that modern voters on the whole like the opportunity to decide issues themselves and are often suspicious of what their representatives are up to. Bowler and Donovan (2002) have provided empirical evidence to support this contention with their recent comparison of public images of government across the US states. In summary, direct and unmediated opportunities to vote seem to increase systemic legitimacy. But does expanded access increase the quality of policy-making? This is more disputable. On the one hand, more popular input in the form of more elections and referendums should increase the likelihood that the government will produce policies that match the people's preferences as defined by the median voter. So, if the goal of a democracy is simply to aggregate people's preferences into a collective decision, then more citizen input should lead to better policy output. On the other
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
271
hand, better decisions are not always the immediately popular ones. Anthony King (1997), for instance, questions whether the perpetual campaign strengthens or weakens the governing process. In addition, representative democracies are often designed to offset the dangers of majority tyranny, or to check popular decisions that might be ill-considered. Hence, representative institutions may include protections for minority opinions (such as super-majority rules and proportional representation electoral systems) or features that slow the deliberative process and require legislation to be scrutinized by separate legislative houses. Lacking these protections, referendum-driven policies might violate minority rights more frequently, balance majority and minority interests less well, or adopt measures that the public will later come to regret. Even worse, if the public is inconsistent and inattentive, the quality of policy-making in direct democracy will suffer in obvious ways. Optimists believe that technology and more education prepare citizens for their civic tasks. Pessimists argue that nothing has changed since the early days of democracy, and that the costs of informed civic involvement are too high to be realized by all but a few voters. If the latter, citizens might not be prepared to make the complex and difficult decisions that direct democracy requires. In short, governments may perform less well when they rely upon the quality of citizen input more than the quality of the elected representatives.
ADVOCACY DEMOCRACY AND SYSTEMIC PERFORMANCE What impact does advocacy democracy have on government legitimacy and performance? On balance, it probably exerts a positive influence, but that is by no means a foregone conclusion. On the plus side, greater transparency and access allow citizens to uncover and punish government malfeasance and, by so doing, provides stronger incentives for government officials to behave well. Also, by making it easier to unearth prior mistakes that might otherwise be buried, they allow for learning the lessons of the past. If they improve performance in these ways, it potentially can aid government legitimacy by allaying citizen fears and suspicions about the governments’ motives and actions. At the same time, there is a distinct risk that the lessons that citizens may learn from revelations about past governmental mistakes is that government is not to be trusted because those who govern are incompetent, or worse. More generally, if transparency is taken to an extreme, it could in some ways diminish government efficiency and possibly weaken legitimacy. In particular, if there is such a thing as functional secrecy, as implied in the traditional state model, there could also be dysfunctional openness. This could take many forms. It could mean enemies obtaining state secrets that undermine national security. It could mean trade secrets leaking out to competing businesses and undermining incentives for innovation. Or it could mean that the political opposition's hand is so strengthened that ruling governments are severely weakened by the discovery and politicization of
272
DALTON, CAIN, AND SCARROW
every mistake made by the incumbent administration. Any of these developments could erode the sense of legitimacy and the quality of decision-making that transparency was supposed to strengthen. Greater citizen participation in the various forms of advocacy democracy can also affect governing. A stronger form of transparency, the right to observe decision-making, gives citizens more information, in a timely fashion and without mediation. Aside from the fact that getting information immediately might enhance the prospect of influencing the final outcome substantially, the presence of observers can influence decision-makers to take observer interests more seriously in their decision calculations. The right to participate pushes citizen influence one step further: the right to comment upon, and in some cases to share in, the decision-making. There are several assumptions here: that (1) observed decision-making will be different from and better than the unobserved kind; and (2) immediate observation and participation will produce better governmental responsiveness. The danger is that these assumptions might not be correct. Observed decision-making might be different, but not better. It might lead to public posturing, hardened positions, and less real negotiation or compromise. If observation influences outcomes (for example, decision-makers giving more weight to the concerns of those who are in the audience or closely monitoring them), then any biases in the representation of interests in the hearing will produce biases in policy outcomes. More fundamentally, if observation is a costly activity, as it most certainly will be, then information that is disseminated from these hearings will still be mediated, but by the media and self-selected groups. It is possible that advocacy democracy trades one kind of information distortion (self-serving government distortion) for another (self-serving interest group or media distortion). Clearly, if citizens are allowed to comment upon or participate in administrative decision-making, the potential for policy skew only increases. Depending on how it works out, this could either increase or diminish government legitimacy. Finally, the right to seek to overturn administrative decisions through greater access to the courts has several important effects. It gives citizens more defined rights than they previously had. It also gives them a new policy venue to shop in, increasing the number of potential veto points and opportunities. On the plus side, this provides more checks against potential government abuse. But, as the case of the US amply illustrates, it has distinct problems as well. Litigation costs are likely to soar, and the prospect of losing in court can become a new source of policy paralysis. In short, as with representation, it is more certain that advocacy democracy will alter behaviour than it is whether those alterations will be a good or a bad development in the end. As we have said before, having more opportunities does not guarantee that all citizens will use those opportunities. And if only some take advantage of advocacy democracy, then it is hard to say what the net effects will be in the long run. It will lead to a new kind and expanded version of pluralism,
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
273
with groups and individuals selectively using the new opportunities to further their causes. Whether this will be superior to the old forms of the state is uncertain at this time.
CONCLUSION Our opening chapter raised the question of whether the changes we investigate in this study add up to something that might be described as a ‘second transformation of democracy’. At the end of this study we are ready to cautiously assert that the sum of the changes across multiple governmental arenas does constitute a significant change in policy processes and in assumptions about who should be included in policy decisions. In saying this we do not mean to imply that democratic models have been static in other periods. Rather, it is our contention that the amount of change over the last several decades has been unusually great, as a variety of governing traditions have changed to accommodate new public pressures for inclusion and information. In many ways, the governing institutions of established democracies have undergone a cumulative process of democratization. This is not to say that all the changes have been beneficial or without problems. Although there is tendency to put a high rhetorical value on democratization, the increase of one type of democratic access may exacerbate old problems of inequality and imperfect information. And, as historians would surely point out, the first transformation of democracy at the beginning of the twentieth century was hardly an unmitigated success. In the 1920s and 1930s some democracies failed as newly enfranchised electorates backed non-democratic answers to economic and geo-political crises. In the democracies that survived, fear of ‘mass society’ led to the rise of elitist models of democracy. Thus, in labelling this set of changes a ‘second democratic transformation’ we are not trumpeting a triumphalist message, nor are we suggesting that the changes we document are likely to continue in a democratizing direction indefinitely. We would argue that the main virtue of seeing these changes across multiple arenas as part of a larger pattern is that it enables us to better focus on the realities and challenges of contemporary democratic life. Recognizing that changes bring risks as well as benefits allows us to begin to develop strategies for coping with potential problems, such as the need to bring expanding interest articulation and diminishing interest aggregation into better balance, or to deal with the potential inequality issues raised by these institutional reforms. Solutions may be as much about cultivating new attitudes and new forms of civic education as about the introduction of new procedures. Greater opportunities for citizen participation also put greater demands on citizens to be educated about how political processes work, and to recognize the limits and dangers of these processes. There may be a need to teach citizens political virtues such as political moderation, coalition building, and compromise, lessons that are routinely learned by new legislators
274
DALTON, CAIN, AND SCARROW
but which may not be taught by special interest groups that motivate members with the purity of their rhetoric. There is also a need for other political actors to adapt to these changes, and draw out the best of these reforms and hopefully mitigate their potential limitations. The democratization process has begun, but it is not finished. A new model of democracy is evolving. The contemporary democratic process requires more of its citizens. It also challenges politicians and bureaucrats to figure out what it means to move past a trustee model of politics without abdicating political leadership. But the result may be a further democratization of advanced industrial democracies, and the betterment of society and politics that this may produce.
NOTES 1
During the inter-war period and especially after the onset of the Great Depression, several European democracies faced severe challenges from the public or elites and eventually collapsed (Linz and Stepan 1978; Laski 1931). Perhaps this makes the post-war era of the 1950s and early 1960s appear even more halcyon. 2 Although not tailored to our specific interests, the 2000 Social Capital Survey 〈http://www.ropercenter.uconn. edu/scc_bench.html〉 demonstrates that participation in direct forms of action—working on a community project, signing petitions, participating in political meetings and neighbourhood associations, and protests—are a common part of the political repertoire of Americans: Activity % active Voted in 1996 presidential election 69 Worked on community project in last year 38 Signed a petition in last year 35 Participated in parent association or other school group 22 Participated in neighborhood association 20 Attended political meeting or rally in last year 16 Participated in a demonstration, boycott or march in last 7 year 3 4 5
Perhaps the notable exception in Dahl's research is his attention to a need for citizen participation in the economic domain (Dahl 1985). Other democratic theorists have a more inclusive definition of democracy that embraces the forms of direct democracy and advocacy democracy that we have presented, but often these discussions lack the focus and rigour of Dahl's theorizing on democracy. See, for example, Held (1996); Fuchs (1996). Debates continue in some countries about which residents should be eligible for citizenship, and how high the barriers for citizenship should be. One such case is Germany,
DEMOCRATIC PUBLICS AND DEMOCRATIC INSTITUTIONS
6 7 8
275
which has a large population of permanent resident aliens who fully participate in German economic and social life, but who are excluded from key aspects of civic life, such as electoral participation. Here the debates have centred on how difficult the naturalization process should be for long-term residents, though some have proposed constitutional changes to permit non-citizen residents to participate in local-level elections. Similar discussions have occurred in other EU member states. Another exception would be citizen juries. This form of participation would be problematic on inclusion and equality criteria, but may do better on understanding, agenda control, and the effectiveness of participation. We appreciate David Meyer's suggestions on the points in this paragraph. Also see Theodore Lowi (1969). This should not be a surprising conclusion, but it is surprising that much of the literature on strong democracy and deliberative democracy stresses only the positive features of these new forms, overlooking the limitations.
References Aarts, Wilma, Hofman, Maaike, and Smink, Carla (2001). Assessing Debate and Participative Technology Assessment ADAPTA Project Final Report. Département d'Economie et de Sociologie Rurales Unité de Recherche de Grenoble. January 〈http://www.inra.fr/Internet/Directions/SED/science-gouvernance/pub/ADAPTA/〉. Aberbach, Joel, Putnam, Robert, and Rockman, Bert (1981). Bureaucrats and Politicians in Western Democracies. Cambridge, MA: Harvard University Press. Abromeit, Heidrun (1998). Democracy in Europe: Legitimising Politics in a Non-state Polity. New York: Berghan Books. Abu-Duhou, Ibtisam (1999). School-Based Management. Paris: UNESCO. Adams, David and Hess, Michael (2001). ‘Community in Public Policy: Fad or Foundation?‘ Australian Journal of Public Administration, 60: 13–23. Agnew, John (1990). ‘Political Decentralization and Urban Policy in Italy: From “State-Centered” to “State-Society” Explanation’. Policy Studies Journal, 18: 768–85. Agranoff, Robert, and Gallarin, Juan Antonio Ramos (1997). ‘Towards Federal Democracy in Spain: An Examination of Intergovernmental Relations’. Publius, 31: 1–39. Ainley, John and McKenzie, Phillip (2000). ‘School Governance: Research on Educational and Management Issues’. International Education Journal, 1/3 〈http://iej.cjb.net/〉. Alderson, Stanley (1975). Yea or Nay? Referenda in the United Kingdom. London: Cassell. Alexander, Herbert E. (1989). ‘Money and Politics: Rethinking a Conceptual Framework’, in Herbert E. Alexander (ed), Comparative Political Finance in the 1980s. Cambridge: Cambridge University Press. Almond, Gabriel and Verba, Sidney (1963). The Civic Culture. Princeton: Princeton University Press. Altbach, Philip and Laufer, Robert (eds) (1972). The New Pilgrims: Youth Protest in Transition. New York: McKay. Amelsvoort, H. W. C. H. van (1995). ‘Decentralization in Education in an International Perspective’. Study commissioned by the Dutch Ministry of Education. Enschede: University of Twente. Andersen, Ida-Elisabeth and Jaeger, Birgit (1999). ‘Scenario Workshops and Consensus Conferences: Towards More Democratic Decision-making’. Science and Public Policy, 26: 331–40. Ansell, Christopher and Fish, M. Steven (1999). ‘The Art of Being Indispensable: Noncharismatic Personalism in Contemporary Political Parties’. Comparative Political Studies, 32: 283–312. Applegate, John (1998). ‘Beyond the Usual Suspects: The Use of Citizen Advisory Boards in Environmental Decisionmaking’. Indiana Law Journal, 73: 903–57. Appleton, Andrew (2001). ‘France: Party-Group Relations in the Shadow of the State’, in Clive S. Thomas (ed.), Political Parties and Interest Groups: Shaping Democratic Governance. Boulder, CO, London: Lynne Rienner Publisher. Archibald, S. (1993). ‘The Early Years of the Freedom of Information Act: 1955 to 1974’. PS: Political Science & Politics, 26: 726–31.
