Activating the Citizen: Dilemmas of Participation in Europe and Canada

Activating the Citizen Dilemmas of Participation in Europe and Canada

Edited by

Joan DeBardeleben and Jon H. Pammett

Activating the Citizen

This page intentionally left blank

Activating the Citizen Dilemmas of Participation in Europe and Canada

Edited by

Joan DeBardeleben Professor, Institute of European, Russian and Eurasian Studies Carleton University, Canada

and

Jon H. Pammett Professor, Department of Political Science Carleton University, Canada

Editorial Matter, Selection and Introduction © Joan DeBardeleben and Jon H. Pammett 2009 All remaining chapters © respective authors 2009 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2009 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN-13: 978–0–230–57594–3 hardback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Activating the citizen : dilemmas of participation in Europe and Canada / edited by Joan DeBardeleben and Jon H. Pammett. p. cm. Includes bibliographical references and index. ISBN 978–0–230–57594–3 (alk. paper) 1. Political participation—Europe. 2. Elections—Europe. 3. Political participation—Canada. 4. Elections—Canada. I. DeBardeleben, Joan. II. Pammett, Jon H., 1944– JN40.A285 2009 323’.042094—dc22 2009013297 10 9 8 7 6 5 4 3 2 1 18 17 16 15 14 13 12 11 10 09 Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne

Contents List of Tables List of Figures List of Appendices List of Abbreviations Notes on the Contributors Acknowledgments

vii ix x xi xii xvii

Introduction Joan DeBardeleben and Jon H. Pammett

Part I

Defining the Participation Gap

1

15

1 Citizen Participation and Democratic Deficits: Considerations from the Perspective of Democratic Theory Mark E. Warren

17

2 The Decline of Political Participation: An Empirical Overview of Voter Turnout and Party Membership Alan Siaroff

41

Part II

61

Voter Turnout: Understanding the Decline

3 Turnout in Electoral Democracies Revisited André Blais and Agnieszka Dobrzynska

63

4

Where Turnout Holds Firm: The Scandinavian Exceptions Jørgen Elklit and Lise Togeby

83

5

New Members, Old Issues: The Problem of Voter Turnout in European Parliament Elections Joan DeBardeleben and Lawrence LeDuc

Part III

Parties as Vehicles of Participation

6 Party Membership and Activism in Comparative Perspective Paul Whiteley

v

106

129 131

vi

Contents

7 Models of Party Organization and Europarties Luciano Bardi and Enrico Calossi 8

Grassroots Participation and Party Leadership Selection: Examining the British and Canadian Cases William Cross and John Crysler

Part IV

Beyond Elections and Parties: Innovations in Activating Citizens

9 Participation and the Good Citizen Jon H. Pammett

151

173

195 197

10

Institutionalizing Participation through Citizens’ Assemblies Jonathan Rose

11

Citizen Involvement in Constitutional Politics: European and Canadian Experiences Lawrence LeDuc

233

Youth Engagement, Civic Education, and New Vehicles of Political Participation Eileen Saunders

257

12

Bibliography Index

214

276 297

List of Tables 2.1 Turnout levels in parliamentary elections, decade averages 2.2 Levels of party membership 3.1 Mean turnout (%) in 451 democratic elections by continents, decades and level of democracy, 1972–2004 3.2 The determinants of electoral participation: socio-economic environment 3.3 The determinants of electoral participation: socio-economic environment and institutional setting 3.4 The determinants of electoral participation: socio-economic environment, institutional setting and party system 4.1 Political interest (national level) by country 4.2 Party policy images, Denmark 2001. Voters’ placement of political parties on three scales 4.3 Is voting a civic duty? Voter attitudes by country and age 4.4 Turnout in recent elections in Denmark 4.5 Socio-demographic turnout determinants among Danish citizens at national and local levels, municipality of Aarhus, Denmark 4.6 Comparison of the relative impact of the main factors on turnout differences in the three Scandinavian countries 5.1 Electoral turnout in European Parliament elections, 1979–2004 5.2 Decline in European Parliament election turnout, by accession cohorts 5.3 Turnout in EU new member states’ national parliamentary elections since the collapse of communism 5.4 Turnout in national and EP votes and accession referendums 6.1 Socioeconomic characteristics of party members and activists 6.2 The contribution of party members and activists to civic engagement 6.3 The multi-level model of participation with party activism and membership variables 7.1 Party models and their characteristics 7.2 Party functions related to participatory aspects 8.1 Characteristics of the leadership selection processes in UK and Canadian political parties vii

43 52 66 67 69 71 93 94 95 97

100 102 108 109 117 119 135 141 144 157 161 184

viii

List of Tables

8.2

Parties’ electoral context at time of adopting direct election of the party leader in UK and Canadian political parties 8.3 Participation rates and member recruitment rules for leadership contests in UK and Canadian political parties, 1994–2008 9.1 Duties of the good citizen, pooled ISSP data 9.2 The good citizen: factor structure 9.3 Correlations between political actions in pooled ISSP data 9.4 Factor structure of participation, pooled ISSP data 9.5 Participatory patterns for dimensions of good citizenship 10.1 Age distribution of Ontario and BC assembly members and population 10.2 Age distribution of Dutch civic forum 10.3 Ontario CA members’ views of what contributed to the success of the assembly

186

189 201 202 206 208 209 218 218 223

List of Figures 3.1 4.1 4.2 4.3 5.1 6.1 11.1 11.2 11.3 11.4

The overall distribution of turnout in 451 democratic elections (1972–2004) Turnout in national elections in Scandinavia and Finland, 1975–2007 Turnout among Danish citizens in the municipality of Aarhus for national elections, 1960 and 2001 Turnout in Danish elections at four levels, 1981–2007 Turnout in European Parliament elections, 1979–2004 Party membership, including party activists, in 36 countries in 2004 Public opinion in the French referendum on the EU constitutional treaty, December 2004–May 2005 Public opinion in the Canadian constitutional referendum, August–October 1992 Public opinion in the Danish referendum on the Euro, March–September 2000 Public opinion in the Quebec sovereignty referendum, March–October 1995

ix

65 89 90 98 106 134 239 243 246 247

List of Appendices 3.1 Democratic elections 1972–2004 and participation rate 3.2 Variables, indicators and sources 6.1 Definition of variables in the multi-level model

x

74 79 147

List of Abbreviations BC CA EC EP EU ISSP MMP MP NAFTA NDP OECD PR SMP UK US

British Columbia Citizens’ assembly European Community European Parliament European Union International Social Survey Programme Mixed-member proportional Member of Parliament North American Free Trade Agreement New Democratic Party Organization for Economic Cooperation and Development Proportional representation Single-member plurality United Kingdom United States

xi

Notes on the Contributors

Editors Joan DeBardeleben is Chancellor’s Professor in the Institute of European, Russian, and Eurasian Studies at Carleton University in Ottawa. She is also Director of the Centre for European Studies. She is the editor and contributing author to numerous volumes, including most recently The Boundaries of EU Enlargement: Finding a Place for Neighbours (2008), Democratic Dilemmas of Multilevel Governance: Legitimacy, Representation and Accountability in the European Union (2007, co-editor with Achim Hurrelmann), and Soft or Hard Borders: Managing the Divide in an Enlarged Europe (2005). Her recent publications deal with EU–Russian relations and with Russian politics, particularly relating to democratization, elections, public opinion, and federalism. Jon H. Pammett is Professor of Political Science at Carleton University in Ottawa. He is also co-Director of the Carleton University Survey Centre and Canadian delegate to the International Social Survey Programme. He is co-author (with Harold D. Clarke, Jane Jenson, Lawrence LeDuc) of Political Choice in Canada (1979) and Absent Mandate (1996), books about voting behaviour in Canadian elections. He is co-editor of and contributing author to numerous books on Canadian politics including (with Christopher Dornan) The Canadian Federal Election of 2008 (2009), as well as several volumes in this series on earlier elections. He has published journal articles on public opinion and voting behaviour in Canada and Russia and has also worked in the field of political education. He is currently engaged in a study of voting turnout in Canada as compared to other countries, and is also writing a book on Canadian elections through history.

Authors Luciano Bardi is Professor of International Relations, European Political Systems, and Comparative Politics at the University of Pisa. His previous appointments were at the European University Institute in Florence and at the University of Bologna. He has been involved in numerous xii

Notes on the Contributors xiii

international research projects including the First MEP Survey (as general coordinator of field work) and a project headed by Richard S. Katz and Peter Mair on party organization (as project director for Italian and transnational parties). He has written extensively in English and Italian on political parties in general and on European Union parties in particular. His most recent publications include Il Parlamento Europeo (2004, co-author with Piero Ignazi); Partiti e sistemi di partito (2006 editor); I partiti italiani. Iscritti, dirigenti, eletti (2007, co-editor with Piero Ignazi and Oreste Massari); and Dimensioni e dilemmi della sicurezza nel Mediterraneo (2008, co-editor with Federica Bicchi and Serena Guiusti). André Blais is Professor of Political Science at the Université de Montréal and holds a Canada Research Chair in Electoral Studies. He is a research fellow with the Centre for the Study of Democratic Citizenship, the Centre interuniversitaire de recherche en économie quantitative (CIREQ), and the Centre for Interuniversity Research Analysis on Organizations (CIRANO), as well as a fellow of the Royal Society of Canada. He is past president of the Canadian Political Science Association and a coinvestigator of the Canadian Election Study since 1988. His research interests are elections, electoral systems, voting, turnout, public opinion, and methodology, and he has published numerous books and articles on these issues. Enrico Calossi is a Doctoral Candidate at the IMT (Institutions Markets Technologies) Lucca Institute for Advanced Studies and he collaborates with the Department of Political Science at the University of Pisa. He has been involved in research on party organizations. He contributed the chapter ‘Rifondazione Comunista e Comunisti Italiani’ to I partiti italiani. Iscritti, dirigenti, eletti (2007, edited by Luciano Bardi, Piero Ignazi and Oreste Massari). He has attended several Italian and international conferences and presented papers on political parties, Italian military missions, and political participation. William Cross holds the Dick and Ruth Bell Chair in Canadian Parliamentary Democracy in the Department of Political Science at Carleton University. He is a leading expert on Canadian and comparative political parties and election campaigns. His work has appeared in many academic journals and his most recent book is Political Parties (2004), part of the Canadian Democratic Audit series. From 2003 to 2005 he served as Director of Research for the New Brunswick Commission on Legislative Democracy.

xiv

Notes on the Contributors

John Crysler is a Doctoral Candidate in Political Science at Carleton University and holds a doctoral fellowship from the Social Sciences and Humanities Research Council. He is currently researching his dissertation, which is a multi-country comparative study on the presidentialization of political parties. Agnieszka Dobrzynska recently joined the Office of Institutional Research at Université de Montréal as a Research and Planning Analyst. A former postdoctoral fellow with the Canada Research Chair in Electoral Studies at the Université de Montréal, she was also a consultant for Elections Canada. Her most recent publications are on the topics of elections and voting behaviour, public opinion, media coverage of politics, electoral systems, and research methodology. Jørgen Elklit is Professor of Political Science at the University of Aarhus. His current primary professional interests relate to electoral systems and election management in both established and emerging democracies. Since 1990 he has also been engaged in advisory missions on elections and democratization in 20 countries in Africa, Asia, and Europe, most recently as Secretary to the Independent Review Commission in Kenya. His other main field of expertise is Danish politics, where he has worked on a number of topics, including political party membership and the current process of reforming Danish local government structure. He has written, contributed to, and edited a number of books and professional articles. Lawrence LeDuc is Professor of Political Science at the University of Toronto. His books include The Politics of Direct Democracy (2003), Comparing Democracies (2002, with Richard G. Niemi and Pippa Norris) and Absent Mandate: Canadian Electoral Politics in an Era of Restructuring (1996, with Harold D. Clarke, Jane Jenson, and Jon H. Pammett). He is a member of the editorial boards of Electoral Studies, the Journal of Elections, Public Opinion and Parties, and the European Journal of Political Research, as well as a member of the IDEA expert group on Direct Democracy. Jonathan Rose is Associate Professor of Political Studies at Queen’s University in Kingston, Ontario, where he teaches and researches Canadian politics and political communication. He is author of Making ‘Pictures in Our Heads’: Government Advertising in Canada (2000), co-editor (with Douglas Brown) of Canada: the State of the Federation (1995), and has published numerous articles on political advertising and the mass media in Canada. In 2006–2007 he served as the Academic Director of the Ontario Citizens’ Assembly on Electoral Reform.

Notes on the Contributors xv

Eileen Saunders teaches in the School of Journalism and Communication at Carleton University in Ottawa. She served as the founding Director of the Arthur Kroeger College of Public Affairs at Carleton University and Associate Dean of the Faculty of Public Affairs between 1997 and 2007. Her research interests are in the fields of youth, media and citizenship, social inequality and gender, and media institutions and policies. She is the author of numerous publications concerning such topics as youth and civic engagement, the representation of youth in mainstream media, public opinion and social inequality, and the access of visible minority groups to news media. Her current research focuses on the potential of social networking sites for civic engagement. Alan Siaroff is Professor of Political Science at the University of Lethbridge. He is the author of Comparing Political Regimes: A Thematic Introduction to Comparative Politics (now in a second revised edition 2008), Comparative European Party Systems: An Analysis of Parliamentary Elections Since 1945 (2000), and various articles and book chapters on democratization, electoral systems, party politics, and political institutions. He has also provided expert advice on electoral reform for government commissions in both New Brunswick and Quebec. Lise Togeby was Professor of Political Science at the University of Aarhus in Denmark. She was the Director of the PhD programme of the department and a member of the Danish Committees on Scientific Dishonesty. From 1997 to 2004 she was Chairman of the Steering Group of ‘the Danish Power and Democracy Study’ (Magtudredningen), a major project sponsored by parliament. During the 2007 spring term she was a member of one of the international panels evaluating all academic research at Uppsala University. Her recent publications include ‘The Political Representation of Ethnic Minorities – Denmark as a Deviant Case’ in Party Politics (2008), and (with P. M. Christiansen) ‘Power and Democracy in Denmark: Still a Viable Democracy’ in Scandinavian Political Studies (2007). Professor Togeby died in October 2008. Mark E. Warren teaches political theory at the University of British Columbia, where he holds the Harold and Dorrie Merilees Chair for the Study of Democracy. He is especially interested in the relationship between new forms of citizen participation and democratic representation, civil society and democracy, political corruption, democracy, and deliberation. He is author of Democracy and Association (2001), editor of Democracy and Trust (1999), and co-editor (with Hilary Pearse) of Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (2008).

xvi Notes on the Contributors

His work has appeared in journals such as the American Political Science Review, the American Journal of Political Science, and Political Theory. Paul Whiteley is Professor of Government at the University of Essex. His current research interests include political parties, electoral behaviour, political economy, and methodology in the social sciences. He is author or co-author of sixteen academic books including The Labour Party in Crisis (1983); Political Control of the Macroeconomy (1986); Pressure for the Poor: The Poverty Lobby and Policy-Making (1987); Labour’s Grassroots: The Politics of Party Membership (1992); True Blues: The Politics of Conservative Party Membership (1994); High Intensity Participation: The Dynamics of Party Activism in Britain (2002); Political Choice in Britain (2004) and Citizenship in Britain: Values, Participation and Democracy (2004). He is the author or co-author of some fifty articles in political science journals. He was director of the Economic and Social Research Council Research Programme on Democracy and Participation from 1998 to 2003 and has been co-director of the British Election Study since 2001. He is a regular contributor to politics programmes on BBC television and radio and has written numerous columns for newspapers such as the Guardian and the Independent.

Acknowledgments This volume is a publication of the Centre for European Studies (CES) at Carleton University in Ottawa. The centre is grateful for financial support to enable completion of this volume from Carleton University and from the European Commission, the latter based on a grant to CES, which was designated an EU Centre of Excellence by the European Commission in 2006. We also acknowledge support from the Social Sciences and Humanities Research Council (SSHRC) of Canada for supporting research that formed the basis of some of the chapters in the volume. We are particularly grateful to our copy-editor, Erin Lyndsey Seatter, who contributed countless hours and unwavering attention in preparing each and every chapter. We express our gratitude to our colleague William Cross for his involvement in conceptualizing the project, as well as to the anonymous reviewers who offered insightful comments and suggestions. We thank all of the authors for the collective effort that went into producing a coherent approach and for their attention to our many suggestions, requests, and inquiries. We dedicate this volume to one of our authors, Lise Togeby, who, sadly, passed away before she could see this work in print. We remember her as a scholar who was dedicated not only to the research endeavour but also to spreading the word about Denmark’s success in maintaining a high level of citizen participation in an era when it is declining elsewhere. We hope that this volume will inspire both reflection and reforms to help bolster the most important democratic resource that our societies possess, namely the committed interest and involvement of average citizens in the political process. Joan DeBardeleben and Jon H. Pammett

xvii

This page intentionally left blank

Introduction Joan DeBardeleben and Jon H. Pammett

A crisis of public confidence in the efficacy of political participation plagues democracies in the Western world. The widely discussed ‘democratic deficit’ affects attendance at the ballot box, political party membership, and other forms of civic activism. Voter turnout has fallen consistently and significantly in national elections in many of the established democracies of Europe and North America and in the new democracies of Central and Eastern Europe. These regions are not alone in suffering from a withdrawal of public confidence and participation. Overall, world voting turnout has declined almost five points in the period after 1990 (IDEA, 2006). Political parties have also suffered membership declines and a reduced capacity to recruit activists, as scepticism towards parties in general has grown in many countries and as party structures have had difficulties in adapting to new challenges (Webb, Farrell, and Holliday, 2002; Gunther, Montero, and Linz, 2002). While many citizens view parties as unresponsive and elections as ineffective in bringing politicians to account, these are still the most fundamental vehicles that citizens have to assert influence over their political leaders. While citizen activity is certainly not absent in other areas, it has not picked up the slack left by a withdrawal of the public from the arena of electoral and party politics. This volume seeks not only to document declines in political participation, but also more importantly to examine explanations and innovative mechanisms to address the declines. While not irrational, the withdrawal of the citizenry from participation in public life has the potential to reduce the legitimacy of public decision making and to inhibit effective responses to collective problems. Abstention from participation in most cases does not signify satisfaction with elite decisions; on the contrary, public opinion data show that attitudes towards politicians and 1

2

Activating the Citizen

political parties are increasingly negative on a range of measures in many countries (Semetko, 2006). Furthermore, if the public is not engaged by politics, people will feel less committed to supporting the decisions that governments make. In addressing key policy problems the willingness of the public to work together and make sacrifices to implement policies will be critical to their success. This need is particularly evident in the environmental arena, where public compliance with regulations may require adapting habits and where high levels of social and political trust are needed to motivate such change. Likewise, in the face of increasing inequality and the retrenchment of the welfare state, a public willingness to share burdens and realize a more equitable distribution of resources will be difficult to attain if citizens do not feel that they have had their say. Activating the citizen, therefore, is essential to putting the ‘public’ back into ‘public life’. A comparative method offers the most promising avenue for understanding the dynamics underlying the ‘participation deficit’. Given the pervasiveness of participation declines, we have plenty of cases to work with to feed such comparisons. More problematic is the dearth of success stories where high levels of involvement have been maintained. But they do exist and they also deserve study as part of a comparative approach. This book takes a two-track approach. On the one hand, we examine differences in how the participation decline plays out in particular political jurisdictions, specifically in European countries, Canada, and the European Union (EU). This track allows us to pay close attention to institutional and systemic factors that may facilitate or depress participation, by comparing aggregate level outcomes and seeking to explain them. As a second track of analysis, our authors, where appropriate, use survey data to try to see what influences individual citizens to act or not act and to determine whether these factors are fairly universal or whether they vary by country, interacting in suggestive ways with institutional features.

Comparisons between countries and EU participation This volume is primarily focused on the countries of the EU. This universe of cases includes a large number of established democracies as well as some new ones; it also permits us to explore linkages between the national level and the EU. We also include Canada in our comparative framework. Why include Canada alongside the range of European cases? The explanation has to do with a number of features that make the Canadian system similar to European ones. First, Canadian political institutions, at both the provincial and national levels, are marked by their

Joan DeBardeleben and Jon H. Pammett 3

European origins and share with most European states the parliamentary system of governance, including attendant features of the political party system. To be sure, there is some diversity within Europe, notably the French semi-presidential system, but even there governments are formed on the basis of parliamentary outcomes, as is typical of the Westminster model. These institutional similarities are important in relation to public participation because they affect the vehicles through which citizens can hold governments and politicians accountable and affect outcomes. Second, like many European democracies, Canada has a multiparty system, and its models of party membership are closer to the European pattern than to the American one. Because of similarities in both of these arenas of institutional structures, innovations in Canada to attempt to address the participation and democratic deficits may be of particular applicability to Europe and vice versa. Not surprisingly, similar changes have been attempted in the Canadian and European contexts, some of which are addressed in the final section of this volume. These include citizens’ assemblies, national referendums, and reform to party leadership selection procedures. Third, methods applied for measuring voter turnout and party membership in Canada fall within the European standard, facilitating comparison with European cases. An additional feature of Canadian political life makes a comparison with the European context particularly fruitful. This is the multi-level character of both systems, which we turn to in the next section. We have made a deliberate decision not to include comparisons with the United States in this book, a decision based on measurement differences in our key variables of voter turnout and political party membership. Differences in how voter turnout is measured could generate invalid comparisons because of the American system of state-based voter registration. Not only is registration voluntary and subject to periodic re-registration requirements, but states differ in their criteria for eligibility, for example for those previously convicted of crimes. These differences between the states are so severe as to motivate American researchers to measure turnout in terms of the percentage of the votingage population, rather than the percentage of those registered to vote. In addition, the American context of political participation is focused around presidential elections (with its complex Electoral College system), which don’t have a corollary for citizens in Canada or most European countries. Participation in off-year elections (those in which the presidency is not up for election) is often considerably lower, as it is in a variety of state and local offices that are on the ballot at the same time. The extensive American system of primary elections to choose the party

4

Activating the Citizen

candidates for all public offices further complicates the situation, since it creates an entirely different context for voter decision making; voters in primaries are sometimes considered members of parties because they participate in choosing the parties’ candidates. Inclusion of the United States in our comparative framework would inevitably have shifted the focus towards the differences between that country, on the one hand, and Europe and Canada, on the other.

Multi-level systems and citizen participation As a federal system with distinct elections, parties, parliaments, and jurisdictions at the national and provincial levels, Canada shares a multilevel character with the EU. Multi-level systems can make voters’ choices more complex, lead to a fragmentation of political party structures, and inevitably generate complex interaction effects between the different levels in the system. Two particular challenges deserve comment. The first one relates to the way citizens make their decisions and whether they think that elections can hold politicians accountable. In the European case, this dynamic works to the detriment of turnout in elections for the European Parliament. Even though most Europeans recognize that the decisions of the EU are very important to their daily lives, they turn out to vote in relatively low and decreasing numbers. There is a disconnect in the mind of the voter between voting and a visible outcome, so elections to the European Parliament continue to have what academics have dubbed a ‘second-order’ character (Reif and Schmidt, 1980). This means that when people cast a vote for the European Parliament they very often are expressing reactions to national-level political contexts, reflecting a judgement about the party in power in their country or a reaction to current domestic debates. European issues are not prominent enough in electoral campaigns or in the choices facing the electorate to elicit high levels of interest. Europe-wide political parties have not emerged to help citizens’ choices or to help citizens understand why elections to the European Parliament matter (Hix and Lord, 1997). The European Parliament seems distant and ineffective as a way for prospective voters to have any influence. The national representatives who are elected through these contests journey to Brussels or Strasbourg and form a bloc with representatives from other countries who speak different languages and are concerned with a wide variety of issues relating to their own national contexts. These representatives then consider legislation that has been constructed by, or heavily conditioned by, the bureaucracy and unelected executive of the EU. At this point, we have travelled quite a

Joan DeBardeleben and Jon H. Pammett 5

long way from what might have been in the mind of the individual voter in, for example, Poland or Slovenia. Low voter motivation to take part in EU contests, given the tenuous links to outputs and the high costs of acquiring information about EU activities involving the parliament, is quite understandable. The dynamic affecting the electorate in Canada is somewhat different, but it is also affected by the multi-level character of the system. Here national votes are of clear primacy in terms of electoral interest, and provincial elections are not ‘second order’ but are generally governed by distinct agendas. But the notable distinction between provincial parties and federal parties makes it difficult for the voter to rely on party identification as a political compass. Furthermore, as Pammett (forthcoming 2009) has demonstrated, the multi-level Canadian system is as likely to provide multiple opportunities for voters to opt out of voting as it is to reinforce opportunities for participation. Many of the characteristics of federal and provincial non-voters are the same, and their reasons for abstaining correlate with each other. Missing a voting opportunity at both levels in a short period of time can help to create permanent nonvoters. Such multiple opportunities are even more present in European countries. In Germany, for example, there could be four elections held in a short period of time: local, land, national, and European Parliament (Steinbrecher, Huber, and Rattinger, 2007). In France, the number of elections is greater, since there are two rounds to the presidential and parliamentary elections, as well as municipal, regional, and cantonal elections (Appleton, 2000). A second dynamic that may affect multi-level systems like Canada and the EU has to do with the shifting and sometimes unclear jurisdictions of various levels of government. Here one might suppose that as more and more competencies are granted to the EU, the capacity of national governments to make a difference is gradually reduced. This dynamic mirrors a more general trend in the face of globalization, where powerful multinational economic actors constrain the ability of governments to control economic forces that directly impact the financial and employment well-being of citizens (Marden, 2003). In the case of the EU, even where governments do act, they are often involved in transposing EU policies into national legislation rather than setting the basic parameters of policy themselves. The existence of the EU does not directly depress turnout in national elections in Europe, but it does contribute to a situation where the decision-making power of national governments is circumscribed. The areas where national governments have ceded decision-making power to the EU include many related to the

6

Activating the Citizen

economy, for coordinated action to develop the European economy was the main initial purpose of establishing the European Economic Community in the first place. Regulations, incentives, and subsidies to the industrial or agricultural sectors are traditionally important to national, state, and even local governments, and the ability to propose such policies in elections has effectively been removed from their control. In short, the more the EU makes a difference to EU citizens, the less ‘first-order’ elections to the national parliaments seem an efficacious tool to directly affect outcomes. Accordingly, domestic issue agendas have shifted away from economic development and towards social policy, including health, education, and welfare, even though the EU is active in attempting to coordinate policies in these areas as well. Citizenship and border issues, including elements of immigration policy, are of national concern, but treaties have established many policies that are out of national control. While it would be overstating the case to say that national and subnational elections in the countries of the EU have had their normal issue agenda pre-empted, it does seem to be true that it has been truncated. People may find voting in these elections less important because the consequences are less significant. In Canada, membership in the North American Free Trade Agreement (NAFTA) has also truncated the power of the national parliament to make effective decisions in the last 20 years (Clarkson, 2002), although to a lesser extent than in the EU case. While NAFTA ostensibly binds all three North American countries to the same economic rules, the dominance of the United States and the ability of that government to dictate economic rules for Canada and Mexico have meant that in all sectors relating to the economy, the die is already cast for the smaller partners. Even if that formal treaty were not in place, the dependence of the Canadian economy on American markets is so well established that there is little public expectation that Canadian governments can affect domestic economic conditions to a significant extent. Another factor plays a role in Canada, as social policy decisions in Canada are constitutionally granted to the provinces with the exception of contributions to their funding. Complicated funding practices engaged in by successive federal governments have further blurred the lines of jurisdiction between the federal and provincial levels on social policy, with the result that neither federal nor provincial governments have been able to establish important new programmes such as child care (with the notable exception of Quebec), despite the tendency of parties to promise action in this area in election campaigns. The election issue agenda in recent Canadian elections has been relatively sparse and small-scale and this in turn may be part

Joan DeBardeleben and Jon H. Pammett 7

of the explanation for low public interest in politics and falling turnout rates.

Political participation and the individual citizen While political systems and institutions can influence levels of participation, individuals’ attitudes and beliefs may also be important in explaining political inactivity. First, how should one interpret political inaction? Is it a sign of satisfaction, alienation, or protest? Some might argue that citizens would participate if they wanted change and that a lower participation rate indicates basic satisfaction with governmental actions and outputs. Most observers, however, see declining participation as more likely reflecting withdrawal or alienation, raising the alarm about the potential negative effects of citizen disengagement. Examining which groups are less engaged and why is important to understanding the meaning and significance of continuing declines in participation. A particularly good example that highlights the importance of analysing individual voter attitudes and behaviour relates to youth. At an aggregate level, we can observe a nearly universal pattern that shows voting rates as strongly associated with age; specifically, young people are markedly less inclined to vote. Using survey methodology, Pammett and LeDuc (2003) estimated that in Canada only about one quarter of eligible voters under 25 years of age actually exercised their right to vote in the 2000 federal election. Furthermore, citizens entering the eligible electorate as far back as the 1970s voted in the 2000 federal election at lower levels than older citizens. The failure of traditional life-cycle factors to bring the participation levels of young people up to average levels could have devastating long-term effects on the participation levels of the entire electorate, as older citizens leave the electorate and are replaced by the new cohorts who are less likely to turn out on a regular basis. Using a different methodology, Elections Canada estimates the voting participation of those under 25 years of age at 37 per cent in 2004 and 44 per cent in 2006, well below the national average (Elections Canada, 2008, p. 9). The association of the turnout decline with age is mirrored in European countries as well, for example in Britain (Clarke, Saunders, Stewart, and Whiteley, 2004). The key to understanding these general patterns lies in the attitudes and motivations of individuals, and the disengagement of the young has in turn raised its own set of explanations. Perhaps youth are just lazy and don’t care about doing their citizen duty because they have had things done for them throughout their lives and have never had to fight for their

8

Activating the Citizen

rights and their material well-being. The ‘me’ generation couldn’t care less about common problems and the resulting political issues around them, some say. The fact that young non-voters often give the explanation that they consider themselves ‘too busy’ to vote seems on its face to give substance to these charges (LeDuc and Pammett, 2006). An alternative interpretation focuses on the inadequate knowledge base on which youth are being expected to act, rendering their failure to participate a result of the pressure of other activities rather than uncaring selfishness (LeDuc, Pammett, and Bastedo, 2008). Young people are not too busy to go a short distance to the polling place to cast a vote; rather, they don’t have the amount of free time that would be required to become fully informed on the important issues in order to make an intelligent choice among the alternatives. Furthermore, issues discussed in elections are often perceived as not relevant to their lives; research indicates that many young people see themselves as deferring action until the future, when they will be established with family obligations and a greater stake in the community. In fact, however, other research suggests that such youthful good intentions are to some extent a delusion. The longer people live, the higher the information costs of catching up and the greater the number of competing activities that continue to defer the process of becoming informed enough to vote. The more voting opportunities missed, the easier it is to continue to abstain. As this example illustrates, examining individual motivations and behaviour patterns is crucial to understanding reasons for the participation decline. Survey data are particularly useful here; comparative survey projects – such as the International Social Survey Programme, carried out in 42 countries including most EU member states and Canada – offer a wealth of information that authors in this volume have used along with materials from other survey research projects.

Mobilization: linking the individual and institutions We are not only concerned with explaining why participation is declining, but also in assessing measures to address these troubling trends. These measures can be of various types, involving, for example, institutional change (e.g., changing voter registration systems or electoral systems), voter education (e.g., advertising or school courses), or increasing political mobilization. Changing electoral institutions or rules is the initial step that often comes to mind when a new outcome is desired, because it offers the promise of a quick fix. If we want a high voting rate,

Joan DeBardeleben and Jon H. Pammett 9

the easiest way to achieve it would be to make voting compulsory, as some countries do. If that is not acceptable, a change in electoral systems is often suggested to improve the representation of all interests in society, thereby changing behaviour through altered incentives. Institutional change, as several of the articles in this book make clear, has its limitations in changing behaviour. Either it comes with undesired side effects or it does not produce the desired outcome. For example, compulsory voting may raise the turnout rate but at the cost of an expensive enforcement mechanism, not to mention the societal debate over the right to non-participation. It is usually very difficult to build coalitions of support for changing electoral systems, and there may be disagreement about the relative tradeoffs of different electoral systems. Just as institutional change may not produce the results expected, it is equally difficult to change the fundamental attitudes that people have about political participation. Attitudes that inhibit participation may be rooted in past experience, either individual or collective, or may reflect habits that are deeply embedded. Research from this volume suggests that political mobilization is particularly important in motivating electoral participation. More than institutional change, mobilization seems a powerful factor in explaining differences and in galvanizing new participation. In some ways, this emphasis takes us back to some of the early studies of participation (e.g., Verba, Nie and, Kim, 1978) whereby the ability of personal resources (time, skill, education) to explain who participated and who didn’t was much stronger in some countries studied than others. In some countries, however, political or social movements propelled into the political arena groups that would not have participated at high levels of their own accord, thus creating a much more even social basis of participation. Mobilization, group activities that organize and lead people to participate, can provide the direct contact with voters that produces behavioural outcomes (Green and Gerber, 2006) Mobilization can also take the form of public education or advertising campaigns, done by electoral agencies or political parties, or events to reach out to target groups of voters (IDEA, 2006). Mobilization is easier if elections and political debates address issues that engage the public and seem relevant to their everyday concerns. One problem with elections to the European Parliament is that they often don’t do that, unlike many national contests. The extent to which elections can involve the public depends on a number of factors, such as the charisma of political leaders, strategic calculations of political parties and politicians, unforeseen events that touch people’s everyday lives, and extensive media coverage. Research also indicates

10

Activating the Citizen

that when elections are more competitive, they are more engaging and elicit higher interest. These are variables that cannot easily be engineered or controlled by electoral commissions or other organizations that may have an interest in increasing participation. However, within these constraints, ‘packaging’ participation opportunities can make a difference, and this has been a strategy consciously used in several initiatives directed particularly at youth, such as student vote programmes in the schools. In addition, reaching out to publics through the Internet or alternate media may be a way to mobilize interest and spur less conventional forms of participation. Mobilization may be particularly effective when it is combined with institutional change or placed within a clear institutional context. An example of a proposal that seeks to make this connection is a reduction in the voting age. This suggestion is important because, as noted earlier, the best way to establish consistently high patterns of turnout in any society is to make sure that its young people vote in the first few opportunities they have after becoming eligible. So important is it to get young people voting that one researcher has advocated reducing the voting age in order to bring initial voting opportunities to the electorate during the high school years, when youth are more likely to be living in the parental household and where school-based activities can stimulate interest and mobilize them to participate (Franklin, 2004, 2006). Lowering the voting age would mean that newly eligible citizens would have contact with mobilizing forces – parents, schools, groups, political parties, media – while they were captive in the school environment, giving a practical method by which institutional change could achieve its objective. Young people would not just be told to go out and vote, but would be pushed and pulled to do so. Peer pressure could also be brought to bear, as well as a range of social networking tools to stimulate interest. Furthermore, the ultimate weapon, guilt, could be readily applied if they did not participate, since they would continue to spend much of their day in close contact with people who were aware of their actions or inactions. This is just one example of a range of suggestions that might be considered to address the participation deficit through a combination of mobilization and institutional adjustment. Other possibilities, which we explore further in this book, have involved more dramatic changes to political institutions. For example, there is a world-wide increase in the number of contentious political questions being put to referendums (LeDuc, 2003), thereby by-passing the normal legislative process, or at least augmenting it. Some referendums do indeed achieve higher levels of participation when compared to

Joan DeBardeleben and Jon H. Pammett 11

elections, but it is not clear that this is consistently true, or that there is a qualitative difference between the two kinds of voting such that referendum voting is an empowering experience for the electorate. Another example involves Internet participation. The Internet has become wellknown as a mobilization tool for some political candidates, but continuing such experiences into sustained political participation has been much more difficult.

The structure of the volume This book is divided into four parts. The first examines the theoretical importance of political participation and the causes of the participation decline in voting and political party membership in Canada and Europe. Mark Warren (Chapter 1) uses the concept of the democratic deficit to measure the gap between the requisites of democratic theory and the mechanisms that are available to the public to act politically. Alan Siaroff (Chapter 2) sets the scene for the ensuing parts of this book by providing a succinct overview of the levels of electoral participation and political party membership and an inventory of eight types of factors that have been found to correlate with these declining levels in Canada and the countries of the EU. Part II is dedicated to a closer examination of voting turnout in Western democracies. André Blais and Agnieszka Dobrzynska (Chapter 3) revisit their article of a decade ago (Blais and Dobrzynska, 1998) to assess the current impact of socioeconomic, institutional, and party system factors in accounting for turnout levels in 91 countries overall and then those of the EU and Canada. Elklit and Togeby (Chapter 4) look at the Scandinavian countries, particularly Denmark and Sweden, to identify factors that maintain overall turnout in these countries at a much higher level than the rest of Europe and that place Denmark in particular at the head of the pack. DeBardeleben and LeDuc (Chapter 5) explore reasons for the lowest turnout extremes, those that exist in European Parliament elections, particularly elections recently held in the newly admitted countries of Eastern Europe. Part III turns the focus to political parties. Paul Whiteley (Chapter 6) uses data from the International Social Survey Programme to show how different those who report belonging to political parties are from the remainder of the population. Those active in parties are model citizens in many ways and embody civic virtues; any decline in party membership will have important effects on the state of social capital. Luciano Bardi and Enrico Calossi (Chapter 7) examine models of party organization

12

Activating the Citizen

and activity to show that Europarties, a name given to the blocs formed in the European Parliament as a result of elections to that body in the member states, are an extreme form of cartel parties, lacking participative structures and failing to connect with the public that is expected to vote for them. William Cross and John Crysler (Chapter 8) concentrate on Canada and Britain, two countries that have taken the lead in expanding the democratization of leadership selection processes in political parties. Part IV is directly concerned with activating the citizen, as it considers some of the steps that have been taken, or at least proposed, to alleviate (or perhaps compensate for) the situation of voting and party decline. In this section, we look, in particular, at innovative approaches to activate citizens that are outside of the traditional frameworks of political party participation and voting. In other words, if electoral turnout and party activism seem to be on a declining trajectory, can citizens be engaged in other ways, which eventually might have a positive impact on these more traditional modes of participation? Because these chapters deal with innovation, they focus on the particular countries or regions where these measures have been tried. Thus, while the Canada–Europe comparative perspective is evident in all of these chapters, it takes a different form in each. Jon Pammett (Chapter 9), using the International Social Survey Programme citizenship data once again, argues that a combination of a service orientation to citizenship and a direct action approach to participation motivates people to think beyond national boundaries and act in ways that expand the conventional participation agenda. Jonathan Rose (Chapter 10) examines the recent Citizens’ Assembly on Electoral Reform in the Canadian province of Ontario as a participatory institution, with comparisons to a similar assembly in British Columbia and one in the Netherlands. Lawrence Leduc (Chapter 11) uses illustrations from European and Canadian constitutional referendums to show how this frequently used device for citizen participation is problematic for achieving results that genuinely incorporate extensive public input. Finally, Eileen Saunders (Chapter 12) puts the spotlight on the political education of young people, many of whom are turning away from conventional vehicles of participation. New networking technologies, she argues, offer some expanded locations and techniques of active citizenship. It is likely that judgements on the efficacy of the changes suggested here, for institutional reform and mobilization enhancement, will be mixed. On the one hand, they are interesting first steps and have some potential for involving segments of society not already actively

Joan DeBardeleben and Jon H. Pammett 13

participating. Furthermore, they could, in principle, have the effect of stimulating revived interest in political parties and elections more generally. For example, if citizen assemblies trigger changes in the electoral system, this could contribute to increased voter turnout. In practice, however, the results have been modest and the scope of these innovative mechanisms may seem on the insubstantial side when compared with the magnitude of the problems of low voter turnout and declining party membership. Nonetheless, we hope that our readers will be stimulated by this book to create more and better ways to activate the citizen.

References Appleton, A. M. (2000) ‘The France that Doesn’t Vote: Nonconsumption in the Electoral Market’ in M. S. Lewis-Beck (ed.) How France Votes (New York: Chatham House), pp. 206–26. Blais, A. and A. Dobrzynska (1998) ‘Turnout in Electoral Democracies’, European Journal of Political Research, 33, 239–61. Clarke, H. D., D. Saunders, M. C. Stewart, and P. Whiteley (2004) Political Choice in Britain (Oxford: Oxford University Press). Clarkson, S. (2002) Uncle Sam and Us: Globalization, Neoconservatism and the Canadian State (Toronto: University of Toronto Press). Elections Canada (2008) Estimation of Voter Turnout by Age Group at the 39th Federal General Election, January 23, 2006 (Ottawa: Elections Canada). Franklin, M. (2004) Voter Turnout and the Dynamics of Electoral Competition in Established Democracies since 1945 (Cambridge: Cambridge University Press). Franklin, M. (2006) ‘You Want to Vote Where Everybody Knows Your Name: Targeting Adults in Need of Electoral Socialization.’ Paper presented at 20th World Congress of the International Political Science Association in Fukuoka, Japan. Green, D. P. and A. S. Gerber (2004) Get Out the Vote! (Washington, DC: Brookings). Gunther, R., J. R. Montero, and J. J. Linz (2002) Political Parties: Old Concepts and New Challenges (Oxford: Oxford University Press). Hix, S. and C. Lord (1997) Political Parties in the European Union (New York: St. Martin’s Press). IDEA (Institute for Democracy and Electoral Assistance) (2006) Engaging the Electorate: Initiatives to Encourage Voter Turnout from Around the World (Stockholm: IDEA). LeDuc, L. (2003) The Politics of Direct Democracy: Referendums in Global Perspective (Peterborough: Broadview). LeDuc, L. and J. H. Pammett (2006) ‘Voter Turnout in 2006: More than Just the Weather’, in J. H. Pammett and C. Dornan (eds) The Canadian Federal Election of 2006 (Toronto: Dundurn), pp. 304–16. LeDuc, L., J. H. Pammett, and H. Bastedo (2008) ‘The Problem of Young Voters: A Qualitative and Quantitative Analysis’. Paper presented at the annual conference of the American Political Science Association in Boston. Marden, P. (2003) The Decline of Politics: Governance, Globalization and the Public Sphere (Aldershot, UK: Ashgate).

14

Activating the Citizen

Pammett, J. H. (forthcoming 2009) ‘Voting Turnout in a System of Multi-Level Governance’ in M. Kanji, A. Bilodeau, and T. Scotto (eds) Four Decades of Canadian Election Studies (Vancouver: University of British Columbia Press). Pammett, J. H. and L. LeDuc (2003) Explaining the Turnout Decline in Canadian Federal Elections (Ottawa: Elections Canada). Reif, K. and H. Schmidt (1980) ‘Nine Second-order National Elections: A Conceptual Framework for the Analysis of European Election Results’, European Journal of Political Research, 8(1), 3–44. Semetko, H. A. (2006) ‘Parties in the Media Age’ in R. S. Katz and W. Crotty (eds) Handbook of Party Politics (London: Sage). Steinbrecher, M., S. Huber and H. Rattinger (2007) Turnout in Germany: Citizen Participation in State, Federal and European Elections since 1979 (Baden-Baden, Germany: Nomos). Verba, S., N. Nie, and J. Kim (1978) Participation and Political Equality: A SevenNation Comparison (Cambridge: Cambridge University Press). Webb, P., D. Farrell, and I. Holliday (2002) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press).

Part I Defining the Participation Gap

This page intentionally left blank

1 Citizen Participation and Democratic Deficits: Considerations from the Perspective of Democratic Theory Mark E. Warren

Are there political problems faced by the developed democracies to which more citizen participation is the answer?1 From the perspective of democratic theory, the answer is clear and long-standing. Governments should be responsive to citizens as a consequence of their participation, through elections, pressure, public deliberation, petitioning, or other conduits. For these forms of participation to function democratically, all potentially affected by the decisions of a government should have the opportunity to influence those decisions, in proportion to their stake in the outcome. From a normative perspective, governments are in democratic deficit when political arrangements fail the expectation that participation should elicit government responsiveness. From an empirical perspective, governments are in democratic deficit when their citizens come to believe that they cannot use their participatory opportunities and resources to achieve responsiveness. From a functional perspective, governments are in democratic deficit when they are unable to generate the legitimacy from democratic sources they need to govern. My aim in this chapter is to clarify the question of democratic deficits as they relate to the participatory elements of democracy. What would it mean to understand the participatory features of the developed democracies as ‘in deficit’? To address this question, I distinguish two kinds of participatory issues within the broader problem of democratic deficits. The first, deficits in formal institutions of electoral democracy, is well theorized and well researched. The second is more recent and distinctive: deficits have appeared in the many new forms of ‘citizen engagement’, which have developed in response to deficits in electoral democracy. Following this distinction, I suggest that each kind of deficit requires distinct, though complementary, approaches. The first kind of deficit 17

18

Activating the Citizen

calls for institutional reforms, such as the redesign of electoral systems, parliamentary institutions, and basic constitutional changes, so that they are more responsive and have greater capacities for information gathering, deliberation, and policy formation. The second kind of deficit calls for what I shall call the ‘retrofitting’ of existing institutions: designing new forms of democracy that supplement and complement the formal institutions of electoral democracy, primarily in those functional policy areas where electoral institutions now have weak capacities to generate democratic legitimacy.