REFERENCES
277
Arnim, Hans Herbert von (1993). ‘Campaign and Party Finance in Germany’, in Arthur B. Gunlicks (ed.), Campaign and Party Finance in North America and Western Europe. Boulder, CO: Westview. —— (2002). ‘Die politische Durchsetzung der Kommunalverfassungsreform der neunziger Jahre’. Die öffentliche Verwaltung, 55: 585–92. Arnold, R. Douglas (1987). ‘Political Control of Administrative Officials’. Journal of Law, Economics and Organization, 3: 279–86. Arterton, Christopher (1987). Teledemocracy: Can Technology Protect Democracy? Newbury Park, CA: Sage Publications. Assembly of European Regions (1998). Declaration on Regionalism in Europe. Strasbourg: Assembly of European Regions. Aubert, Jean-François (1978). ‘Switzerland’, in David Butler and Austin Ranney (eds), Referendums. Washington, DC: American Enterprise Institute. Australia. Commonwealth Ombudsman (2001). Annual Report 2000–2001. Canberra: Commonwealth Ombudsman's Office. Australian Broadcasting Corporation (1999). Fact Sheet: Allocation of Free Broadcasting Time to Political Parties During Election Campaigns 〈http://www.abc.net.au/〉. Avril, Pierre (1994). ‘Regulation of Political Finance in France’, in Herbert E. Alexander and Rei Shiratori (eds), Comparative Political Finance among the Democracies. Boulder, CO: Westview. Bahrenberg, Gerhard (1989). ‘West Germany: From Decentralization in Theory to Centralization in Practice’, in Robert Bennett (ed.), Territory and Administration in Europe. London: Pinter Publishers. Bailey, Peter (1995). ‘“Righting” the Constitution without a Bill of Rights’. Federal Law Review, 23: 1–36. Bang, Henrik (2001). Governance, Governmentality and Democracy. Manchester: University of Manchester Press. Banisar, David (2002a). ‘Freedom of Information Around the World, Update 2002’ 〈http://www.privacyinternational.org/issues/foia/intl-foia-update-202.pdf〉 accessed 14 August 2002. —— (2002b). ‘Freedom of Information and Access to Government Information Around the World’ 〈http://www. freedominfo.org/survey/survey.pdf〉 accessed 14 August 2002. Barber, Benjamin (1984). Strong Democracy. Berkeley: University of California Press. Barnes, Samuel, Kaase, Max et al. (1979). Political Action. Beverly Hills, CA: Sage Publications. Barzelay, Michael (1992). Breaking Through Bureaucracy: A New Vision for Managing in Government. Berkeley: University of California Press. Batley, Richard and Stoker, Gerry (eds) (1991). Local Government in Europe. Houndmills: Macmillan Educational. Battini, Stefano (1999). ‘The Safeguarding of Public Service Users in Italy and the “Carta dei Servizi Pubblici”‘, in Luc Rouban (ed.), Citizens and the New Governance. Amsterdam: IOS Press. Beck, Lynn and Murphy, Joseph (1999). ‘Parental Involvement in Site-Based Management: Lessons from One Site.’ International Journal of Leadership in Education, 2: 82–102. Beck, Ulrich (1997). The Reinvention of Politics: Rethinking Modernity in the Global Social Order (trans. M. Ritter). Cambridge: Polity Press. Beetham, David (ed.) (2001). International IDEA Handbook on Democracy Assessment. The Hague: Kluwer Law International.
278
REFERENCES
Beierle, Thomas and Konisky, David (2001). ‘What Are We Gaining from Stakeholder Involvement? Observations from Environmental Planning in the Great Lakes’. Environment and Planning C: Government and Policy, 19: 515–27. —— and Long, Rebecca (1999). ‘Chilling Collaboration: The Federal Advisory Committee Act and Stakeholder Involvement in Environmental Decisionmaking’. Environmental Law Reporter, 29: 10399. Belloc, Hilaire and Chesterton, Cecil (1911). The Party System. London: Stephen Swift. Bennett, Colin J. (1997). ‘Understanding Ripple Effects: The Cross-National Adoption of Policy Instruments of Bureaucratic Accountability’. Governance, 10: 213–33. Bentley, Michael (1987). Politics Without Democracy. Oxford: Basil Blackwell. Benz, Arthur (1987). ‘Decentralization in the Federal Republic of Germany—A Case of Pragmatic Adaptation’, International Review of Administrative Sciences, 53: 467–81. Berry, Jeffrey (1999). The New Liberalism: The Rising Power of Citizen Groups. Washington, DC: Brookings Institution Press. ——, Portney, Kent, and Thomson, Ken (1993). The Rebirth of Urban Democracy. Washington, DC: Brookings Institution Press. Betlem, Gerrit (1994). ‘Environmental Locus Standi in the Netherlands’. Review of European Community and International Environmental Law, 4: 238. Bickel, Alexander (1962). The Least Dangerous Branch. New York: Bobbs and Merrill. Bille, Lars (1994). ‘Denmark: The Decline of the Membership Party’, in Richard S. Katz and Peter Mair (eds), How Parties Organize. London: Sage. —— (1996). ‘Denmark’. European Journal of Political Research Political Data Yearbook, 30: 315–19. —— (1997). Partier I Forandring. Odense: Odense Universitetsforlag. —— (2001). ‘Democratizing a Democratic Procedure: Myth or Reality?‘ Party Politics, 7: 363–80. Black, Charles (1960). The People and the Court. Englewood Cliffs: Prentice Hall. Blais, André (2000). To Vote or Not to Vote: The Merits and Limits of Rational Choice Theory. Pittsburgh: University of Pittsburgh Press. Blatner, Keith, Carroll, Matthew, Daniels, Steven, and Walker, Gregg (2001). ‘Evaluating the Application of Collaborative Learning to the Wenatchee Fire Recovery Planning Effort’. Environmental Impact Assessment Review, 21: 241–70. Blom-Hansen, Jens (1999). ‘Policy Making in Central-Local Government Relations: Balancing Local Autonomy, Macroeconomic Policy, and Sectoral Policy Goals’. Journal of Public Policy, 19: 237–64. Blondel, Jean, Sinnott, Richard, and Svensson, Palle (1998). People and Parliament in the European Union: Participation, Democracy, and Legitimacy. New York: Oxford University Press. Bobbio, Norberto (1987). The Future of Democracy (trans. Roger Griffen). Minneapolis: University of Minnesota Press. Bobbitt, Philip (1982). Constitutional Fate. New York: Oxford University Press. Bogart, W. A. (1994). Courts and Country: The Limits of Litigation and the Social and Political Life of Canada. Toronto, New York, and London: Oxford University Press. Bogdanor, Vernon (1981). The People and the Party System. Cambridge: Cambridge University Press. —— (1994). ‘Western Europe’, in David Butler and Austin Ranney (eds), Referendums Around the World. Washington, DC: The American Enterprise Institute.
REFERENCES
279
Borre, Ole (1988). ‘The Danish General Election of 1987’. Electoral Studies, 7: 75–8. Bottari, Norberto (2000). ‘Autonomy and Decentralisation: Between Hopes and Illusions. A Comparative Study of Reforms in Five European Countries’. Paper presented at AERA annual meeting, New Orleans. ERIC Document Reproduction Service. Bowler, Shaun and Donovan, Todd (eds) (2001). Demanding Choices: Opinion, Voting and Direct Democracy. Ann Arbor: University of Michigan Press. —— and —— (2002). ‘Democracy, Institutions and Attitudes about Citizen Influence on Government’. British Journal of Political Science, 32: 371–90. ——, —— and Caroline Tolbert (eds) (1998). Citizens as Legislators: Direct Democracy in the United States. Columbus: Ohio State University Press. Boyer, Patrick (1992). Direct Democracy in Canada. Toronto: Dundurn. Brademas, James D. (1993). Citizen Advisory Committees: A Research Study of Eight Midwestern States. Illinois Periodicals Online 〈http://www.lib.niu.edu/ipo/ip931223.html〉. Brennan, Sir Frank (1991). ‘Courts, Democracy and the Law’. Australian Law Journal 65/32: 34–41. Broder, David (2000). Democracy Derailed: Initiative Campaigns and the Power of Money. New York: Harcourt. Bruce, Andrew (1924). The American Judge. New York: Macmillan. Budd, L. (1997). ‘Regional Government and Performance in France’. Regional Studies, 31: 187–92. Budge, Ian (1996). The New Challenge of Direct Democracy. Cambridge: Polity Press. ——, Robertson, David, and Hearl, Derek (1987). Ideology, Strategy, and Party Change: Spatial Analysis of Postwar Election Programmes in 19 Democracies. Cambridge: Cambridge University Press. Budge, Ian et al. (2001). Mapping Policy Preferences. Estimates for Parties, Electors and Governments, 1945–1998. Oxford: Oxford University Press. Bueso, Laura Diez (2000). ‘Spain's Parliamentary Ombudsman Scheme’, in Roy Gregory and Philip Giddings (eds), Righting Wrongs: The Ombudsman in Six Continents. Amsterdam: IOS Press. Butler, David and Ranney, Austin (eds) (1994a). Referendums around the World: The Growing Use of Democracy? Washington, DC: American Enterprise Institute. ——, —— (1994b). ‘Practice’, in Butler, David and Ranney, Austin (eds), Referendums Around the World. Washington, DC: American Enterprise Institute. CADA (Commission d'accès aux Documents Administratifs) (2001). Rapport d'activité 1999–2000. Paris: La Documentation Française. Caiden, George (1983). International Handbook of the Ombudsman. Westport, CT: Cambridge University Press. Cain, Bruce, Ferejohn, John and Fiorina, Morris (1987). The Personal Vote: Constituency Service and Electoral Independence. Cambridge: Harvard University Press. —— and Gerber, Elisabeth (eds) (2000). Voting at the Political Fault Line: California's Experiment with the Blanket Primary. Berkeley: University of California Press. Carey, John, Niemi, Richard, and Powell, Lynda (2000). Term Limits in the State Legislatures. Ann Arbor: University of Michigan Press. Carlson, Ann E. (1998). ‘Standing for the Environment’. UCLA Law Review, 45: 931–1004. Carre de Malberg, R. (1922). Contribution à la théorie générale de l’état. Paris: Sirey.
280
REFERENCES
Carrillo, Ernesto (1997). ‘Local Government and Strategies for Decentralization in the “State of Autonomies”‘. Publius, 27: 39–64. Carter, Elisabeth L. (2002). ‘Proportional Representation and the Fortunes of Right-Wing Extremist Parties’. West European Politics, 25: 127–48. Carty, R. K. and Blake, Donald E. (1999). ‘The Adoption of Membership Votes for Choosing Party Leaders: The Experience of Canadian Parties’. Party Politics, 5: 211–24. Castles, Francis (1999). ‘Decentralization and the Post-war Political Economy’. European Journal of Political Research, 36: 27–53. Caul, Miki L. and Gray, Mark M. (2000). ‘From Platform Declarations to Policy Outcomes: Changing Party Profiles and Partisan Influence Over Policy’, in Russell J. Dalton and Martin P. Wattenberg (eds), Parties Without Partisans. New York: Oxford University Press. Centre d’études et de documentation sur la démocratie directe (2002). 〈http://c2d.unige.ch/c2d/direct_demo_federal/table_resume.html〉. Chandler, David (2000). ‘Active Citizens and the Therapeutic State: The Role of Democratic Participation in Local Government Reform’. Policy and Politics, 29: 3–14. Chandler, J. A. (ed.) (1993). Local Government in Liberal Democracies: An Introductory Survey. New York: Routledge. Christensen, Jorgen (2000). ‘The Dynamics of Decentralization and Recentralization’. Public Administration, 78: 398–408. Christensen, Tom and Lӕgreid, Per (1999). ‘New Public Management—Design, Resistance, or Transformation? A Study of How Modern Reforms are Received in a Civil Service System’. Public Productivity and Management Review, 23: 169–94. Christensen, Tom, Lӕgreid, Per and Wise, Lois R. (2002). ‘Transforming Administrative Policy’. Public Administration, 80: 153–78. Christin, Thomas, Hug, Simon, and Sciarini, Pascal (2002). ‘Interests and Information in Referendum Voting: An Analysis of Swiss Voters’. European Journal of Political Research, 37: 759–76. Ciaurro, Gian Franco (1989). ‘Public Financing of Parties in Italy’, in Herbert E. Alexander (ed.), Comparative Political Finance in the 1980s. Cambridge: Cambridge University Press. Cichowski, Rachel A. (1998). ‘Integrating the Environment: The European Court and the Construction of Supranational Policy’. Journal of European Public Policy, 5: 387–405. —— (2001). ‘Judicial Rulemaking and the Institutionalization of EU Sex Equality Policy’, in A. Stone Sweet, W. Sandholtz, and N. Fligstein (eds), The Institutionalization of Europe. Oxford: Oxford University Press. —— (2002). ‘Litigation, Mobilization and Governance: The European Court and Transnational Activists’. Ph.D. dissertation. Irvine: University of California. Clark, David (2000). ‘Citizens, Charters and Public Sector Reform in France and Britain’. Government and Opposition, 35: 152–69. Clarke, Michael and Stewart, John (1993). ‘From Traditional Management to the New Management in British Local Government’. Policy Studies Journal, 21: 82–94. Cohen, Jean and Arato, Andrew (1992). Civil Society and Political Theory. Cambridge, MA: MIT Press.
REFERENCES
281
Cole, Alistair (2001). ‘The New Governance of French Education?‘ Public Administration, 79: 707–24. Cole, Richard, Hissong, Rodney, and Arvidson, Enid (1999). ‘Devolution: Where's the Revolution’. Publius, 29: 99–112. Colomer, Josep M. (1998). ‘The Spanish “State of Autonomies”: Non-institutional Federalism’. West European Politics, 21: 40–52. Commission of the European Union (2001). White Paper on Good Governance. Brussels: Commission of the European Union. —— (n.d.). Secretariat General's Report on Openness and Access to Documents, 1994–1996 〈http://europa.eu.int/comm/ secretariat_general/sgc/acc_doc/index_en.htm#〉. Converse, Philip (1964). ‘The Nature of Belief Systems in Mass Publics’, in D. Apter (ed.), Ideology and Discontent. New York: Free Press. Coote, Anna and Mattinson, Deborah (1997). Twelve Good Neighbours: The Citizen as Juror. London: Fabian Society. Corkill, David (1999). ‘Portugal's 1998 Referendums’. West European Politics, 22:186–93. Corriveau, Yves (1995). ‘Citizen Rights and Litigation in Environmental Law NGOs as Litigants: Past Experiences and Litigation in Canada’, in S. Deimann and B. Dyssli (eds), Environmental Rights: Law, Litigation and Access to Justice. London: Cameron May. Council of Europe (1993). Local Referendums. Strasbourg: Council of Europe Press. —— (1996). Yearbook of the European Convention on Human Rights, the European Commission and European Court of Human Rights. Strasbourg: Council of Europe Press. —— (1997). Yearbook of the European Convention on Human Rights, the European Commission and European Court of Human Rights. Strasbourg: Council of Europe Press. —— (1998). Regionalisation and Its Effects on Local Self-Government. Strasbourg: Council of Europe Publishing. Council of State Governments (1971). The Book of the States: 1970–1971. Lexington, KY: Council of State Governments. —— (2001). The Book of the States 2000–2001. Lexington, KY: Council of State Governments. Courtney, John C. (1978). ‘Recognition of Canadian Political Parties in Parliament and in Law’. Canadian Journal of Political Science, 11: 33–60. Cousins, Margaret (1999). ‘Democratisation by Stealth: Changing Processes in New Zealand Local Government 1989–1997’. Political Science, 50/2: 225–36. Cox, G. (1997). Making Votes Count. Cambridge: Cambridge University Press. Cree, Nathan (1892). Direct Legislation by the People. Chicago: A. C. McClurg & Company. Crewe, Ivor (1981). ‘Electoral Participation’, in David Butler, H. R. Penniman, and A. Ranney (eds), Democracy at the Polls. Washington, DC: American Enterprise Institute. —— and Denver, D.T. (eds) (1985). Electoral Change in Western Democracies. New York: St Martin's Press. Croly, Herbert (1914). Progressive Democracy. New York: Macmillan. Cronin, Thomas (1984). Direct Democracy: The Politics of Initiative, Referendum and Recall. Cambridge, MA: Harvard University Press.
282
REFERENCES
Crosby, Ned (1999). ‘Using the Citizens Jury Process for Environmental Decision Making’, in Ken Sexton et al. (eds), Better Environmental Decisions. Washington, DC: Island Press. Crozier, Michael, Huntington, Samuel, and Watanuki, Joji (1975). The Crisis of Democracy. New York: NYU Press. Curtis, Gerald L. (1988). The Japanese Way of Politics. New York: Columbia University Press. Dahl, Robert (1956). A Preface to Democratic Theory. Chicago: University of Chicago Press. —— (1971). Polyarchy. New Haven: Yale University Press. —— (1982). Dilemmas of Pluralist Democracy: Autonomy vs. Control. New Haven: Yale University Press. —— (1985). A Preface to Economic Democracy. Berkeley: University of California Press. —— (1989). Democracy and Its Critics. New Haven: Yale University Press. —— (1998). On Democracy. New Haven: Yale University Press. Dahrendorf, Ralf (2000). ‘Afterword’, in Susan Pharr and Robert Putnam (eds), Disaffected Democracies. Princeton, NJ: Princeton University Press. Dalton, Russell (1984). ‘Cognitive Mobilization and Partisan Dealignment in Advanced Industrial Democracies’. Journal of Politics, 46: 264–84. Dalton, Russell (1996). Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies (2nd edn). New York: Chatham House Publishers. —— (1999). ‘Political Support in Advanced Industrial Democracies’, in Pippa Norris (ed.), Critical Citizens. Oxford: Oxford University Press. —— (2000). ‘Value Change and Democracy’, in Susan J. Pharr and Robert D. Putnam (eds), Disaffected Democracies. Princeton: Princeton University Press. —— (2002). Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies (3rd edn). New York: Chatham House Publishers. —— (2003). Democratic Challenges, Democratic Choices: The Decline in Political Support in Advanced Industrial Democracies. Oxford: Oxford University Press. ——, Buerklin, Wilhelm, and Drummond, Andrew (2001). ‘Public Attitudes toward Direct Democracy’. Journal of Democracy, 12:141–53. ——, Flanagan, Scott C., and Beck, Paul Allen (eds) (1985). Electoral Change in Advanced Industrial Democracies. Princeton: Princeton University Press. ——, Flanagan, Scott C. and Kuechler, Manfred (eds) (1990). Challenging the Political Order: New Social and Political Movements. New York: Oxford University Press. ——, Flanagan, Scott C. and Wattenberg, Martin (eds) (2000). Parties without Partisans: Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press. Davis, Kenneth Culp and Richard J. Pierce (1994). Administrative Law Treatise. Boston: Little Brown and Company. Day, Patrick (2000). A History of Broadcasting in New Zealand, ii: Voice and Vision. Auckland: Auckland University Press. De Vries, Michiel S. (2000). ‘The Bureaucratization of Participation’. International Review of Administrative Sciences, 66: 325–48. Delbrück, Hans (1914). Regierung und Volkswille. Berlin: Georg Stilke Verlag. Della Porta, Donatella and Vanucci, Alberto (1999). Corrupt Exchanges: Actors, Resources, and Mechanisms of Political Corruption. New York: Walter de Gruyter. Delli Carpini, Michael and Keeter, Scott (1996). What Americans Know about Politics and Why it Matters. New Haven, CT: Yale University Press. Denmark. Folketingets Ombudsmand (2001). Summary Annual Report 2000. Copenhagen.
REFERENCES
283
Deploige, Simon (1898). The Referendum in Switzerland (trans. C. P. Trevelyan). London: Longmans, Green, and Company. Deschouwer, Kris (1994). ‘The Decline of Consociationalism and the Reluctant Modernization of Belgian Mass Parties’, in Richard S. Katz and Peter Mair (eds), How Parties Organize. London: Sage. Devlin, Nancy, Maynard, Alan, and Mays, Nicholas (1999). ‘New Zealand's New Health Sector Reforms: Back to the Future?‘. British Medical Journal, 322: 1171–4. Diamond, Larry, Linz Juan, and Lipset, Seymour Martin (eds) (1988). Democracy in Developing Countries. Boulder, CO: L. Rienner. Dicey, Albert (1894). ‘The Referendum’. The National Review, March: 65–72. Dienel, Peter C. and Renn, Ortwin (1995). ‘Planning Cells: A Gate to ‘Fractal’ Mediation’, in Ortwin Renn, Thomas Webler, and Peter Wiedemann (eds), Fairness and Competence in Citizen Participation: Evaluating Models for Environmental Discourse. Boston: Kluwer Academic. Doyle, Michael (2001). ‘The Freedom of Information Act in Theory and Practice’. Unpublished master's thesis. Baltimore: Johns Hopkins University. Drysch, Thomas (1993). ‘The New French System of Political Finance’, in Arthur B. Gunlicks (ed.), Campaign and Party Finance in North America and Western Europe. Boulder, CO: Westview. —— (1990). Discursive Democracy: Politics, Policy and Political Science. Cambridge: Cambridge University Press. Dryzek, John (1996). Democracy in Capitalist Times. New York: Oxford University Press. Duverger, Maurice (1954). Political Parties. New York: John Wiley. Dyson, Kenneth (1980). The State Tradition in Western Europe. Oxford: Martin Robertson. van der Eijk, Cees, Franklin, Mark, and Marsh, Michael (1994). ‘Referendum Outcomes and Trust in Government: Popular Support for Europe in the Wake of Maastricht’. West European Politics, 18: 101–17. Eldersveld, Samuel (1964). Political Parties: A Behavioral Analysis. Chicago: Rand-McNally. Elklit, Jorgen and Reynolds, Andrew (2000). The Impact of Election Administration on the Legitimacy of Emerging Democracies: A New Research Agenda (Kellogg Institute Working Papers in International Studies No. 281). South Bend, IN: University of Notre Dame. Elster, John (1998). Deliberative Democracy. New York: Cambridge University Press. Epp, Charles R. (1998). The Rights Revolution: Lawyers, Activists, and Supreme Courts in Comparative Perspective. Chicago: University of Chicago Press. Epstein, Leon (1967). ‘Political Parties in Western Democratic Systems’, in Roy Macrieudis (ed.), Political Parties: Contemporary Trends and Ideas. New York: Harper and Row. —— (1980). Political Parties in Western Democracies. New Brunswick, NJ: Transaction Books. Ewing, Keith (1987). The Funding of Political Parties in Britain. Cambridge: Cambridge University Press. Farrell, David (1994). ‘Ireland: Centralization, Professionalization and Competitive Pressures’, in Richard S. Katz and Peter Mair (eds), How Parties Organize. London: Sage. —— (1996). ‘Campaign Strategies and Tactics’, in Lawrence LeDuc, Richard Niemi, and Pippa Norris (eds), Comparing Democracies. Thousand Oaks, CA: Sage. —— (2001). Electoral Systems: A Comparative Introduction. London and New York: Palgrave.
284
REFERENCES
Farrell, David (2002). ‘Campaign Modernization and the West European Party’, in K. R. Luther and F. Müller-Rommel (eds), Political Parties in the New Europe. Oxford: Oxford University Press. —— and Schmitt-Beck, Rüdiger (eds) (2002). Do Political Campaigns Matter? London and New York: Routledge. —— and Webb, Paul (2000). ‘Political Parties as Campaign Organizations’, in Russell Dalton and Martin Wattenberg (eds), Parties Without Partisans. Oxford: Oxford University Press. Feldman, Eric A. (2000). The Ritual of Rights in Japan: Law, Society and Health Policy. Cambridge: Cambridge University Press. Ferdinand, Peter (1998). ‘Building Democracy on the Basis of Capitalism: Towards an East Asian Model of Party Funding’, in Peter Burnell and Alan Ware (eds), Funding Democratization. Manchester: Manchester University Press. Finland. Parliamentary Ombudsman of Finland (2001). Annual Report 2000. Helsinki. Fishkin, James S. (1991). Democracy and Deliberation: New Directions for Democratic Reform. New Haven, CT: Yale University Press. —— (1995). The Voice of the People: Public Opinion and Democracy. New Haven, CT: Yale University Press. Fligstein, Neil, and Stone Sweet, Alec (2002). ‘Of Polities and Markets: An Institutionalist Account of European Integration’. American Journal of Sociology, 107: 1206–43. Font, Javier (2000). ‘Managing Value and Devolution: The Experience of Regional Government in Spain’. International Review of the Administrative Sciences, 66: 451–62. Franklin, Aimee (2001). ‘Serving the Public Interest: Federal Experiences with Participation in Strategic Planning’. American Review of Public Administration, 31: 126–38. Franklin, Mark (2003). Voter Turnout and the Dynamics of Electoral Competition. New York: Cambridge University Press. ——, Mackie, Tom, Valen, Henry et al. (1992). Electoral Change: Responses to Evolving Social and Attitudinal Structures in Western Countries. New York: Cambridge University Press. Freedom House (2000). Democracy's Century: A Survey of Global Political Change in the 20th Century. New York: Freedom House. Fuchs, Dieter (1996). ‘Wohin geht der Wandel der demokratischen Institutionen in Deutschland?‘ Research paper of the Wissenschaftszentrum Berlin. Führ, Martin and Roller, Gerhard (1991). ‘Appendix’, in M. Führ and G. Roller (eds), Participation and Litigation Rights of Environmental Associations in Europe. Frankfurt: Peter Lang. ——, Gebers, Betty, Ormond, Thomas, and Roller, Gerhard (1995). ‘Access to Justice: Legal Standing for Environmental Associations in the European Union’, in D. Robinson and J. Dunkely (eds), Public Interest Perspectives in Environmental Law. London: Chancery Law Publishing Inc. Fung, Archon (2001). ‘Accountable Autonomy: Toward Empowered Deliberation in Chicago Schools and Policing’. Politics & Society, 29: 73–103. —— (2003). Designing Democracy. Princeton: Princeton University Press. —— and Wright, Erik Olin (2001). ‘Deepening Democracy: Innovations in Empowered Participatory Governance’. Politics & Society, 29: 5–41. Gallagher, Michael (1996). ‘Conclusion’, in Michael Gallagher and Pier Vicenzo Uleri (eds), The Referendum Experience in Europe. Basingstoke: Macmillan.
REFERENCES
285
——, Laver, Michael, and Mair, Peter (1995). Representative Government in Modern Europe (2nd edn). New York: McGraw-Hill. —— and Marsh, Michael (eds) (1988). Candidate Selection in Comparative Perspective: The Secret Garden of Politics. London: Sage. —— and Uleri, Pier Vincenzo (eds) (1996). The Referendum Experience in Europe. Basingstoke: Macmillan. Galligan, Brian (1987). Politics of the High Court: A Study of the Judicial Branch of Government in Australia. Brisbane: University of Queensland Press. Gamaleri, Gianpiero (1995). ‘Italy and the 1994 Elections: Media, Politics and the Concentration of Power’, in Yasha Lange and Andrew Palmer (eds), Media and Elections: A Handbook. Düsseldorf: European Institute for the Media. Geißel, Brigitte and Angela, Oels (2000). ‘Exploring the Diffusion of Local Agenda 21 Initiatives in Germany and in United Kingdom’. Paper presented at the International Workshop Diffusion of Environmental Policy Innovations, Berlin 〈http://www.fu-berlin.de/ffu/akumwelt/abstractsdiff.htm〉. Gentile, Pierre and Kriesi, Hanspeter (1998). ‘Contemporary Radical-Right Parties in Switzerland: History of a Divided Family’, in Hans-Georg Betz and Stefan Immerfall (eds), The New Politics of the Right. New York: St Martin's Press. Geras, Norman (2002). Men of Waugh. Manchester: Geras Publishing. Gerber, Elisabeth (1999). The Populist Paradox: Interest Group Influence and the Promise of Direct Legislation. Princeton, NJ: Princeton University Press. Giddens, Anthony (1991). Modernity and Self-Identity: Self and Society in the Late Modern Age. Stanford: Stanford University Press. Gidlund, Gullan M. (1991). ‘Public Investment in Swedish Democracy: Gambling with Gains and Losses’, in Matti Wiberg (ed.), The Public Purse and Political Parties. Jyväskylä: Gummerus. —— (1994). ‘Regulation of Party Finance in Sweden’, in Herbert E. Alexander and Rei Shiratori (eds), Comparative Political Finance Among the Democracies. Boulder, CO: Westview. Goldsmith, Mike (1992). ‘Local Government’. Urban Studies, 29: 393–410. Gomien, D. (1995). Judgments of the European Court of Human Rights, Reference Chart (Human Rights Information Centre). Strasbourg: Council of Europe Press. Goncalves, Maria (2002). ‘Implementation of EIA Directives in Portugal: How Changes in Civic Culture are Changing Political and Administrative Practice’. Environmental Impact Assessment Review, 22: 249–69. Goodman, Carl F. (2001). ‘The Somewhat Less Reluctant Litigant: Japan's Changing View Towards Civil Litigation’. Law and Policy in International Business, 32: 769–811. Goodnow, Frank (1900). Politics and Administration. New York: Macmillan. Goodwyn, Lawrence (1976). Democratic Promise: The Populist Movement in America. New York: Oxford University Press. Gordon, Liz and Whitty, Geoff (1997). ‘Giving the Hidden Hand a Helping Hand: The Rhetoric and Reality of Neoliberal Education Reform in England and New Zealand’. Comparative Education, 33: 453–67. Gore, Al (1993). The Gore Report on Reinventing Government. Washington, DC: GPO. Gray, Mark (2003). ‘In the Midst of Fellows: The Social Context of the American Turnout Decision’. Ph.D. thesis. Irvine: University of California. —— and Caul, Miki (2000). ‘Declining Voter Turnout in Advanced Industrial Democracies: 1950–1997’. Comparative Political Studies, 33: 1091–122.