The concept of a democratic deficit The term democratic deficit originated in discussions about the political integration of the European Union (EU) (Bellamy and Castiglione, 2000). As a problem frame, the EU debate does not directly generalize to the developed democracies, as the EU is not a state but rather a federation that emerged as a consequence of market integration under the direction of the European Commission. The claim that the EU was in democratic deficit reflected not a democratic past that was eroding, but rather the growing democratic expectations that came with political integration, combined with institutions – the European Parliament in particular – that can and should be measured according to democratic norms. Measured in these terms, the EU was (and is) found wanting, with expectations evolving more quickly than the institutions. In the early 1990s, ‘permissive consensus’, which enabled the elite-brokered rule that underpinned the EU, was thrown into question by the Danish and French referendums on the Maastricht Treaty (Norris, 1997). It is unclear whether representative institutions can generate the legitimacy needed to replace the permissive consensus. Turnout for European Parliament elections is low, and parties focused on Europe-wide issues have yet to form, leaving candidates to focus on national issues. Much policy continues to be formed by relatively independent regulatory commissions. There is little worry that the EU might become an illegitimate tyranny: Its structure is relatively ‘safe’ in that it is well checked by national governments, judicial review, and elections (Moravcsik, 2002). But it is not clear whether the EU can find the democratic legitimacy to push forward with a deeper integration – since it lacks the responsive flexibility that a confident and connected people will give its government (Follesdal and Hix, 2006; Pammett, this volume). Although the institutional differences between the EU and the established democracies do not allow direct comparison, at a broader level

Mark E. Warren

19

the EU serves as a limiting metaphor for a more general condition. It is in this sense that the democratic deficit metaphor is apt, as it suggests a set of problems that are well short of crisis. Neither the EU nor the established democracies – the vast number of which are EU members – are failing. Over the last several decades, there has been little to bear out the 1975 predictions of the Trilateral Commission’s Crisis of Democracy, which expressed the fear that an excess of political demand would cause democratic institutions to revert to authoritarianism. The problems of the developed democracies today are less dramatic, consisting of a widespread citizen malaise with respect to the formal institutions and a common view held by citizens that political institutions underperform. The deficit concept suggests that we think about democratic malaise structurally, as a misalignment between citizen capacities and demands, and in terms of the capacities of political institutions to aggregate citizen demands and integrate them into legitimate and effective governance. From this perspective, the EU challenges look much like those common to the developed democracies. The list is well known. As it must protect national identities and interests as well as minorities, the EU cannot function in a majoritarian manner and thus lacks the capacity for decisiveness. It handles some protections through federal structures, others though subsidiarity, and still others through independent judicial review. Most developed democracies are evolving along similar trajectories. In addition, the EU functions to regulate trade and market integration, responding less to citizens than to market forces, which in itself requires regulatory processes that are relatively insulated from ‘politics’ – including democratic politics. It is likewise for the developed democracies. Finally, over the last half century or so, governing has become increasingly technical and complex: No citizen can hope to know more than a tiny fraction of the business of government. The result, of course, is that government is very much a matter for experts, who at best attempt to uphold the public trust and at worst govern as disconnected technocrats. Our received models of electoral democracy have had the advantage of clear lines of direction from, and accountability to, majorities or majority coalitions. Outside of questions about whether majoritarianism should be equated with democracy (it should not), the electoral machinery of democracies is an increasingly poor fit with the complex, pluralistic, multi-level business that governing has become. Societies now seem to be developing more rapidly than their institutions of government – in terms of citizen expectations and values, complexity and demands for sophisticated performances, pluralism, and levels and scales

20

Activating the Citizen

of organization. Under these circumstances, the most common and least costly form of citizen participation – voting for representatives – has less functional value. Yet these developments do not make citizen participation less important for governing. It is because democracies build in responsiveness and accountability to the people that they have reflexive capacities to form collective wills. Democracies enable their societies to benefit from evolving consensus where possible. Where consensus remains elusive, democracies transform conflict into public discourse and argument, where it can serve a creative rather than destructive role. At the limit, democratic deficits undermine the capacities of democratic political systems to evolve and reform into ever more effective and legitimate agents of citizens. The notion of a democratic deficit calibrates the problem: The misalignment in the established democracies is not a crisis. Rather, there are long-term problems, which, if left unattended, are likely to gradually undermine the legitimacy and capacities of governments.

What should we expect? Participation and trust To what extent can democratic deficits be narrowed through more citizen participation? This dimension of the question also requires perspective, since the same trends that open deficits may also limit participatory responses. Those who first emphasized the complexity of governing and its strategic consequences – Max Weber and Joseph Schumpeter – concluded that effective citizen participation was limited to choosing an elite in competitive elections. In the face of governing structures that depend upon professionals and favour concentrated power, citizens can at best play a passive and mostly retrospective role in checking elites who abuse their offices. These expectations have empirical relevance even today. In a study of citizen views of participation in the United States, Hibbing and Theiss-Morse (2002) argue that citizens do not like politics and prefer that politicians do their jobs while citizens get on with their lives. On average, citizens are averse to conflict. They are interested neither in the constant engagement necessary to communicate their preferences (when they have them), nor in disciplining politicians to attend to them. Rather, they want to trust politicians to look after the public good – though, of course, their trust is often disappointed, which in turn reinforces disaffection. Those who believe more participation is the answer to democratic deficits, Hibbing and Theiss-Morse argue, are waging an uphill battle.

Mark E. Warren

21

Although Hibbing and Theiss-Morse overstate their case (Dalton, 2007), such studies are important for calibrating expectations for participation in a healthy democracy (Van Deth, 2000; Moravcsik, 2002). No citizen can attend to, let alone master, every decision that affects him or her. Nor, given the range of possible activities and satisfactions in today’s societies, should we expect citizens to choose attentiveness to politics – particularly conflict-oriented politics – over competing forms of engagement: family, friends, occupations, hobbies, recreation, and entertainment. In the developed democracies, with their high degrees of institutional and system differentiation, political organization does not encompass these other domains and so citizens are faced with trade-offs. From an economic perspective, the political resources any citizen is able to marshal will be scarce. A good democracy should enable citizens to optimize their political resources. Ideally, citizens should be able to focus their lowest cost resource – voting – on choosing representatives who will fight most of their battles and protect most of their interests. They may join associations that fight other battles. In both cases, citizens judge whether their interests align closely enough with their representatives – whether formally elected or informally selected – that they can trust them as political proxies (Warren, 1996, 2000). To participate in these ways, citizens do not need to participate directly in decision making, but they do need to know something about their representatives’ trustworthiness, based on judgements about alignments of interests and values. They need to participate to the extent that they can ensure that they are, in fact, being represented. In many areas of government – typically those overseen by executive bureaucracies or quasi-judicial commissions – citizens may simply decide that public officials share their interests and will uphold the trust placed in them by the public. They may decide that their representative proxies will alert them when and if their trust appears misplaced. Under these circumstances, citizens can allocate their highcost political resources to the few areas where political conflicts matter to them – where they have reason to mistrust government or their representatives – and where their investment of knowledge, time, and attentiveness may make a difference. In this way, a healthy democracy enables citizens to divide their relationships to government between those of active participation and those of trust based on informed deference. In turn, this division of labour enables citizens to deploy their political resources to those issues that are most contested and where trust is least warranted. The political institutions of a democracy should strive to underwrite these participatory expectations.

22

Activating the Citizen

Even measured against these more modest expectations, however, political institutions in established democracies are in deficit. The easiest form of participation, voting, has been stagnant at best in established democracies over the last several decades (Franklin et al., 2004) and in EU elections has been at markedly lower levels than in national votes. Citizens are now more likely to distrust their political institutions – particularly legislatures – than a few decades ago. They are more likely to judge that government performance has deteriorated. They are less inclined to identify with political parties, which remain the key institutions for translating public opinion into government (Pharr and Putnam, 2000; Rosenblum, 2008). As in other advanced democracies, voting turnout in Canada has declined in recent decades, reaching a record low at the federal level of 59.1 per cent of registered voters in 2008 (Elections Canada, 2008). And, in Canada as in the EU, elite-brokered constitutional referendums have failed to pass muster with the voters ( Johnston, Blais, Gidengil, and Nevitte, 1996; LeDuc, this volume). Other developments also suggest that political institutions based on electoral representation are failing to generate the legitimacy necessary for many government functions. Innovations driven by gridlocked government and poorly performing programmes began a few decades ago, as evidenced by the rapid proliferation of ‘public engagement’ devices, including, for example, citizen juries and panels, advisory councils, stakeholder meetings, professional review boards with lay members, representations at public hearings, public submissions, citizen surveys, deliberative polling, deliberative forums, focus groups, and advocacy group representations (Gastil and Levine, 2005; Fung, 2006a; Smith, 2005, 2009). These developments are not in themselves evidence of democratic deficits in electoral institutions, but they do suggest that much necessary political work generated within complex, pluralized democracies is not accomplished through formal electoral democracy. While the general patterns of disaffection are clear, the causes are not. There is disagreement as to whether the political disengagement of citizens should be attributed to the poor performance of political institutions (Pharr and Putnam, 2000, Dalton, 2004); the increasing capacities of new generations of better educated, more informed, less deferential citizens to be critical of those institutions (Nevitte, 1996; Inglehart, 1997; Norris, 1999; Dalton, 2004); a broader civic phenomenon of declining participation in the social groups and networks that are vital to fostering norms of trust and reciprocity (Putnam, 1995, 2000); or a fundamental popular distaste for the conflict-ridden messiness of politics and a

Mark E. Warren

23

general disinterest in public policy debates (Hibbing and Theiss-Morse, 2001, 2002).

Electoral representation deficits Not every actionable diagnosis requires that causes be pinpointed precisely. Deficits in electoral representation are a case in point: Because electoral representation forms the most visible connective tissue between the people and decision makers, issues here are likely to be interconnected with most of the possible causes of democratic deficits. While I do not address causalities here, their relevance to normative democratic theory depends on how we cast the normative expectations. What should electoral representation achieve, normatively speaking? All forms of electoral representation share three formal features that specify the extent to which they have democratic content. Through elections, representatives are authorized to represent those who inhabit geographical constituencies. Electoral representation is held to be egalitarian and inclusive owing to the universal franchise; every member of an electoral unit, excluding those unfit or not yet fit to exercise the responsibilities of citizenship, is entitled to one vote. Subsequent elections function to hold representatives accountable for their performance while in office (Pitkin, 1967; Mansbridge, 2003; Urbinati and Warren, 2008; Warren, 2008). These three features include normative criteria: Institutions that are inadequate or incomplete in one or more of these dimensions are less than democratic – they are in deficit from a democratic perspective. In practice, democratic deficits appear in each of these three dimensions of representation (see, for example, Fung, 2006a). With respect to authorization, for example, citizens may have unstable preferences, which are neither adequately formed by the electoral process nor communicated by the blunt instrument of the vote. Elected leaders often claim mandates based on very thin evidence, since voting is information poor. Single-member plurality (SMP) systems compound the information problem by limiting the choices of voters, usually to a maximum of two viable parties. The stagnation of voting itself diminishes the significance of electoral authorization. In the 2008 federal election in Canada, for example, the Conservatives were able to form a government based on a 36.6 per cent share of the vote. But because turnout was only 59.1 per cent, the government was authorized by a mere 22.2 per cent of eligible voters. The narrowness of the mandate is carried over into the parliament by the mechanics of the Westminster system,

24

Activating the Citizen

which provides few incentives for parties to form majority coalitions. The mandate problem is somewhat mitigated by European-style proportional representation systems, which do provide parties with coalition incentives. Another problem is that the roles of political parties in the democratic authorization process are increasingly ambiguous. In theory, parties put forward platforms for which they seek authorization; in doing so, they enable voters to authorize substantive policy preferences. Yet broad declines in party identification suggest that party elites should be less confident that votes are signalling approval or disapproval of a party’s proposed policies and positions – a deficit that is particularly pronounced in Canada, where voters are four to five times more likely to declare an absence of partisan ties than Americans or Britons (Clarke and Stewart, 1998; Cross and Crysler, this volume). While levels of partisan identification vary across European countries, at the EU level citizens find it hard to view parliamentary elections as a contest between competing political parties, partly because of the weakness of Europe-wide party groups and partly because these elections are often understood as ‘second order’ politics, that is, commentaries on national politics rather than on EU-level issues (LeDuc, 2007, pp. 142–3). Indeed, advocacy groups are increasingly displacing political parties as key conduits of public opinion, though we have yet to understand and assess their connections to electoral authorization (Dalton, Cain, and Scarrow, 2003; Whiteley, this volume). As Justin Greenwood notes (2007, pp. 180–3), at the EU level the European Commission has actively promoted the formation of a range of Europe-wide groups or associations, in part to gain the benefit of expert input, but also to reinforce the EU’s ‘input legitimacy’ by creating a systematic and transparent process of interaction with a range of civil society interests. Because of the weakness of elections and political parties as ‘authorizing’ vehicles in the EU, however, interaction with advocacy groups may take on an even greater role there than in most national states. With respect to inclusion, citizens are formally equal by virtue of the universal franchise. In practice, however, electoral systems add several layers of exclusion. The first layer of exclusion resides in constituency definition: Leaving aside problems of drawing boundaries, electoral institutions represent ‘the people’ only insofar as they are residents of a particular territory. Attributes not tied to geographical residence (for example, sex, occupation, lifestyle, class, religion, etc.) are not formally represented (Rehfeld, 2005). The second layer of exclusion is inherent in

Mark E. Warren

25

the design of electoral systems, simply because they function to produce collective governing capacity under conditions of conflict. But different systems produce different degrees of exclusion. SMP systems in particular exclude minorities. In some cases (Canada at the federal level, for example), SMP systems exclude majorities as well, concentrating power in the hands of representatives of a plurality of voters. In contrast, consensus-based PR systems tend on average to be more inclusive (Lijphart, 1999). But all electoral systems produce legislative bodies that fail the tests of descriptive representation (and thus exclude the perspectives of those most disadvantaged within society), though some systems (party-list PR) tend to do better than others on this measure. A third layer of exclusion concerns the locus of decision making in complex societies. Even without the first two layers of exclusion, collective decision making has tended to slip into the hands of administrators over the last several decades – not owing to any conspiracy or consequence of bureaucratic power, but simply owing to a lack of legislative capacity to set more than broad goals for the more complex work of administration. As well, in contemporary, differentiated, market-capitalist societies, governments are responsive to markets, which tends to undermine popular control of gevernment (Lindblom, 2002). As a strategic consequence, some government agencies are effectively insulated from democratic control (central banks, for example), and some government policies are responsive not to votes or other clearly democratic inputs, but rather to those who control productive resources. The development of the EU in response to economic integration is a case in point (Bellamy and Castiglione, 2000). More generally, the fact that powerful EU institutions, such as the European Commission and the European Council, are only indirectly responsible to the public at large raises the possibility that these institutions are more responsive to well-organized interests than to the public. The problem is not limited to indirectly responsible institutions. Even elected representative institutions respond better to intense and wellorganized special interests than to latent interests, unorganized interests, and public goods. Because pressure groups tend to represent those with the resources to organize and who care intensely about a single issue, constituency communication may systematically disadvantage public-will formation around common goods. That said, the Commission’s efforts to increase transparency and establish criteria to help assure that groups represent their claimed constituencies are reforms that support inclusion (Greenwood, 2007). With respect to accountability, the picture is also mixed. On the one hand, the sheer amount of electing has increased steadily in the EU,

26

Activating the Citizen

Canada, and other OECD (Organization for Economic Cooperation and Development) countries over the last several decades – so much so that while voter turnout has been declining in most elections, the number of times citizens vote (their number of trips to the polls) has actually increased over the last several decades (Dalton and Gray, 2003). On the other hand, while there are many ways and means of introducing accountability, the vote is often a weak mechanism. Voters are often inattentive, information is incomplete, and other forms of power permeate the system, including actors with capacities to provide or withhold economic resources and administrative officials who have knowledge that representatives cannot match. Citizens often demand contradictory things from government, such as first-rate health care and schools combined with low taxes. And while elections may serve to aggregate preferences into competing packages, they are not very good instruments for stimulating deliberative consideration of collective goals and trade-offs. Still, there are important differences in the kinds of accountability electoral systems can provide. Westminster systems, for example, specialize in strong retrospective accountability, owing to strong institutional connections between electoral parties and the power of the majority party in parliament to govern. But the costs to inclusion and other forms of accountability are high. In Canada, it is unusual for a government to represent a majority of voters, owing to a combination of regionally strong parties and SMP. Only three times in Canada’s history have governments based their power on a majority of voters. When this fact is combined with the Westminster-style concentration of power in the Office of the Prime Minister, the typical effect is to sever representative linkages to the majority of citizens. The system fails in the deliberative dimension of accountability as well: Legislatures should have a deliberative element, which serves a democratic function by displaying before the public reasons for decisions and decision making (Urbinati, 2006). Because Westminster systems concentrate power in the Office of the Prime Minister, decisions are announced and then defended. In a pure Westminster model, parliament itself is a weak policy-making body, empowered only to say ‘yes’ or ‘no’ and lacking the power or capacity to formulate decisions in deliberation with the affected publics. Because Westminster systems concentrate power, there are few incentives for power holders to seek out information or deliberate policy options even in the semi-public forums of parliamentary committees, let alone the public forum of the legislature (Peters, 1997). Accountability to citizens is after the fact; when governments miscalculate – as they often do without inclusive linkages to citizens – they find themselves left with angry

Mark E. Warren

27

or disaffected publics. Of course, this is an all-other-things-being-equal kind of diagnosis: These effects are mitigated, for example, by minority government, federalism, senate reform, and parliamentary reform – though many of these characteristics, adopted in the absence of other reforms, are just as likely to produce stalemate as performance, further eroding legitimacy. Those EU countries that use PR systems present a different kind of accountability problem. Styles of decision making tend to be more consensual and deliberative, no doubt a structural consequence of the coalition building required by PR systems as well as systems with effective dual chambers (Lijphart, 1999; Steiner, Bachtiger, Sporndli, and Steenbergen, 2004). For the same reasons, however, it is difficult for citizens to hold any particular party responsible for policies that are often brokered not only among coalition partners, but also between chambers, or (in some cases) between executives and legislatures. The trade-offs between accountability and institutional complexity are characteristic of the EU as well. As DeBardeleben and Hurrelmann point out, in a complex system like that of the EU, it is difficult to hold any particular institution or individual accountable ‘since it is difficult to trace political acts to identifiable agents’. Referring to multi-level governance systems, which include the EU and federal states like Canada, DeBardeleben and Hurrelmann also note that blame shifting by authorities at various levels in the system ‘can be used as a political tool by political actors in the face of unpopular policies or policy failure’ (2007, p. 7). Finally, electoral representation in itself limits the deliberative dimensions of accountability (Mansbridge, 2004). Because elected representatives function within a context that combines public visibility and adversarial relations, they must weigh the strategic and symbolic impact of speech. Thus, representative institutions have limited capacities for deliberation, which requires a suspension of the strategic impact of communication in favour of persuasion and argument. Because of electoral cycles, representative institutions have limited capacities to develop and improve public policies over a long period of time. And because representatives must attend to vested interests, representative institutions have limited capacities for innovation and experimentation. In short, there are good reasons to conclude that, however necessary, the traditional and recognizable forms of democratic representation – elected officials convened in representative assemblies such as legislatures, parliaments, and councils – are no longer sufficient to carry out the normative purposes of democratic representation, at least not as stand-alone institutions.

28

Activating the Citizen

Democratic deficits in the participatory response Although these observations are not new, they do underscore the degree to which politics – and democratic demand – has flowed into venues outside of electoral representation. The developments are striking. As Dalton, Cain, and Scarrow (2003) demonstrate, political changes that open venues outside of electoral representation have developed rapidly in all the OECD countries with long-standing democratic systems. On the demand side, citizens increasingly expect to have a voice in matters that affect them, with younger cohorts leading the way, even as they participate less in formal democracy (Inglehart and Welzel, 2005; Dalton, 2007, Pammett, this volume; Saunders, this volume; cf. Hibbing and Theiss-Morse, 2002). On the supply side, decision makers have increasingly identified democratic deficits as an issue in all of the developed democracies, in practice if not in name. Public sector decision makers increasingly find the authority granted to them by election, appointment, or expertise insufficient for their decision making and governance responsibilities. In most areas of public policy, decision makers have found that standard administrative techniques based on legislative mandates are sub-optimal. They often generate unexpected opposition from stakeholders; fail to maximize the effective and efficient use of public resources; lack the legitimacy necessary for public acceptance and cooperation; fall short of substantive goals, such as health, public safety, and individual development; or fail to achieve normative ideals, such as distributive justice. Performances that publics judge to be clumsy, inefficient, wasteful, unjust, or unfair can undermine citizens’ confidence in public sector organizations, and democracy is impoverished when the citizenry lacks collective agents for public purposes. Over the last few decades, we have seen an enormous inventiveness in addressing these deficits through new institutions designed, in one way or another, to involve citizens in decision making (Fung 2006b; Parkinson, 2006; Gastil, 2008; Smith, 2009). The array of processes and practitioners is now extensive, including, to name a few, referendums, public hearings, public submission processes, client polling, deliberative polling, town hall meetings, citizen juries, citizen assemblies, participatory budgeting, numerous techniques for dialogue, neighbourhood councils, and issues forums. Citizen participation and engagement are typically understood by administrators as one among many strategies for gaining advice, co-opting pressures, and improving services, and in this way they seek to increase the legitimacy of their policies (Brown, 2006). The frameworks of engagement usually have administrative rather

Mark E. Warren

29

than political origins. And administrators are typically seeking citizen input rather than citizen empowerment in decision making. At the same time, the administrative contexts have perhaps masked the essentially political nature of these developments. We have not really grappled systematically with the question as to what more citizen engagement in this sense would mean for the democratic system as a whole. But we must: these developments are likely to continue, and insofar as democracy is deepening, much of the action is occurring here rather than within formal representative institutions (Cain, Dalton, and Scarrow, 2003; Warren, 2002, 2003; Fung, 2006b; Smith, 2009). At the same time, these new arenas of democratization can cause their own democratic deficits. Referendums, while inclusive of the entire electorate, provide numerous opportunities for interest-group mischief (Pharr and Putnam, 2000). Moreover, as the European constitutional referendums of 2005 demonstrated, they may produce outcomes that are hard to interpret because they force voters into yes/no responses on issues that are inherently complex and multi-dimensional. Many other kinds of new opportunities for citizen participation are less inclusive because they are based on self-selection and therefore tend to favour those who are better educated and wealthier, generating the paradox that increasing citizen opportunities for participation may increase political inequality (Cain, Dalton, and Scarrow, 2003). The same dynamic of self-selection allows the most intense interests and loudest voices to dominate, leading to the under-representation of those who are less organized, less educated, and have fewer resources (Mansbridge, 1983). And for similar reasons, participatory venues can increase the neglect of public goods or increase the unjust distribution of their burdens by empowering local resistance by well-organized groups. Forms of participation that simply aggregate existing opinion (for example, public submissions and hearings) contribute little to the deliberative formation of preferences and policy. And participation without power can lead to more disaffection, as citizens go through the exercise of engaging, only to have decisions taken elsewhere and for reasons unrelated to citizen input (Abelson and Eyles, 2002; Irvin and Stansbury, 2004). Under these circumstances, more participation can actually decrease democratic legitimacy. Finally, participatory venues are replete with representative claims by individuals and groups, on behalf of any number of interests, identities, and ideals. We have little understanding of what these kinds of representative claims add to (or subtract from) democracy, in spite of their growing presence and importance (Rubenstein, 2007; Urbinati and Warren, 2008).

30

Activating the Citizen

What should be done? These considerations should help disaggregate the question about whether more citizen participation will reduce democratic deficits. There are at least two general areas we need to think about, corresponding to two distinct kinds of deficits. The first has to do with deficits in the formal institutions of representative democracy. Democratic deficits in electoral representation are likely to involve incremental reforms of existing institutions, in part because they are not functioning so poorly that they are generating broad constituencies for wholesale changes, and in part because democratizing strategies are subject to limits of scale, complexity, and governability. Our expectations here should probably focus on other goods necessary to a well-functioning democracy such as more responsiveness, high capacities for inclusive deliberation of public matters and decisions, and better performance. The second area of deficit is in administration and governance, where electorally based political institutions have inherent incapacities. The potential for more citizen participation is greatest here, owing to the close relationships between policies and interests enabled by their disaggregation and delivery (Warren, 2002). I shall thus distinguish the reform of representative institutions from the retrofitting of these institutions – that is, upgrading their democratic capacities by supplementing them with new democratic devices, primarily in the areas of administration and governance. In both cases, we shall need to imagine institutional changes that build on broader social changes – including increasing advocacy, changing norms of citizenship, and declining deference to authority – while also enhancing system performance. Reforming institutions It is easier to generalize about democratic deficits in the developed democracies than about institutional reforms, since the institutional causes of deficits vary with the kind of system. Some systems, Denmark’s for example, exhibit few of the deficits evident in the other consolidated democracies (Elklit and Togeby, this volume), probably because of that country’s history of supplementing electoral democracy with policy-specific ‘network governance’ (Sørensen and Torfing, 2003). Others, such as Italy’s, suffer from gridlock induced by inclusion. In the case of Canada, the institutions suffer from overly concentrated executive powers and parliaments with weak deliberative and policy-making capacities, combined with gridlock in federal–provincial relations. EU institutions are sui generis, although they share some of

Mark E. Warren

31

these institutional problems of national multi-level systems such as Canada’s. Reforms can be thought of as targeting three kinds of institutions: electoral systems, parliamentary bodies, and constitutions. Electoral reform From the perspective of democratic deficits, reforms should function to make parliaments more inclusive and deliberative, increase citizen input into policy making, and improve and stabilize policy outcomes (Steiner et al., 2004; Peters, 1997; Lijphart, 1999). In the Canadian case, changing electoral systems from an SMP to a PR system such as mixed-member proportional or single transferable vote should result in reducing the capacities of prime ministers’ offices to make policy in the absence of broad input and deliberation, since powers are more likely to be shared among coalition partners. In addition, fixed election terms are likely to reduce the strategic powers of executives. Both reforms should increase the inclusiveness of electoral representation, which would in turn stabilize and probably improve policy outputs. Although these reforms might produce parliaments that are more inclusive and deliberative, the impact of electoral reform on citizen participation is likely to be modest at best. In New Zealand, for example, although voter efficacy and perceptions of government responsiveness initially increased, they subsequently declined (Banducci, Donovan, and Karp, 1999). Japanese voters have become disillusioned with their new electoral system’s dual candidacy rules, which allow candidates who have narrowly missed out on election in a single-member district to return to the Diet via the party list. The new rules function very much like the old rules, in that they still protect incumbents, suppress turnover, and encourage personalistic candidate behaviour (McKean and Schiner, 2000). The lesson may be that those changes that are politically achievable are those that are not likely to have a dramatic impact on political elites and other vested interests. But even well-designed electoral reforms are unlikely to close democratic deficits. Nor should we expect them to: Given the complexity and scale of government within pluralized contexts populated by multiple powers and actors, it seems unlikely that the standard model of representative democracy – voters elect representatives who develop policy guidelines and direct administrators to execute them – can ever again be adequate, if indeed it ever was. Nonetheless, even if electoral reforms are not sufficient, they may be necessary – particularly in SMP systems – to keep democratic deficits from growing.

32

Activating the Citizen

Parliamentary reforms Once voters elect their representatives, the bodies within which they serve should function to form their interests and values into public wills. The design of parliamentary bodies makes a difference. In the Westminster cases (Canada and the United Kingdom), parliaments should be strengthened as policy-making bodies in order to balance the powers of prime ministers’ offices, increase effective representation, and develop greater capacities for deliberation. These kinds of capacities can be increased through a number of reforms, including relaxing party discipline and strengthening parliamentary committees (Special Committee on Reform of the House of Commons, 1985; Martin, 2002; Commission on Legislative Democracy 2004). In addition, stronger freedom of information and sunshine legislation would reduce the capacities of executives to use information for strategic reasons, while increasing the numbers of informed participants in political processes – though these are likely to be groups and parties rather than individual citizens (Cain, Egan, and Fabbrini, 2003). Efforts within the EU to strengthen the role of the European Parliament would be another example of this type of reform. Since the parliament was made a popularly elected body in 1979, its role has been gradually extended. With the adoption of the Reform Treaty (Treaty of Lisbon), the parliament will take on a further enhanced role in legislation alongside the council, and the Commission would be made formally responsible to it. In addition, the parliament would also elect the president of the Commission based on a recommendation of the Council (Benz, 2008). These treaty changes, akin to constitutional changes in a national polity, are also efforts to address the EU’s democratic deficit. Constitutional reforms From a democratic perspective, there may be new roles for checks and balances, which can increase the inclusiveness and responsiveness of political systems by multiplying veto players, as long as mediating institutions exist to avoid conflicting mandates and gridlock, as in the case of Germany. In Canada and the United Kingdom, changing upper houses into democratically elected or otherwise democratically legitimate bodies would introduce effective bicameralism, which should also increase the inclusive and deliberative qualities of policy making (Department for Constitutional Affairs, 2003; Gibson, 2004). As in the EU, judicial review should continue to be strengthened and developed, in order to pluralize and regularize points of access to the political system, while dispersing

Mark E. Warren

33

powers of judgement (Cichowski and Stone Sweet, 2003). On average, these reforms should increase the inclusiveness as well as the quality of policy making. A stronger judiciary and rights-based regime should increase the power and standing of citizens with respect to government, which is in turn likely to show up as increased advocacy, particularly within the arenas of administration and services.

Political party reforms Finally, like other institutions within established political systems, political parties have real but constrained potentials for reform. Dramatic changes in party organization of the kind that would respond to declining citizen identification are most likely to follow from changes in electoral systems and parliamentary reforms – changes that would alter the environments and incentive structures to which parties respond. The constraints differ by system, making it difficult to generalize about party reform across democracies. But even though institutional contexts vary by country, parties face social and cultural developments across the developed democracies that are much more comparable. Citizens are less deferential to political elites, more plural in their identities, and more likely to attend to values (Norris, 1999; Dalton, 2004). They are increasingly postmodern in Inglehart’s (1997) sense. From the citizen’s perspective, many representative functions of parties are being displaced by public interest groups. At the same time, the bases of political authority have been shifting. It is not enough to win elections and assume power; it is not even enough to be held to account by an opposition. Governments (and the oppositions that hope to succeed them) must continuously explain themselves to the public, with every piece of legislation and with every action and inaction. From one perspective, governing appears to be a permanent campaign. From another perspective, however, the bases of political authority have become increasingly discursive and public in nature. Merely winning elections no longer confers the authority necessary to govern. For these reasons, all parties have struggled over the last few decades to become more internally democratic as well as more attentive to organized advocacy groups, social movements, and the media, while retaining a cohesion sufficient to form platforms, field candidates, and win elections (Kittilson and Scarrow, 2003). Ideally, parties would begin to view themselves as elements of a deliberative political system that generates public legitimacy for them and their candidates through public arguments – a trend currently most advanced in American staging

34

Activating the Citizen

of numerous pre-election debates among candidates, organized and moderated by citizen groups and the media (Rosenblum, 2008). Still, because parties are creatures of the electoral system and constitutionally determined functions in government or opposition, there are limits to their reform, particularly with regard to their functions in connecting society and government. Parties can translate the voices of organized interests and social movements into legislative bargaining positions, particularly in strong party systems. They are not equipped, however, to represent latent and disaffected interests, nor can they engage publics directly affected by policies on an ongoing basis. Their interests are, in the end, episodic (determined by electoral cycles) and strategic (oriented towards winning elections). Parties succeed when they put together winning coalitions, which means that their concerns and platforms operate at a relatively high level of abstraction when compared to the highly differentiated responses to highly complex societies that responsive governing requires, and of the kind provided, say, by single-issue associations, interest groups, and social movements. Parties are essential for democracy, but their current difficulties are unlikely to be significantly mitigated by internal reforms. Retrofitting institutions: supplementary democracy Institutions can be designed to be more sensitive to information, more deliberative, and more formally inclusive. But for these potentials to be realized, there must be connective tissue between institutions and society. The connective tissue needs to perform the political work of defining and engaging the publics affected by policies. As suggested above, over the last few decades all democratic systems have experimented with new, supplementary conduits for engaging citizens, gaining information, and generating informed public opinion. What is distinctive about these experiments is that they have little to do with organized party politics or formal political institutions. Most are functional and segmented by policy area (Ansell and Gingrich, 2003; Hajer and Wagenaar, 2003; Sørensen and Torfing, 2003; Fung, 2006a, 2006b). We might think about these new forms of citizen participation as ‘supplementary democracy’ – consisting in devices and venues that, in effect, retrofit formal political institutions with new capacities to gather interests, organize latent public opinion, and, in particular, provide governments with guidance that is not oriented towards strategic electioneering. As also suggested above, however, many of these new forms suffer from their own democratic deficits. Each has strengths and weaknesses: They are more or less inclusive, more or less deliberative, and

Mark E. Warren

35

more or less costly. Each form performs different kinds of political work, from co-opting obstruction to bringing informed publics into existence for future issues. They may generate information; they may produce more just outcomes; they may produce legitimacy; they may institutionalize new forms of learning. Each kind of process has costs: time and money. But many may also generate alienation, provide venues for NIMBYism, and produce outcomes that are substantially more unjust than professional public servants might produce if sheltered from public pressure. Compared to our extensive and increasingly sophisticated knowledge of formal political institutions, understanding these new forms of citizen engagement is still in its infancy (cf. Fung, 2006b; Parkinson, 2006; Gastil, 2008; Smith, 2009). We have more questions than answers: What kinds of processes are appropriate for what kinds of issues? What kinds of processes are likely to generate better rather than worse outcomes – more legitimacy, justice, or effectiveness, say – given the characteristics of the issues and the constraints of time and money? We know pieces of the answers. For example, combining experts with lay citizens over time within a deliberative context can overcome many of the constraints of technical complexity. We know that processes that depend on citizen self-selection will bias the process towards organized, high-resource interests and that random selection can produce a closer approximation of informed public opinion. But we don’t know how to begin with an issue and a set of goals and then design a democratic process appropriate for these particular goals and constraints. We do, however, have discrete pieces of knowledge and beginnings of middle-level theories that we now need to develop into broader theories and generalizations (Fung, 2006b; Parkinson, 2006; Gastil, 2008; Smith, 2009). That it is possible in principle to narrow democratic deficits through the careful design of supplementary institutions is suggested by research on the British Columbia Citizens’ Assembly, an institution created by the British Columbia government in 2004 to produce a referendum question on electoral reform (British Columbia Citizens’ Assembly, 2004; see also Warren and Pearse, 2008; Rose, this volume). The citizens’ assembly – an experiment repeated in Ontario and the Netherlands – included two key innovations. The first, random selection, avoided interest-group domination of the venue. The second, extensive learning and deliberation over a period of ten months, developed citizen expertise and a near-consensus recommendation. A survey indicated that the citizens of British Columbia placed an extraordinarily high level of trust in the

36

Activating the Citizen

assembly and its recommendation (Cutler and Johnston, 2007), a consequence of an institutional design that closely matched the purposes and qualities of the issue. The case of the British Columbia Citizens’ Assembly suggests that it is possible in principle to design supplementary forms of democracy in ways that directly address and affect democratic deficits. A more general understanding of supplementary democracy is, however, still in its infancy. But we should get started, since it is likely that the longterm solutions to democratic deficits will not only reform our existing institutions, but also retrofit them with new and innovative forms of democracy.

Note 1. I am grateful to Joan DeBardeleben for providing many of the examples from the European Union as well as paragraphs on pages 23–5, 27 and 32 and suggested wording elsewhere. I thank Hilary Pearse and Laura Montanaro for their research assistance.

References Abelson, J. and J. Eyles (2002) Public Participation and Citizen Governance in the Canadian Health System, Discussion Paper 7 (Ottawa, ON: Commission on the Future of Health Care in Canada). Ansell, C. and J. Gingrich (2003) ‘Reforming the Administrative State’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 164–91. Banducci, S. A., T. Donovan, and J. A. Karp (1999) ‘Proportional Representation and Attitudes about Politics: Results from New Zealand’, Electoral Studies, 18(4), 533–55. Bellamy, R. and D. Castiglione (2000) ‘The Uses of Democracy: Further Reflections on the Democratic Deficit in Europe’ in E. O. Eriksen and J. E. Fossum (eds) Integration through Deliberation? On the Prospects for European Democracy (London: UCL Press), pp. 65–84. Benz, A. (2008) ‘The Treaty of Lisbon – A Step towards a More Effective and Democratic European Union’, EUCE Canadian Network Newsletter, 2(1), 1, 3, http:// www.euce-network.carleton.ca/newsletter/V2(1)-EUCENewsletter.pdf British Columbia Citizens’ Assembly on Electoral Reform (2004) Making Every Vote Count: The Case for Electoral Reform in British Columbia, http:// www.citizensassembly.bc.ca/resources/final_report.pdf Brown, M. B. (2006) ‘Citizen Panels and the Concept of Representation’, Journal of Political Philosophy, 14, 203–25.

Mark E. Warren

37

Cain, B. E., R. J. Dalton, and S. E. Scarrow (eds) (2003) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press). Cain, B. E., P. Egan, and S. Fabbrini (2003) ‘Toward More Open Democracy: The Expansion of Freedom of Information Laws’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 115–39. Cichowski, R. A. and A. Stone Sweet (2003) ‘Participation, Representative Democracy, and the Courts’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 192–220. Clarke, H. D. and M. C. Stewart (1998) ‘The Decline of Parties in the Minds of Citizens’, Annual Review of Political Science, 1, 357–8. Commission on Legislative Democracy (2004) Final Report and Recommendations, http://www.gnb.ca/0100/FinalReport-e.pdf Cutler, F. and R. Johnston (2008), ‘Deliberation, Information, and Trust: The BC Citizens’ Assembly as Agenda-Setter’ in M. E. Warren and H. Pearse (eds) Designing Democratic Renewal: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press), pp. 166–91. Dalton, R. J. (2004) Democratic Challenges, Democratic Choices: The Erosion of Political Support in the Advanced Industrial Democracies (Oxford: Oxford University Press). Dalton, R. J. (2007) The Good Citizen: How a Younger Generation Is Reshaping American Politics (Washington, DC: Congressional Quarterly Press). Dalton, R. J., B. E. Cain, and S. E. Scarrow (2003) ‘Democratic Publics and Democratic Institutions’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 250–75. Dalton, R. J. and M. Gray (2003) ‘Expanding the Electoral Marketplace’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 23–43. DeBardeleben, J. and A. Hurrelmann (2007) ‘Introduction’ in J. DeBardeleben and A. Hurrelmann (eds) Democratic Dilemmas of Multilevel Governance: Legitimacy, Representation and Accountability in the European Union (Basingstoke: Palgrave Macmillan), pp. 1–14. Department for Constitutional Affairs (2003) Constitutional Reform: Next Steps for the House of Lords, http://www.dca.gov.uk/consult/holref/index.htm Deth, van J. (2000) ‘Interesting but Irrelevant: Social Capital and the Saliency of Politics in Western Europe’, European Journal of Social Research, 37, 115–47. Elections Canada (2008) ‘Table 4: Voter Turnout for the 2008, 2006, 2004, and 2000 General Elections,’ Fortieth General Election 2008: Official Voting Result, http://www.elections.ca/scripts/OVR2008/default.html Follesdal, A. and S. Hix (2006) ‘Why There Is a Democratic Deficit in the EU: A Response to Majone and Moravcsik’, Journal of Common Market Studies, 44, 533–62. Franklin, M. N., C. van der Eijk, D. Evans, M. Fotos, W. Hirczy de Mino, M. Marsh, and B. Wessels (2004) Voter Turnout and the Dynamics of Electoral Competition

38

Activating the Citizen

in Established Democracies since 1945 (Cambridge: Cambridge University Press). Fung, A. (2006a) ‘Democratizing the Policy Process’ in R. Goodin, M. Moran, and M. Rein (eds) The Oxford Handbook of Public Policy (Oxford: Oxford University Press), pp. 669–85. Fung, A. (2006b) ‘Varieties of Participation in Complex Governance’, Public Administration Review, 66, 66–75. Gastil, J. (2008) Political Communication and Deliberation (Los Angeles: Sage). Gastil, J. and P. Levine (eds) (2005) The Deliberative Democracy Handbook: Strategies for Effective Civic Engagement in the Twenty-first Century (San Francisco: JosseyBass). Gibson, G. (2004) ‘Challenges in Senate Reform: Conflicts of Interest, Unintended Consequences, New Possibilities’, Public Policy Sources, 83, http:// www.fraserinstitute.ca/admin/books/files/ChallengesInSenateReform.pdf Greenwood, J. (2007) ‘Organized Civil Society and Input Legitimacy in the EU’ in J. DeBardeleben and A. Hurrelmann (eds) Democratic Dilemmas of Multilevel Goveranance: Legitimacy, Representation and Accountability in the European Union (Basingstoke: Palgrave Macmillan), pp. 177–94. Hajer, M. A. and H. Wagenaar (eds) (2003) Deliberative Policy Analysis: Understanding Governance in the Network Society (Cambridge: Cambridge University Press). Hibbing, J. R. and E. Theiss-Morse (2001) ‘Process Space and American Politics’, American Political Science Review, 95, 145–53. Hibbing, J. and E. Theiss-Morse (2002) Stealth Democracy: Americans’ Beliefs about How Government Should Work (Cambridge: Cambridge University Press). Inglehart, R. (1997) Modernization and Postmodernization: Cultural, Economic and Political Change in 43 Societies (Princeton, NJ: Princeton University Press). Inglehart, R. and C. Welzel (2005) Modernization, Cultural Change, and Democracy: The Human Developmental Sequence (Cambridge: Cambridge University Press). Irvin, R. and J. Stansbury (2004) ‘Citizen Participation in Decision Making: Is It Worth the Effort?’, Public Administration Review, 64(1), 55–65. Johnston, R., A. Blais, E. Gidengil, and N. Nevitte (1996) The Challenge of Direct Democracy: The 1992 Canadian Referendum (Montreal: McGill-Queen’s University Press). Kittilson, M. C. and S. Scarrow (2003) ‘Political Parties and the Rhetoric and Realities of Democratization’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 59–80. LeDuc, L. (2007) ‘European Elections and Democratic Accountability: The 2004 Elections to the European Parliament’ in J. DeBardeleben and A. Hurrelmann (eds) Democratic Dilemmas of Multilevel Goveranance: Legitimacy, Representation and Accountability in the European Union (Basingstoke: Palgrave Macmillan), pp. 139–57. Lijphart, A. (1999) Patterns of Democracy: Government Forms and Performance in Thirty-six Countries (New Haven, CT and London: Yale University Press). Lindblom, C. E. (2002) The Market System: What It Is, How It Works, and What To Make of It (New Haven, CT: Yale University Press). Mansbridge, J. (1983) Beyond Adversary Democracy (Chicago: University of Chicago Press).

Mark E. Warren

39

Mansbridge, J. (2003) ‘Rethinking Representation’, American Political Science Review, 97, 515–28. Mansbridge, J. (2004) ‘Representation Revisited: Introduction to the Case against Electoral Accountability’, Democracy and Society, 2(1), 12–13. Martin, P. (2002) ‘The Democratic Deficit’, Policy Options, 24(1), 10–12. McKean, M. and E. Schiner (2000) ‘Japan’s New Electoral System: La Plus Ça Change . . .’, Electoral Studies, 19, 447–77. Moravcsik, A. (2002) ‘In Defence of the “Democratic Deficit”: Reassessing Legitimacy in the European Union’, Journal of Common Market Studies, 40, 603–24. Nevitte, N. (1996) The Decline of Deference: Canadian Value Change in Cross-National Perspective (Peterborough, ON: Broadview Press). Norris, P. (1997) ‘Representation and the Democratic Deficit’, European Journal of Political Research, 32, 273–82. Norris, P. (ed.) (1999) Critical Citizens: Global Support for Democratic Governance (Oxford: Oxford University Press). Parkinson, J. (2006) Deliberating in the Real World: Problems of Legitimacy in Deliberative Democracy (Oxford: Oxford University Press). Peters, B. G. (1997) ‘The Separation of Powers in Parliamentary Systems’ in K. von Mettenheim (ed.) Presidential Institutions and Democratic Politics: Comparing Regional and National Context (Baltimore, MD: The Johns Hopkins University Press), pp. 76–83. Pharr, S. J. and R. D. Putnam (2000) Disaffected Democracies: What’s Troubling the Trilateral Countries? (Princeton, NJ: Princeton University Press). Pitkin, H. (1967) The Concept of Representation (Berkeley: University of California Press). Putnam, R. D. (1995) ‘Bowling Alone: America’s Declining Social Capital’, Journal of Democracy, 6(1), 65–78. Putnam, R. D. (2000) Bowling Alone: The Collapse and Revival of American Community (New York: Simon & Schuster). Rehfeld, A. (2005) The Concept of Constituency: Political Representation, Democratic Legitimacy, and Institutional Design (Cambridge: Cambridge University Press). Rosenblum, N. (2008) On the Side of the Angels: An Appreciation of Parties and Partisanship (Princeton, NJ: Princeton University Press). Rubenstein, J. (2007) ‘Accountability in an Unequal World’, Journal of Politics, 69, 616–32. Smith, G. (2005) Beyond the Ballot: 57 Democratic Innovations from around the World (London: The Power Inquiry). Smith, G. (2009) Democratic Innovations: Designing Institutions for Citizen Participation (Cambridge: Cambridge University Press). Sørensen, E. and J. Torfing (2003) ‘Network Politics, Political Capital, and Democracy’, International Journal of Public Administration, 26, 609–34. Special Committee on Reform of the House of Commons (1985) Report of the Special Committee on Reform of the House of Commons (Ottawa, ON: Government of Canada). Steiner, J., A. Bachtiger, M. Sporndli, and M. R. Steenbergen (2004) Deliberative Politics in Action: Analysing Parliamentary Discourse (Cambridge: Cambridge University Press).

40

Activating the Citizen

Trilateral Commission (1975) The Crisis of Democracy, Task Force Report #8 (New York: New York University Press). Urbinati, N. (2006) Representative Democracy: Principles and Genealogy (Chicago: University of Chicago Press). Urbinati, N. and M. E. Warren (2008) ‘The Concept of Representation in Contemporary Democratic Theory’, Annual Review of Political Science, 11, 387–412. Warren, M. E. (1996) ‘Deliberative Democracy and Authority’, American Political Science Review, 90, 46–60. Warren, M. E. (2000) ‘Democratic Theory and Trust’ in M. E. Warren (ed.) Democracy and Trust (Cambridge: Cambridge University Press), pp. 210–34. Warren, M. E. (2002) ‘What Can Democratic Participation Mean Today?’, Political Theory, 30(5), 678–702. Warren, M. E. (2003) ‘A Second Transformation of Democracy?’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 223–49. Warren, M. E. (2008) ‘Citizen Representatives’ in M. Warren and H. Pearse (eds) Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press), pp. 50–69. Warren, M. E. and H. Pearse (eds) (2008) Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press).