286
REFERENCES
Gregory, Roy and Giddings, Philip (eds) (2000). Righting Wrongs: The Ombudsman in Six Continents. Amsterdam: IOS Press. Greve, Carsten and Jespersen, Peter Kragh (1999). ‘New Public Management and its Critics: Alternative Roads to Flexible Service Delivery to Citizens?‘, in Luc Rouban (ed.), Citizens and the New Governance. Amsterdam: IOS Press. Greve, Michael S. (1989). ‘The Non-Reformation of Administrative Law: Standing to Sue and Public Interest Litigation in West German Environmental Law’. Cornell International Law Journal, 22: 197–244. Grofman, Bernard and Lijphart, Arend (eds) (1986). Electoral Laws and their Political Consequences. New York: Agathon. Gunlicks, Arthur (ed.) (1981). Local Government Reform and Reorganization: An International Perspective. Port Washington, NY: Kennikat. —— (1988). ‘Campaign and Party Finance in the West German “Party State”’. Review of Politics, 50: 30–48. Habermas, Jürgen (1987). The Theory of Communicative Action, ii (trans. Thomas McCarthy). Boston: Beacon Press. —— (1990). Moral Consciousness and Communicative Action (trans. Christian Lenhardt and Sherry Weber Nicholsen). Cambridge, MA: MIT Press. —— (1992a). Faktizität und Geltung. Beiträge zur Diskurstheorie des Rechts und des demokratischen Rechtsstaates. Frankfurt: Suhrkamp. —— (1992b). ‘Drei Normative Modelle der Demokratie: Zum Begriff Deliberativer Politik’, in Herfried Münkler (ed.), Die Chancen der Freiheit. Munich/Zürich: Piper. —— (1996). Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (trans. William Rehg). Cambridge, MA: MIT Press. —— (2001). The Postnational Constellation: Political Essays (trans. Max Pensky). Cambridge, MA: Polity Press. Hagen, Martin and Kubicek, Herbert (eds) (2000). One-Stop Government in Europe: Results from 11 National Surveys. Bremen: University of Bremen Press. Hager, Carol (1995). Technological Democracy: Bureaucracy and Citizenry in the German Energy Debate. Ann Arbor: University of Michigan Press. Halligan, John (1997). ‘New Public Sector Models: Reform in Australia and New Zealand’, in Jan-Erik Lane (ed.), Public Sector Reform. London: Sage. Hand, Geoffrey, Georgel, Jacques, and Sasse, Christoph (1979). European Electoral Systems Handbook. London: Butterworths. Handley, Robin (1986). Public order, petitioning and freedom of assembly. Journal of Legal History 7: 123–55. Handley, Matt (2002). ‘Why Crocodiles, Elephants, and American Citizens Should Prefer Foreign Courts: A Comparative Analysis of Standing to Sue’. The Review of Litigation, 21: 97–135. Hansen, Karin (2001). ‘Local Councillors: Between Local “Government” and Local “Governance” ’. Public Administration, 79: 105–23. Hanson, E. Mark (2000). ‘Democratization and Educational Decentralization in Spain: A Twenty Year Struggle for Reform’. Education Reform and Management Publication Series, 1: 1–62. Hart, H. L. A. (1994). The Concept of Law. Oxford: Clarendon. Hartz, Louis (1969). The Founding of New Societies. New York: Harcourt.
REFERENCES
287
Hay, David (2002). ‘Public Sector Decentralization, Accountability and Financial Reporting in New Zealand’. Journal of Public Budgeting, Accounting and Financial Management, 13/2: 133–56. Held, David (1987). Models of Democracy. Stanford: Stanford University Press. —— (1995). Democracy and Global Order. Cambridge: Polity Press. —— (1996). Models of Democracy (2nd edn). Stanford: Stanford University Press. Hendriks, Carolyn (2002). ‘Institutions of Deliberative Democratic Processes and Interest Groups: Roles, Tensions and Incentives’. Australian Journal of Public Administration, 61: 64–75. Hendriks, Frank and Tops, Pieter (1999). ‘Between Democracy and Efficiency: Trends in Local Government Reform in the Netherlands and Germany’. Public Administration, 77: 133–53. Hertogh, Marc (2000). ‘The National Ombudsman of the Netherlands’, in Roy Gregory and Philip Giddings (eds), Righting Wrongs: The Ombudsman in Six Continents. Amsterdam: IOS Press. Hibbing, John and Theiss-Morse, Elizabeth (2002). Stealth Democracy: Americans’ Beliefs about How Government Should Work. New York: Cambridge University Press. Hiebert, Janet (2002). Charter Conflicts. Montreal: McGill University Press. Higgins, Donald (1986). Local and Urban Politics in Canada. Toronto: Gage. Hine, David (1993). Governing Italy: The Politics of Bargained Pluralism. Oxford: Oxford University Press. Hirschl, Ran. (2000). ‘Looking Sideways, Looking Backwards: Judicial Review v. Democracy in Comparative Perspective’. University of Richmond Law Review, 43: 414–41. Hirschman, Albert (1970). Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge, MA: Harvard University Press. Holberg, Erik and Stjernquist, Nils (1988). Vår Författning (7th edn). Stockholm: Nordstedts. Holmstrom, Barry (1995). ‘Sweden’, in C. N. Tate and T. Vallinder (eds), The Global Expansion of Judicial Power. New York: NYU Press. Holtz-Bacha, Christina and Kaid, Lynda Lee (1995). ‘Television Spots in German National Elections: Content and Effects’, in Lynda Lee Kaid and Christina Holtz-Bacha (eds), Political Advertising in Western Democracies. Thousand Oaks, CA: Sage. Hopkin, Jonathan (2001). ‘Bringing the Members Back In: Democratizing Candidate Selection in Britain and Spain’. Party Politics, 7: 343–61. Horowitz, Gad (1968). Canadian Labour in Politics. Toronto: University of Toronto Press. Humes, Samuel and Eileen, Martin (1969). The Structure of Local Government. The Hague: International Union of Local Authorities. Hunold, Christian (2001). ‘Corporatism, Pluralism, and Democracy: Toward a Deliberative Theory of Bureaucratic Accountability’. Governance, 14/2: 151–67. Hunt, C. D. (1992). ‘The Changing Face of Environmental Law and Policy’. University of New Brunswick Law Journal, 42: 87. Huntington, Samuel (1981). American Politics: The Promise of Disharmony. Cambridge: Harvard University Press. —— (1993). The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press. Hutchcroft, Paul D. (2001). ‘Centralization and Decentralization in Administration and Politics: Assessing Territorial Dimensions of Authority and Power’. Governance: An International Journal of Policy and Administration, 14: 23–53.
288
REFERENCES
IDEA (Institute for Democracy and Electoral Assistance) (2002). Voter Turnout from 1945 to Date 〈http://www.idea. int/vt/index.cfm〉. Inglehart, Ronald (1990). Culture Shift. Princeton, NJ: Princeton University Press. —— (1997). Modernization and Postmodernization: Cultural, Economic and Political Change in 43 Societies. Princeton, NJ: Princeton University Press. —— (1999). ‘Postmodernism Erodes Respect for Authority, but Increases Support for Democracy’, in Pippa Norris (ed.), Critical Citizens. Oxford: Oxford University Press. Ingram, Helen and Smith, Steven (1993). Public Policy for Democracy. Washington, DC: Brookings Institution. Initiative and Referendum Institute (2002). Initiative Usage by Decade 〈http://www/iandrinstitute.org/〉. Jackman, Robert and Miller, Ross (1995). ‘Voter Turnout in the Industrial Democracies During the 1980s’. Comparative Political Studies, 27: 467–92. Jacobs, Kerry (1997). ‘The Decentralisation Debate and Accounting Controls in the New Zealand Public Sector’. Financial Accountability and Management, 13: 331–44. Johnston, Ron J. and Pattie, Charles J. (1993). ‘Great Britain: Twentieth Century Parties Operating under Nineteenth Century Regulations’, in Arthur B. Gunlicks (ed.), Campaign and Party Finance in North America and Western Europe. Boulder, CO: Westview. Kapsi, Maria (2000). ‘Recent Administrative Reforms in Greece: Attempts towards Decentralization, Democratic Consolidation and Efficiency’. Paper presented at the Socrates Kokkalis Graduate Student Workshop, JFK School of Government, 11–12 February. Karsten, Sjoerd (1999). ‘Neoliberal Education Reform in the Netherlands’. Comparative Education, 35: 305–17. Kate, Jan ten and Koppen, Peter van (1995). ‘The Netherlands: Toward a Form of Judicial Review’, in C. N. Tate and T. Vallinder (eds), The Global Expansion of Judicial Power. New York: NYU Press. Katz, Richard (1997). Democracy and Elections. New York: Oxford University Press. —— and Kolodny, Robin (1992). ‘The USA: The 1990 Congressional Campaign’, in David Farrell and Shaun Bowler (eds), Electoral Strategies and Political Marketing. London: Macmillan. —— and Mair, Peter (eds) (1992). Party Organizations: A Data Handbook on Party Organizations in Western Democracies, 1960–90. London: Sage. —— and Mair, Peter (1995). ‘Changing Models of Party Organization and Party Democracy: The Emergence of the Cartel Party’. Party Politics, 1: 5–28. ——, —— (1992). ‘The Membership of Political Parties in European Democracies, 1960–1990’. European Journal of Political Research, 22: 329–45. Kautsky, Karl (1898). Der Parlamentarismus, die Volksgesetzgebung und die Sozialdemokratie. Stuttgart: Dietz. Kawashima, Shiro (1995). ‘A Survey of Environmental Law and Policy in Japan’. North Carolina Journal of International Law and Commercial Regulation, 20: 231–72. Keating, Michael (1998). The New Regionalism in Western Europe: Territorial Restructuring and Political Change. Cheltenham, England: Edward Elgar. —— (1999). ‘Asymmetrical Government: Multinational States in an Integrating Europe’. Publius, 29: 71–86.
REFERENCES
289
Kellough, J. Edward (1998). ‘The Reinventing Government Movement: A Review and Critique’. Public Administration Quarterly, 22: 6–20. Kempf, Udo and Mille, Marco (1999). ‘The Role and the Function of the Ombudsman: Personalised Parliamentary Control in Forty-eight Different States’, in Linda Rief (ed.), The International Ombudsman Anthology. The Hague: Kluwer Law International. Kennett, Geoffrey (1994). ‘Individual Rights, the High Court and the Constitution’. Melbourne University Law Review, 19: 581–614. Kettl, Donald (1997). ‘The Global Revolution in Public Management: Driving Themes, Missing Links’. Journal of Policy Analysis and Management, 16: 446–62. —— (2000). ‘The Transformation of Governance: Globalization, Devolution, and the Role of Government’. Public Administration Review, 60: 488–97. Keyssar, Alexander (2000). The Right to Vote: The Contested History of Democracy in the United States. New York: Basic Books. King, Anthony (1997). Running Scared: Why America's Politicians Campaign too Much and Govern too Little. New York: Martin Kessler Books. King, Cheryl S., Feltey, Kathryn M., and O'Neill Susel, Bridge (1998). ‘The Question of Participation: Toward Authentic Public Participation in Public Administration’. Public Administration Review, 58: 317–27. Kirchheimer, Otto (1966). ‘The Transformation of the Western European Party Systems’, in Joseph LaPalombara and Myron Weiner (eds), Political Parties and Political Development. Princeton, NJ: Princeton University Press. Kitschelt, Herbert P. (1989). The Logics of Party Formation: Ecological Politics in Belgium and West Germany. Ithaca, NY: Cornell University Press. —— (1994). The Transformation of European Social Democracy. New York: Cambridge University Press. —— and McGann, Anthony J. (1997). The Radical Right in Western Europe: A Comparative Analysis. Ann Arbor: University of Michigan Press. Klee, Gudrun (1993). ‘Financing Parties and Elections in Small European Democracies: Austria and Sweden’, in Arthur B. Gunlicks (ed.), Campaign and Party Finance in North America and Western Europe. Boulder, CO: Westview. Klevit, Alan (1972). ‘City Councils and their Functions in Local Government’. The Municipal Yearbook 1972. Washington: International City Management Association. Klijn E. H. and Koppenjan, J.F.M. (2000). ‘Politicians and Interactive Decisionmaking: Institutional Spoilsport or Playmakers?‘ Public Administration, 78: 365–87. Klingemann, Hans-Dieter, Hofferbert, Richard I, and Budge, Ian (1994). Parties, Policies, and Democracy. Boulder, CO: Westview. Kloti, Ulrich (2001). ‘Consensual Government in a Heterogeneous Polity’. West European Politics, 24: 19–34. Knaap, Gerrit, Matier, Debra, and Olshansky, Robert (1998). ‘Citizen Advisory Groups in Remedial Action Planning: Paper Tiger or Key to Success’. Journal of Environmental Planning and Management, 41: 337–54. Knutsen, Oddbjørn (1998). ‘Expert Judgements of the Left-Right Location of Political Parties in West European Countries: A Comparative Longitudinal Study’. West European Politics, 21: 63–94. Kolodny, Robin and Katz, Richard S. (1992). ‘The United States’, in Richard S. Katz and Peter Mair (eds), Party Organizations. London: Sage.
290
REFERENCES
Koole, Ruud (1989). ‘The “Modesty” of Dutch Party Finance’, in Herbert E. Alexander (ed.), Comparative Political Finance in the 1980s. Cambridge: Cambridge University Press. —— (1994a). ‘Dutch Political Parties: Money and the Message’, in Herbert E. Alexander and Rei Shiratori (eds), Comparative Political Finance among the Democracies. Boulder, CO: Westview. —— (1994b). ‘The Vulnerability of the Modern Cadre Party in the Netherlands’, in Richard S. Katz and Peter Mair (eds), How Parties Organize. London: Sage. Kornhauser, William (1959). The Politics of Mass Society. Glencoe, IL: Free Press. Kymlicka, Will (1995). Multicultural Citizenship: A Liberal Theory of Minority Rights. Oxford: Oxford University Press. Landfried, Christine (1994). ‘Political Finance in West Germany’, in Herbert E. Alexander and Rei Shiratori (eds), Comparative Political Finance among the Democracies. Boulder, CO: Westview. Lane, Jan-Erik (1997). ‘Introduction’, in Jan-Erik Lane (ed.), Public Sector Reform. Thousand Oaks, California: Sage. —— (2000). ‘The Ombudsman in Denmark and Norway’, in Roy Gregory and Philip Giddings (eds), Righting Wrongs: The Ombudsman in Six Continents. Amsterdam: IOS Press. LaPalombara, Joseph and Weiner, Myron (eds) (1966). Political Parties and Political Development. Princeton, NJ: Princeton University Press. Laski, Harold (1931). Democracy in Crisis. Durham, NC: University of North Carolina Press. Lawson, Kay (1980). ‘Political Parties and Linkage’, in Kay Lawson (ed.), Political Parties and Linkage: A Comparative Perspective. New Haven: Yale University Press. Laycock, David (1990). Populism and Democratic Thought in the Canadian Prairies. Toronto: University of Toronto Press. Lecky, William (1896). Democracy and Liberty, i. New York: Longmans. LeDuc, Lawrence (2001). ‘Democratizing Party Leadership Selection’. Party Politics, 7: 323–41. —— (2002). ‘Opinion Change and Voting Behaviour in Referendums’. European Journal of Political Research, 37: 711–32. ——, Niemi, Richard G., and Norris, Pippa (eds) (1996). Comparing Democracies: Elections and Voting in Global Perspective. Thousand Oaks, CA: Sage. Lee, Jaekyung (2000). ‘School Reforms in England, Japan, Korea and the U.S.; Policy Variation and Educational Convergence’. Paper presented at AERA Annual Meeting, New Orleans. ERIC Document Reproduction Service. Lenaghan, Jo (1999). ‘Involving the Public in Rationing Decisions: The Experience of Citizens Juries’. Health Policy, 49: 45–61. Leroux, Therese, Hirtle, Marie, and Fortin, Louis-Nicolas (1998). ‘An Overview of Public Consultation Mechanisms Developed to Address the Ethical and Social Issues Raised by Biotechnology’. Journal of Consumer Policy, 21: 445–81. Lidstrom, Anders (1999). ‘Local School Choice Policies in Sweden’. Scandinavian Political Studies, 22: 137–56. Lijphart, Arend (1994). Electoral Systems and Party Systems: A Study of Twenty-Seven Democracies 1945–1990. New York: Oxford University Press. —— (1997a). ‘The Difficult Science of Electoral Systems: A Commentary on the Critique by Alberto Penadés’. Electoral Studies, 16: 73–7. —— (1997b). ‘Unequal Participation: Democracy's Unresolved Dilemma’. American Political Science Review, 91: 1–14.
REFERENCES
291
—— (1999). Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries. New Haven, CT: Yale University Press. Linton, Martin (1994). Money and Votes. London: Institute for Public Policy Research. Linz, Juan and Stepan, Alfred (eds) (1978). The Breakdown of Democratic Regimes. Baltimore: Johns Hopkins University Press. Lipset, Seymour Martin (1981). ‘The Revolt against Modernity’, in P. Torsvik (ed.), Mobilization, Center-Periphery Structures and Nation-building. Bergen: Universitetsforlaget. —— (1990). Continental Divide: The Values and Institutions of the United States and Canada. New York: Routledge. —— and Rokkan, Stein (1967). ‘Cleavage Structures, Party Systems, and Voter Alignments’, in. M. Lipset and Stein Rokkan (eds), Party Systems and Voter Alignments. S. New York: Free Press. Loveland, Ian (1999). ‘Incorporating the European Convention on Human Rights into UK Law’. Parliamentary Affairs, 52: 113–27. Lowell, Lawrence (1913). Public Opinion and Popular Government. New York: Longmans, Green and Company. Lowi, Theodore (1969). The End of Liberalism: Ideology, Policy, and the Crisis of Public Authority. New York: Norton. Lowndes, Vivien, Pratchett, Lawrence, and Stoker, Gerry (2001a). ‘Trends in Public Participation: Part I–Local Government Perspectives’. Public Administration, 79: 205–22. ——, ——, —— (2001b). ‘Trends in Public Participation: Part II–Citizens’ Perspectives’. Public Administration, 79: 445–55. Lucardie, Paul and Voerman, Gerrit (1997). ‘The Netherlands’. European Journal of Political Research, 32: 447–49. Luhmann, Niklas (1998). Observations on Modernity. Stanford: Stanford University Press. Lupia, Arthur (1994). ‘Shortcuts versus Encyclopedias’. American Political Science Review, 88: 63–76. —— and McCubbins, Mathew D. (1998). The Democratic Dilemma: Can Citizens Learn What They Need To Know? Cambridge: Cambridge University Press. Luthardt, Wolfgang (1994). Direkte Demokratie. Baden-Baden: Nomos Verlagsgesellschaft. Lutz, Donald (forthcoming). Popular Sovereignty, Separation of Powers and Other Principles of Constitutional Design. McAdam, Doug, Tarrow, Sidney G., and Tilly, Charles (2001). Dynamics of Contention. New York: Cambridge University Press. McAllister, Ian (1996). ‘Leaders’, in Lawrence LeDuc, Richard Niemi, and Pippa Norris (eds), Comparing Democracies. Thousand Oaks, CA: Sage. McCarthy, Adam (2001). A Silent Voice? The Role of Party Members in Candidate and Leadership Selection in Contemporary West European Political Parties. Research monograph. Manchester: University of Manchester. MacCormick, Neil (1978). Legal Reasoning and Legal Theory. Oxford: Clarendon. McCrackan, W. D. (1893). ‘The Swiss Referendum: The Ideal Republican Government’. The Cosmopolitan, 15: 329–33. McCubbins, Mathew D., Noll, Roger G., and Weingast, Barry R. (1987). ‘Administrative Procedures as Instruments of Political Control’. Journal of Law, Economics and Organization, 3: 243–77. MacDonald, J. Ramsay (1909). Socialism and Government. London: Independent Labour Party.
292
REFERENCES
McGarity, Thomas (2002). Public Participation in Risk Regulation: Symposium on Public Participation in Risk Management 〈http://www.piercelaw.edu/risk/pubprtn.htm〉. McGinn, Noel and Welsh, Thomas (1999). Decentralization of Education: Why, When, What, and How? Paris: UNESCO. McGrew, Anthony (1997). The Transformation of Democracy: Globalization and Territorial Democracy. Cambridge: Polity Press. McKean, Margaret A. (1981). Environmental Protest and Citizen Politics in Japan. Berkeley: University of California Press. McLean, Ian (1989). Democracy and New Technology. Cambridge: Polity Press. Maine, Henry Sumner (1886). Popular Government. New York: Henry Holt and Company. Mair, Peter (1995). ‘Political Parties, Popular Legitimacy, and Public Privilege’. West European Politics, 18: 40–57. —— (2000). ‘Populist Democracy vs. Party Democracy’. Paper presented at the European Consortium of Political Research Joint Sessions, Copenhagen, 14–19 April. Mansbridge, Jane (1980). Beyond Adversary Democracy. New York: Basic Books. —— (1997). ‘Social and Cultural Causes of Dissatisfaction with U.S. Government’, in Joseph S. Nye, Jr., Philip D. Zelikow, and David C. King (eds), Why the People Don't Trust Government. Cambridge, MA: Harvard University Press. March, James G. and Olson, Johan P. (1989). Rediscovering Institutions. New York: Macmillan. Marletti, Carlo and Roncarolo, Franca (2000). ‘Media Influence in the Italian Transition from a Consensual to a Majoritarian Democracy’, in Richard Gunther and Anthony Mughan (eds), Democracy and the Media: A Comparative Perspective. Cambridge: Cambridge University Press. Marquand, David (1979). Parliament for Europe. London: J. Cape. Marsh, Norman (ed.) (1987). Public Access to Government-Held Information. London: Stevens and Son. Masajima, Toshiyuki (1999). ‘Evaluating Administrative Reform: An Insider's Account’. Social Science Japan Journal, 2: 215–28. Mateo, Juan Ferrer (1991). ‘Improving Access to Administration in Spain’, in Richard Batley and Gerry Stoker (eds), Local Government in Europe: Trends and Developments. New York: St Martin's Press. Matthews, Forest David (1999). Politics for People: Finding a Responsive Voice (2nd edn). Urbana: University of Illinois Press. Mavcic, Arne (2002). A Tabular Presentation of Constitutional Judicial Review Around the World 〈http://www.us-rs.si/en/ index.html〉. Médiateur de la République (2001). 2000 Yearly Report. Paris: La Documentation française. Mehr Demokratie e.V. March (2002). 〈http://www.mehr-demokratie.de〉. Meny, Yves and Surel, Yves (eds) (2002). Democracies and the Populist Challenge. New York: St Martins Press. Merryman, John Henry (1985). The Civil Law Tradition. Stanford, CA: Stanford University Press. Meyer, David and Tarrow, Sidney (eds) (1998). The Social Movement Society: Contentious Politics for a New Century. Lanham, MD: Rowman & Littlefield. Micheletti, Michele (1998). ‘The Democratic Audit of Sweden’. Viewpoint Sweden, 18. New York: Swedish Information Service.
REFERENCES
293
Michels, Robert (1958). Political Parties. Glencoe, IL: Free Press. Mira, Álvaro Xosé López (1999). ‘Portugal: The Resistance to Change in the State Model’. Regional and Federal Studies, 9/2. Montero, Alfred P. (2001). ‘After Decentralization: Patterns of Intergovernmental Conflict in Argentina, Brazil, Spain and Mexico’. Publius, 31: 43–66. Moring, Tom (1995). ‘The North European Exception: Political Advertising on TV in Finland’, in Lynda Lee Kaid and Christina Holtz-Bacha (eds), Political Advertising in Western Democracies. Thousand Oaks, CA: Sage. Morris, Dick (2000). The New Prince. New York: Renaissance Books. Mulgan, Richard (1994). Politics in New Zealand. Auckland: Auckland University Press. Murchinson, Kenneth M (1995). ‘Environmental Law in Australia and the United States: A Comparative Overview’. British Columbia Environmental Affairs Law Review, 503: 549–73. Nassmacher, Karl-Heinz (1989). ‘Structure and Impact of Public Subsidies to Political Parties in Europe: the Examples of Austria, Italy, Sweden and West Germany’, in Herbert E. Alexander (ed.), Comparative Political Finance in the 1980s. Cambridge: Cambridge University Press. Nathan, Richard (1996). ‘The Devolution Revolution: An Overview’, in Rockefeller Institute Bulletin 1996. Albany, NY: Nelson A. Rockefeller Institute of Government. Netherlands. National Ombudsman of the Netherlands (2001). Annual Report 2000 (House of Representatives Document, 27 645, nos 1–2). The Hague. Neumann, Sigmund (1956). Modern Political Parties. Chicago: The University of Chicago Press. New Zealand. Office of the Ombudsman (2001). Report of the Ombudsmen 2000, Nga Kaitiaki Mana Tangata 〈http:// www.ombudsmen.govt.nz/download%20Annual%20Report/AR2000-2001.pdf〉. Nikos, Michalopolous (2001). ‘Trends of Administrative Reform in Europe: Towards Administrative Convergence’. International Public Management Review, 2: 239–53. Nohlen, Dieter (1984). ‘Changes and Choices in Electoral Systems’, in Arend Lijphart and Bernard Grofman (eds), Choosing an Electoral System. New York: Praeger. Norris, Pippa (ed.) (1999a). Critical Citizens: Global Support for Democratic Governance. Oxford: Oxford University Press. —— (ed.) (1999b). ‘Introduction: The Growth of Critical Citizens?‘, in Pippa Norris (ed.), Critical Citizens: Global Support for Democratic Governance. Oxford: Oxford University Press. —— (ed.) (2002). Democratic Phoenix: Reinventing Political Activism. New York: Cambridge University Press. Norton, Alan (1994). International Handbook of Local and Regional Government. Aldershot: Edward Elgar. Norway Parliamentary Ombudsman (2001). Annual Report 2000 (Document No. 4). Oslo. Nozari, Fariborz (1991). ‘Sweden’, in Ruth Levush (ed.), Campaign Financing of National Elections in Foreign Countries. Washington, DC: Law Library of Congress. Nye, Joseph S., Zelikow, Philip D., and King, David C. (1997). Why People Don't Trust Government. Cambridge, MA: Harvard University Press. O'Brien, David (1996). To Dream of Dreams: Religious Freedom and Constitutional Politics in Postwar Japan. Honolulu, HI: University of Hawaii Press.
294
REFERENCES
O'Brien, David and Ohkoshi, Yasuo (2001). ‘Stifling Judicial Independence from within: The Japanese Judiciary’, in P. Russell and D. O'Brien (eds), Judicial Independence in the Age of Democracy. Charlottesville, VA: University of Virginia Press. Oberholtzer, Ellis (1891). ‘Law-making by Popular Vote; or the American Referendum’. Annals of the American Academy of Political and Social Sciences: 324–44. —— (1900). The Referendum in America. New York: Charles Scribner's Sons. O'Connor, Sandra Day (1986). ‘Reflections on Preclusion of Judicial Review in England and the United States’. William and Mary Law Review, 27: 643–62. OECD (Organization for Economic Cooperation and Development) (1995). Decision-making in 14 OECD Education Systems. Paris: OECD. —— (1996a). Integrating People Management into Public Service Reform. Paris: OECD. —— (1996b). Regulatory Reform: A Country Study of Australia. Paris: OECD. —— (1997). Managing Across Levels of Government. Paris: OECD. —— (1998). Education at a Glance. Paris: OECD. —— (1999). OECD Review of Regulatory Reform in the Netherlands. Paris: OECD. —— (2000a). Government of the Future. Paris: OECD. —— (2000b). Regulatory Reform in Denmark. Paris: OECD. —— (2000c). ‘Public Sector Pay and Employment Data’ 〈http://www.oecd.org/xls/M00008000/M00008481.xls〉. —— (2001a). Citizens as Partners: OECD Handbook on Information, Consultation and Public Participation in Policy-Making. Paris: OECD. —— (2001b). Issues and Developments in Public Management (Country Case Studies). Paris: OECD. —— (2001c). Regulatory Reform in Ireland. Paris: OECD. —— (2001d). Regulatory Reform in Italy. Paris: OECD. —— (2001e). Government-Citizen Relations Profile: The United Kingdom. Paris: OECD. —— (2001f). Government-Citizen Relations Profile: France. Paris: OECD. —— (2001g). Government-Citizen Relations Profile: Finland. Paris: OECD. —— (2001h). Government-Citizen Relations Profile: Denmark. Paris: OECD. —— (2001i). Details of tax revenues (various countries). Vol. 2001, release 01. Paris: OECD. —— (2002a). Regulatory Policies in OECD Countries: From Interventionism to Regulatory Governance. Paris: OECD. —— (2002b). Participatory Decision-making for Sustainable Consumption. Paris: OECD. —— (2002c). Regulatory Reform in Canada: Government Capacity to Assure High Quality Regulation. Paris: OECD. Offe, Claus (1996). Modernity and the State: East, West. Cambridge, MA: MIT Press. O'Neill, Michael (1998). ‘Re-imagining Belgium: New Federalism and the Political Management of Cultural Diversity’. Parliamentary Affairs, 51: 241–59. Opfer, V. Darlene and Denmark, Vicki (2001). ‘Sorting Out a Sense of Place: School and School Board Relationships in the Midst of School-Based Decision Making’. Peabody Journal of Education, 76: 101–18. Orlansky, Dora (2000). ‘Decentralization Politics and Policy’, in Stuart Nagel (ed.), Critical Issues in Cross-National Public Administration. Westport, CT: Quorum Books. Ormond, Thomas (1995). ‘Access to Justice for Environmental NGOs in the European Union’, in S. Deimann and B. Dyssli (eds), Environmental Rights. London: Cameron May.