2 The Decline of Political Participation: An Empirical Overview of Voter Turnout and Party Membership Alan Siaroff

This chapter considers two important indicators of citizen participation in Europe and Canada: voter turnout and party membership. These two indicators capture different dimensions of citizen involvement in politics, and both register significant declines on both sides of the Atlantic. Voter turnout taps broad citizen involvement in political decision making, as it is a relatively ‘easy’ form of participation that is accessible to the entire adult citizenry of a country. Party membership taps a higher level of citizen involvement, generally involving the more politically attentive and mobilized strata of the population. By examining these two levels of political involvement, using a broad range of empirical data from the 1970s until the present period, we gain a picture of the nature of the participation gap in advanced Western democracies. This chapter will also examine the major explanations offered in the literature for both trends, as well as the interrelationship between them. The analysis draws on data from the 27 current European Union (EU) member states, the three main non-EU western European states (Iceland, Norway, and Switzerland), the EU itself (its European Parliament elections through 2004), and Canada. There are clear variations across these countries and, at least in terms of voter turnout, these variations have increased in recent times. Before commencing the examination of voter turnout, some comments on interlinkages between voting and party activism deserve mention. Both lower voter turnout and declining membership are parallel indicators of declining political participation, the first electoral and the second party-related. Lower political participation is thus the broader and ultimately key issue. In and of itself, low(er) voter turnout is argued to have some modest effects on election outcomes; that is, with 41

42

Activating the Citizen

‘complete’ (full) turnout, both small parties and non-incumbent parties would do better (Bernhagen and Marsh, 2007). However, neither end of the ideological spectrum would expect to do better consistently with higher turnout levels; contrary to popular opinion, there is no generalized bias against leftist parties due to low turnout (Lutz and Marsh, 2007). Party membership itself reflects party–citizen ties, and thus countries with higher party membership can be seen as having more institutionalized party systems. Party system institutionalization is certainly desirable, since it produces greater stability and accountability. It can also be hypothesized to lead to higher voter turnout (see below). Moreover, where the party system is not just institutionalized but based around ideological/programmatic differences, then turnout tends to be even higher, as is seen in Chapter 4 of this volume by Elklit and Togeby. Overall, however, the key point is not just the effects, or appeals, of specific parties – it is the legitimacy of the political system in general. Where citizens cannot be bothered, or inspired, to play an active political role, there is little positive endorsement of the political order and the specific policy outputs of the government of the day. Consequently, a victorious political party cannot rightfully claim that citizens have the same enthusiasm for its mandate in a situation of low turnout (or low party membership) as in one of high turnout. Of course, this assumes that the political parties are abiding by democratic principles. As Weimar Germany showed, high turnout can go hand in hand with strong support for extremist parties. Fortunately, though, such parties are rarely strong in Europe and Canada today, so nowadays one can, for the most part, see higher citizen participation as both an indicator of a healthy democracy and of the legitimacy of the political system. Therefore, methods to increase participation, such as those examined in Part IV of this book, are an important element in addressing the phenomenon of the ‘democratic deficit’, which is much discussed in Europe and Canada.

Voter turnout In considering voter turnout, we explore three related questions. First, in which countries has turnout been high or low, and why? Second, when and where has turnout fallen? Third, why has there been this general decline in voter turnout? Table 2.1 shows the average voter turnout (vote participation as share of the electorate, that is, registered voters) for democratic parliamentary

43 Table 2.1: Turnout levels in parliamentary elections, decade averages (% of registered voters) 1970s

1980s

1990s

2000s*

Austria Belgium Bulgaria Canada Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Iceland Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Norway Poland Portugal Romania Slovakia Slovenia Spain Sweden Switzerland United Kingdom

92.3 93.0

91.6 93.9

74.6 80.6

73.3 95.2

88.4

86.7

78.1 76.5 90.9 80.4

73.9 71.9 87.3 83.5

90.4 76.5 92.3

89.4 72.7 89.0

89.5 94.0 83.5 81.6

88.1 95.4 83.5 83.1

85.4

78.0

72.6 90.4 52.3 75.0

73.4 89.1 48.2 74.1

83.8 91.5 72.7 68.3 92.2 82.8 84.3 64.7 67.4 68.5 79.7 79.7 67.0 86.4 67.3 85.5 77.9 64.1 87.4 96.2 76.0 76.9 47.7 65.2 76.2 85.2 79.8 77.6 85.4 43.8 74.7

81.4 91.4 61.2 62.3 90.4 61.2 86.0 59.6 67.3 60.3 78.4 75.2 69.0 85.7 64.8 81.8 66.8 52.1 91.7 94.5 79.8 76.2 46.9 63.6 61.9 62.4 65.5 73.2 81.1 47.0 60.4

Average

82.8

82.0

76.0

70.9

European Parliament country averages of which original six members

65.9 74.9

64.0 70.5

55.5 64.5

47.8 63.2

∗ This

table is current through the end of June 2008. Tables exclude presidential contests because they are or have been central only in a few of the countries covered, for example, France, Portugal, Finland, and Poland. Sources: International Institute for Democracy and Electoral Assistance (2002); Parties and Elections in Europe (online database), http://www.parties-and-elections.de; Elections Canada, Voter Turnout at Federal Elections and Referendums, 1867–2006, Elections Canada http://www.elections.ca/home.asp; Baimbridge (2005).

44

Activating the Citizen

elections in each of the past four decades across the 32 jurisdictions.1 In terms of cross-national variations, two groups of countries stand out in a positive sense. In the first group are those countries that have maintained very high turnout, defined as 85 per cent or more. This group consists of Belgium, Cyprus (since the 1980s), Denmark (although it was a bit below the cutoff in the 1990s), Iceland, Luxembourg, and Malta. In the second group are those countries that have maintained high turnout, defined here as between 75 and 85 per cent. This group consists of Austria, Germany, Greece, Italy, the Netherlands, Norway, and Sweden. Indeed, some countries in this second group used to have very high turnout, specifically Austria, Germany, Italy, and Sweden. Overall, though, all of the countries in these two groups stand out from the rest in a positive sense. Conversely, very low turnout (defined as being below 55 per cent) has been a feature of Switzerland, even in earlier decades. Very low turnout is also now the situation for elections to the European Parliament. Low turnout (defined as being from 55 to 65 per cent) now exists in Canada, France, and the United Kingdom (UK) and also in most of post-communist Europe (with Poland2 and now Lithuania having very low turnout). The turnout situation in post-communist Europe needs to be clarified. Generally the first post-communist elections had high or even very high turnout. Subsequent elections, however, saw much lower turnout as the novelty of democratic voting wore off and frustrations with the democratic system and its outputs (or lack of these) occurred (see Kostadinova, 2003). Finally, the residual category of moderate turnout (neither strikingly high nor low) can be applied to turnout levels of 65 to 75 per cent. A small number of countries fit here: Finland, Ireland, Portugal (most of the time), and Spain, as well as Hungary and Latvia in post-communist Europe. In terms of when and where turnout has declined, the key temporal change clearly occurred between the 1980s and the 1990s. That is, for most longstanding democracies, turnout was high(er) in the 1970s and 1980s (with little difference between these two decades), and low(er) in the 1990s and 2000s (sometimes with the 2000s being even lower than the 1990s). This pattern has not been completely universal, of course. In those countries with very high turnout, turnout has remained consistently high or fallen off only slightly. Stable turnout levels over the long haul have also occurred in Spain. In the UK, the sharp decline in turnout did not occur until this decade. Overall, as seen in Table 2.1, the difference between the first two decades and the recent two decades is unquestionable.

Alan Siaroff 45

One explanation for change is generational, that is, turnout decline has arisen because citizens who have come of voting age in recent decades are voting less than earlier generations. This strong generational change was shown by Blais, Gidengil, Nevitte, and Nadeau (2004) to exist for Canada. Specifically, those born in the 1960s have substantially lower turnout (by 10 per cent) than those born earlier; those born in the 1970s have even lower turnout (by a further 10 per cent). Both of these younger groups thus factor into the lower turnout since the 1990s shown in Table 2.1. Moreover, generational effects are much more central to explaining recent declines in Canadian turnout than any life cycle effects, since ‘the age composition of the electorate has not changed substantially over time’ (Blais et al., 2004, p. 227). These findings have largely been duplicated for Finland by Wass (2007). This generational pattern does not seem likely to reverse itself, at least through non-voters choosing to become voters later in life (a life cycle effect). As Franklin (2004) notes, three elections (that is, the first three in which one is eligible to vote as a young adult) are key to establishing a habit of voting – or of not voting (Franklin, Lyons, and Marsh, 2004). Citizens who have established a habit of non-voting cannot be expected to change this easily. A central question is why fewer citizens do not bother to vote in the first place, especially when compared to earlier periods. To answer this we need to look to factors that are different today from 20 to 30 years ago. First, though, let us analyse the cross-national variations in turnout independent of temporal changes.

Explanations of cross-national turnout levels There is a vast and growing literature on comparative voter turnout (for example, Blais and Carty, 1990; Blais and Dobrzynska, 1998; Franklin, 2002; Siaroff and Merer, 2002; Geys 2006). This academic literature suggests several sets of explanatory factors for why some countries have had higher (or lower) turnout than others. The first theoretical area is socioeconomic. In this regard, factors such as a high level of development, small size, and high population density have been found to affect turnout positively (Blais and Carty, 1990; Blais and Dobrzynska, 1998; see also Blais and Dobrzynska, this volume). It is hypothesized that in more developed countries, people are more informed and have more resources (including time) to devote to politics; thus, turnout will be higher. Indeed, countries with very high or high turnout are generally quite developed, in terms of income per capita or broader measures of development such as the United Nations’ Human Development Index.

46

Activating the Citizen

Yet there are some highly developed nations with very low turnout (Switzerland) or moderate turnout (Finland). It is also hypothesized that turnout will be higher in smaller countries, since citizens will feel more powerful as individuals in relation to others and the state (see Dahl and Tufte, 1973). A related assumption is that turnout will be higher in more densely populated urban settings, as individuals will feel closer to their neighbours as fellow citizens and thus be more likely to take an interest in collective matters. Certainly many of the countries with high or very high turnout are small in terms of population: Austria, Belgium, Denmark, Greece, Norway, Sweden, and especially Cyprus, Iceland, Luxembourg, and Malta (the latter four each having less than a million people and each having very high turnout). Most of these high or very high turnout countries also have a high population density and high level of urbanization. On the other hand, large populations have not prevented high turnout in Germany and Italy. Finally, in terms of demographic factors, Geys (2006) suggests that since ‘cohesion increases group solidarity (and “social pressure”)’ (p. 644), more homogeneous societies should have higher turnout. Most high or very high turnout countries are ethnically homogeneous. Iceland would appear to be the ‘purest’ case here, but ethnic homogeneity also characterizes the Scandinavian countries, Germany, Austria, Italy, Greece, and (Greek) Cyprus. The one case of a very heterogeneous society with very high turnout is Belgium. That said, turnout is not impressive in the other countries that have high ethnic fragmentation (Canada, Estonia, Latvia, Spain, and Switzerland) or moderate ethnic fragmentation (Bulgaria, the Czech Republic, France, Lithuania, Romania, Slovakia, and the UK).3 The next three theoretical explanations regarding voter turnout involve aspects of the party system and thus are also relevant to our later discussion of declining levels of party membership. The first of these is party system fragmentation/concentration. Blais and Dobrzynska (1998) hypothesized that ‘the greater the number of parties, the more choice electors are offered, and the higher the turnout’ (p. 248); they provide further evidence in this volume. If we define a high number of parliamentary parties as six or more (thus leading to an ‘extreme multiparty system’), then there are many countries in this study where voters are offered a great choice of parties. Yet although there are such countries where there is also high or very high turnout (Belgium, Denmark, Netherlands, Norway, and Sweden), there are even more such countries where there is less than high turnout (Bulgaria, Estonia, Finland, Latvia, Lithuania, Poland, Slovakia, and Slovenia). That said, almost all of the latter

Alan Siaroff 47

group are post-communist countries and, as we shall see, other factors are at work there. Related to aspects of party system fragmentation are those of party system competitiveness. Here the issue, or debate, is whether or not turnout is higher in close elections (Blais and Dobrzynska, 1998, also this volume). This measure is more relevant when one is comparing specific elections than longstanding national patterns. It is also relevant whether an election produces a single party majority government. ‘Turnout will be lower when one party wins an outright majority, since the result is predictable’ (Siaroff and Merer, 2002, p. 921). Malta clearly speaks against this argument, since it is a very high turnout country with consistent single-party majorities; moreover, going back to the previous point, Malta is also a pure two-party system. Austria in the 1970s also had three straight single-party majorities yet very high turnout (of course, these were all very narrow majorities). Greece normally has single-party majorities but also high turnout. On the other hand, all of the other countries where single-party majorities occur normally (the UK), frequently (Canada), or occasionally (France, Lithuania, Portugal, and Spain) are countries with turnout problems. Moreover, Johnston, Matthews, and Bittner (2007) show that declining competitiveness was central to the decline in turnout in Canada in the 1990s. Although not specifically emphasized in the literature on comparative voter turnout, one can and should discuss here the issue of party system institutionalization, that is, the extent to which parties are stable and rooted in society. Mainwaring and Scully (1995) distinguish ultimately between institutionalized and ‘inchoate’ party systems (in Latin America) in terms of whether or not the parties are durable, rooted in society, and ideologically consistent with respect to each other; whether electoral volatility is low; and whether parties and elections have broad legitimacy. Of the individual countries in this study, five – Bulgaria, Estonia, Latvia, Lithuania, and Poland – definitely seem to have inchoate party systems, as evidenced most notably by rapid changes in what the major parties are from election to election (for example, having a party that did not even exist during the previous election come first or second in the next election). These five countries are also known for low or very low turnout (at least, after the first post-communist election). What, though, of party system institutionalization at the level of the EU? First of all, the EU does not have parties but rather transnational party groups composed of like-minded national parties. Campaigning in European Parliament elections is done primarily by these national parties. Moreover, research by Hix (2002) shows that it is the positions

48

Activating the Citizen

of national parties that most determine how members of the European Parliament vote, which is not surprising, since it is the national parties that control candidate selection. Thus, even if most member states have institutionalized party systems and most national parties are rooted in their societies, one cannot conclude the same about the transnational party groups. Consequently, the EU can also be deemed to have an inchoate party system, or perhaps more correctly no real European party system at all in terms of voters. In this vein, one can also note the very limited causality going from European elections to EU policies; as Mair (2007) stresses, the EU ‘is not responsive in terms of elections and parties’ (p. 163). As Table 2.1 showed, turnout for elections to the European Parliament is now very low (and was never very high), and this fits with the lack of party system institutionalization (or even the lack of a party system) in the EU. The next overall theoretical area regarding voter turnout involves institutional aspects of the vote itself. The first aspect here is the extent of disproportionality in the election results, with the assumption that ‘unfair’ (disproportionate) results discourage voter turnout. Disproportionality is associated with single-member systems, and indeed we have noted that turnout is now low in Canada, France, and the UK – the three countries in this analysis using single-member electoral systems. Conversely, proportional representation is assumed to increase voter turnout. Blais and Carty (1990) suggest that turnout will be higher under proportional systems since voters will not feel that their vote is wasted, more parties will likely run (but see the aforementioned findings on the number of parties and turnout), and parties will have an incentive to campaign everywhere (see also Blais and Dobrzynska in this volume). The opposite hypotheses all apply to single-member systems. As noted, single-member systems have low turnout, but proportional systems do not all have high turnout; rather, proportional systems have the full range of turnout. In fact, the two countries in this analysis with the lowest turnout (Poland and Switzerland) both have proportional systems. The other institutional variable is compulsory voting, perhaps the central ‘solution’ to increasing turnout (Lijphart, 1997). Here I follow the analysis of Gratschew (2001) and distinguish between the strict enforcement of compulsory voting found in Belgium, Cyprus, and Luxembourg and the weak enforcement of compulsory voting found in Greece and Italy. It is hypothesized that it is only the strict enforcement of compulsory voting that will make an appreciable difference, due to the monetary fines and/or other punishments levied. Indeed, all three countries with strictly enforced compulsory voting are in the very high turnout category,

Alan Siaroff 49

so it can be argued that compulsory voting definitely ‘works’; in other words, it is a sufficient (but not necessary) causal factor producing or maintaining very high turnout, as in Australia. The seventh set of causal factors involves the importance of the national parliament and thus its elections for a country’s political system. As scholars such as Blais and Dobrzynska (1998) argue, other elected bodies could weaken the power of the national parliament. The absence of such bodies in Malta has been noted by Hirczy (1995). (Multiple elections also lessen parties’ ability and will to mobilize for any one specific election; see Cox, 1999.) Rather than assessing this issue for each and every other relevant actor on its own, one can rephrase the question as, ‘Overall, how constrained is a party if it has a majority in a given national legislature?’ Scholars often measure simple majoritarianism/political centralization (or, conversely, consensus democracy, institutional power sharing, or veto powers). Perhaps the best known such analysis is that of Lijphart (1984, 1999), who in fact uses two dimensions (executive-parties and federal-unitary). Ultimately what is being distinguished here is the difference between the need for a supermajority/broad agreement versus simple majority rule without any (or with few) constitutional/institutional checks on the government (McGann and Latner, 2006). In the latter case of simple majority rule, where the government can implement its agenda without compromise, it matters much more who is in power. Likewise, changes in government normally lead to substantive policy changes. Consequently, it makes more sense to vote, since this makes more of a difference than in systems where power is fragmented and shared. Overall, based on various potential institutional constraints – interstate federalism (relevant regional governments), bicameralism, judicial review, presidential power (of an elected president), referendums, concordance democracy/consociationalism, and constitutional rigidity – one can consider countries such as Belgium, Canada, Germany, and Switzerland to have supermajoritarianism due to some combination of political decentralization, institutional power sharing, constraints on the executive, and/or difficulty in changing the constitution. At the other extreme, countries such as Denmark, Greece, Iceland, Malta, Norway, and the UK are clearly simple majority (or plurality) systems. Certainly the countries where power is centralized and basically unconstrained tend to have higher turnout. Conversely, Switzerland is viewed as the country with the most constraints on the executive, including extremely common direct democracy; consequently, the very low turnout in Swiss parliamentary elections is no surprise. This is also very much the case for

50

Activating the Citizen

the EU, where although there is no direct democracy, power is certainly diffused (Lijphart, 1999). Control over the European Parliament gives little control over the executive (the European Commission) once it is confirmed and basically no control in an ideological sense (for example, electing a conservative majority to the European Parliament does not lead to a conservative executive). The eighth and last overall area of explanatory factors involves national political-cultural history, especially in terms of democracy. It has already been noted that post-communist countries have noticeably lower turnout. Perhaps the main analyst of this phenomenon is Richard Rose, who notes the higher levels of distrust of parties and government in the region (as compared with western Europe), which he attributes to the ‘legacy of repressive and corrupt communist rule’ (Rose, 2004, p. 6). If the emphasis here is on communist rule, then this obviously distinguishes the post-communist EU members from the other countries in this analysis. However, if the emphasis is more broadly on the legacy of repressive, autocratic rule, then one can invert this and argue that the longer a nation has been a democracy – with universal suffrage – the more prone its adult population is to vote. This can be measured in terms of the year in which a nation achieved its current status of being an (electoral) democracy with universal suffrage. Beyond the post-communist countries, this factor is of note for the southern European countries of Greece, Portugal, and Spain (which did not become current democracies until the 1970s) and for Switzerland (which did not achieve female and thus universal suffrage until 1971). As for the EU, since there were no pan-European elections until 1979, this year should be used for its ‘transition to democracy’. This distinction of how long a country has been a democracy with universal suffrage seems extremely significant in terms of overall turnout patterns. Not only does it speak to the generally low turnout of the new post-communist democracies, but it also relates to the very low turnout in Switzerland and the European Parliament elections. Conversely, with the exception of Greece, all of the countries with high or very high turnout have been democracies with universal suffrages for many decades. In summary, the academic literature suggests various socioeconomic, party political, and institutional variables that affect voter turnout in a cross-national comparative sense. In particular, turnout definitely does appear to be high or very high where there is a small population, ethnic homogeneity, an institutionalized party system, strictly enforced compulsory voting, concentrated winner-take-all political power (as opposed to ‘supermajoritarianism’), and/or many decades of democracy with

Alan Siaroff 51

universal suffrage. Of these points, strictly enforced compulsory voting seems to be the single most crucial one, not least because it can ‘override’ the other factors. One sees this in the case of Belgium, where strictly enforced compulsory voting has maintained very high turnout despite strong ethnic fragmentation and a supermajoritarian political system. Conversely, an inchoate party system seems to be the factor to most ‘avoid’ since turnout is low or very low in such systems without exception. Certain countries are particularly interesting in terms of turnout. Switzerland is the most supermajoritarian of the cases here and has had universal suffrage only since 1971. The EU has the equivalent of an inchoate party system (see Bardi and Calossi, this volume); as well as being highly supermajoritarian, it has been ‘democratic’ since only 1979. For the post-communist countries, one should note in particular that these are all relatively new democracies, several of them have inchoate party systems, and none of them has strictly enforced compulsory voting. Finally, some comments can also be made about Malta, which has the highest turnout without compulsory voting. As noted in the literature, there is nothing ‘unique’ about Malta beyond its very high levels of key causal factors (see the conclusions of Hirczy, 1995 and Siaroff and Merer, 2002). That is, given its very competitive two-party system, clear party political institutionalization, and low disproportionality and political system that is essentially the antithesis of supermajoritarianism, extremely high turnout in small, urbanized Malta is to be expected.

Party membership and declining political mobilization Political and social mobilization has been found to increase electoral participation (Crewe, 1981). We can look at political mobilization (outside of elections) in terms of party membership as a share of the electorate and examine various explanations offered for the patterns of decline. This leads to a consideration of other related forms of political mobilization, most notably, unionization of the workforce, which has also seen a downward trend in most of the countries we are examining. Table 2.2 shows national levels of party membership for 1997 to 2001, the most recent period for which cross-national data exist. The table also provides historical data from Mair and van Biezen (2001) on party membership for the periods of the late 1970s–1980 and the late 1980s–1991 for various countries. The primary source here for contemporary party membership (1997 to 2001) is Weldon (2006). For those countries he did not include, the sources are as follows: for Bulgaria,

52

Activating the Citizen

Table 2.2:

Levels of party membership (% of the electorate)

Austria Belgium Bulgaria Canada Cyprus Czech Republic Denmark Estonia European Union (of 25) Finland France Germany (West) Greece Hungary Iceland Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Norway Poland Portugal Romania Slovakia Slovenia Spain Sweden Switzerland United Kingdom Average

1977–1980

1988–1991

1997–2001

28.48 8.97

23.71 9.15

7.30

5.88

15.74 5.05 4.52 3.19

13.52 2.98 3.89 6.33 2.11

5.00 9.66

4.86 9.10

4.29 15.35

3.19 13.13

4.28

5.08

1.20 8.41 10.66 4.12

2.07 8.00 7.98 2.63

17.66 6.56 6.48 2.00 11.50 3.93 5.13 1.17 3.30 9.65 1.61 2.94 6.45 2.31 16.87 2.97 4.04 1.50 1.93 8.59 23.80 2.54 7.16 1.13 3.91 2.50 4.10 6.68 3.33 5.54 5.92 1.91 5.78

Sources: See text.

Spirova (2005); for Canada, William Cross (personal communication); for Cyprus, Christophorou (2006) provides the 2001 values for the two main parties and the rest are estimates; for Estonia, calculations are made from Mikkel (2006) to get the 1999 values for the five major parties of that period; and for Lithuania, calculations are made from Ramonaite˙ (2006) to get the 2000 values. For two countries, Canada and Iceland, the figures are adjusted to remove inflated values due to leadership contests/party

Alan Siaroff 53

primaries; that is, the normal base value is determined. For Iceland (which has the highest party membership in Weldon’s data), the official values can certainly be deceptive, since once one joins a political party one remains a member until death; moreover, the parties collect no annual dues. Consequently, Icelandic assessments provided by Kristjánsson (2004) are used that measure those who actually consider themselves to be party members and remove those who joined a party only to vote in a primary. Finally, no credible complete data for Latvia and Romania have been found and thus best guesses are used, largely extrapolated from the values of neighbouring countries. As the data show, for several countries – Belgium, Finland, Italy, Norway, and Spain – party membership dropped significantly in the 1990s, paralleling the aforementioned analysis of voter turnout. However, for other countries the decline was ongoing, and indeed for Greece and especially Spain, party membership has actually been going up over time – albeit from low (Greece) and very low (Spain) start levels. Party membership is in general a parallel indicator of (declining) political participation alongside voter turnout. If party membership is thus more of a parallel than a causal variable with respect to voter turnout, does it arise from the same multiple theoretical factors? Not exactly – here the list is much shorter. First and foremost, as Weldon (2006) argues and clearly demonstrates, (logged) population size relates strongly to national party membership level. Certainly, the countries currently with the highest per capita party membership have small (Austria and Finland) or tiny (Cyprus, Iceland, and Malta) populations. The other socioeconomic variable that seems to affect party membership is population density. In theory, joining a party (or any group) is easier in a more densely populated setting. Lastly, whether a country is a longstanding democracy or a new democracy also seems to affect party membership, since without democracy with universal suffrage there would be little, or limited, incentive for broad party membership. Yet none of these causal factors has clearly changed for the worse over time. Countries have grown bigger over time in terms of population, but small countries still remain small. Population density has indeed been increasing. Lastly, the longstanding democracies have become even more longstanding. Thus in terms of theoretical explanations for the decline in political participation (both voter turnout and party membership), three broad factors should be noted. The first is the decline in mobilization efforts undertaken by parties and related actors – in other words, the weakening of attempts by political actors to get people to vote (and/or become members). This thesis has been argued by Gray and Caul (2000), who

54

Activating the Citizen

contrasted the situations in the 1950s and the 1990s for 18 advanced industrial democracies. They focus on the shrinking vote share of labour parties broadly defined (including communists, socialists, and social democrats) and the decrease in union density (trade union membership), as well as the mobilization capacity of leftist forces, since they ‘have traditionally had the biggest impact on mobilizing peripheral voters’ (Gray and Caul, 2000, p. 1101). They note that in most countries, labour parties’ share of the vote has declined (as of 1997). Moreover and importantly for this analysis, for many countries the decline did not begin until the 1980s – reinforcing the importance of this as the decade of change. As for unionization density, they note that this declined in fewer cases: just over half. Yet such a decline also relates statistically to lower turnout over time. More detailed data on trends in union density can be found in the work of Visser (2006). His study of 24 industrial countries is much less equivocal than that of Gray and Caul: Union density rates in the early part of this decade are generally lower (often much lower) than in 1970, and also much lower in the EU overall. The only exceptions here are the four small European economies of Belgium, Denmark, Sweden, and Finland, where from 1970 to 2003 union density grew by 10 per cent or more (most sharply by 23 per cent in Finland). As Visser emphasizes, ‘these four happen to be the only ones in which unions are involved in the administration and execution of unemployment insurance’ (2006, pp. 45–6) and thus can deny unemployment benefits to non-union members. With these exceptions, though, a decline in unionization and related political mobilization is indeed a general pattern. The second factor worth noting – even though it does not appear to have been specifically linked to declining turnout in the comparative literature – is the decline in political polarization in industrial democracies, leading to less policy difference between a government of (or led by) party ‘A’ versus one of party ‘B’. This point, though, has been made for the UK by Heath and Taylor (1999): On rational choice grounds, if there is a great deal of difference between the parties, then the outcome is likely to make more difference to the country’s political future and hence there is more reason to turn out and vote. Conversely, if there is little difference between the parties, then the election result will not make much difference either. We would, therefore, expect to see the highest turnouts when the parties are furthest apart ideologically from each other. Strictly speaking, the standard rational choice model of turnout is a multiplicative one,

Alan Siaroff 55

and we would expect to see the highest turnout when there was both a large difference between the parties and a close race. (pp. 178–9) Certainly the evidence supports their argument: Post-war Britain’s highest turnouts (and the only ones above 80 per cent) were in the elections of 1950 and 1951, both of which involved major policy differences between Labour and the Conservatives and a close national race. More recently, the 1992 British election also combined both factors (at least in terms of the campaign; it was a close race), and it too had relatively high turnout (Heath and Taylor, 1999). In terms of measuring ideological (and consequently policy) differences between parties, two important cross-national studies of expert assessments of left–right party positions have been done on the industrial societies. The first is that of Castles and Mair (1984), which provided data on 17 countries. The second study is that of Huber and Inglehart (1995), which looked more broadly at 42 countries – including all of those in Castles and Mair. In both cases national experts were asked to score the various parties, and the average scores and standard deviations were then determined. The fact that these studies were only a decade apart might seem limiting in terms of longitudinal changes, but fortunately they fall on either side of the 1980s versus 1990s change in turnout emphasized above. One slightly complicating factor is that Castles and Mair used an 11-point scale (ranging from 0 to 10) whereas Huber and Inglehart used a 10-point scale (ranging from 1 to 10). Thus, while in both cases 10 is the most right-wing score possible, there is a 1-point different in the most left-wing score possible, which presumably would affect the placement of all leftist parties. That said, the relevant information for our purpose is that in various countries the main leftist party has clearly moved closer to the centre. So in Austria the Socialist Party scored 3.0 on the Castles and Mair index and 4.75 on the Huber and Inglehart index. Likewise the British Labour Party went from 2.3 on the first index to 4.43 on the second, and the French Socialist Party similarly went from 2.6 to 4.13. In some countries like Denmark though, little has changed in terms of polarization (and indeed turnout). Moreover, these measures do not take into account further changes from the mid-1990s onwards, in particular how the British Labour Party under Tony Blair took an even more centrist position as New Labour. In summary, the decline in polarization in many countries can certainly be argued to have affected lower election turnout. The third factor leading to declining political participation has been cultural change in the advanced industrial societies. In this regard,

56

Activating the Citizen

as Dalton (2004) shows, there has arisen a growing distrust of political authorities and scepticism about political institutions – but no decrease in support for democracy. Thus, what has arisen in the advanced industrial democracies is a broad public malaise specifically concerning politicians, political parties, and parliaments. In most countries, confidence in politicians and government has fallen steadily over the past few decades. Although there is a theoretical debate about whether voter turnout is aided by a supportive, trusting population (who will turn out to vote for its respective traditional party) or an alienated population (who will turn out to vote the incumbents out or cast protest votes), Dalton’s analysis of cross-national data shows that in almost every country it is the politically trusting who vote more often. For example, he points out that the British (self-reported) data show voting to be 87 per cent among the most trustful on a political trust scale, versus only 69 per cent among the least trusting. These attitudinal changes have led to crossnational declines not just in voter turnout but also in party identification (and presumably party membership), as well as demands for institutional reforms such as new electoral systems, term limits, and direct democracy. Overall, these changes can be related to the growth of post-materialism. One can thus build on this conclusion and hypothesize that the limited or non-existent decline in political participation in the southern European countries of Greece and Spain relates to the smaller share of post-materialists there compared with northern Europe.

Conclusions: what can be done? We have identified several factors that help explain the decline in voter turnout and political party membership over time and across countries. What kinds of measures, by governments or other actors, might halt or reverse these trends in countries such as Canada and the member states of the EU? Some explanatory factors are not subject to quick change. For example, polities obviously cannot, in short order, increase the length of their experience with democracy based on universal suffrage. At the other extreme, the one conspicuously quick, albeit highly controversial, solution is to implement strictly enforced compulsory voting. Another effective change would be to institutionalize party systems where these are inchoate, but this is a particular challenge for new party systems, as it is a long-term process. Concentrating political power will certainly increase parliamentary turnout, but such radical institutional change is most unlikely; for example, Germany or Canada are hardly likely to become unitary states. Nor can we assume that elections will

Alan Siaroff 57

become more competitive. Finally, barring any lasting depression or other socioeconomic change, the number of post-materialists is likely to keep increasing, with a presumably parallel increasing distrust of politicians. A more plausible future development to the benefit of turnout levels would be a re-emergence of more polarized politics with clearer choices and/or greater political mobilization. Both of these factors put the onus on political parties to better mobilize voters and activists to reverse the current participatory decline.

Notes 1. This analysis excludes presidential elections because these are or have been politically central in only a few of the countries covered, most notably Finland, France, Poland, and Portugal. 2. Although turnout in Poland jumped up considerably in its 2007 election, the actual level (53.8 per cent) still was very low. 3. In most cases, these conclusions are drawn using Fearon’s (2003) cross-national data on ethnic fractionalization.

References Baimbridge, M. D. (2005) ‘Euphoria to Apathy: EP Turnout in the New Member States’ in Juliet Lodge (ed.) The 2004 Elections to the European Parliament (Basingstoke: Palgrave Macmillan), pp. 45–56. Bernhagen, P. and M. Marsh (2007) ‘The Partisan Effects of Low Turnout: Analyzing Vote Abstention as a Missing Data Problem’, Electoral Studies, 26(3), 548–60. Blais, A. and K. Carty (1990) ‘Does Proportional Representation Foster Voter Turnout?’, European Journal of Political Research, 18, 167–81. Blais, A. and A. Dobrzynska (1998) ‘Turnout in Electoral Democracies’, European Journal of Political Research, 33, 239–61. Blais, A., E. Gidengil, N. Nevitte, and R. Nadeau (2004) ‘Where Does Turnout Decline Come From?’, European Journal of Political Research, 43(2), 221–36. Castles, F. G. and P. Mair (1984) ‘Left-Right Political Scales: Some “Expert” Judgements’, European Journal of Political Research, 12, 73–88. Christophorou, D. (2006) ‘Party Change and Development in Cyprus (1995– 2005)’, South European Society and Politics, 11(3–4), 513–42. Cox, Gary W. (1999) ‘Electoral Rules and the Calculus of Mobilization’, Legislative Studies Quarterly, 24, 387–419. Crewe, I. (1981) ‘Electoral Participation’ 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), pp. 216–63. Dahl, R. A. and E. R. Tufte (1973) Size and Democracy (Stanford, CA: Stanford University Press).

58

Activating the Citizen

Dalton, R. J. (2004) Democratic Challenges, Democratic Choices: The Erosion of Political Support in Advanced Industrial Democracies (New York: Oxford University Press). Fearon, J. D. (2003) ‘Ethnic and Cultural Diversity by Country’, Journal of Economic Growth, 8, 195–222. Franklin, M. N. (2002) ‘The Dynamics of Electoral Participation’ in L. LeDuc, R. G. Niemi, and P. Norris (eds) Comparing Democracies 2: New Challenges in the Study of Elections (London and Thousand Oaks, CA: Sage), pp. 148–68. Franklin, M. N. (2004) Voter Turnout and the Dynamics of Electoral Competition in Established Democracies since 1945 (New York: Cambridge University Press). Franklin, M. N., P. Lyons, and M. Marsh (2004) ‘The Generational Basis of Turnout Decline in Advanced Democracies’, Acta Politica, 39(2), 115–51. Geys, B. (2006) ‘Explaining Voter Turnout: A Review of Aggregate-Level Research’, Electoral Studies, 25, 637–63. Gratschew, M. (2001) ‘Compulsory Voting’, http://www.idea.int/vt/compulsory_ voting.cfm Gray, M. and M. Caul (2000) ‘Declining Voter Turnout in Advanced Industrial Democracies, 1950 to 1997: The Effects of Declining Group Mobilization’, Comparative Political Studies, 33(9), 1091–122. Heath, A. and B. Taylor (1999) ‘New Sources of Abstention?’ in G. Evans and P. Norris (eds) Critical Elections: British Parties and Voters in Long-Term Perspective (London: Sage), pp. 164–80. Hirczy, W. (1995) ‘Explaining Near-Universal Turnout: The Case of Malta’, European Journal of Political Research, 27, 255–72. Hix, S. (2002) ‘Parliamentary Behavior with Two Principals: Preferences, Parties, and Voting in the European Parliament’, American Journal of Political Science, 46(3), 688–98. Huber, J. and R. Inglehart (1995) ‘Expert Interpretations of Party Space and Party Locations in 42 Societies’, Party Politics, 1(1), 73–111. International Institute for Democracy and Electoral Assistance (2002) ‘The International IDEA Database: Voter Turnout from 1945 to 2001’ in R. López Pintor and M. Gratschew (eds) Voter Turnout since 1945: A Global Report (Stockholm, Sweden: International IDEA), pp. 117–69. Johnston, R., J. Scott Matthews, and A. Bittner (2007) ‘Turnout and the Party System in Canada, 1988–2004’, Electoral Studies, 26(4), 735–45. Kostadinova, T. (2003) ‘Voter Turnout Dynamics in Post-Communist Europe’, European Journal of Political Research, 42, 741–59. Kristjánsson, S. (2004) ‘Iceland: Searching for Democracy along Three Dimensions of Citizen Control’, Scandinavian Political Studies, 27(2), 153–74. Lijphart, A. (1984) Democracies: Patterns of Majoritarian and Consensus Government in Twenty-One Countries (New Haven, CT and London: Yale University Press). Lijphart, A. (1997) ‘Unequal Participation: Democracy’s Unresolved Dilemma’, American Political Science Review, 91, 1–14. Lijphart, A. (1999) Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries (New Haven, CT and London: Yale University Press). Lutz, G. and M. Marsh (2007) ‘Introduction: Consequences of Low Turnout’, Electoral Studies, 26(3), 539–47. Mainwaring, S. and T. R. Scully (1995) ‘Introduction: Party Systems in Latin America’ in Mainwaring and Scully (eds) Building Democratic Institutions: Party Systems in Latin America (Stanford, CA: Stanford University Press), pp. 1–34.

Alan Siaroff 59 Mair, P. (2007) ‘Political Parties and Party Systems’ in P. Graziano and M. P. Vink (eds) Europeanization: New Research Agendas (Basingstoke: Palgrave Macmillan), pp. 154–66. Mair, P. and I. van Biezen (2001) ‘Party Membership in Twenty European Democracies, 1980–2000’, Party Politics, 7(1), 5–21. McGann, A. J. and M. Latner (2006) ‘Consensus Without Veto-Players: Testing Theories of Consensual Democracy’, Center for the Study of Democracy, University of California, Irvine, CA, Paper 06-03, http://repositories.cdlib.org/csd/ 06-03 Mikkel, E. (2006) ‘Patterns of Party Formation in Estonia: Consolidation Unaccomplished’ in S. Jungerstam-Mulders (ed.) Post-Communist EU Member States: Parties and Party Systems (Aldershot: Ashgate), pp. 23–49. Ramonaite, ˙ A. (2006) ‘The Development of the Lithuanian Party System: From Stability to Perturbation’ in S. Jungerstam-Mulders (ed.) Post-Communist EU Member States: Parties and Party Systems (Aldershot: Ashgate), pp. 69–90. Rose, R. (2004) Europe Expands, Turnout Falls: The Significance of the 2004 European Parliament Election (Stockholm, Sweden: International Institute for Democracy and Electoral Assistance). Siaroff, A. and J. Merer (2002) ‘Parliamentary Election Turnout in Europe since 1990’, Political Studies, 50(5), 916–27. Spirova, M. (2005) ‘Political Parties in Bulgaria: Organizational Trends in Comparative Perspective’, Party Politics, 11(5), 601–22. Visser, J. (2006, January) ‘Union Membership Statistics in 24 Countries’, Monthly Labor Review, 38–49. Wass, H. (2007) ‘The Effects of Age, Generation and Period on Turnout in Finland 1975–2003’, Electoral Studies, 26(3), 648–59. Weldon, S. (2006) ‘Downsize My Polity? The Impact of Size on Party Membership and Member Activism’, Party Politics, 12(4), 467–81.

This page intentionally left blank

Part II Voter Turnout: Understanding the Decline

This page intentionally left blank

3 Turnout in Electoral Democracies Revisited André Blais and Agnieszka Dobrzynska

When and where is turnout highest and lowest and why? In a paper published in 1998 in the European Journal of Political Research, we examined the impact of the socio-economic environment, political institutions, and the party system on turnout in 324 democratic national lower house elections held in 91 countries between 1972 and 1995. Our initial findings indicated that a great number of factors had a relatively strong effect on electoral participation. Namely, we found that ‘turnout is likely to be highest in a small, industrialized, densely populated country, where the national lower house election is decisive, voting is compulsory and the voting age is 21, having a PR system with relatively few parties and a close electoral outcome’ (Blais and Dobrzynska, 1998, p. 251). The existing literature on comparative voter turnout stresses the impact of institutional variables. However, as Blais (2006) notes, ‘many of the findings in the comparative cross-national research are not robust.’ We need to corroborate patterns observed in well-established democracies as the democratic world is greatly expanding. In the present study, we verify the robustness of our initial findings. The decision to vote or not to vote is of course an individual one that is affected by a whole set of attitudes and perceptions (see Blais, 2000; Clarke et al., 2004, chapter 8). But, as Franklin (2004) has forcefully argued, those individual decisions are equally shaped by systemic and contextual factors. These are the factors that are the focus of this chapter. Ideally, we would want to incorporate both individual level and contextual level variables and interaction effects between the two into a comprehensive model of turnout but that is an agenda for future research. We expand and improve our initial analysis of the impact of contextual variables in two ways. First, we re-examine our original results using a new data set that covers the period until 2004. Second, we 63

64

Activating the Citizen

ascertain whether the same patterns hold without the inclusion of regional (continental) dummy variables as well as inside and outside Europe. We use the same basic approach. We distinguish three blocs of variables that are plausible determinants of turnout: the socioeconomic environment, the institutional setting, and the party system. We start with the first bloc, examine which socio-economic factors influence turnout, then explore the impact of a variety of institutions controlling for socioeconomic environment, and finally consider the party system and its independent effect on electoral participation. The sample includes all democratic national lower house elections in the world held between 1972 and 2004 (see Appendix 3.1).1 We re-examine the hypotheses and the findings presented in our initial study. We make two modifications. First, the initial study included continental dummy variables to control for ‘the impact of other unmeasured factors which may be specific to a given geographic area’ (Blais and Dobrzynska, 1998, p. 243). We are sensitive to the risks involved in incorporating too many control variables (Achen, 2002), especially when it is not clear what these variables measure. We have thus decided to drop the continental dummy variables. At the same time, we are particularly concerned with the robustness of our findings, especially given the observation that some of the standard results in the literature do not seem to hold as well in non-established democracies (Blais, 2006). We perform separate analyses for Europe, where almost half our cases are situated, and the rest of the world.2 Looking separately at Europe is particularly interesting since the continent has a long history of elections as compared to the rest of the world, it includes the world’s most successful democracies in terms of durability, and it has a wide variety of electoral institutions and party systems. Furthermore, turnout has recently declined in most places (see Chapter 2). Chapter 2 provides a global overview of turnout in Europe and Canada and discusses eight sets of factors that can explain why turnout is higher or lower in some countries: socio-economic, party system fragmentation, competitiveness, party system institutionalization, disproportionality, compulsory voting, the importance of national parliament, and democratic history. In this chapter, we will be looking at basically the same variables, but regrouped differently in three blocs. The difference is that we will be looking at the impact of these contextual factors outside as well as inside Europe, that we will explicitly test a multivariate model, that we will be particularly concerned with comparing our new results

André Blais and Agnieszka Dobrzynska 65

with those previously reported, and that we will confine ourselves to electoral participation (while Chapter 2 also covers party membership).

Variations in turnout Our goal is to explain why turnout is higher (or lower) in some countries and/or at some times than in others. The first step is to look at variations in turnout. Appendix 3.1 shows turnout (as a percentage of registered electors) in each of the 451 elections covered in our study. Mean turnout is 75 per cent, the median is 77 per cent. The highest turnout ever registered is 99.3 per cent, in the 1972 Fiji election, and the lowest is 21.1 per cent, in the 1992 Mali election. In Europe, the records are the 1996 election in Malta (97.2 per cent) and the 1995 Swiss election (42.2 per cent). Variations are pretty substantial. Figure 3.1 shows the overall distribution. There is no concentration around the mean (or median). There are as many cases in the 80 per cent range or in the 90 per cent range as in the 70 per cent range. It should be noted that very low turnout below 50 per cent, as well as turnout above 95 per cent, occurs extremely rarely. Table 3.1 presents mean turnout by continents, decades, and level of democracy. Turnout is highest in Oceania, followed by Europe, and lowest in Africa.3 We observe an important declining pattern that started in the 1990s. Mean turnout went down in that period by about five points,

Number of elections

80

60

40

20

0 0

20

40

60

80

100

Turnout (%) Figure 3.1: The overall distribution of turnout in 451 democratic elections (1972–2004)

66

Activating the Citizen

Table 3.1: Mean turnout (%) in 451 democratic elections by continents, decades and level of democracy (1972–2004) 70’s

80’s

90’s

00’s

Entire period

Europe North America South America Africa Asia Oceania

84 73 70 63 73 92

81 73 76 72 78 88

75 70 76 67 73 84

71 60 78 67 66 87

78 69 76 67 73 87

Established democracies New democracies

84 71

82 75

78 70

74 66

80 71

All countries

80

79

74

69

75

Note: Turnout is measured as the percent of those registered who cast a vote.

in Europe as elsewhere, and has declined further in the first years of the twenty-first century (the same pattern is noted in Chapter 2). The exceptions are South America where turnout has remained quite stable, and Oceania, where recent turnout is similar to that observed in the 1980s. As for level of democracy4 , turnout is higher in established democracies but the same declining pattern emerges in established as well as new democracies. The main source of this decline appears to be generational replacement, as turnout is much lower among new generations (Miller and Shanks, 1996; Lyons and Alexander, 2000; Blais et al., 2004; Clarke et al., 2004; Dalton, 2007). We may note that turnout used to be slightly above 70 per cent in Canada and that it is now slightly below 60 per cent. Such a turnout is typical in the North American (including Central American) continent and is about 10 percentage points lower than in Europe. The time trend in Canada, a decline of about 10 points, is similar to that observed in North America, Europe, and, indeed, the whole world.