REFERENCES
295
Orren, Gary (1997). ‘Fall From Grace: The Public's Loss of Faith in Government’, in Joseph S. Nye Jr., Philip D. Zelikow, and David C. King (eds), Why People Don't Trust Government. Cambridge, MA: Harvard University Press. Osamu, Koike and Wright, Deil S. (1998). ‘Five Phases of Inter-Governmental Relations in Japan: Policy Shifts and Governance Reform’. International Review of Administrative Sciences, 64: 203–18. Osborne, David and Gaebler, Ted (1992). Reinventing Government. Reading, MA: Addison-Wesley. Owens, Jon (2001). ‘Comparative Law and Standing to Sue: A Petition for Redress for the Environment’. The Environmental Lawyer, 7: 321–77. Pacelle, Richard L., Jr. (1991). The Transformation of the Supreme Court's Agenda. Boulder, CO: Westview. Painter, Martin (2001). ‘Multi-level Governance and the Emergence of Collaborative Federal Institutions in Australia’. Policy and Politics, 29: 137–50. Palerm, Juan (2000). ‘An Empirical-Theoretical Analysis Framework for Public Participation in Environmental Impact Assessment’. Journal of Environmental Planning and Management, 43: 581–600. Paltiel, Khayyam Zev (1989). ‘Canadian Election Expense Legislation, 1963–85: A Critical Appraisal or Was the Effect Worth It?‘, in Herbert E. Alexander (ed.), Comparative Political Finance in the 1980s. Cambridge: Cambridge University Press. Panebianco, Angelo (1988). Political Parties: Organization and Power. Cambridge: Cambridge University Press. Parker, Douglas (1995). ‘Standing to Litigate “Abstract Social Interests” in the United States and Italy: Reexamining “Injury in Fact”‘. Columbia Journal of Transnational Law, 33: 259–318. Parker, Simon (2002). ‘Bedford Wants a Mayor’. Guardian Unlimited, 22 February 〈http://politics.guardian.co.uk/ localgovernment/story/0,9061,655454,00.html〉. Parskevopoulos, Christos J. (1998). ‘Social Capital and the Public-Private Divide in Greek Regions’. West European Politics, 21: 154–78. Parsons, Talcott (1971). The System of Modern Societies. Engelwood Cliffs, NJ: Prentice-Hall. Pascoe, Susan and Pascoe, Robert (1998). ‘Education Reform in Australia: 1992–1997’. Education Reform and Management Publication Series, 1/2. Washington, DC: World Bank. Patapan, Haig (2000). ‘Politics of Interpretation’. Sydney Law Review, 22: 245–72. Pateman, Carole (1970). Participation and Democratic Theory. Cambridge: Cambridge University Press. Pauger, Dietmar (1983). ‘Das Stadt- und Gemeinderecht in Österreich und Möglichkeiten der Bürgermitbestimmung and Bürgerbeteiligung’. Österreichische Gemeinde-Zeitung, 49: 522–33. Pederson, Mogens (1996). ‘Euro-Parties and European Parties: New Arenas, New Challenges and New Strategies’, in Svein Anderson and Kjell Eliassen (eds), The European Union: How Democratic Is It? London: Sage. Pelinka, Anton (1994). ‘Der Aufstieg des Plebiszitären’, in Anton Pelinka (ed.), EU-Referendum: Zur Praxis direkter Demokratie in Deutschland. Vienna: Signum Verlag. Pelinka, Anton and Greiderer, Sylvia (1996). ‘Austria: The Referendum as an Instrument of Internationalisation’, in Michael Gallagher and Pier Vicenzo Uleri (eds), The Referendum Experience in Europe. Basingstoke: Macmillan.
296
REFERENCES
Penadés, Alberto (1997). ‘A Critique of Lijphart's “Electoral Systems and Party Systems” ‘. Electoral Studies, 16: 58–71. Pennings, Paul and Hazan, Reuven Y. (2001). ‘Democratizing Candidate Selection: Causes and Consequences’. Party Politics, 7: 267–75. Pernthaler, Peter (ed.) (1980). Direkter Demokratie in den Ländern und Gemeinden. Vienna: Wilhelm Braumüller. Perris, Lyall (1998). ‘Implementing Education Reforms in New Zealand: 1987–1997’. The Education Reform and Management Series, 1: 1–32. Peters, B. Guy (2001). The Future of Governing. Lawrence, KS: University of Kansas Press. Petts, Judith (2001). ‘Evaluating the Effectiveness of Deliberative Processes: Waste Management Case Studies’. Journal of Environmental Planning and Management, 44: 207–26. Pharr, Susan J. and Putnam, Robert D. (eds) (2000). Disaffected Democracies: What's Troubling the Trilateral Countries? Princeton: Princeton University Press. Pickard, Susan (1998). ‘Citizenship and Consumerism in Health Care: A Critique of Citizens’ Juries’. Social Policy and Administration, 32: 226–44. Pierce, Richard J. (1999). ‘Is Standing Law or Politics?‘. North Carolina Law Review, 77: 1741–65. Pierre, Jon, Svåsand, Lars, and Widfeldt, Anders (2000). ‘State Subsidies to Political Parties: Confronting Rhetoric with Reality’. West European Politics, 23: 1–24. ——, and Widfeldt, Anders (1994). ‘Party Organizations in Sweden: Colossuses with Feet of Clay or Flexible Pillars of Government?‘, in Richard Katz and Peter Mair (eds), How Parties Organize: Change and Adaptation in Party Organizations in Western Democracies. London: Sage. Piven, Frances Fox and Cloward, Richard (1989). Why Americans Don't Vote. New York: Pantheon. Poguntke, Thomas (1998). ‘Party Organizations’, in Jan van Deth (ed.), Comparative Politics: The Problem of Equivalence. London: Routledge. —— (2000). Parteiorganisation im Wandel. Gesellschaftliche Verankerung und organisatorische Anpassung im europäischen Vergleich. Wiesbaden: Westdeutscher Verlag. —— (2002). ‘Party Organizational Linkage: Parties Without Firm Social Roots’, in Kurt Richard Luther and Ferdinand Müller-Rommel (eds), Political Parties in the New Europe. Oxford: Oxford University Press. Poiares Maduro, Miguel (1998). We, the Court: The European Court of Justice and the European Economic Constitution. Oxford: Hart. Polakiewcz, Jorg (1996). ‘The Application of the European Convention on Human Rights in Domestic Law’. Human Rights Law Journal, 17: 405–11. Pollitt, Christopher and Geert, Bouckaert (2000). Public Management Reforms: A Comparative Analysis. Oxford: Oxford University Press. Popkin, Samuel (1991). The Reasoning Voter. Chicago: University of Chicago Press. Port, Kenneth L. (1996). Comparative Law: Law and the Legal Process in Japan. Durham, NC: Carolina Academic Press. Powell, G. Bingham, Jr. (2000). Elections as Instruments of Democracy: Majoritarian and Proportional Visions. New Haven, CT: Yale University Press. Powell, Walter W. and DiMaggio, Paul J. (eds) (1991). The New Institutionalism in Organizational Analysis. Chicago: University of Chicago Press.
REFERENCES
297
Pratchett, Lawrence (1999). ‘New Fashions in Public Participation: Towards Greater Democracy?‘ Parliamentary Affairs, 52: 616–34. Preston, Brian J. (1991). ‘Public Enforcement of Environmental Laws in Australia’. Journal of Environmental Law and Litigation, 6: 39–57. Preuss, Ulrich (1995). Constitutional Revolution: The Link Between Constitutionalism and Progress (trans. Deborah Lucas Schneider). Atlantic Highlands, NJ: Humanities Press. Pujas, Véronique (2000). ‘The Funding of Political Parties and Control of the Media: Another Italian Anomaly’, in Mark Gilbert and Gianfranco Pasquino (eds), Italian Politics: The Faltering Transformation. New York: Berghan Books. —— and Rhodes, Martin (1999). ‘Party Finance and Political Scandal in Italy, Spain and France’. West European Politics, 22: 41–63. Punnett, R. M. (1992). Selecting the Party Leader: Britain in Comparative Perspective. Hemel Hempstead: Harvester Wheatsheaf. Putnam, Robert (1993). Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press. —— (2000a). Bowling Alone. New York: Simon and Schuster. —— (ed.) (2002b). Democracies in Flux. Oxford: Oxford University Press. Quagliariello, Gaetano (1996). Politics without Parties. Aldershot: Avebury. Qualter, Terence H. (1970). The Election Process in Canada. Toronto: McGraw-Hill. Ranney, Austin (1981). ‘Candidate Selection’, in D. Butler, H. R. Penniman, and A. Ranney (eds), Democracy At the Polls: A Comparative Study of Competitive National Elections. Washington, DC: American Enterprise Institute. Rawls, John (1993). Political Liberalism. New York: Columbia University Press. Reeves, Dory (1995). ‘Developing Effective Public Consultation: A Review of Sheffield's UDP Process’. Planning Practice and Research, 10: 199–213. Regional Environmental Center for Central and Eastern Europe (1998). Doors to Democracy: A Pan-European Assessment of Current Trends and Practices in Public Environmental Participation in Environmental Matters. Szentendre, Hungary: REC. Reif, Karlheinz (ed.) (1984). European Elections 1979/81 and 1984: Conclusions and Perspectives from Empirical Research. Berlin: Quorum. Renner, Tari and DeSantis, Victor (1993). ‘Contemporary Patterns and Trends in Municipal Government Structures’. The Municipal Yearbook 1993. Washington, DC: International City/County Management Association. Rhodes, Martin (1997). ‘Financing Party Politics in Italy: A Case of Systemic Corruption’. West European Politics, 20: 54–80. Rhodes, R.A.W. (1999).’Traditions and Public Sector Reform: Comparing Britain and Denmark’. Scandinavian Political Studies, 22: 341–70. Ricolfi, Luca (1997). ‘Politics and the Mass Media in Italy’. West European Politics, 20: 135–56. Rippe, Klaus Peter and Schaber, Peter (1999). ‘Democracy and Environmental Decision-making’. Environmental Values, 8: 75–88. Rittinghausen, Moritz (1877). Die direkte Gesetzgebung durch das Volk (4th edn). Cologne. Robertson, James R. (1997). The Canadian Electoral System. Ottawa: Library of Parliament 〈http://www.parl.gc.ca/ information/library/PRBpubs/bp437-e.htm〉. Robertson, K. G. (1982). Public Secrets: A Study in the Development of Government Secrecy. New York: St Martin's Press.
298
REFERENCES
Rommelfanger, Ulrich von (1988). Das konsultative Referendum: eine verfassungstheoretische, -rechtliche und-vergleichende Untersuchung. Berlin: Duncker and Humblot. Rose-Ackerman, Susan (1999). Corruption and Government: Causes, Consequences, and Reform. Cambridge: Cambridge University Press. Rosenbluth, Frances and Thies, Michael F. (2002). ‘Politics in Japan’, in Gabriel A. Almond, G. Bingham Powell, Jr., Russell Dalton, and Kaare Strom (eds), Comparative Politics Today (updated 7th edn). New York: Addison Wesley Longman. Rosenstone, Steven and Hansen, Mark (1993). Mobilization, Participation, and Democracy in America. New York: Macmillan. Rouban, Luc (1999). Citizens and the New Governance: Beyond New Public Management. Amsterdam: IOS Press. Russell, Peter H. (1992). ‘The Growth of Canadian Judicial Review and the Commonwealth and American Experiences’, in D. Jackson and C. N. Tate (eds), Comparative Judicial Review and Public Policy. Westport, CT: Greenwood Press. Rüther, Günther (ed.) (1996). Repräsentative oder plebiszitäre Demokratie – eine Alternative? Grundlagen, Vergleiche, Perspektiven. Baden-Baden: Nomos. Saarikoski, Heli (2000). ‘Environmental Impact Assessment (EIA) As Collaborative Learning Process’. Environmental Impact Assessment Review, 20: 681–700. Sandel, Michael J. (1996). Democracy's Discontent: America in Search of a Public Philosophy. Cambridge, MA: Harvard University Press. Scalia, Antonin (1983). ‘The Doctrine of Standing as an Essential Element of the Separation of Powers’. Suffolk University Law Review, 17: 881–203. Scannell, Yvonne (1995). Environmental and Planning Law in Ireland. Dublin: Round Hall Press. Scarrow, Susan E. (1996). Parties and their Members: Organizing for Victory in Britain and Germany. Oxford: Oxford University Press. —— (1997). ‘Party Competition and Institutional Change: The Expansion of Direct Democracy in Germany’. Party Politics, 3: 451–72. —— (1999). ‘Parties and the Expansion of Direct Democracy: Who Benefits?‘ Party Politics, 5: 341–62. —— (2001). ‘Direct Democracy and Institutional Design: A Comparative Investigation’. Comparative Political Studies, 34: 651–65. —— (2002). Perspectives on Political Parties. New York: Palgrave. ——, Webb, Paul, and Farrell, David M. (2000). ‘From Social Integration to Electoral Contestation: The Changing Distribution of Power Within Parties’, in Russell J. Dalton and Martin P. Wattenberg (eds), Parties Without Partisans. New York: Oxford University Press. Schattschneider, E. E. (1942). Party Government. New York: Rinehart. Schedler, Petra and Glastra, Folke (2001). ‘Communicating Policy in Late Modern Society: On the Boundaries of Interactive Policy Making’. Policy & Politics, 29: 337–49. Schiffers, Reinhard (1971). Elemente direkter Demokratie im Weimarer Regierungs-system. Düsseldorf: Droste Verlag. Schlesinger, Arthur M. Jr. (1999). The Cycles of American History. New York: Houghton Mifflin. Schmidt, Vivien A. (1990). ‘Unblocking Society by Decree: The Impact of Governmental Decentralization in France’. Comparative Politics, 22: 459–81.