The socio-economic environment We start with the most distant set of factors that could affect turnout, the socio-economic environment. We focus on the two socio-economic variables that are most closely associated with turnout, as revealed in the previous study: GDP per capita and size of population (see Appendix 3.2 for a more detailed description of variables).5 Those two variables were consistently significant in our initial study. We expect

André Blais and Agnieszka Dobrzynska 67 Table 3.2: The determinants of electoral participation: socio-economic environment (OLS regressions, robust estimates) All countries Variables

B

(Error)

GDP per capita (log) 15.83∗∗∗ Size of population (log) −1.46 Switzerland −35.40∗∗∗ Constant

17.83

N R2

Outside Europe B

B

(Error)

(3.09) 13.07∗∗ (4.94) 20.54∗∗∗ (1.30) −.67 (1.50) −3.66 (1.84) – −37.06∗∗∗ (12.18) 25.24

451 .22

(Error)

In Europe

(18.96) 244 .10

7.20

(6.42) (2.37) (1.85) (30.12)

207 .38

∗

= significant at the .10 level = significant at the .05 level ∗∗∗ = significant at the .01 level ∗∗

economic development to make people more politically sophisticated and engaged and this should foster turnout (Powell, 1982).6 We assume that this relationship is non-linear (logarithmic), which means that the most important difference is between countries with some modest level of economic development and all other countries, and that over a certain level of development, turnout does not increase substantially. As for size of population, we expect turnout to be higher in smaller countries where social and political activities are easier to organize and mobilize. Again, we examine the logarithmic relationship, since the main difference should be between smaller countries and all other countries. We also include through all our analyses a dummy variable for Switzerland. As Franklin (2004) persuasively argues, Switzerland is clearly an outlier, and its exceptionally low turnout can be imputed to the presence of a pact reached by the four major parties to share the collective executive, which means that there is no longer a connection between the outcome of legislative elections and the formation of the government. This makes electoral outcomes more or less irrelevant, and contributes to an excessively low turnout. The findings are presented in Table 3.2. They confirm that the level of economic development is strongly correlated with turnout; but size of population, which came out as an important variable in the 1998 study, never reaches statistical significance here (though it always has the correct sign). Those patterns apply for all countries as well as outside and inside Europe.

68

Activating the Citizen

With respect to the first bloc, the socio-economic environment, our findings are mixed. We are quite pleased to report that our initial conclusion that turnout is strongly affected by economic development holds well. Greater level of economic development facilitates electoral participation though it must be kept in mind that the only relevant contrast is between poor countries and all others; it does not matter whether a country is rich or not, provided it is not poor. At the same time our new results force us to admit that it is not clear whether a larger population size contributes to a lower turnout.

Political institutions At the second stage we re-examine the impact of political institutions on electoral participation. The general idea is that turnout should be higher when and where institutional arrangements make the act of voting easier or make the outcome of the election particularly meaningful. However, the general findings on the impact of institutions are not very consistent (Blais, 2006), and we need to pay particular attention to the robustness of our results. We include five institutional factors that were revealed to be important in our previous study: compulsory voting, low voting age, new democracy, decisiveness and proportional representation. Our expectations are as follows. In those countries where the law makes it compulsory to vote, the cost of not voting is obviously higher and as a consequence abstention is lower.7 On the opposite side, turnout should be lower when and where the law establishes a low voting age,8 as younger voters are usually less interested in politics and less prone to vote (Wolfinger and Rosenstone, 1980; Blais et al., 2004). Proportional representation9 should foster turnout for three main reasons (Powell, 1986; Jackman, 1987; Blais and Carty, 1990; Franklin, 1996; Blais and Dobrzynska, 1998). First, because in a PR system the percentage of seats a party gets tends to be about the same as its vote share, the system may appear to be fair, and that may make people more inclined to participate. Second, proportional representation increases the number of parties contesting an election and this leads to a broader array of interests to be represented in parliament. Third, the multi-member districts utilized under PR may make the election more competitive and as a consequence political parties have more incentives to mobilize voters nationwide. Mobilization should increase turnout. The decisiveness of elections pertains to the amount of power that is bestowed on those elected. The election is more decisive where the

André Blais and Agnieszka Dobrzynska 69

national lower house is more powerful. With an elected upper house, with an elected president or in a federal country with sub-national elections, the lower house has to share power with other institutions. The decisiveness of the lower house election is thus reduced if the election for the upper house, the president or the sub-national government is not held simultaneously (see Appendix 3.2 for a detailed description of variables). When the lower house election is just one among many equally important ones, it is less likely to be perceived as very important. Consequently, we expect turnout to be higher in more decisive elections (Franklin, 2004). The last institutional factor to be considered pertains to age of democracy. In our sample we can distinguish ‘established’ and ‘new’ democracies.10 The assumption is that pro-democratic attitudes are weaker in new democracies. Moreover, in many new democracies, citizens’ first experience with democracy has been unfavourable (Inglehart 2003). As a consequence we expect turnout to be lower in new democracies. In our previous study which covered the 1972–1995 period, all those institutional variables had important effects.11 The new findings are presented in Table 3.3. Compulsory voting seems still to be the most

Table 3.3: The determinants of electoral participation: socio-economic environment and institutional setting (OLS regressions, robust estimates) All countries Variables

B

(Error) ∗∗

GDP per capita (log) 7.00 Size of population (log) −1.14 Switzerland −35.24∗∗∗ Compulsory voting 14.90∗∗∗ New democracy −6.00∗∗ Decisiveness 8.20∗∗ Low voting age −4.59∗∗ PR 5.92∗∗∗ Constant

47.59

N R2 ∗

= significant at the .10 level = significant at the .05 level ∗∗∗ = significant at the .01 level ∗∗

451 .43

(3.36) (1.10) (2.26) (3.70) (2.32) (3.71) (1.94) (1.98) (13.80)

Outside Europe B

(Error)

In Europe B

(Error)

∗∗

9.35 (4.46) −.90 (6.56) .27 (.97) −1.09 (1.95) – −37.04∗∗∗ (2.49) 22.41∗∗∗ (3.14) 5.24∗ (2.57) −1.63 (3.12) −11.19∗∗∗ (2.86) 14.66∗∗ (5.89) 8.44∗ (4.67) −1.59 (2.33) −6.54∗∗∗ (2.03) 3.89 (2.44) 10.08∗∗∗ (2.26) 23.18 (19.80) 80.67 (27.96) 244 .35

207 .65

70

Activating the Citizen

powerful incentive when fostering turnout (see also Chapter 2). However, its impact is weaker in Europe. Decisiveness also matters both inside and outside Europe (the same relationship is noted in Chapter 2). The coefficient of the proportional representation dummy variable has the correct sign across analyses but it is weak and not statistically significant outside Europe. The same pattern emerges with respect to low voting age and new democracy: both variables are associated with lower turnout in all countries and in Europe but they are not significant outside Europe. Clearly, some of the previously analysed institutional variables do not matter as much in countries outside Europe, and thus our initial findings are not very robust. It should be noted that the level of economic development has a much weaker effect after we introduce institutional variables. In particular, this variable does not hold in Europe anymore. In order to make sense of this result we have to take into account another variable, that is, level of democracy, which is strongly related to the level of economic development. In the third column of Table 3.3 we can observe that the coefficient for ‘new democracy’ in Europe is highly significant and is almost twice as big as for the sample of all countries while the result is null for ‘GDP per capita’. This means that in Europe the main difference in turnout is between new and established democracies. While according to Table 3.2 level of the economic development fosters electoral participation, in Europe this is so only because richer countries are generally established democracies. Outside Europe in contrast, it appears that the most crucial factor is the level of economic development, not age of democracy.

Party system In the last stage of our analysis we add variables that concern party system and reassess their independent additional impact on turnout. Two aspects of the party system have been assumed to affect turnout. The first is competitiveness as measured in terms of closeness.12 The basic hypothesis, inspired by the rational choice model, is that the closer the outcome of the election is, the higher the probability individual votes could decide the outcome (see Blais, 2000; Geys, 2006). As for number of parties two opposite effects are possible.13 First, the presence of more parties means that voters have more choice, and this should bolster turnout. Second, the task of making up one’s mind whom to vote for becomes more complex when there are many parties to choose from. Furthermore, the more parties there are, the higher the probability of coalition governments is. And election outcomes are less decisive

André Blais and Agnieszka Dobrzynska 71

when they lead to the formation of coalition governments, as the latter depend on deals between the parties about which voters have no (direct) say (Downs, 1957; Jackman, 1987). In short, the number of parties could have both a positive and a negative effect, and so it is difficult to come up with a clear prediction. Table 3.4 presents the results. We can confirm our previous finding that turnout tends to decrease when the number of parties increases. Initially, in our first study, this variable was clearly significant. However, with our new dataset we can see that this result is not very robust. The negative impact of number of parties is clear in Europe but outside Europe this variable is not statistically significant and for all countries it is only marginally significant. As for closeness, the coefficient does have the expected negative sign (the greater the gap between the leading and the second party, the lower turnout is) but, contrary to our initial findings, the effect is not statistically significant. With a new data set that expands the period over ten years the impact of the closeness on electoral participation becomes more ambiguous.

Table 3.4: The determinants of electoral participation: socio-economic environment, institutional setting and party system (OLS regressions, robust estimates) All countries Variables

B

(Error)

GDP per capita (log) 7.25∗∗ (3.24) Size of population (log) −.26 (1.16) Switzerland −33.99∗∗∗ (2.21) Compulsory voting 14.98∗∗∗ (3.43) (2.21) New democracy −5.87∗∗ Decisiveness 9.24∗∗ (3.49) Low voting age −5.07∗∗∗ (1.80) PR 7.19∗∗∗ (2.09) Closeness −.06 (.06) Number of parties (log) −4.15∗ (2.18) Constant 49.90 (13.34) N R2 ∗

= significant at the .10 level = significant at the .05 level ∗∗∗ = significant at the .01 level ∗∗∗

451 .44

Outside Europe B

(Error)

8.99∗∗ .24

(4.46) (1.15)

In Europe B

(Error)

– 22.16∗∗∗ (3.24) −1.46 (3.16) 14.92∗∗ (5.93) −1.85 (2.28) 3.67 (2.87) −.06 (.07) .07 (2.55) 25.42 (19.52)

.62 (5.35) .19 (1.41) −33.90∗∗∗ (2.28) 7.67∗∗ (2.95) −10.85∗∗∗ (2.18) 10.10∗∗∗ (3.40) −7.02∗∗∗ (1.59) 11.23∗∗∗ (1.99) −.05 (.09) −8.78∗∗∗ (2.30) 84.32 (20.09)

244 .35

207 .71

72

Activating the Citizen

Conclusion We have revisited our original study on the impact of the socio-economic environment, political institutions, and the party system on turnout, which was published ten years ago. We have enlarged our data set, which now covers all democratic elections from 1972 until 2004, and we have performed separate analyses inside and outside Europe. Our re-examination of the expanded data set yields very robust findings with respect to two variables: compulsory voting and the decisiveness of elections. In those countries where the electoral law makes voting compulsory and where the lower house is relatively powerful, the incentive to vote is clearly stronger. Our initial results concerning these two variables are clearly confirmed with the new data set, and the pattern holds both inside and outside Europe. With respect to the socio-economic environment, there is one intriguing finding. On the one hand, the level of economic development seems to matter a lot outside Europe but in Europe the crucial variable appears to be the contrast between new and established democracies. The implication is that turnout is substantially lower than one could expect in Eastern Europe. We have no ready explanation for why it is so but clearly this requires further investigation. One pattern clearly emerges from our reassessment: many of the previously examined factors do not matter as much in countries outside Europe, and this throws some doubt on the robustness of our initial findings. Low voting age, proportional representation, new democracy and number of parties are all significant variables in Europe but not outside Europe. The implication is that as the process of democratization goes on, and as the number of cases outside Europe increases, and as our studies include more and more of these cases, we may be prone to come up with weaker patterns. This is certainly not what we were hoping for. We had no theoretical reason, ex ante, to suppose that the inclusion of more non-European cases would lead to more ambiguous findings. The reasons why proportional representation and closeness could boost turnout are not specific to the European context, they should hold everywhere. And more cases should help to obtain more statistically significant patterns. The question, then, is this: why this contrast between Europe and the rest of the world? One possibility is that the contrast is between established and non-established democracies. According to that interpretation, politics tends to be more volatile in new democracies and the patterns that we observe among established democracies will eventually

André Blais and Agnieszka Dobrzynska 73

set in as new democracies mature. That may be the case. But there is another possibility. It may be that the interpretations that political scientists have proposed since the seminal work of Powell (1980, 1982) to explain cross-national variations in turnout were elaborated in good part to account, ex post, for well-known differences within Europe, and that these interpretations just do not work very well in other settings. The inconsistent findings presented here should induce us to revisit our theories about turnout, as well as to revisit the conventional wisdom that turnout depends first and foremost on institutional factors. It is important in this respect to come back to the observation made at the very beginning about the overall decline in turnout that has taken place since the 1990s. Turnout has declined in the absence of institutional change, and elections have not become less competitive recently. The most plausible explanation for the recent turnout decline, as we have indicated, above, is generational replacement, even though the exact reasons why new generations are less inclined to vote remain nebulous. If this interpretation is valid (and we think it is), the implication is that models of turnout should include variables measuring the proportion of different cohorts in the electorate (see Blais and Rubenson, 2007). Institutions matter, but they may not matter as much as we have been led to think, and other kinds of variables need to be taken into account. One way of illustrating the limits of an institutional account of turnout variations is to see what predictions the typical contextual models yield for the Canadian case. If we were to use the model shown in the last column of Table 3.4, which includes all European elections and provides quite a good overall fit of the data with an R squared of .71, to predict turnout in the 2000 and 2004 elections, the prediction would have been a turnout of 71 per cent in both cases, which corresponds to mean turnout in Europe in recent elections. Yet, as we know, turnout in those two years was 61 per cent. The implication is that the lower turnout that we observe in Canada relative to Europe cannot be explained by the set of factors included in our model. Either electoral institutions do not matter as much as we thought or they affect turnout in more complex ways than we anticipated.

Appendix 3.1: Democratic elections 1972–2004 and participation rate Country and election date

Participation Country and (%) election date

Participation Country and (%) election date

Participation Country and (%) election date

Participation (%)

Europe Austria 75 Austria 79 Austria 83 Austria 86 Austria 90 Austria 94 Austria 95 Austria 99 Austria 02 Belgium 74 Belgium 77 Belgium 78 Belgium 81 Belgium 85 Belgium 87 Belgium 91 Belgium 95 Belgium 99 Belgium 03 Bulgaria 91 Bulgaria 94 Bulgaria 97 Bulgaria 01 Croatia 00 Croatia 03 Czech Republic 96 Czech Republic 98 Czech Republic 02 Denmark 73

92.9 92.2 92.6 90.5 86.1 81.9 86.0 80.4 80.5 90.3 95.1 94.8 94.6 93.6 93.4 92.7 91.2 90.6 96.3 83.9 75.2 58.9 66.6 76.5 61.7 76.3 74.0 57.9 88.7

France 73 France 78 France 81 France 86 France 88 France 93 France 97 France 02 Germany 72 Germany 76 Germany 80 Germany 83 Germany 87 Germany 90 Germany 94 Germany 98 Germany 02 Greece 74 Greece 77 Greece 81 Greece 85 Greece 89 Greece 93 Greece 96 Greece 00 Greece 04 Hungary 90 Hungary 94 Hungary 98

81.3 83.2 70.9 78.5 66.2 68.9 68.0 60.3 91.1 90.7 88.6 88.4 84.3 77.8 79.0 82.2 79.1 79.5 81.1 81.5 83.8 84.5 83.0 76.3 75.0 76.5 65.1 68.9 56.7

Italy 83 Italy 87 Italy 92 Italy 94 Italy 96 Italy 01 Latvia 95 Latvia 98 Latvia 02 Lithuania 96 Lithuania 00 Lithuania 04 Luxembourg 74 Luxembourg 79 Luxembourg 84 Luxembourg 89 Luxembourg 94 Luxembourg 99 Luxembourg 04 Malta 76 Malta 81 Malta 87 Malta 92 Malta 96 Malta 98 Malta 03 Moldova 98 Moldova 01 Netherlands 72

89.0 90.5 87.3 86.1 82.9 81.4 71.9 71.9 71.2 52.9 58.2 46.1 90.1 88.9 88.8 87.3 88.3 86.5 91.7 95.0 94.6 96.1 96.0 97.2 95.4 95.7 69.1 70.0 83.5

Portugal 80 Portugal 83 Portugal 85 Portugal 87 Portugal 91 Portugal 95 Portugal 99 Portugal 02 Romania 96 Romania 00 Slovakia 94 Slovakia 98 Slovakia 02 Slovenia 92 Slovenia 96 Slovenia 00 Slovenia 04 Spain 79 Spain 82 Spain 86 Spain 89 Spain 93 Spain 96 Spain 00 Spain 04 Sweden 73 Sweden 76 Sweden 79 Sweden 82

85.4 78.6 75.4 72.6 67.8 66.3 61.0 62.8 76.0 65.3 75.4 84.2 70.1 85.9 73.7 70.4 60.1 68.1 79.8 70.6 69.9 77.0 78.1 68.7 77.2 90.8 91.8 90.7 91.4

Denmark 75 Denmark 77 Denmark 79 Denmark 81 Denmark 84 Denmark 87 Denmark 88 Denmark 90 Denmark 94 Denmark 98 Denmark 01 Estonia 95 Estonia 99 Estonia 03 Finland 72

88.2 88.0 85.6 83.2 88.4 86.7 85.7 82.8 84.3 85.9 87.1 68.9 57.4 58.2 81.4

Hungary 02 Iceland 74 Iceland 78 Iceland 79 Iceland 83 Iceland 87 Iceland 91 Iceland 95 Iceland 99 Iceland 03 Ireland 73 Ireland 77 Ireland 81 Ireland 82 (Feb.) Ireland 82 (Nov.)

73.5 91.4 90.3 89.3 88.6 90.1 87.5 87.4 84.1 87.7 76.6 76.3 76.2 73.8 72.8

Netherlands 77 Netherlands 81 Netherlands 82 Netherlands 86 Netherlands 89 Netherlands 94 Netherlands 95 Netherlands 98 Netherlands 03 Norway 73 Norway 77 Norway 81 Norway 85 Norway 89 Norway 93

88.0 87.0 81.0 85.8 80.1 78.7 79.1 73.2 80.0 80.2 82.9 82.0 84.0 83.2 75.8

Finland 75

73.8

Ireland 87

73.3

Norway 97

78.0

Finland 79 Finland 83 Finland 87 Finland 91 Finland 95 Finland 99 Finland 03

75.3 75.7 72.1 68.3 68.0 65.3 69.7

Ireland 89 Ireland 92 Ireland 97 Ireland 02 Italy 72 Italy 76 Italy 79

65.5 68.5 66.1 62.6 93.2 93.4 91.1

Norway 01 Poland 91 Poland 93 Poland 97 Poland 01 Portugal 76 Portugal 79

75.0 43.2 52.1 47.9 46.2 83.3 87.5

Sweden 85 Sweden 88 Sweden 91 Sweden 94 Sweden 98 Sweden 02 Switzerland 75 Switzerland 79 Switzerland 83 Switzerland 87 Switzerland 91 Switzerland 95 Switzerland 99 Switzerland 03 United Kingdom 74 (Feb.) United Kingdom 74 (Oct.) United Kingdom 79 United Kingdom 83 United Kingdom 87 United Kingdom 92 United Kingdom 97 United Kingdom 01

89.9 86.0 86.7 88.1 81.4 80.1 52.4 48.1 48.9 46.1 46.0 42.2 43.2 45.4 78.9

Mali 92 Mauritius 95 Mauritius 00 Namibia 94 Namibia 99 Sao Tome & Principe 91

21.1 79.7 80.9 76.0 62.8 76.7

72.9 76.3 72.8 75.4 77.8 71.5 59.4

Africa Benin 91 Benin 99 Benin 03 Botswana 74 Botswana 79 Botswana 84

51.7 70.1 55.9 33.0 58.0 76.0

Botswana 89 Botswana 94 Botswana 99 Botswana 04 Cape Verde 91 Cape Verde 01

68.2 76.6 77.1 76.2 75.3 54.1

Gambia 72 Gambia 77 Gambia 92 Ghana 00 Lesotho 02 Malawi 94

79.5 82.0 55.8 61.2 66.7 80.0

(Continued)

Appendix 3.1 : Continued Country and Participation Country and election date (%) election date Sao Tome & Principe 94 Sao Tome & Principe 98

52.1 64.7

Sao Tome & Principe 02 South Africa 94

Participation Country and (%) election date 66.3

South Africa 99 South Africa 04

Participation Country and (%) election date

Participation (%)

89.3 76.7

Zambia 91

45.4

67.9 71.4 73.3 67.0 59.0 60.6 59.9 88.4 88.1 75.7 71.9

South Korea 96 South Korea 00 South Korea 04 Taiwan 98 Taiwan 01 Taiwan 04 Thailand 01 Turkey 77 Turkey 87 Turkey 91

63.9 57.2 60.0 68.1 66.2 59.2 69.8 70.0 93.3 83.9

82.5 83.7 89.0 91.7 87.2 85.2 85.2 88.3 84.8

New Zealand 02 Papua New Guinea 82 Papua New Guinea 97 Samoa 96 Samoa 01 Vanuatu 83 Vanuatu 98

77.0 51.7 65.7 86.3 82.5 98.7 39.9

86.9

Asia Cyprus 81 Cyprus 85 Cyprus 91 Cyprus 01 India 84 India 89 India 96 India 98 India 99 Israel 73 Israel 77

95.8 94.6 94.3 91.8 63.5 58.2 57.9 62.0 59.7 78.6 79.2

Israel 81 Israel 84 Israel 88 Israel 92 Israel 96 Israel 99 Israel 03 Japan 72 Japan 76 Japan 79 Japan 80

78.5 78.8 79.7 77.4 79.3 78.7 67.8 71.7 73.4 68.0 74.6

Japan 83 Japan 86 Japan 90 Japan 93 Japan 96 Japan 00 Japan 03 Mongolia 96 Nepal 91 South Korea 88 South Korea 92 Oceania

Australia 72 Australia 74 Australia 75 Australia 77 Australia 80 Australia 83 Australia 84 Australia 87 Australia 90

95.4 95.4 95.4 95.0 94.4 94.6 94.2 93.8 95.5

Australia 93 Australia 96 Australia 98 Australia 01 Australia 04 Fiji 72 Fiji 82 Fiji 99 New Zealand 72

95.8 95.8 95.2 94.9 94.6 99.3 86.0 89.4 89.1

New Zealand 75 New Zealand 78 New Zealand 81 New Zealand 84 New Zealand 87 New Zealand 90 New Zealand 93 New Zealand 96 New Zealand 99

South America Argentina 85 Argentina 87 Argentina 89 Argentina 91 Argentina 93 Argentina 95 Argentina 97 Argentina 99 Bolivia 85 Bolivia 89 Bolivia 93

82.2 83.6 84.2 89.7 79.7 81.0 78.2 78.6 82.0 73.7 72.2

Bolivia 97 Bolivia 02 Brazil 86 Brazil 90 Brazil 94 Brazil 02 Chile 93 Chile 01 Columbia 74 Columbia 78 Columbia 82

71.4 72.1 85.0 84.6 82.2 68.9 91.0 86.6 57.1 33.0 41.0

Columbia 86 Columbia 91 Ecuador 84 Ecuador 86 Ecuador 88 Ecuador 90 Ecuador 98 Peru 80 Peru 85 Peru 01 Surinam 77

43.6 31.9 71.2 74.0 77.7 67.7 47.3 80.1 80.5 81.4 77.8

Surinam 00 Uruguay 89 Uruguay 94 Uruguay 99 Uruguay 04 Venezuela 73 Venezuela 78 Venezuela 83 Venezuela 88

69.8 88.7 91.4 91.7 89.6 96.5 87.5 87.8 81.7

North America Antigua & Barbuda 84 Antigua & Barbuda 89 Bahamas 77 Bahamas 82 Bahamas 87 Bahamas 92 Bahamas 97 Bahamas 02 Barbados 76 Barbados 81 Barbados 86 Barbados 91 Barbados 94 Barbados 99 Barbados 03

61.1

Canada 84

75.7

El Salvador 85

43.3

St. Kitts & Nevis 93

66.4

60.7

Canada 88

75.5

El Salvador 97

89.2

St. Kitts & Nevis 95

68.4

89.9 89.8 87.9 91.9 91.7 90.0 74.1 71.6 76.7 63.7 60.3 63.1 56.9

Canada 93 Canada 97 Canada 00 Canada 04 Costa Rica 74 Costa Rica 78 Costa Rica 82 Costa Rica 86 Costa Rica 90 Costa Rica 94 Costa Rica 98 Costa Rica 02 Dominica 80

69.6 67.0 61.2 61.2 79.9 81.2 78.6 81.8 81.8 81.1 70.0 68.8 80.2

El Salvador 00 El Salvador 03 Grenada 76 Grenada 90 Grenada 95 Grenada 99 Grenada 03 Honduras 85 Honduras 89 Honduras 97 Jamaica 72 Jamaica 76 Jamaica 80

38.1 41.0 65.3 68.4 61.8 56.7 57.4 84.0 76.0 71.8 78.9 85.2 86.1

St. Kitts & Nevis 00 St. Lucia 79 St. Lucia 82 St. Lucia 87 (Apr. 6) St. Lucia 87 (Apr. 30) St. Lucia 92 St. Lucia 97 St. Lucia 01 St.Vincent & Grenadines 79 St.Vincent & Grenadines 84 St.Vincent & Grenadines 89 St.Vincent & Grenadines 94 St.Vincent & Grenadines 98

64.2 68.0 65.8 60.7 64.7 62.8 66.1 53.7 63.9 88.8 72.4 65.6 67.4 (Continued)

Appendix 3.1 : Continued Country and Participation Country and election date (%) election date

Participation Country and (%) election date

Participation Country and (%) election date

Belize 84

75.0

Dominica 85

74.6

Jamaica 89

78.4

Belize 89 Belize 93 Belize 98 Belize 03

72.6 74.6 90.1 79.5

66.6 65.2 60.2 69.5

Jamaica 93 Jamaica 97 Jamaica 02 Mexico 00

Canada 72

77.2

56.2

Canada 74

71.0

Canada 79

75.7

Canada 80

69.3

Dominica 90 Dominica 95 Dominica 00 Dominican Republic 86 Dominican Republic 90 Dominican Republic 98 Dominican Republic 02 El Salvador 72

Participation (%) 69.2

67.7 65.4 56.8 57.2

St.Vincent & Grenadines 01 Trinidad & Tobago 76 Trinidad & Tobago 81 Trinidad & Tobago 86 Trinidad & Tobago 91

Mexico 03

41.7

Trinidad & Tobago 95

63.3

53.0

Panama 99

75.9

Trinidad & Tobago 00

63.1

51.1

St. Kitts & Nevis 84

77.7

56.7

St. Kitts & Nevis 89

66.4

55.8 56.4 65.5 65.8

Democratic elections excluded from the analysis because of missing participation data: El Salvador 74; Fiji 77; Kiribati 91, 94, 98, 02, 03; Liechtenstein 93f; Mali 02; Marshall Islands 91, 95, 99, 03; Micronesia 91, 93, 95, 99, 01; Nauru 76, 77, 80, 83, 92; Palau 04; Papua New Guinea 02; Solomon Islands 84; Tuvalu 85, 89, 93, 98; Vanuatu 02, 04. Democratic elections excluded from the analysis because of other missing data: Andorra 93, 97, 01; Antigua 04; Argentina 03; Bangladesh 91, 96; Benin 95; Burkina Faso 78; Cape Verde 95; Chile 97; Cyprus 96; Czechoslovakia 90, 92; Dominican Republic 78, 82; Ecuador 92, 94, 96; Ghana 04; Guyana 97, 01; India 77, 80, 04; Kiribati 83, 87; Lebanon 72; Liechtenstein 89, 93o, 97, 01; Lithuania 92; Madagascar 93, 98; Mauritius 82, 87, 91; Monaco 93, 98, 03; Mongolia 00, 04; Namibia 04; Nauru 87, 89, 95, 97, 00; Nigeria 83; Palau 96, 00; Panama 94, 04; Papua New Guinea 77, 87, 92; Philippines 87, 95, 98, 01, 04; Samoa 91; San Marino 93, 98, 01: Solomon Islands 80, 89, 93, 97; Sri Lanka 77; St. Kitts and Nevis 04; Tuvalu 77, 81; Vanuatu 87, 91, 95, 98; Venezuela 98. Jamaica 83 excluded because of boycott (participation rate: 2.7%).

Appendix 3.2 : Variables, indicators and sources Variables

Indicators

Sources

Turnout

Percentage of those registered on the electoral list who cast a vote

Blais and Dobrzynska (1998); the turnout data was updated using the following websites: IFES Election Guide, http://www.electionguide.org/; Center on Democratic Performance, Election Results Archive (ERA), http://www.binghamton.edu/cdp/era/countries/; International IDEA Voter Turnout Website, http://www.idea.int/vt/; NationMaster.com: Democratic Statistics by country, http://www.nationmaster.com/cat/demdemocracy

GDP per capita (log)

Real GDP per capita (Chain series) in international dollars, 2000 constant prices

Penn World Table, PWT 6.2

Size of population (log)

Total population

World Bank, World Tables

Switzerland

A dummy variable which equals 1 when the election was held in Switzerland

Compulsory voting

A dummy variable which equals 1 when voting is compulsory with sanctions

Same as for turnout*

New democracy

A dummy variable which equals 1 when the election was held in the country which did not obtained continuously a score of 1 or 2 on political rights from 1972 until 2005

Freedom House, http://www.freedomhouse.org

(Continued)

Appendix 3.2 : Continued Variables

Indicators

Sources

Decisiveness

Scale from 0 to 1, depending on the presence and the timing of subnational elections in federations, upper house direct elections in bicameral countries and presidential direct elections: 1 = no such elections or other elections (subnational, upper house, or presidential) are held simultaneously; 0.5 = one other election (subnational, upper house, or presidential) held nonsimultaneously; 0 = two other elections (subnational, upper house, or presidential) held non-simultaneously A dummy variable which equals 1 when the voting age requirement is 18 years old or lower A dummy variable which equals 1 when the election was held under the proportional representation rule, including mixed corrective systems The difference in vote shares between the leading and the second parties The number of parties running in the election which obtained at least 1% of votes

Same as for turnout*

Low voting age PR Closeness Number of parties (log) ∗ The

data were updated using the same sources as Blais and Dobrzynska (1998). Additionally we had access to the following websites: Political Database of the Americas, http://pdba.georgetown.edu/; African Elections Database, http://africanelections.tripod.com/sn.html; Le site de référence des élections politiques http://www.election-politique.com/

Same as for turnout* Same as for turnout* Louis Massicotte (2004) Same as for turnout* Same as for turnout*

André Blais and Agnieszka Dobrzynska 81

Notes 1. All in all, during the covered period, they were 561 democratic elections in 108 countries. The data on all variables used in our analyses were available for 451 elections in 88 countries. 2. Our sample includes 207 European out of 451 democratic elections. 3. Note that North America is defined according to the UNESCO Statistical Yearbook and includes what is sometimes called Central America. 4. The data comes from Freedom House ratings of political rights the year the election was held. Countries were given a score of 1 (the highest score) on political rights and others a score of 2. We define a new democracy as a country that did not obtain the highest score on political rights throughout the whole period studied here (1972–2005). 5. We excluded all other socio-economic factors used in the previous study that did not keep the statistical significance through all three stages of analysis. 6. A more appropriate measure of economic development was available for this study: GDP per capita as compared to the previously used GNP per capita. This new measure comes from a set of national accounts economic time series, The Penn World Table, version 6.2. The indicator is real GDP per capita measured in international dollars, 2000 constant prices. For more details see Heston et al. (2006). 7. Blais et al. (2003) found that compulsory voting affects turnout, provided that there is a penalty for failing to vote. Accordingly, we improved our measure of compulsory voting distinguishing countries where voting is compulsory and penalties are imposed for failure to comply as compared to all other countries (where there is no obligation to vote or where there are no penalties for failure to comply). 8. Low voting age is defined in our study as 18 years old or lower (the lowest voting age in the sample is 16 years old and is found in Brazilian elections). 9. Proportional representation systems include corrective mixed systems and they are compared to all other systems (plurality, majority and mixed parallel). For a typology of electoral systems, see Blais and Massicotte (2002). 10. See note 4. 11. Turnout was about 11 points higher in countries with compulsory voting, 6 points higher when lower house elections are most decisive, 3 points higher in PR than in plurality, majority or mixed systems, and 4 points lower when voting age is 18 or lower. 12. Closeness corresponds to the gap in vote shares between the leading and second parties. 13. Number of parties is measured in terms of the actual number of parties running in the election.

References Achen, Christopher H (2002) ‘Toward a New Political Methodology: Microfoundations and ART,’ Annual Review of Political Science 5, 423–50. Blais, André (2000) To Vote Or Not To Vote? The Merits and Limits of Rational Choice (Pittsburgh: University of Pittsburgh Press). Blais, André (2006) ‘What Affects Turnout?’ Annual Review of Political Science, 9, 111–25.

82

Activating the Citizen

Blais, André and Ken Carty (1990) ‘Does Proportional Representation Foster Voter Turnout?’ European Journal of Political Research, 18, 167–81. Blais, André and Agnieszka Dobrzynska (1998) ‘Turnout in Electoral Democracies’, European Journal of Political Research, 33, 239–61. Blais, André, Elisabeth Gidengil, Neil Nevitte, and Richard Nadeau (2004) ‘Where Does Turnout Decline Come From?’ European Journal of Political Research, 43, 221–36. Blais, André and Louis Massicotte (2002) ‘Electoral Systems’, in Lawrence LeDuc, Richard G. Niemi, and Pippa Norris (eds), Comparing Democracies 2: New Challenges in the Study of Elections and Voting (London: Sage). Blais, André, Louis Massicotte, and Agnieszka Dobrzynska (2003) Why is Turnout Higher in Some Countries than in Others? (Ottawa: Elections Canada). Blais, André and Daniel Rubenson (2007) ‘Turnout Decline: Generational Value Change or New Cohorts’ Response to Electoral Competition?’ Montreal: Typescript. Clarke, Harold, David Sanders, Marianne Stewart, and Paul Whiteley (2004) Political Choice in Britain (Oxford: Oxford University Press). Dalton, Russell J. (2007) The Good Citizen: How a Younger Generation is Reshaping American Politics (Washington, DC: CQ Press). Downs, Anthony (1957) An Economic Theory of Democracy (New York: Harper). Franklin, Mark (1996) ‘Electoral Participation’, in Lawrence LeDuc, Richard G. Niemi and Pippa Norris (eds), Comparing Democracies: Elections and Voting in Global Perspective (California: Sage). Franklin, Mark (2004) Voter Turnout and the Dynamics of Electoral Competition in Established Democracies since 1945 (Cambridge: Cambridge University Press). Geys, Benny (2006) ‘Explaining Voter Turnout: a Review of Aggregate-Level Research,’ Electoral Studies, 25, 637–63. Heston, Alan, Robert Summers, and Bettina Aten (2006) Penn World Table Version 6.2. Center for International Comparisons of Production, Income and Prices at the University of Pennsylvania. Inglehart, Ronald (2003) ‘How Solid is Mass Support for Democracy – And How Can We Measure It?’ PS, Political Science & Politics, 36, 51–7. Jackman, Robert W. (1987) ‘Political Institutions and Voter Turnout in Industrial Democracies’, American Political Science Review, 81, 405–24. Lyons, William and Robert Alexander (2000) ‘A Tale of Two Electorates: Generational Replacement and the Decline of Voting in Presidential Elections’, Journal of Politics, 62, 1014–34. Miller, Warren E. and J. Merrill Shanks (1996) The New American Voter (Cambridge, MA: Harvard University Press). Powell, G. Bingham (1980) ‘Voting Turnout in Thirty Democracies’, in Richard Rose (ed.), Electoral Participation (Beverly Hills: Sage). Powell, G. Bingham (1982) Comparative Democracies: Participation, Stability and Violence (Cambridge, MA: Harvard University Press). Powell, G. Bingham Jr. (1986) ‘American Voter Turnout in Comparative Perspective’, American Political Science Review, 80, 17–43. Rubenson, Daniel, André Blais, Patrick Fournier, Elisabeth Gidengil, and Neil Nevitte (2007) ‘Does Low Turnout Matter? Evidence from the 2000 Canadian Election’, Electoral Studies, 26, 589–97. Wolfinger, Raymond E. and Steven J. Rosenstone (1980) Who Votes? (New Haven: Yale University Press).

4 Where Turnout Holds Firm: The Scandinavian Exceptions Jørgen Elklit and Lise Togeby

The decline in voter participation in many countries has brought about a conspicuous revival of interest in turnout. The downward trend is generally most pronounced among younger voters, making this development even more worrisome; however, the pattern of decline has not been without exceptions. In all three Scandinavian countries, voter turnout remains relatively high and the overall relative decline has been moderate. Nonetheless, important differences can be observed between the Scandinavian countries. In this chapter, we explore reasons for the Scandinavian exceptionalism but also identify important differences between the Scandinavian countries, which help to shed further light on conditions that facilitate high voting rates. The most recent national elections involved turnout percentages of 77 (Norway 2005), 82 (Sweden 2006), and 87 (Denmark 2007), well above most other European countries, such as Finland (65 per cent in 2007) or the United Kingdom (61 per cent in 2005), and also above Canada (65 per cent in 2006). However, Norway and Sweden are part of the global trend of declining turnout, whereas turnout in Denmark has been remarkably stable at its high level for many years (Elklit, Møller, Svensson, and Togeby, 2005; Elklit, Svensson, and Togeby, 2005; Gray and Caul, 2000; Bernhagen and Marsh, 2007). This pattern makes it attractive from an analytical perspective to focus on the three countries: They share many cultural, social, economic, and institutional traits, but they are also characterized by differences between themselves and in relation to their European neighbours. The recent revival of interest in turnout has generated better comparative databases with an increasing number of countries and variables. Strangely enough, these improvements have not resulted in

83

84

Activating the Citizen

corresponding improvements in the ability to explain variations in turnout between countries. For example, many well-established propositions about the impact of institutional variables have received mixed support when more countries were included in the analyses. In Chapter 3 of this volume, Blais and Dobrzynska conclude their analysis of turnout in an impressive number of electoral democracies by stressing that one may doubt the robustness of some of their earlier findings (such as Blais and Dobrzynska, 1998) and by indicating a need to take a fresh look at some questions that still have not been answered satisfactorily. Despite the enlarged data set used in the analysis, it looks as if Blais is still ‘more impressed by the gaps in our knowledge’ (Blais, 2006, p. 122); it seems as though many results depend on the type of countries under scrutiny (see also Fornos, Power, and Garand, 2004) and also the time period. This chapter focuses on a small number of similar countries, rather than undertaking a large multi-country study. Considered from a distance, the Scandinavian countries are quite similar. Upon closer scrutiny, however, differences become more evident. By deliberately choosing these countries, we were able to hold a number of important variables constant and focus on the important differentiating variables. This approach also allows us to include more ‘soft’ variables, which are difficult to use in studies with many diverse cases. In this manner, we also gain a better understanding of the general factors influencing turnout. In this chapter, we first examine differences between countries, primarily the differences between the three Scandinavian countries, but also differences between Scandinavia and other European countries. Second, we consider the differences between various types of elections in Denmark, that is, national, local, and European Parliament elections, arguing that the factors accounting for the differences between countries probably also account for the differences between the types of elections in a single country. Three groups of factors appear particularly relevant in explaining aggregate level variations in countries as similar as the three Scandinavian nations: 1) political institutions, which explain turnout variations by the regulatory framework governing the elections in combination with the importance of the elections in question; 2) political mobilization, which explains turnout variations by the differential mobilization efforts of political actors; and 3) political culture, which explains turnout variations in terms of differences in the patterns of attitudes, norms, or trust found in the electorate. By concentrating on these three groups of factors, we disregard other aggregate level factors known to influence turnout such as socioeconomic level, country size, and democratic

Jørgen Elklit and Lise Togeby 85

history (Powell, 1982; Blais and Dobrzynska, 1998; see also Blais and Dobrzynska’s useful update in this volume; Blais, 2006; Bernhagen and Marsh, 2007). We do this intentionally, as the three Scandinavian countries are quite similar as regards such other factors.

How to explain turnout variations between countries and different types of elections The three above-mentioned types of explanatory factors form the general organizing framework throughout the chapter. In this section, we discuss the different explanations in more detail. Political institutions Comparative studies have repeatedly demonstrated the importance for turnout rates of a number of aspects of electoral institutions. In a recent review of the literature, however, André Blais (2006) concludes that the results of such studies are much more mixed than we used to think and that the validity of many generalizations is restricted to the old European democracies (see, again, the update by Blais and Dobrzynska in this volume). One of the most important factors, at least in the case of the old democracies, is the use of compulsory voting. The impact of compulsory voting depends on how strictly the legislation is enforced and the cost of non-compliance. Countries with mandatory voting regulations generally have higher turnout (Jackman, 1987; Franklin, 1996; Blais and Dobrzynska, 1998). Another important factor is how voters get on the electoral roll (voter registration list). Turnout is lower when voters themselves must register than in cases of automatic registration, particularly in relation to the total number of potential voters (voting age population) (Lijphart, 2000; Norris, 2002; Norris, 2004). Automatic registration lowers the cost of participating in an election. The strength of the impact of registration on the aggregate turnout level is disputed, but few deny that the lack of an automatic registration system contributes to the low turnout rate in countries such as the United States (cf. Franklin, 1996). The electoral system also influences turnout levels. Again, one ought to avoid overstating the impact of this factor, but the tendency is nevertheless clear (although Latin American countries may offer an exception). Countries using a proportional electoral system generally have higher turnouts than countries using majoritarian or plurality systems (Jackman, 1987; Jackman and Miller, 1995; Franklin, 1996; Blais

86

Activating the Citizen

and Dobrzynska, 1998, and this volume; Fornos et al., 2004). Furthermore, proportional systems have their lists of candidates organized in different ways, which may also impact turnout; open list systems, where voters have a strong impact on candidate selection, probably have a stronger mobilizing effect than closed party list systems, because voters can vote personally (cast a preferential vote) for a person from their party of preference. This obviously increases the possible rewards of voting. Finally, some – but not all – studies have indicated that turnout increases if the various rules surrounding the election facilitate voting, for example, if the election is held on a Sunday or if it is possible to vote in advance of the formal voting day (Norris, 2002; Franklin, 1996, 2004; Blais, 2006). In addition to these electoral institutions, turnout may also be influenced by the importance of the election. The more powerful the representative body, the greater the possible rewards of voting and, therefore, the higher the expected turnout. Consequently, we should expect higher turnouts in parliamentary systems than in presidential systems and higher turnouts in unicameral systems than in bicameral systems (Jackman, 1987; Blais and Dobrzynska, this volume). This is also the idea behind the distinction between first- and second-order elections: First-order elections (the most important elections at the national level) are generally associated with higher turnouts than second-order elections (all other elections) (Franklin, 1996; Norris, 2004). Most theories about institutional impact are based on comparative analyses of countries. We expect, however, that the same theories are applicable in studies of differences between elections in the same country, for example, differences between national and local elections. If this expectation is met, it can be interpreted as additional support for the theories. Political mobilization In 1960, Stein Rokkan and Angus Campbell published their seminal comparative analysis of political participation in Norway and the United States, which demonstrated higher turnout in Norway as well as a weaker relationship between turnout and traditional variables that measure socioeconomic status. They explained this difference by referring to the contrasting strength of the labour movements in the two countries. In Norway, the labour movement was successful in mobilizing blue-collar workers in trade unions and the Labour Party, which resulted in high turnout among blue-collar workers. In the United States, the situation was very different. More generally, we might draw the conclusion that

Jørgen Elklit and Lise Togeby 87

collective political mobilization, especially the mobilization of groups with comparatively limited individual resources, is important for obtaining a high level of voter turnout (see also Verba, Nie, and Kim, 1978). In accordance with these analyses and the general argument, Gray and Caul (2000) hypothesize that the decline in turnout in many advanced industrial democracies in recent decades ought to be explained in light of the declining mobilizing capacity of labour movements. One may add that political institutions may create stronger or weaker incentives and ‘possibility spaces’ for collective mobilization. It is easier for the potential agents of mobilization to imagine the results of their efforts in a country with a proportional electoral system than in a country with a less inclusive electoral system, especially if the proportional electoral system works with open lists. A case in point is the relatively high turnout among immigrants in local government elections in Denmark and the corresponding high representation of immigrants in local councils (Togeby, 2008). While Rokkan and Campbell (1960) and Gray and Caul (2000) focus on the collective mobilization of specific social groups, Rosenstone and Hansen (1993) additionally emphasize the general mobilization transpiring in connection with a single election. They define mobilization more broadly as ‘the process by which candidates, parties, activists and groups induce people to participate’ (Rosenstone and Hansen, 1993, p. 25). Elections and election campaigns may be more or less competitive and more or less politicized. As well, they may attract more or less attention. It almost goes without saying that more competitive elections with stronger media coverage ought to bring about a higher turnout than less competitive elections with less intensive media coverage (see also Franklin, 2004). Differences in political mobilization, both collective mobilization and mobilization in connection with individual elections, may explain differences between countries, differences between the same type of election held at different times in the same country, and differences between different types of elections in the same country. Political culture In general, the social and material costs associated with casting one’s vote are relatively limited, at least in systems with a genuinely secret vote, just as the individual rewards are limited. That explains why theories based on rational choice and theories based on resource considerations do not perform particularly well in explaining variations in turnout at the individual level. Instead, we have witnessed a growing interest in

88

Activating the Citizen

the impact of psychological factors at the individual level and cultural factors at the aggregate level. Most interest has been attached to the sense of civic duty (e.g., Blais, 2000), and a number of studies have suggested that a decline in the sense of civic duty can explain the decline in turnout (Rattinger and Krämer, 1995; Bjørklund, 1999; Teorell and Westholm, 1999; Blais, 2000). Differences in turnout at the national level may be explained by differences in the sense of civic duty, just as turnout differences in different kinds of elections may be accounted for by differences in the sense of civic duty experienced in connection with these elections. Moreover, turnout could also be influenced by the level of trust in the representative body in question. If voters have trust in the democratic system or in the parliament, they are probably also more likely to think that the representatives are responsive and that their vote matters, which may increase turnout. At the individual level, the evidence is not particularly convincing (Norris, 2002), but Bernhagen and March (2007) show that turnout is influenced at the aggregate level by the level of satisfaction with democracy. We expect that the differences between countries as well as the differences between the types of elections are influenced by the trust of the electorate in the elected body. One might argue that cultural differences, as such, do not explain much and that political culture might merely pose a mediating factor for more important explanatory factors. Nevertheless, it remains interesting to see if differences in voter turnout can be connected to significant differences in the overall political culture surrounding an election.