REFERENCES
299
Schmitt, Carl (1988). The Crisis of Parliamentary Democracy (trans. Ellen Kennedy). Cambridge, MA: MIT Press. Schofield, Mark A. and Thompson, David S. (1994). ‘Access to Justice and the Right to a Healthful Environment in Canada: Public Participation in Environmental Decision Making’. Review of European Community and International Environmental Law, 3: 231. Schumpeter, Joseph (1943). Capitalism, Socialism, and Democracy. New York: Harper. Schwartz, Herman (1994). ‘Small States in Big Trouble: State Reorganization in Australia, Denmark, New Zealand, and Sweden in the 1980s’. World Politics, 46: 527–55. Sclove, Richard. 1996. ‘Town Meetings on Technology’. Technology Review, 99/5: 24–31. —— (1997). ‘Telecommunications & the Future of Democracy: Preliminary Report On The First U.S. Citizens’ Panel’. Loka Institute 〈http://www.loka.org〉. Sear, Chris (2001). Parliamentary Pay and Allowances: Current Rates. Research Paper 01/87. House of Commons Library: Parliament and Constitution Centre 〈http:// www. parliament.uk/commons/lib/research/rp2001/rp01-087. pdf〉. Seidenfeld, Mark (2000). ‘Empowering Stakeholders: Limits on Collaboration as the Basis for Flexible Regulation’. William and Mary Law Review, 41: 411. Selborne, Earl of (1911). ‘The Unionist Party and the Referendum’. The Oxford and Cambridge Review, July: 3–32. Selle, Per (1997). ‘Parties and Voluntary Organizations: Strong or Weak Ties?‘, in Karre Strom and Lars Svansand (eds), Challenges to Political Parties: The Case of Norway. Ann Arbor: The University of Michigan Press. Shafer, Byron E. (1983). Quiet Revolution: The Struggle for the Democratic Party and the Shaping of Post-reform Politics. New York: Russell Sage Foundation. Shapiro, Martin (1988). Who Guards the Guardians: Judicial Control of Administration. Athens: University of Georgia Press. —— (2002). ‘Judicial Delegation Doctrines: The US, Britain, and France’. West European Politics, 25: 173–99. —— and Stone Sweet, Alec (2002). On Law, Politics, and Judicialization. Oxford: Oxford University Press. Sharpe, L. J. (1988). ‘The Growth and Decentralization of the Modern Democratic State’. European Journal of Political Research, 16: 365–80. Shugart, Matt and Wattenberg, Marty (eds) (2001). Mixed Member Electoral Systems: The Best of Both Worlds? Oxford: Oxford University Press. Simpson, Alan (ed.) (1992). Referendums: Constitutional and Political Perspectives. Wellington: Victoria University. Singer, Michael (1982). ‘The Constitutional Court of the German Federal Republic: Jurisdiction over Individual Complaints’. International and Comparative Law Quarterly, 31: 331–56. Smith, Anthony (1981). ‘Mass Communications’, in David Butler, H. R. Penniman, and A. Ranney (eds), Democracy at the Polls. Washington, DC: American Enterprise Institute. Smith, Gordon (1989). ‘Core Persistence: Change and the People's Party’. West European Politics, 12: 157–68. Smith, Graham and Wales, Corrine (2000). ‘Citizens’ Juries and Deliberative Democracy’. Political Studies, 48: 51–65. Sniderman, Paul, Brody, Richard, and Tetlock, Philip (1991). Reasoning and Choice. New York: Cambridge University Press.
300
REFERENCES
Starke, F. (1993). Krise ohne Ende? Parteiendemokratie vor neuen Herausforderungen. Cologne: Bund Verlag. Stewart, John (1996). ‘Innovation in Democratic Practice in Local Government’. Policy & Politics, 24: 29–41. Stokes, Donald (1962). ‘Popular Evaluations of Government’, in H. Cleveland and H. D. Lasswell (eds), Ethics and Bigness. New York: Harper. Stone, Alec (1992). The Birth of Judicial Politics in France: The Constitutional Council in Comparative Perspective. Oxford: Oxford University Press. Stone Sweet, Alec (1999). ‘Judicialization and the Construction of Governance’. Comparative Political Studies, 31:147–84. —— (2000). Governing with Judges: Constitutional Politics in Europe. New York: Oxford University Press. —— and Brunell, Thomas (1999). The Alec Stone Sweet and Thomas L. Brunell Data Set on Preliminary References in EC Law, 1958–98. Florence, Italy: Robert Schuman Centre for Advanced Studies, European University Institute. —— and Caporaso, James (1998). ‘From Free Trade to Supranational Polity: The European Court and Integration’, in W. Sandholtz and A. Stone Sweet (eds), European Integration and Supranational Governance. Oxford: Oxford University Press. Strom, K. (2000), ‘Parties at the Core of Government’, in Dalton Russ and Marty Wattenberg (eds), Parties Without Partisans. Oxford: Oxford University Press. Suksi, Markku (1993). Bringing in the People: A Comparison of Constitutional Forms and Practices of the Referendum. Dordrecht: Martinus Nijhoff. Sullivan, J. W. (1892). Direct Legislation by the Citizenship through the Initiative and Referendum. New York: Twentieth Century Publishing Company. Sundberg, Jan (1994). ‘Finland: Nationalized Parties, Professionalized Organizations’, in Richard S. Katz and Peter Mair (eds), How Parties Organize. London: Sage. Svåsand, Lars (1991). ‘State Subventions for Political Parties in Norway’, in Matti Wiberg (ed.), The Public Purse and Political Parties. Jyväskylä: Gummerus. —— (1992). ‘Norway’, in Richard S. Katz and Peter Mair (eds), Party Organizations. London: Sage. —— (1994). ‘Change and Adaptation in Norwegian Party Organizations’, in Richard S. Katz and Peter Mair (eds), How Parties Organize. London: Sage. Sweden. Parliamentary Ombudsman (2001). 2000 Annual Report. Stockholm: Sverge Rikstag 〈http://www.riksdagen. se/debatt/0001/forslag/jo1/jo1.pdf〉. Swyngedouw, Marc (1998). ‘The Extreme Right in Belgium: Of a Non-existent Front National and an Omnipresent Vlaams Blok’, in Hans-Georg Betz and Stefan Immerfall (eds), The New Politics of the Right. New York: St Martin's Press. Taagepera, Rein (1998). ‘Effective Magnitude and Effective Threshold’. Electoral Studies, 17: 393–404. Takao, Yasuo (1998). ‘Participatory Democracy in Japan's Centralization Drive’. Asian Survey, 38: 950–67. Tannenwald, Robert (1998). ‘Implications of a Balanced Budget Act of 1997 for the “Devolution Revolution”‘. Publius: The Journal of Federalism, 28/1: 23–48. Teske, Nathan (2000). ‘A Tale of Two TAGS: Dialogue and Democracy in the Superfund Program’. American Behavioral Scientist, 44: 663–77. Teubner, Günter (1983). ‘Substantive and Reflexive Elements in Modern Law’. Law and Society Review, 17: 239–85.
REFERENCES
301
Thatcher, Mark and Stone Sweet, Alec (2002). ‘Theory and Practice of Delegation to Non-Majoritarian Institutions’. West European Politics, 25: 1–22. Thomas, Clive S. (ed.) (2001). Political Parties and Interest Groups: Shaping Democratic Governance. Boulder, CO: Lynne Rienner Publisher. Thompson, Dennis F. (1995). Ethics in Congress: From Individual to Institutional Corruption. Washington, DC: The Brookings Institution. Tiebout, C. (1956). ‘A Pure Theory of Local Expenditures’. Journal of Political Economy, 64: 416–24. Toohey, Mr Justice John (1993). ‘A Government of Laws, and Not of Men?‘. Public Law Review, 4: 158–69. Toonen, Theo A. J. and Raadschelders, Jos C. N. (1997). ‘Public Sector Reform in Western Europe’. Paper presented at the Conference on Comparative Civil Service Systems, Indiana University, Bloomington, IN. Treschel, A. and Pascal, Sciarin (1998). ‘Direct democracy in Switzerland: Do Elites Matter?‘ European Journal of Political Research, 33: 99–124. Troper, Michel (1994). Pour une Théorie Juridique de L’état. Paris: Leviathan. United Kingdom. Parliamentary Ombudsman (2001). Annual Report 2000–01. London: Stationery Office. United States Census (2001). Statistical Abstract of the United States, 2001 (Table 414). Washington, DC: Bureau of the Census. United States. Department of Justice (2001). Annual FOIA Reports FY 2000 〈http://www.justice.gov/foia/fy00. html〉 accessed 14 August 2002. —— (2002). Freedom of Information Case List, May 2002. Chapter: Law Review Articles 〈http://www.justice.gov/ foia/lawrevart.pdf〉 accessed 13 August 2002. Uleri, Pier Vincenzo (1996). ‘Introduction’, in Michael Gallagher and Pier Vicenzo Uleri (eds), The Referendum Experience in Europe. Basingstoke: Macmillan. Upham, Frank K. (1987). Law and Social Change in Postwar Japan. Cambridge, MA: Harvard University Press. Van Deth, Jan (2000). ‘Interesting But Irrelevant: Social Capital and the Saliency of Politics in Western Europe’. European Journal of Political Research, 37: 115–47. Vandelli, Luciano (1992). ‘The New Local Government Law’, in Robert Leonardi and Fausto Anderlin (eds), Italian Politics: A Review, No. 6. London: Pinter. Verba, Sidney and Nie, Norman (1972). Participation in America. New York: Harper & Row. ——, ——, and Kim, Jae-on (1978). Participation and Political Equality. New York: Cambridge University Press. ——, Schlozman, Kay, and Brady, Henry (1995). Voice and Equality: Civic Volunteerism in American Politics. Cambridge, MA: Harvard University Press. Villadsen, Soren (1993). ‘Another Century for Local Democracy? Decentralization, Deregulation and Participation in Scandinavia in Times of European Integration’. International Journal of Urban and Regional Research, 17: 42–56. Walberg, Herbert, Paik Susan, Komukai Atsuko, and Freeman Karen (2000). ‘Decentralization: An International Perspective’. Educational Horizons, Winter: 155–66. Wallis, Joe and Dollery, Brian (2001). ‘Local Government Policy Evolution in New Zealand and the Ex-Post Emergence of Consensus or Rival Advocacy Coalitions’. Public Administration, 79: 533–60. Warren, Mark E. (1996). ‘Deliberative Democracy and Authority’. American Political Science Review, 90: 46–60.
302
REFERENCES
Warren, Mark E. (1999a). ‘What is Political?‘ Journal of Theoretical Politics, 11: 207–31. —— (1999b). ‘Democratic Theory and Trust’, in Mark E. Warren (ed.), Democracy and Trust. Cambridge: Cambridge University Press. —— (2001). Democracy and Association. Princeton, NJ: Princeton University Press. —— (2002a). ‘Deliberative Democracy’, in Geoffrey Stokes and April Carter (eds), Democratic Theory Today. Cambridge: Polity Press. —— (2002b). ‘What Can Democratic Participation Mean Today?‘ Political Theory, 30: 678–702. Wattenberg, Martin (1991). The Rise of Candidate Centered Politics. Cambridge, MA: Harvard University Press. —— (2002). Where Have All the Voters Gone? Cambridge, MA: Harvard University Press. Webb, Paul, Farrell, David, and Holliday, Ian (eds) (2002). Political Parties in Advanced Industrial Democracies. Oxford: Oxford University Press. Wehling, Hans-Georg (1993). ‘Kommunalverfassungen im Modellvergleich’. Alternative Kommunal Politik, 14: 30–3. —— (1994). ‘Deutschland’. Der Bürger im Staat, 44: 183–9. Weiler, Joseph H. H. and Lockhart, Nicolas J. S. (1995a). ‘“Taking Rights Seriously” Seriously: The European Court and its Fundamental Rights Jurisprudence’. Common Market Law Review, 32: 51–94. ——, —— (1995b). ‘“Taking Rights Seriously” Seriously: The European Court and its Fundamental Rights Jurisprudence’. Common Market Law Review, 32: 579–627. Weingast, Barry R. (1995). ‘The Economic Role of Political Institutions: Market-Preserving Federalism and Economic Growth’. Journal of Law, Economics, and Organization, 111–31. Weir, Stuart and Beetham, David (1998). Political Power and Democratic Control in Britain: The Democratic Audit. London: Routledge. Welan, Manfred (1986). ‘Regierungssystem und direkte Demokratie in Österreich’, in Frederich Koja and Gerald Stourzh (eds), Schweiz––Österreich: Ähnlichkeiten und Kontraste. Vienna: Hermann Böhlaus Nachf. Whitely, Paul and Seyd, Patrick (2002). High Intensity Participation: The Dynamics of Party Activism in Britain. Ann Arbor: University of Michigan Press. Williams, George (1999). Human Rights under the Australian Constitution. Melbourne: Oxford University Press. Williams, Richard C., Harold, Barbara, Robertson, Jan, and Southworth, Geoff (1997). ‘Sweeping Decentralization of Educational Decision-making Authority: Lessons from England and New Zealand’. Phi Delta Kappan, 78: 626–32. Witte, Els (1992). ‘Belgian Federalism: Towards Complexity and Asymmetry’. West European Politics, 15: 95–117. Wolf, Adam (2000). ‘Trends in Public Administration—A Practitioner's View’. International Review of Administrative Sciences, 66: 689–96. Wolfe, Amy, Kerchner, Nichole, and Wilbanks, Tom (2001). ‘Public Involvement on a Regional Scale’. Environmental Impact Assessment Review, 21: 431–48. Wollmann, Hellmut (2001). ‘Direkte Demokratie in den ostdeutschen Kommunen— Regelungsschub und Anwendungspraxis’, in Hans-Ulrich Derlien (ed.), Zehn Jahre Verwaltungsaufbau Ost—eine Evaluation. Baden-Baden: Nomos. Wright, Vincent (1994). ‘Reshaping the State: The Implications for Public Administration’. West European Politics, 17: 102–38.