Explaining high turnout levels in national elections in Scandinavia In Norway and Sweden, there is a growing concern about declining turnout in national elections. In contrast to Denmark, where the turnout rate has been quite stable over the last 60 years, the turnout rates in the other two countries have declined close to 10 percentage points over recent elections (see Figure 4.1, which for the sake of comparison also includes Finland). Compared to most other Western democracies, however, turnout rates remain high in Norway and Sweden and the decline has actually been less pronounced than in many other countries. In any case, Denmark, Norway, and Sweden are today among the countries with the highest turnout rates.

Jørgen Elklit and Lise Togeby 89

100 90 80 70 60 50 40 30 20 10 0 1970

1975 Finland

1980

1985

1990

Denmark

1995

2000

Norway

2005

2010

Sweden

Figure 4.1: Turnout in national elections in Scandinavia and Finland, 1975–2007 (% of total electorate) Note: Numbers from Finland include voters living abroad. Sources: Official election statistics from the election authorities in Finland, Denmark, Norway, and Sweden.

A comparison between the Scandinavian countries and a couple of larger industrialized countries in western Europe, Germany and the United Kingdom, demonstrates a striking similarity between the average drop in turnout in national elections during a period from the mid-1970s to the mid-1980s and during a period from the mid-1990s to the latest national elections. The drop in Germany and the United Kingdom was close to 10 percentage points, as it was in Sweden (and Finland). Incidentally, the same over-time drop has been registered in Canada (LeDuc and Pammett, 2006). In Norway, the drop was almost 6 percentage points, whereas in Denmark it was almost absent: a modest 1 percentage point. A similar average drop of 10 percentage points in a number of countries over three decades, irrespective of the turnout level in the first period, is striking (see Table 2.1 in this volume), but so are the less pronounced drop in Norway and the stable level in Denmark over the same extended period. As in most other countries, the decline in voter turnout in Norway and Sweden has been most pronounced among the younger generations of voters (Bennulf and Hedberg, 1999; Bjørklund, 2002). Against this background, we find it remarkable that age group differences have not changed in Denmark over the last 40 or so years and that the turnout in the youngest generation remains surprisingly high (see Figure 4.2),

90

Activating the Citizen

100 90 80 70 60 50 40 30 20 10 0

1960 18–22 23–24 25–29 30–39 40–49 50–59 60–69 70–74

2001 75⫹

Figure 4.2: Turnout among Danish citizens in the municipality of Aarhus for national elections, 1960 and 2001 (% of total electorate) Sources: For 1960, Jeppesen and Meyer (1964); for 2001, Elklit, Møller, Svensson and Togeby (2005).

especially as the voting age in the same period has been lowered from 23 to 18 years (cf. Franklin, 2004; Blais and Dobrzynska, this volume). In Denmark, a stable relationship between age and turnout thus accompanies the stability in the overall level of turnout. The main reason for this appears to be an undisturbed early and rapid socialization of new generations to vote in national elections from Election 1, that is, the first election subsequent to having reached voting age (cf. Franklin, 2004). The data used for Figure 4.2 are for the total counts of both years based on the markings in the voter registers used by the election authorities in Aarhus, Denmark’s second largest city, covering more than 210,000 voters in 2001 (Elklit, Møller, et al., 2005). The 1960 data are from the first academic study of voter turnout in Denmark (Jeppesen and Meyer, 1964). Rokkan and Campbell (1960) demonstrated that turnout in Norway had a weaker relationship with socioeconomic status factors than was the case in the United States. The Scandinavian countries, with their strong labour movements, were characterized by greater participation equality than in the United States. That is still the case, even if studies in Sweden (Bennulf and Hedberg, 1999) have demonstrated that increasing turnout differences between social groups accompany the declining Swedish turnout. In order to demonstrate the Scandinavian feature of only modest turnout differences between social groups, we have examined the relationship between education and turnout in Denmark in the national

Jørgen Elklit and Lise Togeby 91

election in 2001, which turns out to be not as strong as expected. As many as 79 per cent of those with only a primary school education turned out to vote, while the turnout level for those with at least a master’s degree was 94 per cent. Other educational levels are found in between these two. The similar turnout levels for groups with different levels of education support our claim that turnout differences between social groups in Scandinavia remain smaller than in many other countries, and perhaps particularly so in Denmark. The labour movements are also now undoubtedly weaker in the Scandinavian countries as compared to the situation examined by Rokkan and Campbell (1960). In Denmark, however, the results indicate that either the labour movement has not lost its mobilizing capacity as much as some claim or other mobilizing agents have taken over. In summary, then, in Denmark we find a high and stable turnout in national elections, a surprisingly high turnout among the youngest generation of voters, and an exceptionally weak relationship between social resources and turnout. The same is true for the other Scandinavian countries, but to a lesser degree. We will now attempt to explain these results by using the framework suggested in the introduction. Political institutions The relevant political institutions in the three Scandinavian countries are much alike and share a number of features distinguishing them from many other old democracies. All three countries are old monarchies with unicameral, powerful parliaments. They all have automatic voter registration and proportional electoral systems (with a 2 to 4 per cent electoral threshold). Preferential voting for individual candidates is an option in Sweden and Denmark at all levels, but in Norway only in local and regional elections. Thus, all three countries have most of the institutional characteristics generally associated with high voter turnout. Furthermore, voters in the Scandinavian countries also have easy access to advance voting by mail. What the countries do not have, however, is compulsory voting. There are also minor differences between the proportional representation electoral systems in the three countries. Foremost among these differences are the rules and regulations governing the selection of candidates from the lists presented by the parties in the various multimember constituencies and how the casting of a personal (preferential) vote impacts this selection. Elklit (2005) describes this variant in the Danish context, and what is evident is that personal (preferential) votes have a stronger and more transparent impact on the selection of candidates in Denmark than in Sweden. It is not difficult to cast a personal

92

Activating the Citizen

vote in either country, but in Sweden, the requirements for personal votes to actually influence the selection of candidates from the lists are rather demanding, as legislation has implemented a couple of thresholds, which might explain the different usage of this facility. If, for instance, a Swedish parliamentary candidate does not win at least 8 per cent of the personal votes in the multi-member constituency in question, then the party’s ordering of the list applies (Valmyndigheten, 2007). For lower level elections, the requirement is 5 per cent of the party’s vote (and at least a certain number of votes). Because of these non-trivial requirements, we would expect personal voting levels to be higher in Denmark than in Sweden. In Denmark, slightly more than 50 per cent of all voters cast personal votes in 2005 as well as in 2007. The average for the last ten parliamentary elections is 48 per cent. Sweden had had this kind of preferential voting in only the two most recent national elections; the level was 22 per cent in 2006, down from 30 per cent in 2002. So Danish voters use this option much more than Swedish voters, which supports our argument that it might offer a partial explanation for the higher and more stable turnout in national elections in Denmark than in Sweden. One can even argue that the level declined in Sweden from 2002 to 2006 because, in the 2002 election, the Swedish version of preferential voting proved less effective in securing the election of specific candidates than some voters expected. It is worth noticing that 50 per cent of all Danish voters express a specific preference for a particular candidate on a list containing a number of candidate names from each party. The level of personal voting is even higher in local government elections, where the corresponding figure for the 2005 local government election was 80 per cent, an increase from the usual average level of 75 per cent. The explanation for this difference is that many voters in local elections cast a personal (preferential) vote for the top candidate on the list, that is, the top person in the local party hierarchy, who is usually also the mayoral candidate of the party in question. Only a minority of the voters have this opportunity in national elections, namely those voters in the specific multi-member constituencies in which the party leader stands for election, while voters in other multi-member constituencies only have lower-level party dignitaries to vote for. Political mobilization As has been the case in many other countries, political parties in Scandinavia have lost a substantial number of their rank-and-file members in recent years. This decline in party membership started earlier in

Jørgen Elklit and Lise Togeby 93 Table 4.1:

Political interest (national level) by country (% of respondents)

Norway Denmark Switzerland Sweden The Netherlands Germany Portugal Spain

‘Much’ or ‘some’ political interest

N

69 68 67 62 60 56 32 28

2,270 1,631 1,599 1,252 1,626 2,958 969 4,035

Source: The CID project ((‘Citizens, Involvement, Democracy’), an international social science project with twelve countries, data collection 2000–2001; van Deth, Montero, and Westholm, 2006).

Denmark than in the other two countries, but the decline accelerated in Norway and Sweden in the 1990s. Today, only around 5 per cent of the voters in the three Scandinavian countries are card-carrying members of a political party (cf. Siaroff, this volume). Simultaneously, there is greater social disparity in party member recruitment than previously was the case. The decline in membership has undoubtedly weakened the ability of the political parties to mobilize voters, especially voters with limited social resources. Consequently, electoral mobilization now depends much more on mass media activity than it did previously. As documented in Table 4.1, political interest is nevertheless high in the Scandinavian countries – and higher than in most other European countries. A high level of political interest ought to be indicative of a relatively high level of political mobilization, so we expect a clear positive relationship between political interest and turnout, both at the individual and aggregate levels. Another and possibly more satisfying individual level indicator of political mobilization and the level of competition between parties is the ability of voters to distinguish between the policy positions of the various political parties. Table 4.2 presents data from the 2001 election study in Denmark, in which voters were asked to place the political parties on a number of issue scales. The table demonstrates that voters are perfectly capable of distinguishing between the seven political parties in the Danish parliament even on a number of different dimensions. It is also noteworthy that the voters use virtually the entire scale when placing the parties, which in our interpretation means that Danish voters perceive the policies of the respective political parties as differing

94

Activating the Citizen

Table 4.2: Party policy images, Denmark 2001. Voters’ placement of political parties on three scales Public spending

Refugee policies

EU policies

Red–Green Alliance (socialist) Socialist People’s Party Social Democrats Social Liberals Conservatives Liberals (Venstre) Danish People’s Party

+54

+52

−73

+44 +23 +11 −44 −50 −51

+39 +4 +17 −52 −54 −95

−24 +63 +54 +62 +69 −66

N

2.026

2.026

2.026

Source: Goul Andersen (2003b). Note: Mean values, recoded to run from −100 to +100; a + in front of a figure means that the voters on average see the party as being in favour of the policy, a − means that the party is against the policy.

strongly and distinguishably from one another. Therefore, we also expect the voters to perceive the election as important and with important consequences, which should increase overall turnout. This expectation is consistent with the results of Crepaz’ (1990) analysis of the impact of party polarization and post-materialism on turnout. The differentiation between the parties is strongest in the case of refugee policy. We also know that the refugee issue has had a prominent position on the voters’ political agenda in Denmark for a number of years (Goul Andersen, 2003b; van der Brugge and Voss, 2007), much longer than has been the case in Norway (Narud and Valen, 2007) or Sweden (Holmberg, 2000), and that attitudes towards immigrants and refugees have a strong impact on voting behaviour in Denmark (Goul Andersen, 2003a). Furthermore, the Danish political party with the strongest anti-immigrant policies, the Danish People’s Party, enjoys its strongest support from blue-collar workers (Andersen and Goul Andersen, 2003). In this context, one might argue that an important element in the high and stable turnout in Denmark is the mobilizing efforts of the Danish People’s Party, which counteracts the decline in the collective mobilization of the traditional labour movement (cf. also Goul Andersen, 2007). The aforementioned weak relationship between education and turnout supports this interpretation (more than three quarters of those with limited education vote in national elections). It is also worth noting that the Danish People’s Party enjoys the same level of support among young blue-collar workers as among middle-aged and older blue-collar

Jørgen Elklit and Lise Togeby 95 Table 4.3: Is voting a civic duty? Voter attitudes by country and age (mean values, on a scale from 0 to 10)

Denmark Sweden Norway West Germany The Netherlands Spain Switzerland Portugal East Germany

All citizens

18–29 years old

N

8.8 8.6 8.2 7.4 7.3 7.2 7.2 7.1 6.9

8.1 7.8 7.5 6.6 6.4 6.4 6.1 6.7 6.2

1,631/327 1,252/283 2,270/466 1,970/385 1,626/299 4,035/1,194 1,599/289 969/315 988/189

Source: The CID project (see Table 4.1).

workers (Andersen and Goul Andersen, 2003). It is therefore quite likely that this party has contributed more than most other parties to getting young blue-collar workers to vote in recent elections, something we do not see to the same degree in other countries. It is probably these differences in mobilization and conflict level that best explain the stable turnout in Denmark, whereas turnout, especially among young people, has declined in Norway and Sweden. Political culture It is generally acknowledged that a relatively high turnout is closely related to a political culture characterized by a strong norm according to which voting is perceived to be a civic duty. Countries obviously differ when it comes to the strength of this norm. Table 4.3 illustrates that the Scandinavian countries, and Denmark in particular, are distinguished by a strong adherence to the norm, even among the younger generations. As such, the level of support for the norm of voting as a civic duty likely also contributes substantially to the high level of turnout in the Scandinavian countries. As Rose (2004, 2005) has demonstrated, however, it is necessary to examine the combined effect of this norm and other factors, primarily the importance of the election. In our view, this is why Danish voters, in spite of their high level of adherence to the norm, do not turn out in elections to the European Parliament in greater numbers than do the voters in other EU member states. The Scandinavian countries are also among the nations whose populations have the greatest trust in parliament. This especially applies to Demark, where 64 per cent of the electorate state that their level of

96

Activating the Citizen

trust is high or relatively high. This compares to 48 to 49 per cent in Norway and Sweden and 38 per cent in the Netherlands, alongside only 28 to 29 per cent in Germany and the United Kingdom. It is highly plausible that the high level of trust in parliament in the Scandinavian countries also contributes to the high turnout. Grönlund and Setälä (2004) demonstrate how trust in the six institutions included in the European Social Survey contributes to explaining voter turnout levels in most European countries. One particularly interesting element in their analysis is that trust in politicians also acts as a significant variable when it comes to explaining turnout. Denmark scores considerably higher on this particular variable than the two other Scandinavian countries, or in comparison to Finland, Luxembourg, the Netherlands, and the Czech Republic, which again score somewhat higher than the other countries included in the study (Grönlund and Setälä, 2004). Political culture in Denmark is characterized by a strong adherence to the norm that voting is a civic duty (at least in national and local elections) as well as high trust in parliament and in politicians. The same is true for Norway and Sweden, though to a lesser degree. In our assessment, a supporting political culture is an important element in explaining the high turnout in the Scandinavian countries, particularly in Denmark.

Different levels of participation in different kinds of elections within a country If we consider elections at other politico-administrative levels than the national, it immediately becomes clear that Danish voters are willing to downgrade their participation in non-first-order elections. Table 4.4 and Figure 4.3 demonstrate this point. There appears to be a hierarchy of elections, with national elections at the top, followed by regional local elections and elections to the European Parliament, and other kinds of elections at the bottom. This hierarchy appears to reflect the importance that voters associate with the respective representative bodies. We suggest that these differences can be accounted for with reference to the same factors that explain the differences between countries. Political institutions Voters generally attach greater importance to national politics than to local politics, so turnout differences should also reflect the perceived benefits of voting at the various politico-administrative levels. Turnout in the elections for the European Parliament is low, even though the entire country is a single constituency for these elections,

Jørgen Elklit and Lise Togeby 97 Table 4.4:

Turnout in recent elections in Denmark (% of total electorate)

Type of election

Year

%

National government Local government Regional government European Parliament School boards

2007 2005 2005 2004 2006

87 70 69 48 45

Parochial church

2004

17

2005–2007

41

2006

Est. 17

Local government consultative councils for senior citizens (voters 60+ years)∗ Local government integration councils∗∗

Comments

Only school districts with contested elections; 83% of all elections were not contested, as only one list was presented. Only parishes with contested elections; 91% of all elections were not contested, as only one list was presented.

∗ No

national statistics available. The figure here (41%) is based on information from 60% of all municipalities, that is, those providing relevant information and that did not conduct senior citizens’ councils elections simultaneously with local government elections in November 2005. Turnout for elections conducted simultaneously with local government elections was 64%, that is, at the level expected for this age group of voters based on their local government election participation rate. ∗∗ No national statistics available. Figures given are from a non-random sample of municipalities. Sources: Official election statistics from Ministry of the Interior/Ministry of Welfare in combination with the authors’ own compilations and calculations.

where all voters have the option of casting a preferential vote for any of the candidates on the ballot, some of which are nationally well-known personalities, including former prime ministers. The level of preferential voting is much higher than in parliamentary elections, but turnout is nevertheless low, probably because many voters traditionally perceive the European Parliament as a body of limited consequence compared to the European Commission and the European Council as well as to the national parliaments. The low turnout in what one might even refer to as ‘third-order’ elections, for example, elections for local school boards, parish councils, councils for seniors, and integration councils, probably reflects the same understanding of these bodies as having limited competencies.

98

Activating the Citizen

100 90 80 70 60 50 40 30 20 10 0 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007

National election Regional government elections

Local government elections European parliament

Figure 4.3: Turnout in Danish elections at four levels, 1981–2007 (% of total electorate) Sources: Official election statistics from Statistics Denmark and Ministry of the Interior/ Ministry of Welfare.

The low turnout in the municipal elections for the consultative councils for seniors may also reflect the fact that elections are conducted using an electoral system called the single non-transferable vote, which has no room for lists of candidates and requires voting for individuals. Consequently, political parties or other political groupings are prevented from presenting candidates in these elections, which cannot but contribute to the low turnout. Political parties are generally also absent from the ballot in the various other third-order elections. Political mobilization Turnout in third-order elections is normally low. In some cases, however, turnout suddenly increases considerably; in most such cases, these increases reflect conflicts between personalities or groups of voters (for example, parents or churchgoers with different religious inclinations). Conflict appears to stimulate interest in the election and, consequently, increases turnout.

Jørgen Elklit and Lise Togeby 99

The same can be said about the presence of political parties. When political parties are not available as signposts that voters can use for navigation when deciding whether to vote, as in many of the low-level elections as noted above, political mobilization becomes more difficult. This is hardly surprising, as the structuring of the vote is a classical function of political parties. We should thus expect a lower turnout when political parties are either deemed absent or do not want to be on the ballot, compared to those cases where elections are contested by all or most of the major political parties. Political parties are more often available at the national level than at other politico-administrative levels. At the same time, political interest among voters is also more focused on the national level than on other levels (for example, the local or EU levels), and many more people find national politics easier to follow than European politics (Goul Andersen, 2004). Another factor differentiating kinds of elections is media coverage. Media coverage is generally most extensive in first-order elections and least extensive in second- and especially third-order elections. Media coverage stimulates interest in elections, and mobilization is generally higher in first-order elections than in second- or third-order elections. Therefore, we also expect social resources and social integration to play a larger role in second- and third-order elections than in first-order elections. Table 4.5 demonstrates that this is the case: Age, marital status, and length of residence in the municipality all have a greater impact on turnout in local elections than in national elections. Getting the vote out among people with fewer resources requires greater mobilization. The differences in turnout between different kinds of elections are probably primarily a result of the differences in mobilization and conflict. Political conflict and political mobilization enhance political participation. Political culture In 2005, at the time of local government elections, Danish voters were asked to what degree they felt committed to vote in elections at different levels. In relation to the parliamentary elections, as many as 80 per cent answered ‘very committed’, while 15 per cent responded ‘fairly committed’. The figures were also high in relation to the local government elections (76 and 17 per cent respectively), though slightly lower than in relation to parliamentary elections. Thus, even though the figures are not very different, the norm is weaker at the local government level than at the national level. It is rather interesting that the norm of voting in the elections for the newly established regional councils (which have replaced the county

100 Activating the Citizen Table 4.5: Socio-demographic turnout determinants among Danish citizens at national and local levels, municipality of Aarhus, Denmark (% of total electorate) Local election 1997

National election 2001

Difference

Age 18–24 years 25–29 years 30–39 years 40–49 years 50–59 years 60–69 years 70–79 years 80+ years

63 65 68 78 81 83 77 53

80 84 84 88 90 90 86 45

17 19 16 10 9 7 9 8

Marital status Single Married

65 82

85 92

20 10

Length of residence in the municipality Less than a year 1–4 years 5–9 years 10 years+

61 68 70 74

82 87 88 88

21 19 18 14 15

All

72

87

N

211,739

226,720

Source: 1997 and 2001 Voter Registry Studies. See Elklit, Møller, Svensson, and Togeby (2005) and references therein.

councils, but with relatively unclear responsibilities and without taxing authority) is weaker than in relation to the other two electoral levels, with figures of only 62 and 22 per cent respectively. However, one can also claim that the norm actually remains quite strong, at least compared to what one would expect based on comparable levels elsewhere. These figures contribute significantly towards validating the claim that Danish voters have integrated the norm of voting as a civic duty more than is the case with voters elsewhere (cf. Rose, 2004) and that this is a key element in Danish political culture. This observation also strengthens the paradox mentioned earlier, that most Danish voters abstain from voting in elections for the European Parliament, even though they adhere strongly to the voting norm. This may have something to do with the lack of trust in EU institutions. We expect that if voters trust a particular institution and feel that the institution is

Jørgen Elklit and Lise Togeby 101

legitimate, turnout for this particular body will be greater than if voters have less trust. A survey has revealed that the percentage of voters with a high level of trust in the national parliament is two-thirds higher than the percentage having a similar level of trust in the European Parliament. It is noteworthy that the relationship between these two levels of high trust is almost the same as the relationship between turnout levels for the two organs (cf. Figure 4.3).

Conclusion Turnout, particularly at the national level, but also at the local government level, holds quite firm in Denmark, contrary to other countries. In explaining this anomaly, we have focused on the intricate interplay of three bundles of causal variables (political institutions, political mobilization, and political culture), which together go a long way towards explaining the difference between Denmark and other countries. Table 4.6 summarizes the findings and the main argument. While the causal variables are powerful in explaining national differences, they are not as persuasive in accounting for Denmark’s high and stable turnout rates in both national and local government elections; as so often in the social sciences, the explanation is a complicated combination of different factors, all of which contribute to some degree, but not necessarily in the same combination on each occasion in which the phenomenon is being analysed! The impact of political institutions and political culture is evident; however, the strongest factor accounting for the differences in turnout when comparing the Scandinavian countries is political mobilization. A high level of competitiveness in elections, especially if a close result is expected, stimulates media interest and makes voters increasingly aware of the value of casting a vote. This also generates greater political mobilization, especially among groups of voters who would generally be expected to have lower levels of participation, for example those with limited education, the socially marginalized, and young people. Recent elections in Denmark have been quite competitive (particularly in 1998, 2001, and 2007), as were the 2006 election in Sweden and the 2005 election in Norway. All of these elections were characterized by an increase in voter turnout. What sets Denmark apart from the other two Scandinavian countries is its combination of political institutions, which allow and invite political mobilization, on the one hand, and mobilizing actions, on the other.

102

Table 4.6: Comparison of the relative impact of the main factors on turnout differences in the three Scandinavian countries

Political institutions

Political mobilization

Political culture

Denmark

Norway

Sweden

+ Proportional representation, automatic voter registration, advance voting, etc.

+ Proportional representation, automatic voter registration, advance voting, etc.

+ Proportional representation, automatic voter registration, advance voting, etc.

+ quite effective personal voting

− no personal voting at national level

+/− weak personal voting

+ A major political party has been able to mobilize some of the young and socially marginalized to vote

+/− Different levels of competitiveness in recent decades because of the (earlier) relatively strong Labour Party

+/− Different levels of competitiveness in recent decades because of the strong Social Democratic Party. 2006 was quite competitive

+ Recent parliamentary elections have been quite competitive − Traditional collective mobilization agents less strong than previously

− Traditional collective mobilization agents less strong than previously

− Traditional collective mobilization agents less strong than previously

+ Strong voting norm

− Less strong voting norm

− Less strong voting norm

Note: + indicates factors increasing turnout; − indicates factors decreasing turnout.

Jørgen Elklit and Lise Togeby 103

References Andersen, J. and J. Goul Andersen (2003) ‘Klassernes forsvinden’ [The Disappearance of the Classes] in J. Goul Andersen and O. Borre (eds) Politisk forandring: Værdipolitik og nye skillelinjer ved folketingsvalget 2001 [Political Change: Value Politics and New Lines of Division in the Danish National Elections 2001] (Aarhus, Denmark: Systime Academic), pp. 207–22. Bennulf, M. and P. Hedberg (1999) ‘Utanför demokratin: Om det minskade valdeltagandets sociale och politiska rötter’ [Outside of Democracy: On the Social and Political Roots of Declining Electoral Participation] in E. Amnå (ed.) Valdeltagande i förandring [Changes in Electoral Participation] Demokratiutredningens forskarvolym, XII (Stockholm: SOU), pp. 75–135. Bernhagen, P. and M. Marsh (2007) ‘Voting and Protesting: Explaining Citizen Participation in Old and New European Democracies’, Democratization, 14(1), 44–72. Bjørklund, T. (1999) Et lokalvalg i perspektiv: Valget i 1995 i lys av sosiale og politiske endringer [A Local Election in Perspective: The 1995 Election in light of Social and Political Changes] (Oslo: Tano Aschehoug). Bjørklund, T. (2002) ‘Den synkende deltakelsen ved lokalvalg – noen årsaker og konsekvenser’ [The Declining Participation in Local Elections – Some Causes and Consequences] in B. Aardal (ed.) Valgdeltakelse og lokaldemokrati [Electoral Participation and Local Democracy] (Oslo: Kommuneforlaget), pp. 69–88. Blais, A. (2000) To Vote or Not to Vote: The Merits and Limits of Rational Choice Theory (Pittsburgh: The University of Pittsburgh Press). Blais, A. (2006) ‘What Affects Voter Turnout?’, Annual Review of Political Science, 9, 111–25. Blais, A. and A. Dobrzynska (1998) ‘Turnout in Electoral Democracies’, European Journal of Political Research, 33, 239–61. Crepaz, M. M. L. (1990) ‘The Impact of Party Polarization and Postmaterialism on Voter Turnout: A Comparative Study of 16 Industrial Democracies’, European Journal of Political Research, 18, 183–205. Elklit, J. (2005) ‘Denmark: Simplicity Embedded in Complexity (or is it the Other Way Round)?’ in M. Gallagher and P. Mitchell (eds) The Politics of Electoral Systems (Oxford: Oxford University Press), pp. 453–71. Elklit, J., B. Møller, P. Svensson, and L. Togeby (2005) Gensyn med Sofavælgerne: Valgdeltagelse i Danmark [The Non-Voters Revisited: Electoral Participation in Denmark] (Aarhus, Denmark: Aarhus University Press). Elklit, J., P. Svensson, and L. Togeby (2005) ‘Why Is Voter Turnout Not Declining in Denmark?’ Paper presented at the APSA Annual Meeting in Washington, DC. Fornos, C., T. J. Power, and J. C. Garand (2004) ‘Explaining Voter Turnout in Latin America, 1980 to 2000’, Comparative Political Studies, 37(8), 909–40. Franklin, M. N. (1996) ‘Electoral Participation’ in L. LeDuc, R. G. Niemi, and P. Norris (eds) Comparing Democracies: Elections and Voting in Global Perspective (Thousand Oaks, CA: Sage), pp. 216–35. Franklin, M. N. (2004) Voter Turnout and the Dynamics of Electoral Competition in Established Democracies Since 1945 (Cambridge: Cambridge University Press). Goul Andersen, J. (2003a) ‘The General Election in Denmark, November 2001’, Electoral Studies, 22(1), 186–93.

104 Activating the Citizen Goul Andersen, J. (2003b) ‘Partiernes image: De borgerlige bedst til at sikre velfælden’ [Party Images: The Right-of-centre Parties Best at Ensuring Welfare] in J. Goul Andersen and O. Borre (eds) Politisk forandring: Værdipolitik og nye skillelinjer ved folketingsvalget 2001 [Political Change: Value Politics and New Lines of Division in the 2001 National Elections] (Aarhus, Denmark: Systime Academic), pp. 151–70. Goul Andersen, J. (2004) Et ganske levende demokrati [A Perfectly Lively Democracy] (Aarhus, Denmark: Aarhus University Press). Goul Andersen, J. (2007) ‘Samfundskonflikter, partier i bevægelse og vælgere med omtanke: Rids af en generel teoretisk ramme’ [Conflicts in Society, Parties in Movement and Prudent Voters: The Outline of a General Theoretical Framework] in J. Goul Andersen et al. (eds) Det nye politiske landskab: Folketingsvalget 2005 i perspektiv [The New Political Landscape: The 2005 National Elections in Perspective] (Aarhus, Denmark: Academíca), pp. 11–55. Gray, M. and M. Caul (2000) ‘Declining Voter Turnout in Advanced Industrial Democracies, 1950 to 1997: The Effects of Declining Group Mobilization’, Comparative Political Studies, 33(9), 1091–122. Grönlund, K. and M. Setälä (2004) ‘Low Electoral Turnout: An Indication of a Legitimacy Deficit?’ Paper presented at the ECPR workshop in Uppsala, Sweden. Holmberg, S. (2000), Välja parti [Choose a Party] (Stockholm: Nordstedt Juridik). Jackman, R. (1987) ‘Political Institutions and Voter Turnout in the Industrial Democracies’, American Political Science Review, 81, 405–23. Jackman, R. and R. Miller (1995) ‘Voter Turnout in the Industrial Democracies During the 1980s’, Comparative Political Studies, 27, 467–92. Jeppesen, J. and P. Meyer (1964) Sofavælgerne: Valgdeltagelsen ved danske folketingsvalg [The Non-Voters: Electoral Participation in the Danish National Elections] (Aarhus, Denmark: Department of Political Science, University of Aarhus). LeDuc, L. and J. H. Pammett (2006) ‘Voter Turnout in 2006: More Than Just the Weather’ in J. H. Pammett and C. Dornan (eds) The Canadian Federal Election of 2006 (Toronto: Dundurn Press), pp. 304–26. Lijphart, A. (2000) ‘Turnout’ in R. Rose (ed.) International Encyclopedia of Elections (Washington DC: CQ Press), pp. 314–22. Narud, H. M. and H. Valen (2007) Demokrati og ansvar: Politisk representation i et flerpartisystem [Democracy and Responsibility: Political Representation in a Multi-party System] (Oslo: N.W. Damm & Søn). Norris, P. (2002) Democratic Phoenix: Reinventing Political Activism (Cambridge: Cambridge University Press). Norris, P. (2004) Electoral Engineering: Voting Rules and Political Behavior (Cambridge: Cambridge University Press). Powell, G. B. (1982) Contemporary Democracies: Participation, Stability and Violence (Cambridge, MA: Harvard University Press). Rattinger, H. and J. Krämer (1995) ‘Wahlnorm und Wahlbeteiligung in der Bundesrepublik Deutschland: Eine Kausalanalyse’ [Electoral Norms and Turnout in the German Federal Republic: A Causal Analysis], Politische Vierteljahresschrift, 36(2), 267–85. Rokkan, S. and A. Campbell (1960) ‘Citizen Participation in Political Life: A Comparison of Data from Norway and the United States of America’, International Social Science Journal, 12(1), 69–99.

Jørgen Elklit and Lise Togeby 105 Rose, L. E. (2004) ‘Normer og roller – borgerpliktens endelikt?’ [Norms and Roles – The End of Civic Duty?] in B. Aardal (ed.) Valgdeltakelse og lokaldemokrati [Electoral Participation and Local Democracy] (Oslo: Kommuneforlaget), pp. 139–75. Rose, L. E. (2005) ‘Normer og stemmegivning: Både indre plikt og ytre press?’ [Norms and Voting: Internal Duty as well as External Pressure?] in J. Saglie and R. Bjørklund (eds) Lokalvalg og lokalt folkestyre [Local Government Elections and Local Democracy] (Oslo: Gyldendal Akademisk), pp. 228–51. Rosenstone, S. J. and J. M. Hansen (1993) Mobilization, Participation, and Democracy in America (New York: Macmillan). Teorell, J. and A. Westholm (1999) ‘Att bestämma sig för att vara med och bestämma: Om varför vi röstar – allt mindre’ [To Decide to Participate in Deciding: On Why We Vote – Less and Less] in E. Amnå (ed.) Valdeltagande i förandring [Changes in Electoral Participation], Demokratiutredningens forskarvolym, XII (Stockholm: SOU), pp. 137–204. Togeby, L. (2007) ‘Immigrant Exceptionalism: Immigrant Turnout in Local Elections in Denmark’. Paper presented at the ECPR workshops in Helsinki. Togeby, L. (2008) ‘The Political Representation of Ethnic Minorities. Denmark as a Deviant Case’, Party Politics, 14(3), 343–61. Valmyndigheten [Swedish Election Authority] (2007) http://www.val.se/ valsystemet/rosta/personrosta/index.html van der Brugge, J. and H. Voss (2007) ‘Årsagen til de socialistiske partiers tilbagegang i perioden 1990–2005’ [The Reason for the Decline of the Socialistic Parties in the Period 1990–2005] in J. Goul Andersen et al. (eds) Det nye politiske landskab: Folketingsvalget 2005 i perspektiv [The New Political Landscape: The 2005 National Elections in Perspective] (Aarhus, Denmark: Academíca), pp. 127–51. van Deth, J. W., J. R. Montero, and A. Westholm (2006) Citizenship and Involvement in European Democracies: A Comparative Analysis (London: Routledge). Verba, S., N. H. Nie, and J. Kim (1978) Participation and Political Equality: A Seven Nation Comparison (Cambridge & New York: Cambridge University Press).

5 New Members, Old Issues: The Problem of Voter Turnout in European Parliament Elections Joan DeBardeleben and Lawrence LeDuc

Voting turnout in European Union (EU) elections has consistently declined since the first vote in l979 (see Figure 5.1); the slide has, in many instances, been even more marked than in member states’ parliamentary votes. The first direct election of the European Parliament (EP) in 1979 registered 65.9 per cent as the mean turnout for the nine countries participating; by 2004 the mean for the 25 member states was only 47.8 per cent.1 With few exceptions, turnout in elections for the EP has also been consistently lower than national turnout levels. In some cases the difference is dramatic. This chapter explores underlying causes of both the low levels of turnout and of the dramatic decline, taking guidance from the more

75 70 65 60 55 50 45 40 Figure 5.1:

1979

1984

1989

1994

1999

2004

Turnout in European Parliament elections, 1979–2004 106

Joan DeBardeleben and Lawrence LeDuc

107

extensive literature on turnout decline in national-level elections. We consider various explanations for the difference between national-level and EU turnout, including those relating to institutions, the context in which elections occur, and individual voter attitudes. We ask the following: In considering the phenomenon at the EU level as compared to the national level, what is similar and what is different? An answer to this question will potentially help to extend the reach of theories and concepts in the literature on voter turnout, increasing their explanatory power in different contexts. In this examination, we also consider the extent to which the decline is a function of EU enlargement, as opposed to a secular decline in the nine countries that participated in the first EP election in 1979. Finally we examine specific problems affecting the new EU member states that joined in May 2004, many of which experienced the lowest levels of turnout in the 2004 EP elections.

The nature of the turnout decline Table 5.1 shows levels of turnout in EP elections, as a percentage of registered voters in all of the member states, indicating the mean value and the decline from the first EP election to the most recent one in 2004. In calculating these means we omit inaugural participation if the first election was held outside of the normal time, on the grounds that conditions were exceptional and therefore not strictly comparable with later contests. This method of calculation actually reduces the mean value of decline indicated for all of these countries (Greece in 1979; Portugal and Spain in 1984; and Austria, Finland, and Sweden in 1994), because turnout in the inaugural vote was higher in all cases except Greece than in any EP election since. There are some exceptions to the pattern of voter decline. Belgium and Luxembourg employ compulsory voting, a factor affecting both national and EP elections. Luxembourg also holds its national and EU elections on the same day, thereby eliminating any turnout variation between the two levels. Denmark is also exceptional, in that it has experienced no significant change since 1979 in either EP or national turnout (a topic addressed by Elklit and Togeby in Chapter 4). The United Kingdom has had generally low turnout levels in all EP elections, but these improved somewhat in 2004 relative to all previous EP votes, while still remaining below 40 per cent. Finally, while Finland did not show a decline in the two regularly scheduled elections that it participated in (1999 and 2004), the EU membership period of both Sweden and Finland is too short to draw any firm conclusions. In both cases turnout seems to have stabilized

108 Activating the Citizen Table 5.1: Electoral turnout in European Parliament elections, 1979–2004 (% of registered voters) Accession Country

1957 1957 1957 1981 1957 1973 1957 1957 1973 1973 1986 1986 1995 1995 1995 2004 2004 2004 2004 2004 2004 2004 2004 2004 2004

1979 1984 1989 1994 1999 2004 Mean Net change: First to 2004

Belgium1 Luxembourg1 Italy Greece1 Germany Ireland France Netherlands Denmark UK Portugal Spain Austria Finland Sweden Malta Cyprus Lithaunia Latvia Hungary Czech Republic Slovenia Estonia Poland Slovakia

91.4 88.9 84.9 78.6* 65.7 63.6 60.7 58.1 47.8 32.2

Mean

65.9 63.6 63.8 53

92.2 87.0 83.4 77.2 56.8 47.6 56.7 50.6 52.2 31.8 72.4* 68.9*

90.7 96.2 81.4 80.1 62.3 68.3 48.8 47.5 47.4 36.6 51.2 54.7

90.7 88.5 74.8 80.4 60.0 44.0 52.7 35.6 52.9 36.4 35.5 59.1 67.7* 57.6* 41.6*

91.0 85.8 70.8 75.3 45.2 50.7 46.8 29.9 50.4 24.0 40.4 64.4 49.0 30.1 38.8

90.8 90.0 73.1 62.8 43.0 58.7 43.1 39.3 47.9 38.9 38.7 45.9 41.8 39.4 37.8 82.4 71.4 48.3 41.3 38.5 28.3 28.3 26.9 20.8 16.9

91.1 89.4 78.1 75.2 55.5 55.5 51.5 43.5 49.8 33.3 41.5 56.0 45.4 34.8 38.3

−0.6 +1.1 −11.8 −14.4 −22.7 −4.9 −17.6 −18.8 +0.1 +6.7 −12.5 −8.8 −7.2 +9.3 −1.0

52.8 47.8

*Excluded from calculation in all cases, as it was an initial election held at a different time than the main EP election. 1 Country has compulsory voting. Source: European Parliament (2004).

at a lower level following their initial EP elections in 1995, which were held at a different time from the general EU parliamentary votes. Table 5.2 offers a different perspective on this data. Here we can see that the addition of new members in 2004 introduces a major distortion in the overall pattern. Many of the countries admitted in 2004 had a much lower rate of participation than the older member states. Whereas the decline in participation level from the first to the most recent EP election is 18 per cent overall, if we examine just the nine countries voting in the first election in 1979, we see only a 7.6 per cent decline over the period. This is very similar to the decline in participation found in domestic parliamentary votes in the same countries. Even as regards

Joan DeBardeleben and Lawrence LeDuc

109

Table 5.2: Decline in European Parliament election turnout, by accession cohorts1 (in %) Mean national turnout values in EP elections:

1979 1984 1989 1994 1999 2004 Net change: First to 2004 for each cohort

Of all member states 65.9 for the given election Of countries voting 65.9 in 19792 Of countries voting in 19843 Of countries voting in 1989/19944 Of countries voting in 1999 Of all 2004 new member state countries Of only central and eastern European new member state countries

63.6

63.8

53.0

52.8

47.8

−18.1

62.0

64.4

59.5

55.0

58.3

−7.6

61.5

64.2

59.6

54.0

56.8

−4.7

63.8

59.2

56.2

56.0

−7.8

52.8

52.7

−0.1

40.3 31.2

1 In

each case the calculation excludes countries where the initial election to the EP was held later than the main EP election due to accession to the EU after the date of that election. 2 Excludes the Greek vote in 1981. See Table 5.1 for list of countries. 3 Excludes the Portuguese and Spanish votes in 1986. See Table 5.1 for list of countries. 4 Excludes the Austrian, Finnish, and Swedish votes in 1995. See Table 5.1 for list of countries. Source: European Parliament (2004).

the old member states, however, where the decline has been at ‘normal levels’ (that is, approximating those affecting parliamentary votes), this similarity does not necessarily mean that the reasons are the same as those affecting domestic contests. The overall lower level of EP electoral turnout compared to national ones still requires explanation. Other member state cohorts (see Table 5.2) show similar or lower levels of decline. These data indicate that the addition of the 10 new member states in 2004, and particularly the eight that emerged from communism, substantially depressed EP voter turnout overall. This may seem a surprising development since, as noted above, other countries experienced higher turnout rates in their first foray into Europe-wide electoral contests. This may suggest some special factors affecting the central and east European new member states, a topic to which we will return later in this chapter.

110 Activating the Citizen

National level turnout and EU turnout: similar or different phenomena? The most commonly referenced explanation for low turnout in EP elections is their second-order nature. Initially proposed by Reif and Schmitt (1980) after the 1979 elections, the second-order elections hypothesis as applied to EP elections has been taken up by other analysts in relation to subsequent votes (Smith, 1996; Marsh, l998: Perrineau, Grünberg, and Ysmal, 2002; LeDuc, 2007). The basic idea is that voters are responding to national-level factors when voting in second-order contests, which could also include other lower-level elections, for example at the provincial or municipal level. Thus, voters will often punish the incumbent national government if they are dissatisfied with the state of affairs, or they may reward smaller parties that they are reluctant to support in national-level elections for fear of wasting their votes. In some specific circumstances the second-order effect may be weaker than others, and there is also considerable variance in the magnitude of such an effect (LeDuc, 2007). For example, parties that have been in government for only a short time often fare better, while those that have held national office for a substantial period of time may be punished more severely by voters in EP elections. Thus, in 2004, governing parties in Spain and Greece did somewhat better than the average, while those in Ireland and Sweden fared more poorly. A vote against the national governing party in an EP election is not necessarily a vote for the main opposition parties. Other scholars have noted that smaller political parties often do better in European elections than in national contests in the same countries (van der Eijk, Franklin, and Marsh, 1996; Marsh, 1998). Such a trend is not always observable in every country or in every EP election. Nevertheless, the tendency for voters to cast a vote against the national governing party in an EP election not uncommonly benefits one or more smaller parties that are well positioned to capture such ‘protest’ votes (Curtice, 1989). Low turnout in second-order elections, however, is one effect that seems to be both pervasive and consistent. The secondary nature of these votes, the argument goes, elicits lower levels of voter interest and thus lower turnout. The second-order election thesis puts a heavy weight on the larger political context and suggests few prospects for change, short of a major shift in perceptions on the part of voters or the relative power and importance of the institutions involved. While the second-order election thesis is a powerful explanatory concept, the approach that we take in this chapter is somewhat different. Here we seek to understand

Joan DeBardeleben and Lawrence LeDuc

111

whether some of the commonly advanced explanations for countrylevel differences in domestic turnout or differential levels of turnout decline in national elections are as persuasive in accounting for the low turnout in EP elections. In other words, we want to determine whether behind the second-order thesis some more fundamental variables may be lurking. Before turning to this question, however, we should note one aspect of the second-order election thesis that is unique to that model and could hardly be included in explanations of differences between countries in national level turnout, at least not in a comparable way. This has to do with the timing of EP elections. EP elections take place at fixed 5-year intervals, so their timing in relation to national election cycles can vary from one country to another and from one election to the next. In 2004, for example, seven counties had held a national election during the previous year.2 In such cases where a national vote has recently taken place, the EP election might be seen as having rather marginal relevance to national politics. In this context, political parties, particularly the larger parties, may tend to be less engaged, the campaigns less intense, and the voters less interested, thereby generating lower levels of citizen interest and perhaps lower turnout (Irwin, 1995). The opposite tends to occur when a national election is expected to occur within a short time after the European elections. In these circumstances, the EP election can be highly relevant to national politics. Political parties, preparing for a national campaign, are active and engaged, and the EP election provides them with opportunities to test issues and themes for the national campaign to follow. More typically, however, the EP elections take place at a point in the cycle distant enough from national elections to be relatively free of such a direct connection to national politics. The EP elections may still have consequences for the governing party, but unlike those occurring just before national elections, such consequences are not immediate. Voters may punish a governing party without removing it from office, promote the cause of an opposition party without giving it power, or make specific issue statements by casting a vote for minor parties. Of course, turnout may also be lower under such conditions, but the low turnout in itself may magnify some of these hypothesized effects, in the process motivating certain types of voters to participate who might otherwise not be inclined to do so, while at the same time inducing others, particularly supporters of the governing party, not to participate. The fact that EP elections are affected by their temporal proximity to national elections has important connotations for the second-order election thesis.

112 Activating the Citizen

Other factors affecting turnout in EP elections may, however, be more similar to forces that explain variations in national-level turnout (Blais and Dobrzynska, 1998; Franklin, 2002, 2004; Blais, 2006). The previous three chapters in this volume explored different approaches to understanding the reasons for declines in election turnout in Europe and Canada. In commencing our analysis of turnout levels in EP elections, we ask three questions. First, do explanations of national differences in turnout in country-level elections help explain differences between countries in EU voting levels and differing levels of decline in EP election turnout between countries? Second, to what extent do explanations advanced to account for variations in turnout for national parliamentary elections also help to explain the lower turnout levels in EP elections (as compared to national votes) and EP turnout decline within particular countries? Finally, what additional factors need to be taken into account that may be peculiar to EU elections? Institutional and contextual factors relating to the nature of elections Regarding the first question, with some refinement, several institutional explanations for differences between countries’ turnout levels in national elections may also help explain differences between countries’ EP turnout levels. Using an extensive database involving 91 countries, Blais and Dobrzynska, in Chapter 3 of this volume, identify compulsory voting and decisiveness of elections as explanatory variables that apply across the board, whereas higher voting age, proportional representation, a smaller number of parties on the ballot, and being an established democracy all affect turnout levels in a positive direction in national elections in Europe, but not elsewhere. In their examination of Scandinavian exceptionalism, and particularly the Danish case, Elklit and Togeby, in Chapter 4, identify high competitiveness of elections as a key factor, one that increases opportunities for political mobilization of the electorate through the media and citizen activity, combined with an electoral system based on proportional representation. In addition, from his review of literature on the topic, Siaroff in Chapter 2 draws attention to two other factors that may encourage lower voter turnout both at the national and EU levels, namely, low levels of institutionalization of political parties and diffusion of power between political institutions (at the national level sometimes manifested as weak parliaments). How adequate are these explanations for understanding differences between countries in terms of turnout in EP elections? In other words, do some of the factors that explain differences in low turnout at the national

Joan DeBardeleben and Lawrence LeDuc

113

level have an equal or perhaps even stronger impact when it comes to explaining differing turnout levels in EP elections? For many institutional and contextual factors, this may seem to be the case. Institutional factors are of two types: those relating to electoral mechanisms and those relating to the larger institutional structure of decision making within the polity. Electoral mechanisms themselves may be important when considering why some countries have lower EP turnout than others; indeed, many types of electoral mechanisms would operate in a similar way in both EP and national elections in a given country. Such examples would be compulsory voting and voting age. Compulsory voting produces a clearly positive impact on turnout in both national and EP elections in those countries that use it. One institutional feature that can differ between the national parliamentary contests in a particular country has to do with the electoral system. Although all EU member states are required to utilize some form of proportional representation in EP elections, there are nevertheless variations in voting rules and procedures. Some countries, for example, use a single national list of candidates, while others employ regional lists. Closed lists are used in some countries, while others permit preferential voting for individual candidates. Ireland continues to use single transferable vote in both European and national elections. Such variations mean that the political opportunities for smaller parties can vary considerably. In a country with a large number of parties and a single national list, it is easier for a small single-issue party to gain a seat. On the other hand, regional parties may fare better in countries that employ regional lists. To the extent that countries use the same electoral systems in EP and national elections, one would hypothesize that the electoral system might affect turnout in a similar way in both contests, although this is not necessarily the case if differences in the party system or name recognition of parties differs between the two votes. In countries where a different type of electoral system is used in domestic parliamentary and EP elections, the electoral system may have stronger differential effects both on election outcomes and on turnout. In France, for example, which utilizes two-ballot plurality in national elections but list proportional representation in European ones, there are substantial differences in election outcomes. Parties such as the Greens or the National Front are all but shut out in national elections, but in EP elections, these parties win seats. In Britain, which continues to use first-past-the-post in national elections, the relative differences in the strength of the parties at the national and EP levels are greatly magnified by the differences in the electoral systems employed at the two levels. Such differences affect turnout in a variety of ways, because they

114 Activating the Citizen

are capable of motivating the participation of supporters of particular parties, while depressing that of others. In terms of larger institutional structures of decision making within the polity, those factors would, on the face of it, affect all EU member states in a fairly similar manner. There is no reason to believe that Italians or Greeks (who have relatively higher EP turnout rates) understand the complex institutional arrangements that characterize the EU’s decision-making processes better than do the Dutch or British (who have lower levels). One would expect this factor to contribute to lower levels of turnout in EP elections generally, although one could speculate that citizens in new member states may be even more baffled by what their vote could mean than citizens in older member states. This would be consistent with the idea that being in an older member state facilitates higher participation, perhaps in part because citizens have become more comfortable with EU structures and how their vote relates to them. One factor that could differ between countries in EP elections is the level of political mobilization. As Elklit and Togeby suggest, the higher level of mobilization of the vote in Denmark applies to EP elections as well as national votes, even though turnout in EP elections is substantially lower than in national ones. However, in terms of explaining differences between countries in EP turnout, mobilization may be important, just as it is in explaining differences in national level turnout. Now we turn to the second question, namely the extent to which explanations advanced to account for variations in turnout for national parliamentary elections also help to explain the lower turnout levels for EP elections within particular countries. Here institutional factors play a more dubious role in explaining these differences. As LeDuc (2007) has documented elsewhere, this difference usually is in the order of 15 to 20 per cent between the EU parliamentary election and the most proximate national election, but rose closer to 25 per cent in 2004 with the addition of so many additional low turnout countries. What explains this difference between elections at the two levels? Some factors that generally seem to impact turnout can be excluded. For example, the mandatory proportional representation electoral system for EU elections would normally be expected to boost, not depress, turnout. Likewise, several institutional features would not differ between national and EU elections, such as voting age, voter registration systems, or other organizational details. Demographic factors, such as the declining rates of participation of young voters and the effects of generational replacement over time, would likewise be relative constants in their effects on both EP and national elections (Franklin, Lyons, and Marsh, 2004;

Joan DeBardeleben and Lawrence LeDuc

115

Pammett and LeDuc, 2004; Wass, 2007). Age of the democratic experience (Blais and Dobryznka’s ‘old democracy/new democracy’ variable) likewise would not play a role in explaining differences between national and EU turnout levels in a given country, although it may account for variations in turnout more generally. Among those factors that might, on the basis of national-level studies, seem to potentially explain differences in turnout between the national and EU levels are the lower degree of institutionalization of the European party system in comparison to national party systems, the minimal ability of the EP to affect policy compared to the greater impact of the national parliamentary decision, and the generally more diffuse nature of decision making in the EU as compared to national decision making. All of these would act to produce a lower degree of perceived competitiveness of EP elections compared to national votes, since the outcome would have less clear implications for the voter. Because the implications of EP elections themselves may seem unclear, and the impact of one’s vote insignificant, voters may view them as an opportunity to comment on developments in national-level politics, as the second-order hypothesis suggests. So behind the second-order election explanation lurk, indeed, some factors that in fact do explain variations or declines in national-level voting in particular countries as well. So far we have only looked at system-level variables. Several analysts, however, have suggested that differences in attitudes of individual voters, when aggregated to a national level, may also account for variations in turnout. When individual attitudes evidence distinct patterns in particular societies, these are sometimes labelled ‘cultural’ factors. In their contribution to this volume, Elklit and Togeby, for example, point to the presence of strong norms of civic duty and relatively high levels of trust in political institutions as factors underlying sustained high turnout in the Danish context. Rose (2004) also indicates that ‘where distrust of parties and of governments is lower, turnout at the European Parliament election is significantly higher’ (pp. 5–7); he also notes higher levels of turnout in smaller countries. Examining data from the 2004 EP elections, Flickinger and Studlar (2007) note that ‘trust in one’s national government and satisfaction with the way democracy works in one’s country are . . . positively and significantly linked to EP turnout’ (p. 389). They find that not only attitudes toward national politics, but EU-level factors such as whether the country hosts an EU institution and the proportion of agricultural workers in the country seem to have an influence on national turnout rates in EP elections by influencing the overall feeling about the EU and its ‘closeness’ to the country involved.

116 Activating the Citizen

Data are available on relative levels of trust in the EU as compared to national institutions. Eurobarometer measured levels of trust in both EU and national institutions in a survey carried out in October and November 2004, shortly after the June 2004 EP election. The data provide two interesting insights. The 2004 survey shows that levels of trust in EU institutions were higher than in national institutions, with 50 per cent indicating that they trusted the EU; this represented an 8 per cent increase from a similar question asked in spring of the same year (European Commission, 2005). Second, among European institutions, the EP, overall, commanded the highest level of trust, with levels rising from 1999 to 2004 from 50 per cent to 57 per cent of the sample. These data would suggest that, on an aggregate level, differences in levels of trust work in the opposite direction than that expected – that is, trust in EU institutions is higher than in national institutions, but turnout is lower, and relatively high levels of trust in the EP do not seem to offset other factors that reduce turnout for EP elections. However, when one looks at specific countries a different pattern emerges. Two countries showing the highest levels of trust, Lithuania and Hungary, had among the highest turnout levels of the new member states (first and third, with Latvia in between). Likewise Greece had both high turnout and high levels of trust. These figures do not document a direct correlation, either in a positive or negative direction, between levels of trust and voting on the individual level, but they do suggest that there may be some relationship and that it may have a different impact in different national contexts.

The new democracies of the EU: turnout in EP elections and accession referendums For post-communist new member states, the decline in national level voting has been steep in most cases (See Table 5.3). While turnout in the first post-communist election was, except in Poland, 75 per cent or more of registered voters, in five of the eight countries involved (Czech Republic, Latvia, Lithuania, Slovakia, and Slovenia) these rates declined by over 20 per cent from that time to the most recent election (before 2008), and in Estonia a close 16.3 per cent. Even in Poland, which had the lowest turnout in its national election in 1989 (62 per cent), an equivalent decline occurred by 2005, perhaps because of the unpopularity of the Kaczynski ´ government. Hungary experienced a recovery in the 2002 and 2006 elections from dips in the 1990s. Nonetheless, as Table 5.3 indicates, the picture is not entirely gloomy, as turnout increased in three

Table 5.3: Turnout in EU new member states’ national parliamentary elections since the collapse of communism (% of registered voters) Country

Lithuania Latvia Hungary Czech Slovenia Estonia Poland Slovakia

Sequence of national parliamentary elections lst

2nd

3rd

4th

75.2 81.2 75.4 96.3 85.9 78.2 62.1 96.3

52.9 89.9* 68.9 84.7 73.7 67.8 43.2 84.7

58.2* 71.9 56.7 76.3 70.4 68.9 52.1* 75.4

46.1 71.9 73.5* 74.0 60.6 57.4 47.9** 84.2*

5th

6th

71.2 64.4** 57.9

61.0

58.2 46.2** 70.1

61.9* 40.6 54.7

7th

64.5*

53.9*

Change from first to most recent −29.1 −20.2 −11 −31.8 −25.3 −16.3 −8.2 −41.6

Years of elections

1992, 96, 2000, 04 90,93,95,98,02,06 90,94,98,02,06 90,92,96,98,02,06 92,96,00,04 90,92,95,99,03,07 89,92,93,97,01,05, 07 90,92,94,98,02,06

*Deviates by more than 2 per cent from a pattern of linear decline in turnout. **A decline from the previous election but an increase from a preceding election in the series. Source: Adapted from International Institute for Democracy and Electoral Assistance, http://www.idea.int/vt/parl.cfm

117

118 Activating the Citizen

countries in the most recent elections (Czech Republic, Estonia, and Poland), suggesting the possibility of a reversal in trend. Still, voter turnout in national level elections and EP elections in new member states is markedly lower than in the 15 ‘older’ member states, a pattern consistent with the findings of Blais and Dobrzynska about differences between old and new democracies. In the June 2004 EP election that included post-communist countries joining the EU in May 2004, the group, on average, showed not only levels of turnout lower than in national elections (see Table 5.4), but also significantly lower rates of turnout than older member states (see Table 5.1), and six of these countries fell at the very bottom of the list. The post-communist countries joining in 2004 had an average turnout rate in the 2004 parliamentary election of only 31.2 per cent, compared to the EU average of 47.8 per cent. This echoes Rose’s (2004) phrase, ‘Europe Expands, Turnout Falls’. An examination of temporal change must await the next EP election, since we cannot now say whether the low EP turnout rates in these countries in 2004 will mirror the temporal decline in national-level voting or whether they are reflective of a different set of influences. However, it can be noted that, for many previous EU entrants, their first election after accession produced a higher turnout level than did those held subsequently (see Table 5.1). Koepke and Ringe (2006) have argued that EP elections do not act like typical second-order contests in these new member states, except in relation to turnout patterns. They coined a new term, ‘pure preference election’, for circumstances in which the electorate uses nonparliamentary contests to vote sincerely rather than strategically, noting that, in contrast to expectations based on the second-order election thesis, ‘voters in CEE [central and eastern Europe] do not cast protest votes against their incumbent national governments in second-order elections’ (p. 341). Rather, when elections promise little in terms of changing government, central and eastern European ‘voters cast ballots based on their general, and relatively stable, positions vis-à-vis their communist past and not on how changes since the last election relate to those positions’ (p. 342). While noting that this explanation remains a hypothesis to be tested in future EP votes, the import of Koepke and Ringe’s work is that political dynamics affecting multi-level voting may be distinct in these new post-communist member states as compared to other EU countries. However, Koepke and Ringe do not explain why turnout in EU elections in these countries lags behind parliamentary levels, just as in other EU countries. Auers (2005), on the other hand, finds EP elections in central and eastern Europe to have many second-order characteristics in addition

Table 5.4: Turnout in national and EP votes and accession referendums (% of registered voters) Turnout rates National parliamentary (election closest after EP election)

Difference in turnout rates Accession European Parliament referendum (14 June, 2004)2 (March–Sept. 2003)1

Accession EP compared to referendum national compared to parliamentary national parliamentary

Accession Yes vote referendum referendum compared to EP election

Lithuania Latvia Hungary Czech Slovenia Estonia Poland Slovakia

46.1 61.0 64.4 64.5 60.6 61.9 40.6 54.7

63.4 72.8 45.6 55.2 61.1 64.1 58.9 52.2

48.4 41.3 38.5 28.3 28.3 26.8 20.9 17.0

17.3 11.8 −18.8 −9.3 0.5 2.2 18.3 −2.5

2.2 −19.7 −25.9 −36.2 −32.3 −35.0 −33.1 −37.8

15.1 31.2 7.1 26.9 32.1 37.2 35.1 35.3

91.1 67.0 83.8 77.3 66.1 66.8 77.5 93.7

Mean

56.7

59.2

31.2

2.4

27.2

27.5

77.9

1 Accession

referenda occurred between 3 March, 2003 (Slovenia) and September 2003 (Latvia and Estonia). EP election occurred on 14 June, 2004. Sources: Adapted parliamentary data from International Institute for Democracy and Electoral Assistance (2004); EU parliamentary election and referendum figures from Rose (2004).

2 The

119

120 Activating the Citizen

to turnout levels. Flickinger and Studlar (2007) also note that turnout in EP elections in new member states may be affected by different dynamics than in the EU-15. Looking at bivariate relationships with turnout for aggregate national-level variables, they conclude that ‘national level concerns (e.g., trust in government, satisfaction with democracy, and links to political parties) mattered for citizens in the new member states but not for citizens of established democracies’ (p. 397), whereas EU-level factors were more important in the EU-15. Tillman (2004), on the other hand, finds that European issues have affected voter choice and party positioning in domestic elections in the new member states as well. In addition to the 2004 EP vote, the post-communist countries of the EU did take part in one other EU vote, namely the accession referendums. These took place before the June 2004 EP election, between March 2003 (in Slovenia) and September 2003 (in Latvia and Estonia). Here, as shown in Table 5.5, turnout rates, while low by comparison with accession votes in previous enlargements (Szczerbiak and Taggart, 2004b), were well above rates for the EP elections that were to follow. Except for Hungary (where the drop was a modest 7 per cent), in all of the other countries, turnout fell dramatically between the accession referendums and the EP elections. Another notable feature of these votes was the high level of support for the ballot question, which, as LeDuc discusses in Chapter 12 of this volume, is fairly unusual in referendums. In three countries (Estonia, Latvia, and Slovenia), two thirds of those voting supported accession; in all five of the other acceding post-communist countries support was over 75 per cent, with a mean national value of 78 per cent. These two characteristics of the accession referendums – relatively higher turnout and strong positive votes – may both be related to political mobilization.3 Those mobilizing the vote, depending on the country, included political leaders (of both government and opposition parties), the electoral commissions, and citizen groups. While most elites supported accession, an active ‘no’ campaign also existed in some countries. The mobilization campaigns were assisted by the cardinal importance of the decision, which had a long build-up that had generated high levels of public awareness. However, one should not overlook the fact that these factors could have depressed turnout; voters might just as easily have considered the outcome to be inevitable, given all the hype that had surrounded the accession process in the previous time period. The assertive campaign to motivate voting was, therefore, most likely critical. In two countries – Poland and Lithuania – turnout for the referendum exceeded turnout for the closest national election by a wide margin. In both cases mobilization campaigns were particularly important, in

Joan DeBardeleben and Lawrence LeDuc

121

part because a 50 per cent participation rate was required to validate the outcome. In Poland, even opposition political parties that supported accession tried to get people to vote for fear that the unpopularity of the government in power at the time might harm the outcome. Another measure to facilitate a successful outcome was extending the voting period to two days, with interim turnout figures reported after the first day (Szczerbiak, 2004). In comparison to parliamentary elections, societal organizations were also eligible to campaign rather than just political parties and voters. In Lithuania, although the outcome was not in doubt, political elites and a variety of other social actors apparently considered a strong showing to be important. Here too, the referendum was extended to two days, and supermarkets gave discounts to those who had voted. The Catholic Church also actively encouraged participation. In Slovakia, in a referendum held before the Polish one, and where 50 percent turnout was also required, concerns about achieving the minimum turnout were reinforced by the fact that in previous referendums turnout had fallen short, invalidating the results (Henderson, 2004). In Slovakia pro-EU groups had experience in mobilizing voters in the 1998 and 2002 parliamentary elections, because they were considered critical to the accession process, given that the previous Me˘ciar government was viewed as obstructive to EU accession. A strong pro-EU consensus threatened to reduce public interest and suppress turnout, meaning that mobilization of the public by elites was seen as especially important (Henderson, 2004). The Hungarian case poses a counterexample. Turnout in the referendum vote was relatively low at just above 45 per cent and considerably lower than the most proximate parliamentary election (although nonetheless higher than the EP election that followed). There was apparently a fear that mobilizing voters could lead to a higher ‘no’ vote. Fowler (2004) points to the low competitiveness of the vote along with the absence of a threshold for validity as important factors depressing turnout. Many voters apparently believed the outcome was already clear and interest levels were low. The importance of mobilization (or lack thereof in the Hungarian case) is consistent with the findings of other authors in this volume, for example Elklit and Togeby, who identify this as an important factor holding up turnout in Denmark. Szczerbiak and Taggart (2004a, 2004b) test a general model to explain turnout in these referendums, concluding that the ‘level of contestation of the European issue’ (2004b, p. 753) and normal turnout levels in a country are the most important factors in explaining turnout levels in the 2003 accession referendums. They conclude that the actual level of

122 Activating the Citizen

resources expended on the referendum campaign did not seem to have a significant impact. These findings suggest that, to the extent mobilization is important, it has to do with the overall level of interest generated by the vote, including but not limited to explicit efforts by elites to stimulate turnout. The Szczerbiak and Taggart analysis focused on explaining differences between turnout in the various referendums rather than on the relatively higher levels there as compared to the EP election that followed, and therefore may underestimate the importance of mobilization as such. A further question relates to whether the same factors that have produced declines in national voter participation in these countries also contribute to the low turnout in EP elections. Auers (2005) notes the exceptionally low level of mobilization efforts in the 2004 EP vote, evidenced in lower campaign spending by political parties. He sees this factor as the key to explaining the lower turnout: ‘the parties did not prioritize the European elections’ (p. 750). The lack of mobilization efforts combined with other factors commonly cited as eliciting low turnout elsewhere, such as the distance between the voter and Brussels and an unclear picture of how the EU works. Auers also points to voter fatigue in some cases, especially Slovakia, due to the large number of votes in temporal proximity. Unlike Koepke and Ringe (2006), Auers sees the outcome as consistent with the general pattern of second-order elections elsewhere, namely that in most cases ‘governing parties . . . fared badly’ (p. 752). Elsewhere in this volume, Blais and Dobyzynska identify ‘being a new democracy’ as an explanatory variable associated with lower turnout. It seems likely that this factor is a surrogate for other variables. First, even by 2004 these democracies were not so new. They had all had over a decade of experience with democratic elections and had seen a drop rather than a steady low level of participation. While southern European countries that joined the EU after previous bouts with authoritarian government were somewhat more established when they joined the EU, nonetheless their previous parliamentary votes prior to accession had not exhibited particularly low turnout rates. It is possible that being a post-communist new democracy (that is, not just any kind of new democracy) creates a different environment for discouraging electoral turnout, given previous mandates to vote in the communist period. Bernhagen and Marsh (2007) find support for the idea that individual characteristics have differing impacts on the propensity to vote in the post-communist countries of central and eastern Europe and western European countries, although the same does not apply to other forms of protest activity. However, if

Joan DeBardeleben and Lawrence LeDuc

123

one looks to such post-communist political cultural factors to explain the ‘new democracy’ variable, one might suppose that those attitudinal characteristics inherited from the communist period that inhibited civic activity should have progressively decreased, not strengthened. However, Bernhagen and Marsh note that non-electoral forms of protest activity also declined between 1990 and 1999 in these countries. Another hypothesis, and perhaps a more convincing one, is that the weight of adopting the ‘acquis communitaire’ during the pre-accession period meant that elected members of parliament, to a large extent, were implementers of legislative requirements that emanated from Brussels, leaving fewer choices in domestic hands. As Raik (2004) notes, referring to Estonia in the pre-accession period, ‘the underlying logic of enlargement/integration promoted bureaucratic, executive-dominated policy making and left little room for democratic politics in applicant countries’ due to the imperative of accepting legislation ‘as quickly as possible’ (p. 589). Policy making became depoliticized, which would serve to make elections less important to outcomes. Pridham (2002) has made a similar point in relation to Slovakia and Romania. He argues that the pre-accession dynamic ‘creates a potential for widening the gap between political elites and the masses’ even though requirements of EU conditionality may have an overall positive impact on democratization processes, particularly at the institutional level. However, the effects are ‘the least evident at the level of civil society, at least during accession’ (p. 971). These types of effects are hard to measure because they may not be evidenced as conscious motivations, but rather may be an underlying disposition, which survey research can tap only with great difficulty, if at all. It is also not known whether they will dissipate, or how long such a process may take. One might expect that as citizens in the new member states come to understand that their countries are EU ‘policy makers’ as well as ‘policy takers’, then the interest in EU matters could rise.

Conclusion The second-order model predicts that turnout in EP elections will be lower than in national elections because voters see EP elections as less relevant to their own concerns than national elections and because political parties are less effective at mobilizing their supporters in such contests. Evidence from all of the EP elections to date has supported this hypothesis, as turnout in every European election since 1979 has been lower than the average found in national elections for the member countries. But there is also considerable variation between countries as

124 Activating the Citizen

well as variation between the same countries over time that cannot be accounted for by the second-order model. In most EU member countries, turnout in EP elections has established a fairly consistent pattern, but in a few others it is found to vary considerably from one EP election to the next. In some instances, these variations over time are explained by contextual factors, such as the position of the EP election in the national elections cycle or the structure of party systems. However, most countries that have established a pattern of low voter turnout in EP elections have tended to maintain that pattern, even where changes have taken place over time. In the Netherlands for example, turnout in EP elections has varied from a high of 58.1 per cent in 1979 (the third lowest among countries voting in that election) to 39.3 per cent in 2004 – more than 40 per cent lower than the turnout recorded in the most recent Dutch national election in 2006. Perhaps of even greater concern than the low turnout rate in EP elections relative to national elections is the persistent pattern of decline. Turnout in EP elections began to drop sharply after 1989, and it registered a new low in 2004 at less than 50 per cent of eligible voters. While part of this decline in the mean level of turnout may be accounted for by lower participation rates in some of the new member countries, it is clear that the decline began long before enlargement. Turnout has also been declining in national elections in many European countries, and we have hypothesized that some of the decline in EP participation rates is being driven by the same factors that account for declining rates of turnout in national elections – generational factors, declining trust and efficacy, electoral competitiveness, or varying levels of political mobilization. But there are also elements specific to EU politics that partly explain both the decline and the low overall rates of participation in EP elections. Evidence also suggests that the dynamics affecting voter turnout in EP elections in the countries of central and eastern Europe may have a somewhat distinctive character, which, until now, has reinforced negative turnout trends in the EU as a whole. One might have hypothesized that turnout would be somewhat higher in those countries voting for the first time in a European election in 2004, as it has been historically for new member countries. But for all but one (Lithuania) of the eight post-communist countries that joined the EU In 2004, this was not the case. Seven of the ten new EU members joining in 2004 had turnout levels below the mean for all EU countries, and five of these (Czech Republic, Slovenia, Estonia, Poland, and Slovakia) recorded turnout levels below 30 per cent. Whether these countries will, like earlier members, establish a persistent pattern of low and declining

Joan DeBardeleben and Lawrence LeDuc

125

turnout can only be known after one or two future EP elections have taken place. Alternatively, some factors that depressed turnout in the first EU election in these countries may be mitigated in the future; for example, political and cultural legacies of the communist period or a sense of resignation in the face of the EU accession requirements. Our analysis has indicated the particular importance of political mobilization, so the manner in which elites in central and eastern Europe take up the challenge of promoting active citizen interest in EU matters could be of critical importance. Whether the more general pattern of decline in both national and EP elections can be reversed remains a question for future research. And, for those scholars who connect the problem of low and declining turnout to the larger issue of a democratic deficit in the EU more generally, the answers to these questions are of major importance.

Notes 1. For the EU, taken as a whole, it was 45.6 per cent. Here, we use the mean of all member countries, because it is more readily comparable with national turnout figures. 2. Greece, Spain, Belgium, Estonia, Finland, Malta, and the Netherlands. 3. The European Parties Elections and Referendums Network at the Sussex European Institute carried out studies of the accession referendums through its EPERN Referendum Briefings, published online at http:// www.sussex.ac.uk/sei/1-4-2-9.html.

References Auers, D. (2005) ‘European Elections in Eight New EU Member States’, Electoral Studies, 24, 741–84. Bernhagen, P. and M. Marsh (2007) ‘Voting and Protesting: Explaining Participation in Old and New European Democracies’, Democratization, 14(91), 44–72. Blais, A. (2006) ‘What Affects Voter Turnout?’, Annual Reviews of Political Science, 9, 111–25. Blais, A. and A. Dobrzynska (1998) ‘Turnout in Electoral Democracies’, European Journal of Political Research, 33, 239–61. Curtice, J. (1989) ‘The 1989 European Election: Protest or Green Tide?’, Electoral Studies, 8, 217–30. Eijk, C. van der, M. Franklin, and M. Marsh (1996) ‘What Voters Teach Us About Europe-Wide Elections: What Europe-Wide Elections Teach Us About Voters’, Electoral Studies, 15, 149–66. European Commission (2005) Eurobarometer 62: Public Opinion in the European Union, http://ec.europa.eu/public_opinion/archives/eb/eb62/eb_62_en.pdf

126 Activating the Citizen European Parliament (2004) European elections 10–13 June, http://www.europarl. europa.eu/elections2004/ep-election/sites/en/ results1306 / turnout_ep / index. html Flickinger, R. S. and D. T. Studlar (2007) ‘One Europe, Many Electorates? Models of Turnout in European Parliament Elections After 2004’, Comparative Political Studies, 40(4), 383–404. Fowler, B. (2004) ‘Hungary: Unpacking the Permissive Consensus’, West European Politics, 27(4), 624–51. Franklin, M. N. (2002) ‘The Dynamics of Electoral Participation’, in L. LeDuc, R. G. Niemi, and P. Norris (eds), Comparing Democracies 2: New Challenges in the Study of Elections and Voting (London: Sage), pp. 148 –168. Franklin, M. N. (2004) Voter Turnout and the Dynamics of Electoral Competition in Established Democracies Since 1945 (New York: Cambridge University Press). Franklin, M. N., P. Lyons, and M. Marsh (2004) ‘The Generational Basis of Turnout Decline in Established Democracies’, Acta Politica, 39, 115–51. Henderson, K. (2004) ‘EU Accession and the New Slovak Consensus,’ West European Politics, 27(4), 652–70. International Institute for Democracy and Electoral Assistance (2004) Parliamentary elections, http://www.idea.int/vt/parl.cfm Irwin, G. (1995) ‘Second Order or Third Rate?: Issues in the Campaign for the Elections for the European Parliament 1994’, Electoral Studies 14(2), 183–99. Koepke, J. R. and N. Ringe (2006) ‘The Second-order Election Model in an Enlarged Europe’, European Union Politics, 7(3), 321–46. LeDuc, L. (2007) ‘European Elections and Democratic Accountability: The 2004 Elections to the European Parliament’, in J. DeBardeleben and A. Hurrelmann (eds), Democratic Dilemmas of Multilevel Governance: Legitimacy, Representation, and Accountability in the European Union (Basingstoke and New York: Palgrave Macmillan), pp. 139–57. Marsh, M. (1998) ‘Testing the Second Order Election Model After Four European Elections’, British Journal of Political Science, 28, 591–607. Pammett, J. H. and L. LeDuc (2004) ‘Four Vicious Circles of Turnout: Competitiveness, Regionalism, Culture and Participation in Canada.’ Paper presented to the Joint Sessions Workshops of the European Consortium for Political Research in Uppsala, Sweden Perrineau, P., G. Grünberg, and C. Ysmal (eds) (2002) Europe at the Polls: The European Elections of 1999 (Basingstoke: Palgrave – now Palgrave Macmillan). Pridham, G. (2002) ‘EU Enlargement and Consolidating Democracy in PostCommunist States’, Journal of Common Market Studies, 40(3), 953–73. Raik, K. (2004) ‘EU Accession of Central and Eastern Europe: Democracy and Integration as Conflicting Logics’, East European Politics and Societies, 18, 567–94. Reif, K. and H. Schmitt (1980) ‘Nine Second-order Elections: A Conceptual Framework for the Analysis of European Election Results’, European Journal of Political Research, 8, 3–44. Rose, R. (2004) Europe Expands, Turnout Falls: The Significance of the 2004 European Parliament Election (Stockholm: International Institute for Democracy and Electoral Assistance). Smith, J. (1996) ‘How European are European Elections?’, in J. Gaffney (ed.), Political Parties in the European Community (London: Routledge), pp. 275–90.

Joan DeBardeleben and Lawrence LeDuc

127

Szczerbiak, A. (2004) ‘History Trumps Government Unpopularity: The June 2003 Polish EU Accession Referendum’, West European Politics, 27(4), 671–90. Szczerbiak, A. and P. Taggart (2004a) ‘Conclusion: Towards a Model of (European) Referendums’, West European Politics, 27(4), 749–77. Szczerbiak, A. and P. Taggart (2004b) ‘The Politics of European Referendum Outcomes and Turnout: Two Models’, West European Politics, 27(4), 557–84. Tillman, E. R. (2004) ‘The European Union at the Ballot Box?: European Integration and Voting Behavior in the New Member States’, Comparative Political Studies, 37(5), 590–610. Wass, H. (2007) ‘The Effects of Age, Generation and Period on Turnout in Finland: 1975-2003’, Electoral Studies, 26, 648–59.

This page intentionally left blank

Part III Parties as Vehicles of Participation

This page intentionally left blank

6 Party Membership and Activism in Comparative Perspective Paul Whiteley

Cross-national research into party membership has been hampered in the past by the lack of comparative data on the voluntary dimension of party organization. With the emergence of comparative data sets such as the World Values Survey, the European Social Survey, and the International Social Survey, this state of affairs is increasingly being remedied. It is now possible to get a clear picture of how voluntary activity in political parties is faring in many contemporary democracies. One important finding from recent research into European political parties is that a long-term decline in membership and grassroots party activism has been occurring over the last quarter of a century (see Siaroff, this volume, as well as Katz et al., 1992; Seyd and Whiteley, 1992; Scarrow, 2000; Mair and van Biezen, 2001; Webb, 2002; Whiteley and Seyd, 2002). This development has important implications for the future of democracy, since it is not clear if political parties can continue to play a key role in sustaining civil society in the absence of a voluntary dimension to party activity. This is because parties that lack volunteers run the risk of becoming highly dependent on the state and in effect becoming absorbed into a state-sponsored cartel (Katz and Mair, 1995; Detterbeck, 2005). If this happens, then political parties are likely to turn into, in the words of one researcher, ‘public utilities’ (van Biezen, 2004). Needless to say, this severely undermines their links to civil society. Interestingly enough, party leaders in a number of countries are aware of this danger and have been revising their party constitutions in order to give members a greater role in the recruitment of candidates, the election of leaders, and the policy-making process more generally (Scarrow, Webb, and Farrell, 2000; Pennings and Hazan, 2001; Seyd and Whiteley, 2002). There are good reasons that political leaders might try to strengthen the voluntary dimension of their party organizations in this way. Party 131

132 Activating the Citizen

members and activists fulfil several important functions, and if these are not supported, then political parties are likely to be significantly weakened as a result. In this chapter, I reflect on the significance of these declines by analysing the contribution to civil society of party membership and, further, of party activism. The data for this exercise come from 36 countries that participated in the 2004 International Social Survey Programme (ISSP). (This is the same data set used by Jon Pammett in his examination of participation and citizenship in this volume.) The countries include the advanced industrial democracies of Europe and North America as well as some ‘near’ democracies in Asia and Latin America. The aim is to assess the effectiveness of the voluntary dimension of party organization, while evaluating if party members actually fulfil the functions attributed to them in civil society. One difficulty of comparative research into voluntary party activity is defining party member. In most democracies, joining a party means paying a subscription to the organization and possibly agreeing to abide by some broad principles supported by the party (Whiteley and Seyd, 2002). However, in some countries, notably the United States, party members are not defined in these terms. In the United States, individuals identify themselves as party supporters in the course of the voter registration procedure, but this does not involve paying dues to a party organization (Rapoport and Stone, 2006). So there are issues arising from the meaning of party membership in cross-national research. The ISSP avoids this problem by asking respondents if they belong to a political party and actively participate in it or if they belong to a party but do not actively participate. This subjective definition allows people to define membership in their own terms. It may not be precisely comparable across countries, but it does capture the extent to which individuals are psychologically engaged with political parties. The numbers of people belonging to parties in the sense used here are therefore somewhat higher than the data on official party members, as outlined in Chapter 2. Nevertheless, for ease of terminology, I will refer to them as members in this chapter. I begin by examining the social characteristics of those who reported belonging to political parties before exploring their role in the political system.

Who are the party members? The pooled data across all of the countries show that 3.4 per cent of respondents belonged to a party and were active, a further 7.2 per cent

Paul Whiteley 133

were inactive members, and 8.6 per cent used to belong to a party. These average figures conceal large variations across countries. Figure 6.1 shows the percentage of people who were party activists or party members in each of the countries in the survey. It can be seen that Poland had the fewest activists and members, closely followed by Hungary, Latvia, and Russia. It is clear that Soviet-style communism left a legacy of weak civil society, and this is most apparent in the figures on voluntary party activity. Britain had just over 10 per cent of respondents who were members or activists and Canada had just under 13 per cent. The high scores for the United States, in part, reflect the definition of party membership in that country, as noted earlier. A rough guide to the health of voluntary party organizations in each country is provided by the figures on former members. It is instructive that there were rather more ex-members than members in the sample as a whole, and former members outnumbered current members and activists in 16 of the countries, including France, Germany, and Sweden. In Britain, about 6 per cent of respondents were ex-members and in Canada it was almost 12 per cent. These figures show that party membership and activism are dynamic phenomena with individuals moving from membership to activism and in the reverse direction over time. It also shows that voluntary party activity is in sharp decline in some countries. The social demographic characteristics of members and ex-members are described in Table 6.1. Some of the figures in this table are quite striking, such as the gender bias; men were much more likely to report belonging to parties than women, and the bias is even more marked in relation to activists. Similarly, the age profile of members shows that respondents under the age of 35 were significantly under-represented among members and even more so among activists. The lack of party activists under the age of 25 is particularly noticeable. With respect to occupational status, members and activists were more likely than nonmembers to be managers or professionals, although in this case the relationships are not strong. With regard to the employment sector, nearly a quarter of party activists worked for the government compared with less than a fifth of non-members. Members and activists were more likely to be self-employed than the rest of the population. In keeping with their professional status, they were more likely to be graduates, and activists were even more highly educated than members. Finally, they were more likely to be married than the rest of the population, suggesting that they have stronger community ties than their counterparts in the general population.

Po H lan un d ga La ry t R via us si So ut C a h hil Ko e Fr rea G an er c m e Au an st y ra J lia Sl apa ov n Po eni rtu a g Sp al Bu ai l n Sl gar ov ia ak B ia C ze D ra ch en zil R ma ep rk u Fi blic nl N Ir and et el h a Sw erl nd itz and Ph erla s illi nd p Sw ine ed s e N ew B n r Ze itai al n a Ta nd iw M an e C xic an o C ada yp N rus or w Ve Is ay ne rae zu l Au ela U U str ni ru ia te gu d a St y at es

Percentages

134

45

40

35

30

25

20

15

10

5

0

Figure 6.1: Party membership, including party activists, in 36 countries in 2004 (% of respondents)

Paul Whiteley 135 Table 6.1: Socioeconomic characteristics of party members and activists (column % within each category) Belong and participate

Belong not participate

Used to belong

Never belonged

Gender Male Female

61 39

53 47

57 43

46 54

Age 18–25 26–35 36–45 46–55 56–65 66+

9 14 22 22 18 15

11 16 19 19 17 19

5 11 18 22 20 23

17 21 20 17 13 14

Occupation Unskilled Skilled or semi-skilled White collar/clerical Technical Managerial Professional

11 28 13 16 10 22

10 35 11 15 11 18

10 35 11 16 10 18

12 44 10 14 8 13

Sector Work for government Public corporation Private firm Self-employed

24 6 44 26

22 7 49 22

22 12 47 20

18 10 56 17

Education Non-graduate Graduate

77 24

82 18

80 20

87 13

Marital status Married Widowed Divorced/separated Single

63 6 11 21

63 6 9 22

66 9 10 15

56 8 7 29

Overall, the evidence in Table 6.1 suggests that party members and activists are higher status individuals, older, and more likely to be male than the rest of the population. Given these differences, one might expect that members and activists differ from the population with respect to their attitudes and beliefs. I consider this issue next.

136 Activating the Citizen

What do they think? Since the earliest research into ‘civic culture’, it has been well understood that democracies rely on citizens having certain attitudes and beliefs that support democratic institutions (Almond and Verba, 1963). Underpinning democracy is a set of norms, values, and practices that bind society together, make democratic government possible, and help individuals to solve collective action problems (Pattie, Seyd, and Whiteley, 2004). These values and practices are rooted in an implicit bargain between the individual and the state, in which citizens claim rights and protections in exchange for accepting various obligations. For example, citizens may demand that the state provide a decent standard of living for all, but in exchange they have to be willing to pay their taxes. If they demand rights without accepting obligations, the system will not be able to deliver and problems of governance will arise. Successful democratic government relies on this implicit bargain. Equally, democracy relies on norms of tolerance, particularly in relation to free speech and freedom of assembly, since it is rooted in the idea that differences can be settled by deliberation rather than force. This raises an interesting question about the extent to which party members contribute to the norms and values that support democracy. In the survey, an analysis of perceptions of eight citizen duties, such as voting in elections, paying taxes, obeying the law, and undertaking voluntary work, shows that party members attached more importance to them than ex-members and non-members. Furthermore, party activists were more likely than members to think these duties were important. This pattern can be seen most clearly in perceptions of the duty to vote. Some 76 per cent of activists attached the highest importance to this duty, compared with 67 per cent of members, 57 per cent of ex-members, and only 49 per cent of non-members. Similar differences existed in relation to perceptions of the duties to keep a watch on the government, help the less privileged in society, and be an active volunteer. Being involved in a political party clearly makes people more aware of their civic duties. Just as citizens should recognize their duties in a democracy, they are entitled to demand their rights from the state. Indeed, an effective performance by the state in delivering various rights is a prerequisite for democracy to work properly. The rights of citizens to participate, to be consulted, to be treated equally, and to have a decent standard of living are crucial in this regard. When one examines variations in attitudes to rights, again there are striking differences between the four groups.

Paul Whiteley 137

In general, party activists attached more importance to rights than members, who in turn thought rights were more important than ex-members and non-members did. This gap is particularly marked in relation to the right to participate, where 66 per cent of activists attached the highest importance to this compared with 58 per cent of members, 56 per cent of ex-members, and 54 per cent of non-members. A similar point can be made about the right to be treated equally by governments. Some 81 per cent of activists attached the highest importance to this compared with 79 per cent of members, 77 per cent of ex-members, and 73 per cent of non-members. It is clear that getting involved in a political party makes people more aware of their rights as well as their obligations as citizens. Another key requirement of effective democracy is tolerance of those with a different point of view. This is easily extended to other citizens, but difficulties arise in relation to tolerance of unpopular minorities such as religious extremists and racists. Again party membership and activism is correlated with political tolerance. Some 44 per cent of activists thought that religious extremists should be allowed to hold public meetings, compared with 42 per cent of members, 31 per cent of ex-members, and only 28 per cent of non-members. A similar pattern exists in relation to toleration of racists. Some 27 per cent of activists thought that racists should be allowed to hold public meetings, compared with 22 per cent of members and ex-members and only 17 per cent of non-members. In each case, party members were more tolerant than non-members and activists were more tolerant still. There was less tolerance of groups wishing to overthrow democracy but nonetheless party activists and members were more tolerant than their fellow citizens. Overall, we can see that party members and activists have more of the attitudes that help to support democracy than is true of their fellow citizens. They are more tolerant and more acknowledging of their civic duties, while at the same time more demanding of their rights as citizens. In light of this discussion, it is interesting to examine their political actions. Do party members participate more than their fellow citizens? I consider this next.

What do they do? In earlier work on party membership in Britain, three different aspects of party activity were identified as being particularly important (Seyd and Whiteley, 1992; Whiteley, Seyd, and Richardson, 1994; Whiteley and Seyd, 2002). The first relates to contact between political elites and party

138 Activating the Citizen

members, both within the party organization and in the wider political system. Contact between party members and other people, whether on a face-to-face basis or via the media, is an important aspect of political participation (Verba and Nie, 1972; Barnes and Kaase, 1979). This is the way in which political elites interact with the mass of citizens and in which much democratic deliberation takes place. Recent research suggests that face-to-face contact in voluntary organizations is an important source of social capital, which helps support democracy (Putnam, 1993, 2000). The second important function of party members relates to campaigning. The essence of campaigning is the attempt by citizens to influence policy making at the local and national levels. This involves activities such as running election campaigns, donating money to parties and other organizations, participating in demonstrations, and persuading others to vote for a given party or to agree with a certain political point of view. It is now well established that local campaigns influence both turnout and voting behaviour in elections, so such campaigning has a key role in the political life of democracies (Denver and Hands, 1997; Johnston and Pattie, 1997; Clarke, Sanders, Stewart, and Whiteley, 2004). The third dimension of party activism relates to representation, that is, acting as a party representative either within the party organization or as an officeholder in the wider society. When party members win local elections, become officeholders in their local party branches, or accept official positions in the community such as justice of the peace, they are fulfilling this function. As long as electoral politics is dominated by parties, this is a key mechanism through which democratic governments find the personnel to fill the many offices of the state. These types of activities are classic examples of high-intensity participation (Whiteley and Seyd, 2002). It is possible to investigate the contact and campaigning dimensions of party activism, but not the representation dimension since the surveys lack indicators of this aspect. The differences between activists and the other groups were huge when it comes to contacts. Thus, 44 per cent of activists had contacted a politician in the previous year, whereas only 16 per cent of members, 13 per cent of ex-members, and 5 per cent of non-members had done so. Similar patterns exist for contacting the media and participating in Internet political forums. Activists were nearly seven times more likely to contact the media than nonmembers and about four times more likely than ex-members. In relation to face-to-face discussions about politics with other people, activists were about five times more likely do this often than non-members and about twice as likely as members and ex-members. Clearly, one of the key components of party activism is being an ‘ambassador in the community’

Paul Whiteley 139

(Scarrow, 1996), that is, a person who gives both a voice and a face to the party within the wider community. There are six indicators of the campaign dimension of party activity in the surveys. In each case the earlier pattern persists, with members and ex-members being more active than non-members, and activists participating most. In the case of signing petitions, for example, activists were about twice as involved as non-members, a rather similar ratio as compared to boycotting products for political or ethical reasons. However, when it comes to participating in demonstrations, activists were five times more likely to do this than non-members, and more than twice as likely than members. Perhaps not surprisingly, the largest gap between activists and the other groups relates to attending political meetings; only one in 20 non-members reported doing this compared with more than half of the activists. Activists were also quite involved in raising money for political organizations and donating money to political parties and political campaigns. This evidence testifies to differences between party activists, party members, ex-members, and citizens in general. There are important differences in their attitudes and their behaviour with respect to the civic cultures of their societies. Clearly party members are much more engaged than non-members and activists are the most engaged of all. However, as the earlier discussion indicated, only about a fifth of the population currently or formerly considered that they belonged to a political party. While they may participate more than citizens in general, this might make little difference to overall levels of public engagement, because these groups are relatively small in number. Thus it is important to examine how much of a difference party members make to civic values and participation in the wider society. This is examined next.

What difference do they make? One way of addressing the question of how much difference party members and activists make to the civic life of democracies is to repeat the earlier analysis, but from a different perspective. Instead of asking, ‘How do members differ from non-members?’ we can ask, ‘How much of a contribution do members make to the civic life of these democracies?’ We know that activists and members make up just over 10 per cent of citizens, so it is interesting to ask what percentage of the citizens who assign great importance to duties and rights and who participate in politics are party members. In this way, we can see the extent to which they contribute to the overall civic culture.