REFERENCES
303
Wuarin, Louis (1895). ‘Recent Political Experiments in the Swiss Democracy’. Annals of the American Academy of Political and Social Science, 6: 361–80. Young, Iris Marion (2000). Inclusion and Democracy. Oxford: Oxford University Press. Zolo, Danielo (1992). Democracy and Complexity: A Realist Approach. Cambridge: Polity Press.
This page intentionally left blank
Index Aboriginal Rights to Self-Government (Canada) l54 ACLU 198 Act Respecting Environmental Rights in Ontario (Canada) 211 Administrative Procedures Act (Japan) 175, 178 Administrative state (reforms) 164–89; citizens’ charter and 166–7; multi-service centers 167–8; neo-liberal 164, 165–76; procedural simplification 168–9; surveys of citizens’ satisfaction 168 adjudication 197–8 advocacy democracy 10, 11, 13, 18, 252, 254–6, 262–5, 267–9, 271–2 Alternativ Movement (Germany) 8 American Jewish Congress 198 American Sierra Club 44 amount of electing 4, 25–7, 35 APÖ (Germany) 8 Applegate, John 178 Arnold, R. Douglas 117 Assemblée Nationale (France) 91 Assembly of European Regions 24 Association of Finnish Regional and Local Authorities 188 Australia 34, 50, 91, 154; administrative reform and 167–9; constitutionalism and 207; courts in 207, 214–15; freedom of information and 119; interest groups and lobbies 214–16; referendums and 51; separation of powers and 195–6 Australian Democrats 68 Austria 31, 34, 62, 92; access to ballot and 90; administrative reform 168, 169; civil law and 201; freedom of information and 120 Austrian Freedom Party 62 Balanced Budget Act (US) 155 Barber, Benjamin 2 Beirle, Thomas 174 Belgium 26, 31, 62; administrative reform and 168; decentralization and 143–5; judicial system in 213; media access and 92 Bennett, Colin J. 118 Berlusconi, Silvio 121 Bill of Rights Act (New Zealand) 207 Black Panthers 136 Blair, Tony 151, 174 Blom-Hansen, Jens 150 Brandt, Willy 8 British Conservative Party 65, 67, 167 British Department of Transportation, Environment and Regions 188
British Liberal Democrats 68 British Official Secrets Act 116 Budge, Ian 57 bureaucracy 164 Burke, Edmund 4, 11 Butler David 49 Canada 5, 91; administrative reform and 169; environment 210–11; freedom of information and 119; judicial systems 195, 196, 198–200, 211 Canadian Broadcasting Corporation 91 Canadian Conservative Party 68 Canadian Environmental Assessment Act 211 Canadian Progressive Conservative Party 71 Caul, Miki 38 Centrelink 168 Chamberlain, Will 136 Chandler, J. A. 35, 178, 180 Charter School Movement (US) 156 citizen involvement 3, 47, 140, 179, 250 citizen juries 179–81, 275
306
INDEX
citizen resources/skills 8, 13, 18, 37, 115 Clarke, Michael 166 Code of Practice on Written Consultation (UK) 174–5 collaborative learning 177 Commission d'access aux documents administratifs (CADA) (France) 128–30, 136, 137 Community Advisory Committees (CACs) 178 Comparative Manifesto Party Project 60–1 Consensus Conferences 180–1 constitutionalism 197–203 cooperative federalism (Australia) 154 corporatism 7, 143, 149–51, 153 courts 18, 192–217, 269, 272 Crewe, Ivor 25 Crosby, Ned 179 Crozier, Michel 225–6 D'Hondt rule 86 Dahl, Robert 2, 8, 228, 256 Dahrendorf, Ralf 2 Danish Board of Technology (DBT) 181 Danish Center Democrats 68 Danish Conservatives 62 Danish Socialist People's Party 69 De Gaulle, Charles 7 DeVries, Michael S. 174 decentralization 18, 25, 140–63; education and 155–9 deliberative accountability 178–9 Delivering Better Government (Ireland) 169 democracy 1; advocacy 10, 11, 13, 18, 252, 254–6, 262–5, 267–9, 271–2; courts and 192–217; definition 19, 226–7; deliberative 179–88; direct 2, 3, 4, 5, 6, 8, 10, 11, 14, 15, 17, 23, 44, 57, 81, 164, 223, 224, 225, 227; elections and 23, 35–40, 44, 55–7, 252–3, 256–7; inclusion and 224, 229–33, 257; individual and cultural transformations and 225; liberal democracy 224; political legitimacy and 195, 269–70; representative 1, 2, 5, 6, 8, 10, 14, 15, 17, 23, 25, 57, 180, 223, 224, 225, 227, 252–253 Democratic Party (US) 9, 62 Democrats' 66(Netherlands) 69 Denmark 26, 62; decentralization in 149; judicial system in 213; new participatory tools in 188 doctrine of parliamentary sovereignty 195–6 Doyle Michael 137 Duverger, Maurice 87 Ecole des Sciences Politiques 7 Elections 23, 35–44, 55–7, 88; access to media and 87–93; ballot access and 87–93; campaign financing and 87–93; electoral laws and 81, 82, 86; electoral institutions 81, 93–5; funding of 10, 17, 23; media and 88; non-party 44–58; parties and 82; platforms 61–75; referendums 29; turnout 1, 38, 39 electoral systems 6, 8, 9, 81; electoral cycles and 26, 29, 33;
electoral fatigue and 38–9 elites 1, 8, 23, 82 Environmental Impact Act (UK) 11, 12, 18 Environmental Protection Act Ontario (Canada) 211 environmentalism 8, 11 Erfurt Programme (Germany) 6 European Charter of Human Rights (ECHR) 200, 205–7 European Court of Justice (ECJ) 12, 200, 203–5 European Parliament 25, 26, 33, 36, 131, 133 European Union 3, 8, 25, 26, 28, 29, 31, 35, 36, 50, 200; courts and 100–203; democratic deficit and 37; freedom of information and 131–4, 138; network governance and 121; supranational governance and 24 Farrell, David 88 Fascist movements 7 Federal Advisory Committee Act (1972) 11 Federal consensual states 152–3 Federal majoritarian states 153–5 Feltey, Kathryn M. 176
INDEX
Finland 10, 31, 69; administrative reform and 168; ballot access and 90; decentralization in 149; freedom of information and 119 Finnish Greens 63, 69 Finnish Participation Project 188 First Reform Bill (US, 1832) 5 Fisheries Act Ontario (Canada) 211 Fleet Street Casuals (UK) 212 Flemish Christian People's Party (CVP) 69 France 7, 26, 91; civil law and 201, 202; decentralization and 145; electoral systems and 86; freedom of information and 119, 128–30, 138; new participatory tools in 188 Franklin, Aimee 174 Free Speech Movement (US) 8 Freedom of Information Act (Japan) 175 Freedom of Information Act (UK) 12, 174 Freedom of information 115, 116–17, 128–30 French Civil Code 208 French Constitutional Court 208 French Socialist Party 62 Gaebler, Ted 154 Gallagher, Michael 49 Germany 6, 7, 8, 86; administrative reform and 168; civil law and 201, 203; deliberative democracy in 179; freedom of information and 118; judicial system in 202, 213; referendums and 51–3, 56, 69 Giddings, Philip 170 globalization 224, 225 Gotha Programme (Germany) 6 government: ; accountability 3, 6, 14, 115, 134, 137, 140–63, 164, 165, 169; confidence in 1, 165, 243; control of information 116, 117–18; responsiveness 164; secrecy 115, 116; transparency 3, 8, 14, 18, 35, 115, 164, 232 Government Performance and Results Act 174 Gray Panthers 8 Gray, Mark 38, 39 Great Society Social Services (US) 11 Green Party (UK) 12 Greenpeace 213 Gregory, Roy 170 Greve, Carsten 167 Havel, Vaclav 4 Held, Daniel 67 Hendriks, Frank 152, 177 Hibbing, John l3 Huntington, Samuel 225–6 Iceland 9 Inclusion 224, 229–33, 257 Initiative and Referendum Institute (US) 10 International Ombudsman Institute 170 Ireland 6, 145; administrative reform and 168–9; judicial system in 213
307
Irish Fine Gael 62 Irish Labour Party 68 Irvine 25, 26 Italy 2, 8, 86; administrative reform and 167, 168, 169; civil law and 201; decentralization and 145; European Union and 121; freedom of information and 120–2, 138; judicial system 213; referendums and 37, 44, 47–54 Japan 30, 31, 32, 62, 86, 208; administrative reforms and 169; decentralization and 145 Japanese Decentralization Promotion Law (1995) 145 Japanese Liberal Democratic Party 71 Jespersen, Peter Kragh 167 judicial review 192, 195–6, 199, 201, 203, 207, 216–17, 269 Karsten Joerd 159 Katz, Richard 64, 65, 87 Kempton, Murray 136 Kettl, Donald 166 King, Martin Luther 136 King, Sheryl S. 176 Labour Party (UK) 12, 56 Lennon, John 136 Lijphart, Arend 141–3, 153, 159 Livingston, Ken 56 Local Agenda 21 (Germany) 188 Local Government (Denmark) 149
308
INDEX
Local Government Act (New Zealand) 151 Local Government Act (UK) 12, 55 Loka Institute 181 Long, Rebecca 174 Maastricht Treaty 202 Mair, Peter 49, 64, 65, 87 Major, John 167 Marbury vs. Madison 196 Marquand, David 131 Marshall Islands 54 McCain, John 13 McGarity, Thomas 178 Meny, Yves 13 Micronesia 54 Morris, Dick 2 NAACP 198 Napoleonic states 142, 147–9 Nathan, Richard 154 National Performance Review (US) 154 national self-determination principle 7 Netherlands 8; access to ballot and 90; administrative reform and 168; corporatism and 143; judicial system 213 New Public Management (NPM) 152, 154, 165–6, 177 new social movements 227 New Zealand 30, 50, 55, 62, 91, 207–8; electoral system and 82; freedom of information and 119; referendums 49 Nikos, Michalopoulos 167 Norway 31, 34, 36, 69, 92; administrative reform and 168; ballot access and 90; decentralization in 149; freedom of information and 119 OECD 2, 3, 8, 11, 14, 25, 47, 59, 81, 86, 89, 118, 165; decentralization and 140–63; freedom of information and 134, 138 Official Secrets Act (UK) 11, 12, 18, 116 ombudsman 169–74 Osborne, David 154 Painter, Martin 154 Palau 54 Palerm, Juan 176 Participation project (Finland) 177 Patriation of the Constitution (Canada) 154 Petition Committee (Germany) 169 Petts, Judith 178 policy formation/administration 9, 10, 11, 18, 35 political opportunities new 24, 26, 27, 35, 179–88 political participation 1, 5, 8, 34, 45, 250; institutional reforms and 3, 4, 5, 6, 8, 12, 13, 14, 19; means of 233–40; new forms of 3, 8, 11, 13, 24, 179–88, 192, 226, 227–8; scope of 240–4; voting turnout 1, 26, 32–3, 233–6, 260, 266–7 political parties 6, 9, 10, 17, 44, 57, 81–9; access of new parties 17, 81, 87, 89; catch-all 64–7; corporatist, groups within 67–75; democratization within 17, 59–66; leadership
66–74; structure of 64–7 Popkin, Samuel 35 populist movements 5, 7, 10, 12 Powell, G. Bingham 38 Pratchett, Lawrence 166 PRI (Mexico) 88 Public Service Management Act (Ireland) 174 RAF (Germany) 8 Ranney, Austin 68 REACH (Ireland) 168 Reeves, Dory 178 Regulatory negotiation 176, 177 Reif, Karlheinz 36 Republican Party (US) 10, 154–5 rights politics 197–208 Rousseau, Jean Jacques 25, 256 Russia 88 Schattschneider, E. E. 59, 82 School Based Management (SBM) 155–6 Schumpeter, Joseph 8, 252, 258 Scottish Parliament 12 Seidenfeld, Mark 177 separation of powers 193, 195–7 sequential consultation (Australia) 177 Service Canada 167–8
INDEX
Sierra Club vs Morton 210 Sieyes, Abbe de 4 Simplifying Norway 168 Social Democratic Party (Germany) 6, 26 Social Democratic Party (Sweden) 62 Social Democratic Party (UK) 12 subsidiarity 231 suffrage 5, 6, 10, 23, 32 Sussel, O'Neill Bridge 176 Sweden 10, 26, 29, 30, 34; decentralization in 149; freedom of information and 118–19 Switzerland 31, 32, 36, 92; ballot access and 90; freedom of information and 118; referendums and 49, 50, 51, 57 Teske, Nathan 178 Thatcher, Margaret 65, 151 Tiebout, C. 140 treaty-based European law 195–6 Treaty of Amsterdam 50, 131 Treaty of Rome 203 Treschel, A. 13 UK Health Service Commissioner 169 UK Parliamentary Ombudsman 169 Unfunded Mandates Reform Act (US) 155 unitary majoritarian states 151–2 United Kingdom 6, 10, 11–12, 53, 54, 56, 254; ballot access and 91; decentralization and 145; freedom of information and 120; judicial system and 195, 207; new participatory tools in 188; secrecy and 116 United States 5, 8, 9, 11, 13, 29, 30, 31, 35, 36, 62, 86; administrative reform and 168–9; deliberative democracy in 179–80; environmental protection and 210–1; freedom of information and 119, 124–8, 138; judicial system and 192, 196, 197–8; political campaigns and 88 US Administrative Procedure Act 174 US Bill of Rights 196, 198, 252 US CIA 126 US Constitution 196 US Department of Justice 124 US Department of Veterans’ Affairs 124 US Equal Employment Opportunity Commission (EEOC) 126 US Federal Advisory Committee Act 178 US Federal Trade Commission 126 US National Aeronautics and Space Administration 126 US Social Services Administration 124 US Supreme Court 196 USAID 135 Vietnam War 8 voting age population (VAP) 32 voting turnout (see participation) VVD (Netherlands) 69 Walberg, Herbert 155
Warren, Mark 2 Watanuki, Joji 225–6 Welsh National Assembly 12 Whaling Association (Japan) 211 women's rights 11, 243 World Development Movement 212 World Wildlife Fund 213 Yukon Territory's Environmental Act (Canada) 211
309