140 Activating the Citizen

First, considering perceptions of duties, activists and members have a modest impact on overall perceptions in their countries. Party members or activists made up 15 per cent of the people who thought that a citizen should keep watch on the government, 21 per cent of those who thought that a citizen should volunteer, and 14 per cent of those who thought that voting was a very important duty. Thus, party members and activists were over-represented in these categories, but their overall contribution to civic values was not large. If the exercise is repeated for perceptions of rights, the results are rather different. In this case, party members and activists were not that much different from citizens in general. When it comes to perceptions of the right to an adequate standard of living, the right to be treated equally, and the right to participate in politics, party members and activists were very similar in their attitudes to the rest of the population. Table 6.2 looks at the contribution of party activists and members to rates of political participation. The story here is very different from that of civic values. As the table shows, activists, members, and exmembers made a much bigger contribution to political participation than their numbers in the population would predict. They did almost half of all the contacting of politicians and more than 40 per cent of contacting the media. They were surprisingly well represented on the Internet, and they played a big role in supporting demonstrations, signing petitions, and attending political meetings. This pattern extends to face-to-face political discussions, with some 40 per cent of the individuals who often participated in political discussions being activists, members, or ex-members. Similarly, just under 40 per cent of respondents who often tried to persuade their friends of a political point of view were activists, members, or ex-members. Clearly, if we removed the voluntary party organizations from the scene in these countries, then rates of political participation would fall quite dramatically. The evidence in these tables is interesting, but if we wish to get an overall assessment of the contribution of party members to the civic life of these countries a multivariate analysis is required. It is important to take into account other factors known to influence participation if the role of party members is to be properly assessed. The strategy will be to model participation with the help of two rival theories in order to control for confounding factors, while at the same time investigating the impact of party activism and membership on participation. In this way, it is possible to identify the unique contribution that political parties make to civic life in these contemporary democracies.

Paul Whiteley 141 Table 6.2: The contribution of party members and activists to civic engagement (row percentages) Percentages involved in contacting and campaigning over past years

Contacted a politician Contacted the media Joined an Internet political forum Signed a petition Boycott products Taken part in demonstration Donated or raised money for a party Attended a political meeting

Belong and participate

Belong, not participate

Used to belong

Never belonged

18 17 16

14 13 10

13 12 12

54 58 62

7 6 13 8

11 9 12 11

10 10 10 10

72 75 66 71

23

18

11

48

Note: The first cell shows that 18% of people who contacted a politician in the previous year were party activists.

Percentages involved in political discussion Often Sometimes Rarely Never

12 3 2 1

13 9 6 4

5 11 7 4

60 78 86 91

Note: The first cell shows that 12% of people who were often involved in political discussions were party activists.

Percentages involved in trying to convince friends of a point of view Often Sometimes Rarely Never

13 5 2 2

12 10 7 5

14 11 8 6

61 75 83 88

Note: The first cell shows that 13% of people who often tried to convince friends of their point of view were party activists.

Modelling the role of party members in explaining participation The first stage of the analysis is to obtain a summary measure of participation that can be used as a dependent variable in the multivariate model. Accordingly, a principal components analysis of the first eight

142 Activating the Citizen

indicators of participation in Table 6.2 was undertaken, and it produced a single overall participation scale.1 This scale is used as the dependent variable in the subsequent analysis. The purpose of the analysis is to determine whether party membership or activism is a good predictor of political participation when other explanations offered in the literature are controlled for. The method of analysis is ordinary least squares regression using the 2004 ISSP data. Party membership and activism are included among the independent variables (as dummy variables, since they are bivariate), along with standard socio-demographic measures and, more importantly, variables associated with two important theories of political participation, namely the cognitive engagement model and the social capital model. In a previous publication, I assessed the strength of these theories in explaining general levels of citizen participation in countries covered by the 2004 ISSP data (Whiteley, 2008); here I extend that analysis by focusing on the particular impact of party membership on political participation. The ability and willingness of citizens to use political information to make informed choices underpins the cognitive engagement theory (Norris, 1999; Dalton and Wattenberg, 2000; Clarke et al., 2004; Dalton, 2004). Education, especially higher education, is particularly important to this theory, since it provides citizens with the skills needed to process large amounts of information. This theory also suggests that expanded, low-cost access to electronic and web-based information sources produces a process of cognitive mobilization in society (Barnes and Kaase, 1979), which should facilitate political participation. Therefore, in relation to this theoretical approach, we include in our model several independent variables that the theory would posit as likely predictors of higher levels of political participation; these are educational attainment (including graduate status), interest in politics, political efficacy, discussion of politics, and, most importantly, media consumption of politics and current affairs (see Appendix 6.1 for details). Overall, if the cognitive mobilization theory is correct, we would expect highly educated, efficacious, interested, and engaged citizens who absorb a lot of political information to be much more politically active than their counterparts in the wider population. To be sure, one would expect party members and activists to exhibit more of these characteristics than the broader population, but with the regression analysis I seek to determine whether party membership or activism in and of itself, apart from the variables associated with the cognitive engagement model, has an independent impact on political participation. In contrast to the cognitive engagement model, which emphasizes individual choice, the social capital model (Coleman, 1990) sees political

Paul Whiteley 143

participation as resulting from a process of socialization involving norms, values, and practices that come from the family and larger social environment. Social capital theory can be considered a structural-based theory, while the cognitive mobilization model can be depicted as a choice-based theory (Pattie et al., 2004). For most writers who deal with social capital, interpersonal trust is a key indicator (Putnam, 1993; Fukuyama, 1995; van Deth, Maraffi, Newton, and Whiteley, 1999). Trust is important because it allows individuals to move beyond their immediate family or community and engage in cooperative activities, including political ones, with strangers. The types of variables that are important for this model are the degree of voluntary activity in organizations, interpersonal trust, and, in the case of political participation, trust in government. These measures are included as independent variables in the multivariate analysis. In addition, some socio-demographic variables (marital status and self-assigned social status) are included as components of the social capital model, the assumption being that high-status individuals with ties to the community will be more likely to participate as a consequence. Some of the variables associated with the social capital model may themselves be associated with differences between party members and non-members in their rates of political participation, but I am primarily interested here in whether party involvement has an independent impact on participation apart from the social capital variables. In addition, the model includes a number of social background variables to act as controls such as gender and age.2 Finally, I include three party-related variables.3 These identify if an individual is a party activist, party member, or ex-member and they are included in order to evaluate the impact of party-related variables on participation. If the party variables remain statistically significant predictors in a model that controls for measures of cognitive engagement and social capital, then it suggests that party involvement has more of an impact on participation than can be explained by the variables in the two models. In addition to examining the influence of individual-level variables on participation, it is important to take into account the contextual effects of the party measures in the different countries. As Figure 6.1 shows, there are wide variations in party membership across the set of countries. It is likely that a country with a relatively large number of party members will be a more congenial environment for participation anyway. If it is easy for a citizen to join a political party because there are many party members around, this is likely to give a boost to the impact of party membership and activism on participation. The ‘opportunity structures’ (Tarrow, 1994) or the institutional environment will favour the recruitment and retention of party members in countries where parties

144 Activating the Citizen Table 6.3: The multi-level model of participation with party activism and membership variables Aggregate level variables Percentage of activists Percentage of party members Percentage of ex-members R-squared

−0.09 0.19*** −0.08* 0.50

Individual level variables Activist dummy variable Member dummy variable Ex-member dummy variable Personal efficacy Interest in politics Discuss politics Media consumption of politics Trust in government Trust in other people Voluntary activity Male Age (older) Age squared Married Educational attainment (higher) Graduate status Self-assigned social status (higher) R-squared

0.14*** 0.09*** 0.09*** 0.09*** 0.13*** 0.16*** 0.12*** −0.02** 0.05*** 0.14*** 0.01 0.06* −0.20*** −0.02*** 0.10*** 0.04*** 0.02*** 0.38

* = significant at the .10 level; ** = significant at the .05 level; *** = significant at the .01 level

have healthy grassroots organizations and this will serve to stimulate wider participation in society. Equally, in countries where party members are few in number, such opportunity structures will serve to inhibit the recruitment and retention of new members and reduce the impact of party volunteers on participation. These contextual effects can be identified with a multi-level modelling strategy (see Hox, 2002; Raudenbusch and Byrk, 2002). In this case, aggregate data on party membership in each country are used to predict individual-level participation in politics, alongside the individual-level cognitive engagement and social capital measures. The results of this modelling exercise appear in Table 6.3. Table 6.3 contains standardized regression coefficients of the multilevel random intercepts model of participation.4 The aggregate- or country-level effects appear first in the table and they explain about half of the variance in participation across the countries. The estimates confirm that the proportion of party members has a positive impact on

Paul Whiteley 145

individual-level participation in these countries. A political environment with many party members will be a congenial one for stimulating individual-level participation. It appears that the number of members rather than the number of activists is the key contextual variable, since the latter has no significant effect on participation. The percentage of ex-members clearly inhibits participation to a weak extent, indicating that in countries where the voluntary party organizations are declining fastest, participation will be undermined as a consequence. This is true, for example, in Canada, where there are many people saying they used to belong to a political party. Turning next to the individual-level predictors, all the cognitive engagement and social capital variables have statistically significant impacts on participation. Political efficacy, interest in politics, discussion of politics with others, and media consumption of politics all have positive impacts on participation. The sizes of the coefficients suggest that discussion of politics and interest in politics are particularly important in this regard. Interpersonal trust and voluntary activity from the social capital model stimulate participation also, although trust in government inhibits participation. The latter suggests that individuals who do not trust government are more likely to get involved in politics than individuals who do. The demographic variables show that gender has no effect on participation but age does have an effect, with older people being more likely to participate than younger people, but the effect declines with age. This is consistent with party activists retiring from politics at some point in their lives. Highly educated individuals and graduates are more likely to participate than the uneducated, and self-assigned social status has a positive impact on participation. Finally, married individuals are less likely to participate than single people, which may reflect their greater family responsibilities. In the latter case the multivariate findings are different from Table 6.1. An important finding is that all of the individual-level party variables have positive and significant impacts on participation. This means that party membership and activism has a larger impact on participation than can be explained by the cognitive engagement and social capital models. It is interesting to note that ex-members are more likely to participate than non-members, even though in the country-level model this effect had the opposite sign. This is not a contradiction, since the individual-level and aggregate-level effects can work at cross purposes. In the individual-level model the mechanism works through individuals’ political experience and characteristics, whereas in the aggregate model

146 Activating the Citizen

it works through the context in which individuals find themselves. Thus, having been a party member at some point leaves a residue of a willingness to participate, which continues to exist even after an individual has left a party. At the same time, a political environment that contains many ex-members will serve to undermine individual participation because it restricts the opportunity structures that facilitate participation.

Conclusions and discussion The evidence in this chapter suggests that party activists and party members play a major role in supporting political participation across the democratic world. If they were absent, rates of participation would be significantly lower than they are at present in these countries. This means that if voluntary party activity is in long-term decline, something which is certainly true in Europe, this will have implications for participation throughout the democratic world. The links between political elites and the mass public will be increasingly attenuated, since party volunteers play a very important role in supporting these links. Activities such as political meetings and demonstrations will be reduced in scope and volume as voluntary party activity declines. Thus, the implications of declining memberships and activism rates go well beyond political parties themselves. As suggested earlier, many political leaders are concerned by this development and are looking to stimulate voluntary activity at the grassroots level with various initiatives. If the major parties become, in effect, public utilities lacking volunteers, this has significant implications for the recruitment and socialization of political leaders across the democratic world. These important functions will increasingly become divorced from the community and become wholly professionalized. In this situation, politicians are more likely to be seen by the ordinary citizen as state bureaucrats than as local elected representatives. Elections and political campaigns will continue, but it is questionable how effective their messages will be if they become indistinguishable from commercial advertising. They will lack the endorsement of ‘ambassadors in the community’. In the two-step model of political communication, citizens respond to messages from political leaders only if they are endorsed by someone they know or trust (Popkin, 1991). If ordinary citizens think of a political party as being roughly equivalent to the power company, it is unlikely that the party’s message will carry much weight. In short, if party members and activists disappear from the scene, civil society will be greatly weakened and democracy undermined.

147 Appendix 6.1: Definition of variables in the multi-level model Variable

Definition

Participation

Factor scores from the principal components analysis of the first eight items in Table 6.2

Efficacy scale

Sum of responses to three statements: 1. ‘People like me don’t have any say about what the government does.’ 2. ‘I don’t think the government cares much what people like me think.’ 3. ‘I think most people in [my country] are better informed about politics and government than I am.’ (coded so that a high score denotes high levels of efficacy)

Discuss politics

Responses to question, ‘When you get together with your friends, relatives, or fellow workers, how often do you discuss politics?’ (coded so that a high score denotes high levels of discussion)

Interest in politics

Responses to question, ‘How interested would you say you personally are in politics?’ (coded so that a high score denotes high levels of interest)

Media usage

Sum of responses to three statements: 1. ‘How often do you read newspapers in the average week for information about politics and current affairs?’ 2. ‘How often do you watch television in the average week for information about politics and current affairs?’ 3. ‘How often do you listen to the radio in the average week for information about politics and current affairs?’ (coded so that a high score denotes high levels of media consumption)

Trust in government

Sum of responses to two statements: 1. ‘Most of the time we can trust people in government to do what is right.’ 2. ‘Most politicians are in politics only for what they can get out of it personally.’ (coded so that a high score means high trust in government)

Trust in other people

Sum of responses to two statements: 1. ‘Generally speaking, would you say that people can be trusted or that you can’t be too careful in dealing with people?’ 2. ‘How often do you think that people would try to take advantage of you if they had the chance, and how often would they try to be fair?’ (coded so that a high response means high interpersonal trust) (Continued)

148 Activating the Citizen Appendix 6.1: Continued Variable

Definition

Voluntary activity

A positive response to the following question: ‘People sometimes belong to different kinds of groups or associations. Please indicate whether you belong and actively participate, belong but don’t actively participate, used to belong but do not anymore, or have never belonged to a voluntary association.’ (excludes political parties as there were separate questions about them)

Note: The aggregate-level variables represent the average percentage of party activists, party members, and ex-members in each country.

Notes 1. The principal component had an eigen value of 3.6, which explained 44.5 per cent of the variance in the data. All of the loadings of the variables on this principal component were greater than 0.60, with the exception of Internet participation, which had a loading of 0.58. 2. Age is included in the model with a quadratic specification. The assumption behind this is that participation increases with age but at a declining rate of change. 3. A dummy variable scores one if respondents have the attribute and zero otherwise. 4. The aggregate-level variables are allowed to influence the intercept of the individual level model. A positive effect means that individual level participation shifts upwards in a particular country in response to the stimulus provided by party activity in that country. For a further description of this type of model see Raudenbush and Byrk (2002).

References Almond, G. A. and S. Verba (1963) The Civic Culture: Political Attitudes and Democracy in Five Nations (Princeton, NJ: Princeton University Press). Barnes, S. and M. Kaase (1979) Political Action (London: Sage). Biezen, I. van (2004) ‘Political Parties as Public Utilities’, Party Politics, 10, 701–22. Clarke, H. D., D. Sanders, M. C. Stewart, and P. Whiteley (2004) Political Choice in Britain (Oxford: Oxford University Press). Coleman, J. (1990) Foundations of Social Theory (Cambridge, MA: Harvard University Press). Dalton, R. J. (2004) Democratic Challenges and Democratic Choices (Oxford: Oxford University Press). Dalton, R. J. and M. P. Wattenberg (2000) Parties without Partisans: Political Change in Advanced Industrial Societies (Oxford: Oxford University Press).

Paul Whiteley 149 Denver, D. and G. Hands (1997) Modern Constituency Electioneering (London: Frank Cass). Deth, J. W. van, M. Maraffi, K. Newton, and P. F. Whiteley (eds) (1999) Social Capital and European Democracy (London: Routledge). Detterbeck, K. (2005) ‘Cartel Parties in Western Europe?’, Party Politics, 11(2), 173–91. Fukuyama, F. (1995) Trust: The Social Virtues and the Creation of Prosperity (London: Hamish Hamilton). Hox, J. (2002) Multilevel Analysis: Techniques and Applications (London: Lawrence Erlbaum Associates). Johnston, R. and C. Pattie (1997) ‘Where’s the Difference? Decomposing the Impact of Local Election Campaigns in Great Britain’, Electoral Studies, 16, 165–74. Katz, R., P. Mair, L. Bardi, L. Bille, K. Deschouwer, D. Farrell, et al. (1992) ‘The Membership of Political Parties in European Democracies, 1960–1990’, European Journal of Political Research, 22, 329–45. Katz, R. S. and P. Mair (1995) ‘Changing Models of Party Organization and Party Democracy: The Emergence of the Cartel Party’, Party Politics, 1, 5–28. Mair, P. and I. van Biezen (2001) ‘Party Membership in Twenty European Democracies, 1980–2000’, Party Politics, 7(1), 5–21. Norris, P. (ed.) (1999) Critical Citizens (Oxford: Oxford University Press). Pattie, C., P. Seyd, and P. Whiteley (2004) Citizenship in Britain: Values, Participation and Democracy (Cambridge: Cambridge University Press). Pennings P. and R. Hazan (2001) ‘Democratizing Candidate Selection: Causes and Consequences’, Party Politics, 7, 267–75. Popkin, S. (1991) The Reasoning Voter: Communication and Persuasion in Presidential Campaigns (Chicago: University of Chicago Press). Putnam, R. (1993) Making Democracy Work: Civic Traditions in Modern Italy (Princeton: Princeton University Press). Putnam, R. (2000) Bowling Alone (New York: Simon and Schuster). Rapoport, R. B. and W. J. Stone (2006) Three’s a Crowd: The Dynamic of Third Parties, Ross Perot and Republican Resurgence (Ann Arbor: University of Michigan Press). Raudenbusch, S. W. and A. S. Bryk (2002) Heirachical Linear Models: Applications and Data Analysis Methods, 2nd edn. (Thousand Oaks, CA: Sage). Scarrow, S. E. (1996) Parties and Their Members (Oxford: Oxford University Press). Scarrow, S. (2000) ‘Parties without Members? Party Organization in a Changing Electoral Environment’ in R. J. Dalton and M. P. Wattenberg (eds) Parties without Partisans: Political Change in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 79–101. Scarrow, S., P. Webb, and D. M. Farrell (2000) ‘From Social Integration to Electoral Contestation: The Changing Distribution of Power within Political Parties’ in R. J. Dalton and M. P. Wattenberg (eds) Parties without Partisans: Political Change in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 129–56. Seyd, P. and P. Whiteley (1992) Labour’s Grassroots: The Politics of Party Membership (Oxford: Oxford University Press). Seyd, P. and P. Whiteley (2002) New Labour’s Grassroots: The Transformation of the Labour Party Membership (Basingstoke: Palgrave Macmillan).

150 Activating the Citizen Tarrow, S. (1994) Power in Movement, Social Movements, Collective Action and Politics (Cambridge: Cambridge University Press). Verba, S. and N. H. Nie (1972) Participation in America (Chicago: University of Chicago Press). Webb, P. (2002) ‘Political Parties in Advanced Industrial Democracies’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press). Whiteley, P. F. (1999) ‘The Origins of Social Capital’ in J. W. Van Deth, M. Maraffi, K. Newton, and P. F. Whiteley (eds) Social Capital and European Democracy (London: Routledge), pp. 25–44. Whiteley, P. F. (2008) ‘What Makes a Good Citizen? Citizenship Across the Democratic World’ in (eds) A. Park, J. Curtice, K. Thomson, M. Phillips, M. C. Johnson, and E. Cleary British Social Attitudes, The 24th Report (London: Sage), pp. 173–201. Whiteley, P. F., P. Seyd, and J. Richardson (1994) True Blues: The Politics of Conservative Party Membership (Oxford: Oxford University Press). Whiteley, P. and P. Seyd (2002) High Intensity Participation–The Dynamics of Party Activism in Britain (Ann Arbor: University of Michigan Press).

7 Models of Party Organization and Europarties Luciano Bardi and Enrico Calossi

This chapter examines the development of political parties at the European level (henceforth referred to as Europarties) in relation to models developed to understand party organizations at the national level, with a particular focus on how such models affect the ability of parties to perform democratic functions (that is, linking citizens to policy-making institutions). Specifically, we argue that Europarties, in their current form, were created as an organizational response to changing systemic and societal conditions at the national and European levels. We conclude that Europarties appear to be elements and/or manifestations of the organizational development of political parties in European democracies’ multi-level political systems rather than organizations specifically created to perform democratic functions at the European Union (EU) level. In this sense, they exhibit deficiencies that are even more pronounced than those of contemporary national parties of the ‘cartel type’ (which are predominant in Europe and arguably in Canada), and thus fail to offer a realistic solution to Europe’s democratic deficit.

Europeanization and Europarty development Europarties have long been considered as potentially important actors in the European Community’s (EC) and subsequently the EU’s development. Roughly at the time the Treaty of Rome became effective, it was Haas (1958) who first suggested that the growth of Europarties provided an essential analytical focus for an assessment of the EC’s democratic development. Scholarly interest in the potential role of EU-specific parties continued and increased after the European Parliament’s (EP) first universal suffrage elections were held in 1979 (Marquand, 1979; Lodge and Herman, 1982; Kirchner, 1985). This interest was based on the 151

152 Activating the Citizen

simple consideration that elections (and political parties, as one of their essential components) are the necessary prerequisite for the development of a democratic system. Most authors limited themselves to studying the structure and functions of party groups in the European Parliament, that is, the party structures directly and visibly produced by EP elections. This focus has resulted so far in the neglect of other important party components, such as central and territorial organizational structures. An absence of attention to these elements in empirical studies of Europarties is hardly a surprise, given that such structures are, for all practical purposes, non-existent. However, scholars have also failed to consider the normative implications of the failure of Europarties to perform vital systemic functions that are critical to democratic governance. This limit has been made more serious and visible by the momentous events that have characterized EU development in the last 20 years, since the Single European Act. The 1987 Single European Act, and at the same time the setting of the date, 31 December 1992, for the completion of the EC’s internal market, effectively started the phenomenon that has come to be known as Europeanization. Technically the progressive encroachment of Europeanlevel decisions on portions of EC/EU member states’ sovereignty began in the early 1950s with the establishment of the European Coal and Steel Community. But for a very long time, the EC/EU’s perceived impact on the life of European citizens was so limited that their ‘permissive consensus’, regularly revealed by opinion polls, was deemed a sufficient source of legitimacy for the continuing process of European integration. It was not until the formal completion of the internal market and the accelerations provided by the subsequent Maastricht and Amsterdam Treaties that European citizens, after having unsupportively enjoyed for many years the vaunted benefits of integration, became aware of its possible costs. Perceptions of the costs of integration concerned not only economic interest groups, such as entrepreneurs (who felt burdened by stricter regulations as well as the removal of protectionist advantages), but also ordinary citizens. Ordinary citizens generally came to blame Europe for increases in taxation and inflation (the latter was especially true in those countries that adopted the Euro as a common currency), job-market deregulation and the ensuing decrease in job security, and, generally, all immigration-related problems. This last set of complaints was greatly increased by the 2004 and 2007 EU enlargements, which added 12 new member states, mostly from the former Communist Bloc. Be it as it may, it can be argued that in the course of the last two decades Europeanization has enormously increased European citizens’

Luciano Bardi and Enrico Calossi

153

demands for more democracy at the EU level to allow their concerns to be addressed. The post-1987 developments rekindled Europeanists’ hopes that EP groups and, more generally, Europarties would be called upon to play a bigger role in EU democratization – in more technical terms, that they would begin to effectively perform important functions, at least those generically grouped under the ‘linkage’ label, in the merging political systems of the EU. It would however be inaccurate to argue that Europeanization had the direct effect of generating public pressure for the strengthening of existing supranational party structures for the purpose of obtaining better citizen representation at the EU level. It is in fact doubtful that many EU citizens were even aware of the existence, let alone the potential for representation, of Europarties. In general, demands for more EU democracy were requests for increases in the powers of the EP or for the introduction of more elements of direct democracy, such as referenda or universal suffrage elections of top EU officials. Many such proposals were voiced as part of the debate leading to the ill-fated constitutional treaty and of the campaigns in those countries where referenda were required for its ratification. Europarties however have come a long way since 1958 and even since 1979. The approval of European Council regulation 2004/2003 on governing political parties at the European level and the rules regarding their funding is a clear demonstration of this fact. The importance of Europarties in shaping a European consciousness and expressing the political will of EU citizens had been recognized in principle in the treaties (for example, Article 191 of Treaty Establishing the European Community), but the new regulation for the first time provided the legal and financial basis for the effective establishment of integrated political parties at the EU level. Although we are still far from having reached any significant form of EU-level ‘party government’, with European Council regulation 2004/2003 the necessary foundations for party-based representation were laid. The statute for European political parties, as the regulation is ordinarily referred to, is a concise document that defines the role of European political parties and the requirements needed to be able to receive funding from the EU. Much space is dedicated to the aspects directly linked to financing, perhaps also because the statute was, in part, justified by the need to use public funds to cover the costs of democratic promotion in the newly entering countries. The statute’s provisions may well facilitate the more effective consolidation of aspects important to Europarty development, such as transnational federations,

154 Activating the Citizen

parliamentary groups, and national parties. In fact, even if the statute practically identifies Europarties with federations, the provisions for their constitutions and for their access to financing create a link with parliamentary groups and national parties. This could be the most important result of the statute from our viewpoint, as the lack of integration among the various components of Europarties has always been considered one of the main reasons for their weakness. Moreover, when announced, the statute was also seen as an important demonstration of national parties’ willingness to forfeit at least part of their primacy at the EU level, as their role was crucial for the approval of the statute in the EU Council of Ministers. Unfortunately, however, due to some provisions in the statute that effectively give EP groups control over Europarty financing and even more importantly make national parties indispensable because of their role in co-financing, Europarty federations remain the weakest of the three components; this effectively maintains national parties as dominant party actors at the European level. This is probably responsible for the continuing inability of Europarties to perform the vital democratic functions that will be discussed in the next sections.

Models of parties and Europarties Models developed by political scientists to analyse the evolution of party organizations at the national level are useful for understanding the functions fulfilled by Europarties. These models draw attention to parties’ organizational characteristics as well as to the manner in which they carry out key functions, including several involving linkages between citizens and political policy-making institutions. In considering basic notions of party functions (King, 1969) we draw on conceptions of parties (mass and to an extent catch-all) that were developed for analysing specific societal and political contexts. For example, mass parties were important in integrating previously disenfranchised classes and otherwise defined societal groups. But parties, although created for the pursuit of the political aims of given societal groups, in the process came to perform functions that were of fundamental importance for the political system as well. In some cases, organizational survival became parties’ main institutional goal in that they adapted to societal transformations so as to strengthen themselves as organizations. But this did not necessarily enhance their ability to perform larger functions that were important for the overall political system. The integration function was the first one to fade, as citizens became more educated and more or less permanently politicized. Parties responded by trying to broaden their electoral

Luciano Bardi and Enrico Calossi

155

appeal beyond the classe gardée, as electoral success had become a crucial objective in a very competitive political market. As illustrated by the catch-all party model, individual members were replaced by interest groups as instruments for grounding parties in society. Later on, as electors became capable of expressing their views and conveying their demands through different means, parties became even less necessary or, according to some, redundant. But by this time they had, in many cases, become powerful organizations that occupied crucial strategic positions in the political system; this allowed them to develop very effective collective strategies to survive and prosper. This last phase was conceptualised in the cartel party model (Katz and Mair, 1995). As parties underwent the transformations that marked the passage from one type to the next they changed their organizational characteristics in ways that reflected the transition from their political goal-oriented origins to their later organizational survival character. Europarties originated and developed entirely in the course of this last phase and the idea that they might be considered as manifestations of the strategies of (national) cartel parties adapting to a changing institutional environment rather than distinct organizations operating in a different (supranational) political system seems plausible. However, given their distinctive origins and institutional context, one should, at least for analytical purposes, consider Europarties as a distinctive phenomenon, rather than just a sub-type of the cartel party. This permits us to assess the manner in which Europarties carry out key functions, by comparing them to those of classic models of party systems. Should it turn out that they have an important role at the EU level, their distinctiveness will be vindicated. Classic party models have been classified and discussed in terms of their genesis, development, and general organizational characteristics. With the approval of the Europarty statute, enough elements exist for the definition of the main organizational characteristics of Europarties. This could in turn permit a partial, but for our purposes sufficient, description of a Europarty model. A good starting point for the identification of the relevant components of the model is Katz and Mair’s (1995) article on the cartel party, which lists a number of party model characteristics. Some of them describe environmental conditions that favour a given party model. The rest, on the other hand, are bona fide organizational components. As we discuss below, our analysis of functions will concentrate on what in general have been termed linkage functions, that is, those organizational characteristics that involve the party’s ability to connect ‘the governed to the governors’ (King, 1969, p. 117). In particular, we will consider Europarties’ ability to encourage

156 Activating the Citizen

citizen participation, interest articulation and aggregation, and vote structuring. Of the 13 characteristics identified by Katz and Mair (1995), the following, albeit somewhat overlapping at times, seem to be relevant for our concerns: 1) principal source of party resources, 2) character of membership, 3) relations between ordinary members and party elites, 4) nature of party work and party campaigning, 5) party channels of communication, and 6) party position between civil society and state. Table 7.1 reproduces brief descriptions of such characteristics for each major party model, from the elite to the cartel party, as formulated by Katz and Mair (1995). We have extended this framework to include the Europarty model. A cursory look at the table permits two preliminary general observations: 1) the Europarty model’s characteristics resemble more closely those of the cartel party than of any other model; 2) in this and in its additional peculiarities, the Europarty model seems poorly suited to the performance of so-called democratic functions. This latter aspect is indicated by the fact that the Europarty seems to be even more removed from civil society (the entity it should help link to the state) than the cartel party, whose main definitional character is to be largely detached from society. This general impression is confirmed by an examination of differences between the various party models. We take as the starting point the mass party1 because it appears to be best suited for the performance of so-called linkage functions: mass parties ‘were the first that explicitly claimed to represent the interests of only one segment of society’ (Katz and Mair, 1995, p. 10). This was, arguably, also their primary goal, so they developed organizational structures that were particularly apt for the performance of linkage functions. Katz and Mair describe mass parties as ‘bridges’ between civil society and the state, but an examination of the individual organizational components makes clear that the mass party was rooted in society in distinctive ways. For example, concerning party financing, mass parties drew most of their resources from members’ fees and donations. Even if the latter were often in the form of free labour rather than cash, these contributions defrayed costs that would have otherwise required substantial cash outlays. This feature of mass parties established a strong linkage between parties and society. The character of membership, Katz and Mair’s second category, was thus defined by the size and homogeneity of the membership, which confirmed the close identification with a specific group or class. Although we know that the bureaucratization of party structures eventually generated oligarchic tendencies (Michels, 1915/1959), the mass party’s legitimizing myth was

Table 7.1: Party models and their characteristics Elite party

Mass party

Catch-all party

Cartel party

Europarty

Time period

19th century

1880–1960

1945–

1970–

1979–

Principal source of party resources

Personal contacts

Members’ fees and contributions

Contributions from a wide variety of sources

State subventions

State subventions

Character of membership

Small and elitist

Large and homogenous; actively recruited and encapsulated; a logical consequence of identity; emphasis on rights and obligations

Open to all (heterogeneous) and encouraged; rights emphasized but not obligations; marginal to identity

Neither rights nor obligations are important (distinction between members and non-members blurred); emphasis on members as individuals rather than as an organized body; members valued for contribution to legitimizing myth

Collective; sum of members of statelevel member-parties; individual members often unaware of their Euro-level membership and therefore of any potential rights (or obligations)

Relations between ordinary members and party elite

The elite are the ordinary members

Bottom up: elite accountable to members

Top down: members are organized cheerleaders for elite

Stratarchy: mutual autonomy

Bottom up: elite accountable to national party elites

Nature of party work and party campaigning

Irrelevant

Labour intensive

Both labour intensive and capital intensive

Capital intensive

Capital intensive

Party channels of communication

Interpersonal networks

Party provides own channels of communication

Party competes for access to non-party channels of communication

Party gains privileged access to state-regulated channels of communication

Access to stateregulated channels of communication through (but more often in competition with) state-level member-parties

Position of party between civil society and the state

Unclear boundary between state and politically relevant civil society

Party belongs to civil society, initially as representative of the newly relevant segment of civil society

Party as competing broker between civil society and state

Party becomes part of state contact with civil society (only through state-level member-parties)

Party limited to state. No direct contact with civil society

Source: Adapted from Katz and Mair (1995).

158 Activating the Citizen

based on the albeit formal control of party elites by ordinary members, a feature also indicative of the emphasis on the link between party structures and the classe gardée. Moreover, the labour-intensive nature of party campaigning and the resort to direct membership action as the only channel of communication with the potential electorates made mass parties a constant and very visible presence in civil society. With the catch-all party, the situation was modified in that individual parties expanded their reach from selected specific societal groups to society at large. The link with society was at least partially maintained as interest groups replaced, for practical purposes, direct membership, and financial resources were obtained from a variety of sources. The more competitive political market created by this situation by necessity diluted the relationship between parties and civil society. In particular, as members were recruited from different societal groups, their sense of identification with the party and their militancy decreased; as a result, although the need to recruit members in a competitive situation induced parties to emphasize members’ rights more than their obligations, party elites became more and more independent (and distant). As party campaigning and work became more diversified in terms of targets and means, a progressive redundancy of militant membership followed. This tendency resulted also from the catch-all parties’ increasing reliance on diversified sources for party financing and channels of communications (various interest groups and sympathetic, albeit conditionally, press and other media). Katz and Mair (1995) sum up this evolution by defining catch-all parties as ‘competing brokers between civil society and the state’ (p. 18), implicitly still capable of providing a linkage between the two, even if in a less determined and exclusive fashion. In their actions to structure the relationship between society and the state, their emphasis is still on society, but their need to diversify their electorates makes it less focused. Cartel parties move away almost completely from civil society. Membership is no longer appreciated for its financial or active labour contributions but simply for its value in helping to maintain the party’s legitimizing myth. At the same time, the boundaries of membership become very permeable, with individuals easily and frequently shifting from card-holding to simple sympathizing and vice versa. With the transition to the cartel party model, state subventions, obtained through a joint legislative effort by all parties in the cartel, became the dominant source of party revenue. In fact, the reliance on state subventions for party financing and on access to state-regulated channels of communication, coupled with the overall capital-intensive nature of

Luciano Bardi and Enrico Calossi

159

party work and campaigning, all contribute to detaching the party elite from the so-called rank and file, as the party as a whole becomes progressively part and parcel of the state. The state is occupied by parties for reasons linked to their own survival. Their collective control of the state gives them all the resources they need. As this is cartel parties’ paramount, if not only, goal, inputs from civil society, if any, are avoided or ignored, and policy making has the main function of justifying the continuing party government character of democratic systems. Linkage is therefore reduced to ‘selling’ state outputs to the public. Europarties seem to take the cartel party model’s representational deficiencies to a new extreme. Historically, Europarties’ connections with society have been practically non-existent. EP party groups, the first European-level manifestations of parties, existed even prior to the first direct elections when the EP was composed of delegates of national parliaments. As such, they were at least twice removed from society, and the faint link afforded by the fact that national delegates were elected and theoretically bearers of instances coming from their electors was made irrelevant by the EP’s weak powers, which made perfunctory any possible intention to represent citizens. This situation was not improved even after direct elections, by successive increases in the EP’s powers, or by the approval of the Europarty statute. The biggest shortcoming of the statute in fact is that it does not address the issue of how to effectively link Europarties to European citizens, beyond the general statement that such linkage is the main reason for their existence. This function is still performed exclusively through the national parties, who therefore remain the principal gatekeepers of EU-level representation, albeit with the limits caused by the prevalence of cartel characteristics in their organizations. Aside from potential competition or comparison with national parties, Europarties’ total separation from civil society makes them unfit to perform linkage and other systemic functions. This is confirmed by an analysis of their particular organizational characteristics. State subventions are the only source of revenue, even more than for the cartel parties. Europarties have no membership of their own and even for legitimization purposes they have to rely on figures collectively provided by the national parties that make them up. Moreover, they are not even formally accountable to a sum of nationally recruited memberships but rather to the leaders of the national member parties, who select EP candidates and evaluate their performance on the basis of national party interests. Not only is party work, in this case, exclusively limited to campaigning and carried out through means (capital intensive and media oriented) that contribute to maintaining a distance

160 Activating the Citizen

between Europarties and society, but party activities also succumb to the second-order nature of European elections that turn them into occasions for testing the popularity of national political leaders. This means that if Europarties perform any functions at all in this phase, which is the one that brings them closest to European citizens, they do so on behalf of their national counterparts.

European parties and their functions at the national and European levels After describing a set of organizational characteristics of a Europarty model, we now consider implications in terms of citizen participation. Here we focus on those functions of political parties that Poguntke (2006) considers the most important, namely those that involve linking state and society, or, echoing King’s (1969) phrase, the ‘relationship in a democracy between governors and governed’ (pp. 117–18). Drawing on the already too numerous lists of party functions in the literature, we focus on three elements that relate to citizen involvement, namely the manner in which parties may help to structure the vote (and thus orient citizen participation), their role in interest articulation and aggregation, and their position as vehicles for more direct forms of citizen participation in politics (see Table 7.2). In this analysis we draw particularly on the work edited by Webb, Farrell, and Holliday (2002) regarding the roles performed by political parties in advanced industrial democracies. Structuring the vote The first function in which parties should play an important role or, we could say, the most important role, concerns everything that is related to elections. Much party communication and education is in some way related to what King (1969) defines as ‘structuring the vote’. In Epstein’s words, ‘structuring the vote is the minimum function of a political party’, a notion that King reiterates in stating that ‘a group or organization that did not attempt to structure the vote in its own favour would not normally be called a party’ (1969, p. 120). For this reason, even groups that do not like to call themselves parties (for example, the self-proclaimed ‘movements’ of the Italian Second Republic or the many populist antiparty and ‘anti-politics’ movements such as the Danish Anti-European Movement) should be considered parties. In most countries, at most times, the main electoral alignment is party alignment. Although King (1969) was thinking about mass parties, the change to catch-all parties may have led him to draw attention to how other

Table 7.2: Party functions related to participatory aspects Mass party

Catch-all party

Cartel party

Europarty

Structuring the vote: Monopoly of vote communication and structuring. Parties mass education provide own channels of communication. They are interested in education

The vote is structured by parties and other actors. Parties compete for access to non-party channels of communication

The vote is less structured by parties and more by other actors. Parties gain privileged access to state-regulated channels of communication

Structuring the vote is left to national parties or to single candidates. Education is performed by single candidates with public funds to create a European conscience. Communication is left to national parties

Articulation and aggregation of interests

Competitive and collaborative relation with interest groups in articulation of interests. Great capability of aggregation of different interests

Competitive and collaborative relation with interest groups in articulation of interests. Great capability of aggregation of different interests

No role in articulating and aggregating interests. Interest groups articulate the interests. European Commission aggregates the interests

Activists are useful to campaign for more voters

Growing rights for both members and voters in internal decisions. Rights are more related to formal aspects than the real party life

There is not direct citizen participation because there is not a European civil society as a whole

Collaborative articulation with specific interest groups. No interest in aggregation of interests

Citizen participation Party seeks participatory behaviour by the members. Party activists are preferred to voters Note: The elite party is not considered in this table.

161

162 Activating the Citizen

actors, such as single candidates, media, interest groups, or broader social movements, also have influence in structuring the vote. King noted the disjunction between the structure of alternatives that political parties present to the electorate and the demands of particular and often important groups of voters. With the increasing distance between party programme and electoral preferences, parties faced a choice: devote more resources to increase the effectiveness of communicating their programmes or adapt their programmes to what was going on in society. Catch-all parties generally inclined to the second option. Rather than maintaining a strict top-down control in structuring electoral issues, they adapted their programmes to changing popular sentiments, recognizing the role of other political actors in articulating popular interests. Cartel parties were also forced to adapt their programmes and messages to changing social priorities and pressure from external groups. Recent research, such as that carried out by Webb, Farrell, and Holliday (2002), suggests that non-partisan forms of communication and education, especially television, have undermined the role of parties. Based on survey data from the United Kingdom, Webb (2002) reports that 95 per cent of voters view news television programmes on at least a weekly basis, whereas, for example, only a quarter of voters were reached by party canvassing during the 1997 electoral campaign. Increasingly parties have to share the function of shaping citizens’ cognitive understanding of politics with a range of non-partisan actors. In Germany, public subsidies given to party foundations increased by ten times from the 1970s to the late 1990s (Scarrow, 2002); nonetheless, parties have still lost their central role in political communication and education as they have also, in many cases, lost control of widely read newspapers and other public media. Deschouwer (2002) concludes that in the Benelux countries, ‘parties are at the mercy of independent media’ and ‘the press is apt to blame parties’ (p. 175). Since Belgian and Dutch parties have totally lost their own party press, they are obliged to communicate through channels they do not control. The same has occurred in other European countries and happened even earlier in the United States. For example, Sundberg (2002) reports that in Scandinavia a media revolution, comprising the birth of commercial (including cable) and satellite broadcasting in the l990s and the widespread use of the Internet, has undermined the state monopoly of political broadcasting and weakened the parties’ agenda-setting capacity. American television is the main source of campaign information even if its influence varies from time to time (Green, 2002). Parties rely largely on non-partisan vehicles of communication, and have increased their effectiveness in using them.

Luciano Bardi and Enrico Calossi

163

This is certainly true in Canada where parties lost their own press long ago and have had many decades to practise their relationships with nonpartisan media. In Canada, the electoral activity of single-issue groups has been important and increased after the demise of the old party system in 1993; in response, parties have used the Election Act to limit the campaign period advertising of interest groups, to require that the press not charge them higher than average advertising prices, and to guarantee that broadcasters provide free time and additional paid time. All these measures are limited to the electoral period to guarantee parties control of the electoral agenda. Once the campaign is over, Canadian cadre-style parties go back to their quiescent stage (Carty, 2002). In conclusion, even if larger parties still maintain their own press organs, they need to compete for favourable coverage in the independent media. This implies that the agenda-setting capacity of political parties has been reduced. Also, the manner in which independent media cover party activities often reduces the legitimacy of parties as actors shaping the political agenda; in fact, independent media often devote attention to parties only during election periods and then they do so by covering parties’ electoral campaigns rather than the issues addressed in the party platforms. This element increases the general perception of parties as organizations active only during electoral campaigns, feeding popular cynicism about parties and politicians. Losing their power in agenda-setting, parties have decreased their capacity to structure the vote. Poguntke (2006) points out that negative depictions of parties and partisan activity have been reinforced by the privatization of mass media. In fact, commercial media seek to increase their audience instead of performing a public service through provision of information. This leads them to privilege news about scandals rather than focusing on policy disagreements between political parties. This tendency is even more evident for the Europarties. Even if the Maastricht Treaty claims that ‘parties at the European level contribute to forming a European awareness’ (while omitting any policy formation role), Europarties do not have the necessary resources to carry out this function. They lack their own partisan media and thus most often limit their activities to publishing leaflets and brochures during European election campaigns. Since elections for the EP are second-order elections and most often contested along national issues by national parties, these materials also make a minimal contribution to the formation of a ‘European awareness’. Even in cases where individual members of the EP can try to bring citizens in contact with EU institutions (for example, by accessing EU funds to organize trips to EU buildings for school children

164 Activating the Citizen

and university students), only a small proportion of the population is affected and these activities have little to do with Europarties and their differing agendas, serving primarily to promote the EU or the EP as institutions. Without a European media system and without the backing of national parties to distribute and advertise information, there is little that EP party groups can do to communicate their policies and behaviour to Europe’s voters (Hix, 2002). Therefore, Europarties do not, in any meaningful sense, perform the function of structuring the vote. In fact, it is doubtful that European citizens, even if they do vote in EP elections, know which Europarty the national party they are voting for belongs to or will join. Returning to King’s (1969) criterion that parties are expected to try to influence the vote in their favour, we could not consider Europarties to be real parties. But they consider themselves to be parties and are considered in this way by all of the other national and European institutions. This may all suggest that what we are seeing here is the emergence of a new party model. Rather than structuring the vote, Europarties seem to have the goal of representing and defending certain interests at the supranational level, and these are, most particularly, the interests of national parties that form the basis of the given Europarty.

Interest articulation and aggregation The articulation of interests refers to the way that parties or other actors express and pursue the political demands of particular social groups. The aggregation of interests refers to the process by which parties or other actors collect the demands from a variety of social groups and create a perceived coherence among different requests that are potentially in tension with each other. This capacity, to articulate and aggregate interests, is one of the most important mechanisms attributed to parties for linking state and society. It is often manifested in the formulation of complex and detailed electoral programmes. Webb, Farrell, and Holliday (2002) note that parties’ capacity to perform the articulation function has been challenged by the rise of interest groups and new social movements. However, parties do, in fact, still maintain some capacity to articulate group interests; in fact, in many instances, interest group activity does not represent a direct challenge to this important party function but may actually complement it. In addition, parties and trade unions have joined forces to articulate some group demands. The aggregation function is particularly interesting. In presidential and candidate-centred political systems, this function is

Luciano Bardi and Enrico Calossi

165

more often performed by single politicians than by parties. In parliamentary systems parties are still central to this function. The aggregation function that parties have traditionally been credited with is particularly important because the particularistic character of many interest groups, social movements, media, and unelected bureaucrats makes it difficult if not impossible for them to perform this role. In the Italian context, for example, Bardi (2002) argues that, with the formation of broad coalitions on the left and right (reinforcing a bipolarization of party systems), parties’ aggregative capacity has probably increased in recent years. Similarly, Webb (2002) shows how the British Labour Party’s adoption of a catch-all strategy has increased its capacity for interest aggregation but at the same time weakened its ability to articulate many particular interests. This has in turn contributed to popular dissatisfaction with the British party system, indicated by the fact that one-third of Britons favour electoral reform in the direction of proportional representation. In 1959, 48 per cent of Britons were affiliated with interest groups; more recently, this figure has risen to 61 per cent, providing further evidence that the articulation role is being performed by groups other than parties. The reduced ability of French parties to articulate interests (Knapp, 2002) is evidenced in the rate of voting abstention among blue-collar workers, which increased from 24 per cent in the 1980s to 38 per cent in 1997. The shift of blue-collar preferences from leftist parties towards the National Front also seems to suggest the plausibility of this hypothesis. At the same time, the French party system has increased its aggregative ability across the political spectrum, as compared to the 1950s when this capacity was limited to the Gaullist side. In relation to the Scandinavian democracies, Sundberg (2002) notes that the entrance of populist protest parties reflects the failure of the classical parties to articulate the interests of a significant portion of the population. However, another interpretation might be that while the emergence of new parties suggests the failure of traditional ones to articulate certain interests, it also shows the success of the party system as a whole in performing its interest articulation and aggregation functions. For example, in Spain, regionalist (nationalist) parties articulate disruptive positions (Holliday, 2002). Similar disruptive tendencies have become stronger in the Canadian political system, as manifested in the 1993 elections. That time, two new parties – the Reform Party and the Bloc Québécois – disrupted the old national three-party system and promised a new kind of politics and a new approach to how interests ought to be articulated, and in so doing they increased the diversity among party elites and expanded the range of

166 Activating the Citizen

choice offered by the party system. But it has not been enough, and not all Canadians find Canadian cadre-style parties particularly satisfying. Therefore, interest groups have increased their activity in at least three ways: 1) mobilizing their members to enlist a local party’s membership to influence the nomination of a candidate, 2) supporting or opposing particular candidates during local constituency campaigns, and 3) launching large-scale mass media campaigns during an election (Carty, 2002). In conclusion, even if the emergence of new issue agendas relating to gender, ethnicity, regionalism, environmentalism, and European integration have complicated the task of interest articulation for major parties, they do retain a central role in aggregating diverse interests. One indicator of how this process takes place is the increased level of penetration of collateral organizations inside parties. Poguntke (2006) reports that the formal linkage between parties and collateral organizations (which are given the right to a seat in party organs) has increased since the 1960s. In this way, catch-all parties, then cartel parties, have tried to maintain their links with societal sectors despite decreasing party membership. How do European parties compare to these patterns that prevail at the national level regarding interest articulation and aggregation? One can state simply that at the European level parties do not play any role in interest articulation. As Hix (2002) reports, interests are articulated by interest groups (through more then 1700 private interests represented in Brussels) and national governments (through the meetings of the Council of Ministers). But, furthermore, at the European level, the aggregation function is also not performed by parties but rather by the European Commission as well as by national governments. These agents are not directly accountable to European voters or to the elected EP. The absence of mechanisms of accountability at the European level in relation to those bodies that carry out key functions of interest aggregation is an important factor underlying complaints about a democratic deficit inside the EU. While the Maastricht Treaty affirms the importance of political parties at the European level in forming a European consciousness, the weakness of the most important institution – the EP – through which parties could play this role undermines the claim. A strengthened EP would certainly help, should this come to realization through the adoption of the Lisbon Treaty or something similar. Until the EP gains more institutional capacity in policy making, however, Europarties will continue not only to fail in their role as interest articulators, but also as aggregators of interests on a European level.

Luciano Bardi and Enrico Calossi

167

Citizen participation The capacity of political parties to promote mass political participation through their member activities or through mobilization of the electorate2 is the area where they deviate most from their intended function. Particular issues relate to the levels of activism and to the rights and powers of party members, activists, and voters. One organizational indicator of this weakness is declining party membership, while another is declining voter turnout (as discussed in Part II of this volume). According to Webb (2002), there is a ‘supply-side’ as well as a ‘demand-side’ explanation for declining membership. In terms of supply-side factors, various social changes have made citizens more reluctant to join parties. In terms of the demand-side, party strategists no longer seek to recruit members as actively as they did in the past. The reluctance of citizens to join parties is related to the decline of social group identities, such as social class identity, which previously related them to particular political parties. Some other authors (for example, Scarrow, 2002) report that, in response to declining participation, some parties seem to have developed participatory incentives for their members. For instance, they have assigned to members new rights in candidate and leadership selection, and sometimes they have allowed increased member involvement in formulation of the party’s policy positions. For these reasons, currently opportunities for political party members to influence party practice seem to be as great as they have ever been, but in reality, in many instances this power is apparent only because party leaders have maintained for themselves veto power over candidate selection and enjoy considerable autonomy in shaping party policy. Taking an example from the national level to illustrate the point, in the United Kingdom ordinary members have increased their influence in the leadership-selection processes, as Cross and Crysler also document in this volume. However, Webb (2002) points out that at the same time elites have maintained or enhanced their autonomy in the policy-making processes, making evident how in this field activist members have lost their power in favour of elites. Bardi (2002) also stresses the manner in which parties in Italy have stimulated the activism of members and sympathizers through holding consultations on various matters related to internal party affairs and policy positions. For example, in the last three years citizens have been allowed to vote in primary elections for the choice of the leaders of the ‘Unione’ coalition and of the Democratic Party. On these occasions the differences between activists, ordinary members, and sympathizers have disappeared and all citizens have been placed on the same level. This fact has surely

168 Activating the Citizen

increased citizen power in leadership selection (even if not in party policy formation), but while this might seem to have reinforced public participation in political party activities, it also has had the effect of reducing the incentive for people to join parties and become committed party activists. Scarrow (2002) reports that in Germany in the 1990s, experiments were undertaken that changed the rights of members in relation to internal party affairs. This included a Social Democratic Party internal ballot for the party chair and the two Free Democratic Party member plebiscites on policy issues. France presents a different case, where party leaders have shown less interest in developing new incentives to encourage participation, but have, in some cases, expanded the rights of members to have a say. French parties have for the most part been characterized by low membership levels, with the exception of the Communist Party, which has long reported memberships of more than 100,000. Knapp (2002) notes that French parties have generally not been interested in fostering political participation. Membership participation in decision making was also limited in the major French parties until the 1990s. Only then did the Communist Party abandon the Leninist principle of democratic centralism and did the Gaullists move away from a leadership culture in favour of more internal democracy. The Socialist Party has generally been more open to internal democracy and to realizing membership rights. Canada’s basic unit of importance is the electoral constituency. At this level party members are organized in local constituency associations, which are practically inactive in non-electoral periods. Even if members could participate in national party conventions, most experienced party members realize that despite convention resolutions, debates, and suggestions, party policy remains a preserve of the professional politicians. For this reason, members’ activity is dedicated to electoral campaigns that are conducted by party members and non-partisan activists, because campaign teams are directly appointed by constituency candidates. Therefore, it is plausible that the only reason to be a member of a Canadian party is to choose a local candidate or the party leader (Carty, 2002). In the 1990s, most parties opened their doors to the direct election of their leaders and candidates (Cross and Crysler, this volume). This has led to great variations in membership depending on the year (for example, in 1998 Conservative membership rose from 18,000 to 90,000). Political participation does not seem to be related to the choice of different policy positions but only to the nomination of different political personnel. As documented elsewhere in this volume by DeBardeleben and LeDuc, elections at the European level (for the EP) have been characterized by

Luciano Bardi and Enrico Calossi

169

low levels of voter turnout, a pattern possibly related to the indirect nature of membership in the European parties. Individuals are members of European parties by virtue of their membership in a national party. Mass political participation in the work of European parties is extremely limited. Because there are very few European party membership groups in the biggest European cities (an exception being the French Socialists in London), political participation in Europarties can generally be calculated only at the elite level. In the 1980s, many party leaders sent their spokespersons to participate in the party leaders’ summits. In the 1990s, with the increasing powers of the EU, summits began to be held immediately before or after European Council meetings and party leaders began to participate in them more frequently. Hix’s (2002) research indicates that, from 1990 to 1994, 70.7 per cent of party leaders’ meetings occurred in proximity to meetings of the European Council, compared to only 15.4 per cent in the period from 1985 to 1989; likewise, the proportion of party leaders participating increased from 66.5 to 72.9 per cent. This pattern suggests that strong reference points for leaders of Europarties are other members of the European elite. On the other hand, these parties have not nurtured an active membership base, but rather have focused their attention on the general electorate and on ordinary (non-active) members. Without a strong base of party activists, the cartel nature of these parties is reinforced. This pattern of cartelization, which Poguntke (2006) identifies with a ‘lack for the modern parties of a clear ideological profile’ (p. 121), likely contributes to the poor participation rate in Europarties. Thus, a pattern has established itself where the membership base of Europarties resides in particular national parties and where the primary aim of Europarties is to foster the participation of national party leaders.

Conclusion In general, our analysis confirms our expectations about Europarties in terms of their organizational characteristics as well as their ability to perform democratic functions for the EU political system. Organizationally Europarties most closely approximate the cartel party. This follows from the fact that Europarties originated from, and still are developing as, organizational offshoots of national European parties, which, in the course of their evolution, in most cases have acquired cartel party characteristics. This organizational dependency of Europarties is clearly visible in their origins as EP party groups. These were created as instruments for the representation of national parties at the European level

170 Activating the Citizen

prior to the EP’s first universal suffrage elections, under circumstances that could not possibly allow them to link European citizens to EC/EU institutions. In later years, even most recently, after the approval of a Europarty statute, national parties preserved their dominance. It is certainly significant that Europarties are officially recognized in the treaties as ‘Political Parties at the European Level’ instead of ‘Parties of the European Union’. The definition in the treaties reflects that Europarties are extensions of pre-existing party structures at the national level. It is quite natural therefore that Europarties have many similarities with cartel parties. This similarity is particularly evident if we take a look at their main organizational features: the principal source of party resources (almost totally state subventions), the nature of party work and party campaigning (strictly capital intensive), party channels of communication (state-regulated channels of communication albeit with very limited access), and their position between civil society and the state (practically no direct contacts with civil society). In this sense, the Europarty model seems to be the quintessential evolution of the cartel party. Europarties go even further if we consider the character of membership and the relation between ordinary party members and party elites. In fact, where cartel parties have relatively small individual memberships, often with undefined rights and obligations, Europarties, although formally allowing individual members, have practically none. Finally, the relations between party elites and ordinary party members are different in the two models. In cartel parties there is a sort of mutual autonomy between the two, whereas in Europarties elites are accountable to national parties. This is due to the fact that relations inside Europarties are mainly relations between elites. The organizational similarity to cartel parties does not bode well for Europarties’ performance of functions at the EU level. As we have seen, cartel parties are the worst equipped of all models for the task. In particular, Europarties do not have any role in fostering citizen participation because they do not have any direct contact with a very difficult to detect European civil society. Moreover, Europarties do not articulate interests and leave to other actors (especially the EU Commission) the aggregation function; they do not structure the vote, a function in which they seem to have so little interest as to leave it to national parties and single candidates. After this discussion it is questionable whether we can even define Europarties as real parties, even if the treaties, politicians, commentators, and scholars continue to do so. The best that can be said on their behalf is that, as plausible extensions of national cartel party organizations, they perform functions that, at least prospectively, may

Luciano Bardi and Enrico Calossi

171

serve to ensure national parties’ survival. The EU has already acquired portions of state prerogatives and power and represents a limited source of state-like funding. As such, national parties need the instruments, their European components, to occupy the EU, as they have occupied, with a good degree of success, the member states. Thus, should European institutions’ prerogatives grow considerably in the future, the extant multi-level party organizations may well shift their centres of gravity increasingly closer to the European level.

Notes 1. The elite party will not be explicitly discussed here. It appears in the table as it was included in Katz and Mair’s original version. 2. We do not address the function of leadership recruitment generally attributed to political parties, even though it is closely linked to issues of political participation, because this would lead us away from our primary concern with the broader linkage function between state and society.

References Bardi, L. (2002) ‘Italian Parties’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 46–76. Carty, R. K. (2002) ‘Canada’s Nineteenth-Century Cadre Parties at the Millennium’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 345–78. Deschouwer, K. (2002) ‘The Colour People’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 151–80. Green, J. C. (2002) ‘Still Functional After All These Years’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 310–44. Haas, E. B. (1958) The Uniting of Europe (Stanford: Stanford University Press). Hix, S. (2002) ‘Parties at the European Level’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 280–309. Hix, S., A. G. Noury, and D. Roland (2007) Democratic Politics in the European Parliament (Cambridge: Cambridge University Press). Holliday, I. (2002) ‘Spain’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press). Katz, R. S. and P. Mair (1995) ‘Changing Models of Party Organisation and Party Democracy: The Emergence of the Cartel Party’, Party Politics, 1, 5–28. King, A. (1969) ‘Political Parties in Western Democracies. Some Sceptical Reflections’, Polity, 2, 111–41.

172 Activating the Citizen Kirchner, A. (1985) The European Parliament: Performance and Prospects (Aldershot: Gower). Knapp, A. (2002) ‘France: Never a Golden Age’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 107–50. Lodge, J. and V. Herman (1982) Direct Elections to the European Parliament: A Community Perspective (London and New Haven, NJ: Macmillan and Humanities Press). Marquand, D. (1979) Parliament for Europe (London: Jonathan Cape). Michels, R. (1915/1959) Political Parties (New York: Dover). Poguntke, T. (2006) ‘Partiti e società nell’Europa Occidentale’ [Parties and Societies in Western Europe] in B. Luciano (ed.) Partiti e sistemi di partito [Parties and Party Systems] (Bologna, Italy: il Mulino), pp. 103–22. Scarrow, S. E. (2002) ‘Party Decline in the Parties State? The Changing Environment of German Politics’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties In Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 77–106. Sundberg, J. (2002) ‘The Scandinavian Party Model at the Crossroads’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 181–216. Webb, P. (2002) ‘Political Parties in Britain: Secular Decline or Adaptive Resilience’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 16–45. Webb, P., D. Farrell and I. Holliday (eds) (2002) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press).

8 Grassroots Participation and Party Leadership Selection: Examining the British and Canadian Cases William Cross and John Crysler

In many Western democracies, there has been considerable recent debate over the effectiveness of participation in parties in producing political and social change. Many argue that political parties are elite-driven, hierarchical organizations in which grassroots members have little chance to influence decision making. One manifestation of this is young activists turning away from political parties towards alternative methods of public participation such as advocacy groups and social movements (Cross and Young, 2008). But there is no doubt that parties occupy a central place in the democratic life of Western democracies, particularly as a linkage between citizens and the state (Katz, 2001). The extent to which parties engage large numbers of local partisans in key decisions is an important measure of how well they perform this linkage role. Much of the criticism of parties is directed at the realm of policy making where it is argued they have little capacity for innovation. There is one area of democratic life, however, where parties remain indisputably dominant – personnel recruitment. As Sartori (1976) pointed out long ago, the defining characteristic of parties is the recruitment of candidates for public office. In parliamentary systems, parties select both local candidates for election and leaders who compete for the premiership. This chapter focuses on the selection of party leaders. Political observers in Europe and Canada have observed a recent trend towards an increased concentration of power in the position of party leader. Poguntke and Webb (2005) have identified what they call a widespread ‘presidentialization’ of politics. This phenomenon is defined by an increase in the power exercised by elected party leadership in the political executive, an increase in leadership power and autonomy within parties, and increasingly leader-centred electoral campaigns. These effects 173

174 Activating the Citizen

produce party leaders with more authority in government, in the party, and with the electorate. Given the increased authority these leaders have, a full assessment of public participation in democratic decision making must consider the ability of grassroots partisans to influence their selection. Perhaps as an acknowledgement of the increased importance of leaders, grassroots partisans have been seeking a more influential role in their selection in recent years. Traditionally in parliamentary practice, a party leader has been chosen by members of the party’s legislative caucus. While this remains the practice in some jurisdictions (for example, New Zealand and Australia), parties in many countries now provide their grassroots members with a meaningful role in the choice. One possible explanation is that parties are opening up their leadership selection contests as a response to increased demand by partisans to have a say in the choice of leader, a demand fuelled by an increase in the leader’s authority. While the broadening of the leadership selectorate has not taken place in all parties, it is a trend that is observable in many countries including Canada, the United Kingdom (UK), the Netherlands, Israel, and Belgium. Among these, it is in Canada and the UK where the development is most entrenched, with all of the major parties now offering their grassroots, extra-parliamentary membership a key role in the choice of leader. The widespread expansion of the leadership selectorate in these countries, coupled with the different models of selection in use, make them useful subjects for study of the genesis and consequences of direct member involvement in choosing party leaders. We use the British and Canadian cases to consider several questions important to party organization and intra-party democracy that arise from this expansion of the leadership electorate.

Party leadership selection in the United Kingdom and Canada Consistent with the norms of responsible government, in Canada and the UK, party leaders initially were chosen by members of the parliamentary party and typically from among their own ranks. This practice fits with the notion of the party leader being ‘first among equals’ within a parliamentary party, essentially the captain of the parliamentary team. This method of selection provides no role of any consequence to those outside of the parliamentary caucus, which made it vulnerable to attack at various stages of the twentieth century. First in Canada in the 1920s, and then towards the end of the century in the UK, all of the principal

William Cross and John Crysler

175

parties abandoned the caucus-dominated practice in favour of expanding the leadership selectorate to include non-parliamentary partisans. While the practices adopted by the parties vary significantly and have changed over time, one characteristic they share is an expansion of the leadership selectorate to include voices from outside the parliamentary party. In some cases this has involved a wholesale handing over of the process to the grassroots membership, while in others it has been less dramatic, with the party maintaining a key role for parliamentarians. In examining the leadership selection practices adopted by the parties, we can identify a set of four key variables that define the different processes adopted: whether the participation of a party’s grassroots supporters in the leadership process is mediated, whether regional considerations are taken into account in the weighting of votes, whether a privileged role is preserved for the parliamentary caucus, and whether particular supporters or constituents of the party are given a privileged role in the selection. These variables affect the role of grassroots partisans in the process, and they also highlight important differences in the practices between the two countries and among the different parties within them. We return to a discussion of these four points after briefly describing developments in Canada and Great Britain.1

The Canadian case The Canadian parties were at the vanguard of the move to expand the party leadership selectorate beyond the parliamentary caucuses. The Liberals were the first to include non-parliamentarians in 1919. After badly losing the previous election, the Liberal parliamentary group was largely reduced to a rump of francophone Québecois members of parliament (MPs). Divided over the issue of conscription in World War I, the party was faring very poorly in English Canada. Realizing the need to choose a leader who could rebuild the party outside of Quebec and the absence of such a figure in the parliamentary caucus, party elders agreed on the need to consult with supporters outside of parliament in making the choice. The result was a ‘leadership convention’ comprised of approximately 1000 party members, which elected Mackenzie King as leader on the third ballot. Impressed by the popularity boost that the 1919 convention seemed to have given the Liberals, the Conservatives used a similar method in 1927 in the selection of R. B. Bennett as their new leader (Courtney, 1973). Participation in these contests had been expanded beyond the parliamentary caucus but had not yet fully reached the grassroots membership. Convention delegates during this

176 Activating the Citizen

period typically held some sort of ex officio position within the party – elected officeholders, past candidates, members of provincial and federal party boards, and local party notables. This method, with modest changes, lasted through the 1960s. As described by Courtney (1995), in Liberal and Conservative contests in the 1970s, the focus of contestation moved from the convention to the constituency level. The number of delegates roughly doubled during this period and the additional delegates were primarily rank-and-file representatives of the constituency parties, chosen at local meetings in which all party members were eligible to vote. Whereas in prior contests the leadership candidates essentially waited until the convention roster was filled and then aimed their campaign at winning over delegates, in the new era they focused their campaign efforts at the grassroots level in an attempt to secure the election of delegates who were pledged to support them at the convention. These campaigns were aptly described by Carty (1988) as ‘trench warfare’. The important point is that local party members were given a more consequential, though still indirect, say in the choice of party leader, except for the relative few chosen as convention delegates. The next Canadian development came in the 1990s when parties began allowing their constituency members to vote directly for their preferred leadership candidate as opposed to voting for convention delegates. As Cross (1996) describes, this was a movement that began with the provincial parties and was well established at that level before being taken up by the national parties. The Bloc Québécois was the first federal party to use direct member voting in 1997 and it was quickly followed by the Progressive Conservatives in 1998. In the Bloc Québécois contest the candidate receiving a majority of votes was the winner. The Progressive Conservative rules were more complex. Every party member was granted a single, direct vote for the leadership; votes were counted at the constituency level; the results of each constituency were weighted to total 100 votes (essentially percentages); the percentages from each constituency were then totalled; and a majority rule was used.2 This system provided no privileged position for members of the parliamentary party; they voted in their home constituency along with all other party members. However, not all voters were treated equally. Because each constituency was assigned an equal number of votes, ballots cast in places where the party had few members counted considerably more towards the final result than did those cast in places where the party had many active partisans.

William Cross and John Crysler

177

A conservative competitor, the Canadian Alliance, selected leaders through a vote of its membership in 2000 and 2002. In these cases, the votes of party members from across the country were totalled and the candidate with the most votes was elected (again a majority threshold was in effect). Unlike the earlier Progressive Conservative contest, there was no weighting of votes by constituency. The result was that voters from western Canada and Ontario, where the party enjoyed substantial electoral support, played a disproportionately strong role in the selection of the leader, as the party had few active partisans in Quebec and the Atlantic provinces. Similar to the Progressive Conservative process, this method provided no special role for parliamentarians or any other group within the party. When the Progressive Conservatives and Canadian Alliance parties merged to form the new Conservative Party of Canada, the method to be used in the inaugural leadership contest was a major point of contention. Reflecting the earlier practices of the two parties, the area of dispute was whether all votes would be counted equally or would be weighted by local constituency. In the end, the new party decided to adopt the earlier Progressive Conservative practice of allocating 100 votes to each constituency. The New Democratic Party (NDP) first allowed its members to vote directly for their preferred leadership candidate in 1995 (Archer and Whitehorn, 1997). The party utilized a bifurcated process of party membership votes in ‘primaries’ followed by a delegated convention. The purpose of the primaries was to determine eligibility for candidacy at the subsequent convention. Primaries were held regionally across the country and a discrete one was held for members of organized labour. The leadership choice was made at a convention of delegates, some of whom represented the party elite (ex officios) some who represented affiliated trade unions, and some who were chosen by members in the constituencies. For its subsequent leadership contest in 2003, the NDP did away with the primaries and the delegated convention. In this contest, all party members were given a vote. The votes were not weighted by constituency, but rather all ballots cast in the local constituencies were converted to a national total out of 75. The remaining quarter of the votes was allocated to ballots cast on behalf of affiliated labour unions. In this way the party provided a privileged position to its traditional supporters in organized labour while providing no such accommodation for the parliamentary party. The federal Liberal party has yet to choose a leader through a direct election of its members. While the party amended its constitution in

178 Activating the Citizen

1990 to allow for a direct vote of members in the selection of its next leader, it quickly changed its mind and in 1992 adopted a hybrid process falling somewhere between direct election and the traditional delegate convention selection method. The Liberal party has held two contests, in 2003 and 2006, using this new process. The logic of the system is something akin to an open list, regional proportional representation electoral system. The actual choice of leader is made at a convention populated by delegates selected at the local level. Local party members cast two ballots, one for their favoured leadership candidate and one for their preferred set of delegates. The number of delegates the candidates win at the local level is determined by their share of the votes cast for leadership candidates (similar to the party vote in a proportional representation system); the actual identity of these delegates is then determined through the second ballot. The result is that each candidate wins a share of convention delegates roughly proportional to the overall leadership preferences of participating party members. Delegates to the convention are pledged on the first ballot to vote for the candidate they committed themselves to in the constituency election. While there is no privileged place for members of the party’s parliamentary caucus in the delegate selection process, they along with other party officials and local and national party officeholders are automatic delegates to the convention. About 20 per cent of all delegates fall in this ‘ex officio’ category.

The British case The British parties were generally slower than their Canadian counterparts to open up the selection process first to their parliamentary parties and later to the constituency membership. Punnett (1992) writes of Conservative leaders in the nineteenth and early twentieth century ‘emerging’ from the ranks of party notables and describes how when Liberal Prime Minister Gladstone retired in 1894, the choice of Lord Rosebery as leader and prime minister was made by Queen Victoria with no apparent consultation with party officials. The first formal changes to the norms of party leadership selection occurred with the rise of the Labour party. Ever since emerging as one of two principal parliamentary parties in 1922, Labour selected its leaders through a formal ballot of the parliamentary party. This can be contrasted with Conservative methods during this period in which the decision was largely arrived at by a small number of parliamentary elites after

William Cross and John Crysler

179

consultation, which Punnett (1992) describes as ‘extremely random and limited in extent’ (p. 261). The first real fissures in the Tory party came after the 1963 leadership contest when there were several potential contenders and significant opposition among backbenchers to the ultimate winner, Alec DouglasHome. While the party leadership claimed to have determined that Douglas-Home was the preferred choice through a series of consultations within and without the parliamentary party, the consultations were highly secretive and subject to manipulation. Following the party’s defeat in 1964, rules for a formal vote among MPs were adopted. While amended several times and extremely complicated in their operation (with different thresholds applying to successive ballots), this basic method remained in place for several decades. The Liberals were the first British party to expand the selectorate beyond the parliamentary party. After being significantly reduced as a parliamentary force throughout the twentieth century, the party found that its small parliamentary group was dominated by members from rural Scotland and Wales while many of its electoral supporters were in urban centres. Needing to reinforce this urban support, the party desired a more representative leadership selectorate than that offered by its elected MPs. For this reason, along with the desire for a more generally participatory process, in 1976 the party used a form of membership vote for the selection of David Steel (Punnett, 1992; Stark, 1996). In this early method, each constituency party was allocated votes according to a formula taking into account the size of the Liberal vote in the past election and the number of party members. The constituency votes were allocated on the basis of preferences determined through a ballot of members and candidates had to be nominated by five MPs or one fifth of the Liberal parliamentary caucus (whichever was less). At the founding convention of the Social Democratic Party in 1982, there was some dissension as to whether party members or MPs would control leadership selection. Members of the party’s parliamentary body (former Labour MPs) preferred to select the leader themselves, while the new party’s activists wanted a say in the selection. A vote of party members narrowly decided that the new leader would be selected through a one-member one-vote postal ballot. MPs, however, maintained significant control through provisions requiring that leadership candidates be MPs and have the support of at least 15 per cent of their parliamentary colleagues. When the Liberals and Social Democrats merged to form the Liberal Democrats in 1988, the new party agreed upon a leadership

180 Activating the Citizen

selection process that included both a membership vote and a key role for the parliamentary party. In 1978, Labour officially recognized the leader of the parliamentary party as the leader of the broader party. With this change came demands that extra-parliamentary supporters be permitted a role in the selection of the leader. This was coupled with a sense that the 1974–79 Labour government and its backbench supporters were ideologically to the right of the party’s activist base and in Punnett’s (1992) phrase ‘had betrayed socialism’ (p. 267). The result was the adoption in 1981 of the ‘electoral college’ method of leadership selection. Essentially, this method identifies three constituent bodies of the party (the parliamentary party, the trade unions, and the constituency party) and gives each a fixed say in the selection of the party leader. The percentage of leadership votes to be controlled by each bloc has been contested frequently. In its first manifestation the electoral college granted 40 per cent of the votes to the trade unions and 30 per cent to each of the parliamentary and constituency parties. In its current form, each of the three groups controls one third of the votes. The way in which constituency votes are determined has also been contentious and subject to reform. Under the initial rules, the candidate each constituency party supported was decided by the majority vote of an elected general committee of about 100 committed activists (Quinn, 2004). In 1989, Labour began requiring balloting of all party members and each constituency’s vote was awarded to the candidate receiving a plurality of the members’ votes. Following the party’s devastating 1992 electoral defeat, the rules were again revised to require that each constituency’s vote be divided to reflect the proportion of the membership vote received by each candidate. The features of the new system, which was used for the first time in 1994 to elect Tony Blair, can be summarized as follows: Ballots are cast through an equally weighted three-part electoral college comprised of MPs and members of the European Parliament, party members, and levy-paying union members; preferential ballots are used; and a candidate must be an MP in the House of Commons with the support of at least 12.5 per cent of the caucus. The Conservatives were more reluctant than the Liberal Democrats and Labour to expand the leadership selectorate beyond the parliamentary party. The party did not do so until 1998 when, in somewhat of a surprise decision, it adopted new procedures giving the final choice in selection of the leader to the extra-parliamentary membership (Alderman, 1999). MPs continue to play a key role insofar as they conduct their own ballot

William Cross and John Crysler

181

to select the two finalists from among whom the party membership, through a one-member one-vote postal ballot, selects the leader. The Conservatives have chosen three leaders since the adoption of these rules. In 2001, the parliamentary party took three ballots to narrow the field from five to the final two candidates, Kenneth Clarke and Iain Duncan Smith. Duncan Smith then won in the membership vote despite having trailed on all three caucus ballots and never having received the support of more than one-third of MPs. The lack of support from the parliamentary party he was elected to lead led some to contend that he had been placed in an untenable position (Brogan, 2003). Following the parliamentary party’s ouster of Duncan Smith in 2003, it settled upon Michael Howard as its choice. Because he was ultimately unopposed on the MPs’ ballot, the party’s executive decided not to have his selection ratified by the membership at large. Some party members objected that this process essentially allowed the parliamentary party to select the leader by reaching a consensus choice through informal discussions and presenting it as a fait accompli not requiring a vote of the membership. For the selection of his successor, Howard attempted to have the party revert back to a process whereby only the MPs would have a vote. The proposal was hotly debated and ultimately defeated in a party conference vote. The party’s 2005 contest was then held under the existing rules and seems to have solidified party support for the new method.

Making sense of the different methods The different characteristics of the direct election methods used by the parties reflect the different democratic values and electoral imperatives of each party. While we often refer to these processes as one-member onevote, an examination of the voting rules illustrates that they are often anything but. We can begin to make sense of these different versions of direct election processes by considering the four key variables identified earlier. Mediated versus unmediated participation Unmediated participation by party members in the leadership decision now appears to be the norm in most parties. Only in the Canadian Liberal party are the preferences of the party members funnelled through convention delegates before being counted in the leadership choice. The decision to adopt this method was reached quite deliberately and reflects the view that the leadership decision should be made in a collective, deliberative forum. The Liberal party believes that it is important for

182 Activating the Citizen

those ultimately selecting the leader to be able to gather together in one place to deliberate collectively over the choice (Cross, 2004). This reflects the party’s brokerage traditions and its ethos of accommodating divergent perspectives (Whitaker, 1977). Party officials have feared that a pure direct vote would be too atomistic and would not allow for the kind of deliberative, accommodative process facilitated by a convention. At its December 2006 leadership convention, the party considered and rejected a motion to adopt an unmediated form of one-member one-vote. Unmediated participation has, however, become the norm in other parties. In the UK Labour party, constituency and trade union members’ votes are no longer mediated through branch elites but rather now count directly in the leadership choice. A similar progression has occurred in the Canadian NDP. The first generation of membership voting in which members’ votes had only a limited, indirect influence on the final choice has been replaced by a process in which members’ votes are determinative of the outcome. It appears that rank-and-file members given a voice in the leadership choice eventually demand that it be unmediated and conclusive. The Canadian Liberal party is an outlier in this regard, but even here the vote at the party’s 2006 convention found considerable support for the elimination of delegated leadership contests.

Weighting votes by constituency None of the British parties weigh members’ votes by constituency. However, when Labour first began using the electoral college, it did assign an equal number of votes to each constituency. When the Canadian parties used conventions to make the leadership choice, the Conservatives and Liberals allocated an equal number of convention votes to each local constituency. The left-of-centre NDP never followed this practice, instead awarding delegates on the basis of the number of party supporters in each constituency (Archer and Whitehorn, 1997). The Liberals and Conservatives continue to weigh members’ votes equally by constituency, even though some constituencies have many times more members than others (Stewart and Carty, 2002). There are two justifications for this practice. First, treating constituencies equally is consistent with the logic of the single-member plurality system under which all of these parties compete in general election campaigns. Singlemember plurality treats each constituency equally and elections are won by accumulating constituency victories and not by amassing a nation-wide vote total. This logic, however, applies to all parties in both countries, so it is intriguing that only two of our six parties continue this

William Cross and John Crysler

183

practice. The more salient reason is that these two cases involve brokerage parties that practise the politics of regional accommodation (Carty and Cross, 2009). The deepest cleavage in the Canadian case is region and both the Liberal and Conservative parties have regional accommodation as one of their principal objectives. This imperative also helps to explain why the short-lived Canadian Alliance did not weigh votes by constituency. This party had a regional support base (primarily western Canada) and did not practise region-based brokerage politics. For the brokerage parties, treating each constituency equally helps ensure that all regional voices have an equitable say in the leadership choice and that no candidate can expect to be competitive without considerable support in most regions. A privileged place for the parliamentary party All three of the British parties ensure that their parliamentarians play a significant role in the choice of their leaders. In the Conservative case, MPs narrow the choice to two before party members enter the fray and, as in 2003, have the opportunity of making the choice themselves by reaching consensus. In Labour, the parliamentary party has one-third of all votes, which is equal to the weight given to the hundreds of thousands of party supporters in the constituencies. Labour MPs also serve as gatekeepers through a requirement that any candidate must have the support of 12.5 per cent of MPs in order to be eligible to run in a declared contest. In the 2007 contest to replace Blair, this provision ensured the acclamation of Gordon Brown, as MP Michael McDonnell, a would-be challenger to Brown, managed to collect only 29 caucus endorsements – 16 fewer than required. The Liberal Democrats have a similar rule requiring that all leadership candidates be supported by at least 10 per cent of their parliamentary party. These requirements reflect the primacy of the parliamentary party in the British tradition. There are no comparable provisions in any of the Canadian parties. A privileged place for particular party supporters In four of our six cases, there are no provisions ensuring that key groups of supporters are represented in the leadership contest. The exceptions are British Labour and the Canadian NDP, both with deep and historic roots in organized labour, including a dependence on it for much of the financing of their electoral efforts. In both parties a set percentage of the leadership vote is guaranteed to these key constituents, one-third in Labour and one-fourth in the NDP. This is a distinction that is not

184 Activating the Citizen Table 8.1: Characteristics of the leadership selection processes in UK and Canadian political parties Unmediated Privileged Privileged Votes member role for role for weighted participation MPs trade unions regionally Liberal Party (Canada) Conservative Party (Canada) New Democratic Party (Canada) Labour Party (UK) Conservative Party (UK) Liberal Democrats (UK)

X X

X X X X X

X X X X

X

related to country but rather is shared by parties of the left with ties to trade unions. The differences we observe in Table 8.1 reflect the varying democratic ethos of each party and the different electoral contexts in which they exist. What is important here is that the methods of direct leadership election have proven flexible enough to accommodate the divergent needs and priorities of the parties while permitting an expansion of the selectorate to include their rank-and-file supporters.

Motivation for change In an examination of Canadian leadership selection processes, Cross (1996) has argued that the decision to expand the leadership selectorate generally results from one or more of three factors: an unrepresentative parliamentary party, a desire to conform with changing societal democratic norms, and a desire to revitalize a party’s supporters when faring poorly electorally. An examination of our cases provides support for this argument. The regional unrepresentativeness of the parliamentary party, as discussed above, helps explain the initial decision to expand the leadership selectorate beyond the parliamentary party by both the Canadian and UK Liberals. All parties adopting direct election have couched their decision in ‘democratic’ terms, arguing that the expansion of the selectorate is consistent with changing democratic norms and is evidence that they are ‘in touch’ with the demands of their grassroots supporters for greater intra-party democracy (Carty and Blake, 1999; Leduc, 2001; Russell,

William Cross and John Crysler

185

2005). A full discussion of this ‘plebiscitary’ style of party democracy can be found in Seyd (1999) and Young and Cross (2002). We focus here on the third factor. In an examination of the points in time at which membership election of leaders in Canada’s provincial parties was adopted, Cross (1996) found that of the first twelve parties to include members directly in the choice of leader, only two were in government when they adopted the new selection rules and none of the twelve was favoured to win the next election. He argues that ‘many of the parties . . . made the switch to direct election following a disappointing general election in which their party won significantly fewer seats than anticipated . . . [and] many of these parties saw direct leadership selection as a way to attract new members and generally to revitalize’ (Cross, 1996, pp. 297–8). We find similar patterns with parties at the national level in the UK and Canada. Including the predecessors of the Liberal Democrats and the Canadian Conservatives, there are ten cases of major parties adopting some form of membership vote between 1976 and 2003. As illustrated in Table 8.2, we find that in all ten of these cases, the decision to adopt a membership vote was made while in opposition. Parties that have taken a turn governing have invariably waited until finding themselves on the opposition benches to adopt these reforms. For example, two of these parties (the Canadian Progressive Conservatives and the British Conservatives) made the decision to adopt a form of direct vote in their first term in opposition following a long run in government. Canada’s Liberals made the decision after a second straight election loss following an almost uninterrupted 16-year run in government. Labour first expanded its leadership selectorate following Thatcher’s pummelling of the party in the 1979 election, and then amended the process, increasing the influence of constituency party members, first in 1989 (on the heels of a third straight electoral defeat) and again in 1993 following an unexpected fourth consecutive defeat to the Tories in 1992. The Canadian Progressive Conservative case well illustrates what seems a clear pattern. When in government from 1984 to 1993, the party showed little interest in amending its leadership selection processes even though many of its provincial branches were experimenting with direct election. A survey of delegates to the party’s 1993 convention found that only one-third supported the idea of leadership selection by a direct vote of party members (Courtney, 1995). Later that same year, the party suffered a defeat of unparalleled proportions, being removed from government and reduced to just two parliamentary seats. In an effort to

186 Table 8.2: Parties’ electoral context at time of adopting direct election of the party leader in UK and Canadian political parties Party

Year of adoption of direct vote

Electoral context

Liberal Party (UK)

1976

A perennial third party, it won 13 seats in the previous election.

Social Democrats (UK)

1982

Dissidents on the right of the Labour Party created this new party in 1981.

Liberal Democrats (UK)

1988

A merger of the Liberal and Social Democratic parties created this party. Both merger partners had used direct election in selecting their respective leaders.

Labour Party (UK)

1989

After having alternated power frequently with the Conservative Party from World War II until Thatcher’s 1979 electoral triumph, Labour suffered its third straight electoral defeat in 1987.

Liberal Party (Canada)

1990

Considered Canada’s ‘natural governing party’, the Liberals had lost a second consecutive election to the Progressive Conservative Party in 1988.

Reform Party (Canada)

1991

At its founding convention in 1987 the party adopted a delegate convention format for selection of its leader, but it changed to a direct vote format in the lead-up to its breakthrough election of 1993. Its successor, the Canadian Alliance, retained the method.

Progressive Conservative Party (Canada)

1995

The party went from a legislative majority to two seats in the 1993 election.

Conservative Party (UK)

1998

After 18 years in government, the party suffered a humiliating defeat to Labour in the 1997 election.

New Democratic Party (Canada)

2002

A perennial third party, the NDP had too few seats to retain official party status after the 1993 election, regained official party status in 1997, and nearly lost it again in 2000.

Conservative Party (Canada)

2003

A merger of the Canadian Alliance and Progressive Conservative parties created this new party from the opposition benches.

William Cross and John Crysler

187

revitalize the grassroots membership, the party subsequently adopted direct leadership election with little internal opposition.

Participation and party leadership selection Given that the motivation for change is often a desire to revitalize a party’s grassroots membership, it is important to consider questions relating to participation rates. Since the 1970s the Canadian parties have used leadership election as a method of recruiting new members. This process has intensified with the adoption of direct membership election. In the course of a typical Canadian leadership contest, party membership numbers increase as much as 100 per cent or more. The greatest preoccupation of leadership candidates during much of the campaign is recruiting would-be supporters to party membership. An example of this is the 2003 Liberal contest in which the party’s membership increased from approximately 300,000 to more than half a million during the course of the formal campaign period (Sokoloff, 2003). The vast majority of these new members were recruited to the party by a supporter of one of the leadership candidates and the available evidence suggests that these new recruits typically left the party shortly after the contest (Cross and Young, 2004). The British parties do not use leadership campaigns to attract new members and there appears to be little mobilization of supporters into the parties during these contests. Instead, in party-building terms, the British parties view the contests as a way of increasing the influence of their existing membership with the objective of increasing the saliency of party membership and activism. Seyd (1999) argues that the increased role of individual members in leadership selection in the Labour Party was part of a broader strategy to make party membership more attractive. The contrasts between the two countries’ approaches to participation in leadership elections are manifested in the contest rules. A key difference is the length of the campaign period. In the British parties, campaigns tend to last no more than a couple of months while in Canada they often last a year or more. Another key difference, as illustrated in Table 8.3, is that in the UK only those who belong to the party at the outset of the leadership contest are eligible to vote, whereas the Canadian parties typically encourage mobilization of new members as an integral part of leadership campaigns. In Canada, the influx of new party members during the course of the campaign means that the views of a party’s committed activists are diluted in importance as they may be swept aside in a tidal wave of new supporters joining a party to support their preferred

188 Activating the Citizen

candidate. This has often been a source of resentment for long-time party members. In the Progressive Conservative’s 1998 contest, the winning candidate, Joe Clark, referred to the supporters of his nearest rival, David Orchard, as ‘tourists’ in the party. Not surprisingly, the different approaches to member recruitment are related to the turnout rates in these contests. As shown in Table 8.3, in the thirteen contests for which we have data, the participation rates range from 25 per cent to 79 per cent of a party’s membership. One partial explanation for this variance is that there is an inverse relationship between the mass recruitment of new party members and the rate of participation. This makes intuitive sense because newly committed members may have a very weak personal attachment to the process, while long-serving members are expected to have a stronger commitment to the party. The data do seem to support this hypothesis, as contests featuring mass recruitment have lower rates of participation. In the Canadian cases the average participation rate is 49.6 per cent, compared to 70.5 in the British cases. For comparative purposes it is useful to think of participation in terms of the percentage of a party’s voters who participate in the leadership contest. This is meaningful, as a party’s share of voters in the prior election gives us some indication of how many partisans they have in the electorate. We are then interested in knowing what percentage of these voters participate in their preferred party’s leadership contest. The numbers among our cases are remarkably consistent, ranging between 1 and 3 per cent. This can only be interpreted as evidence that very few of any party’s partisans participate in leadership contests. In the UK, this may be partially attributable to the rules that limit participation in the leadership contest to existing members of the party. However, in Canada, the parties actively encourage partisans to join after the call of the leadership campaign for the purpose of participating in the leadership vote. The data presented above illustrate that relatively few heed this call.

Conclusion When considering how to increase citizen participation in the democratic process, examination of the internal operations of political parties is a key component. Parties are at the centre of democratic politics. They dominate election campaigns, organize legislatures, and recruit and, to a large extent, select elected representatives. The selection of leaders, who become candidates for the prime ministership, is one of the most consequential tasks undertaken by parties. In an era when power is increasingly

Table 8.3: Participation rates∗ and member recruitment rules for leadership contests in UK and Canadian political parties, 1994–2008 Party (year)

Canadian parties Liberal Party (2006) Conservative Party (2004) Liberal Party (2003) New Democratic Party (2003) Canadian Alliance (2002) Canadian Alliance (2000) Progressive Conservative Party (1998) UK parties Liberal Democrats (2007) Liberal Democrats (2006)

Number of days allowed for member recruitment**

Number of voters

95 84 115 155 78 84 149

100,000 (approx.) 92,000 133,000 44,707 88,228 120,557 47,089

15 16

41,465 52,036

Percentage of members voting

As a percentage of supporters in prior general election

50 (approx.) 36 25 54 71 59 52

2.0 N/A 2.5 3.1 2.7 3.7 1.9

64 71

0.7 0.9 (Continued)

189

190

Table 8.3: Continued Party (year)

Conservative Party (2005) Conservative Party (2001) Liberal Democrats (1999) Labour (1994) Country averages Canadian averages UK averages ∗ In

Number of days allowed for member recruitment**

Number of voters

Percentage of members voting

As a percentage of supporters in prior general election

0 0 17 0

198,884 256,797 51,021 172,356

78 79 62 69

2.3 2.7 0.9 1.5

109 8

89,369 128,760

49.6 70.5

2.7 1.5

some cases we are dependent upon media reports and cannot be fully certain of the accuracy of the numbers. is calculated by measuring the number of days between the opening of nominations or start of the official campaign and the date by which one must be a party member in order to be eligible to vote. We have recorded ‘0’ for cases in which the eligible to vote date is prior to the beginning of the contest.

∗∗ This

William Cross and John Crysler

191

concentrated in the centre, the party leader has significant authority. Thus, we believe it is appropriate that parties have expanded the selectorate for the choice of leader beyond their parliamentary caucuses to include their grassroots activists. Our examination of this change and the selection processes adopted by the parties allows us to make several observations. The first is that parties are far more likely to revise their selection process to include rank-and-file members when they are electorally weak. This is consistent with Michels’ (1966) observation that the closer a party is to government the greater the concentration of power in the party’s elite. Second, we observe that parties have adopted very different methods of leadership selection that reflect their particular democratic ethos and electoral imperatives. We identify three broad distinctions: The British parties are concerned with preserving a significant role for the parliamentary party, the Canadian parties with maintaining regional balance in the selection, and the parties of the left with preserving a role for organized labour. In each case, the parties are able to craft rules that permit them to include their members in the leadership choice without having to abandon other principles that are important to them. Despite the frequent trumpeting of the openness and accessibility of their new leadership selection rules, all of the parties we have examined here have relatively low rates of voter participation. In almost all cases, fewer than 1 in 30 of a party’s electoral supporters have voted in the leadership contest. The question of why so few partisans participate in these important contests, and what steps the parties may take to increase participation rates, is worthy of further investigation. It is worth noting that American leadership contests (presidential primaries), which are organized by the state and considerably more accessible than the Canadian and British party-run affairs, attract a significantly higher percentage of the parties’ supporters to the polls, often in the 25 per cent range. Given the important consequences of party leadership selection in parliamentary democracies, perhaps the time has come for considering state regulation of these contests to make them more accessible. Finally, some have argued that an expansion of the leadership selectorate to include rank-and-file members serves to enhance the power of the leader (Mair, 1994). The argument is that by turning over leadership selection to a diffuse and disparate membership, there is no permanent body to whom the leader is regularly accountable. While more empirical work is needed to determine the validity of this thesis, it is worth noting that leaders chosen through membership votes in the UK and Canada

192 Activating the Citizen

have resigned their positions as leader when faced with significant opposition from the parliamentary caucus (Charles Kennedy in the UK and Stockwell Day in Canada). While the caucus may not have the formal authority to select (and in some cases to remove) leaders, the dictates of the parliamentary system make it very difficult, if not impossible, for a leader to continue in his or her position for any length of time in the face of significant parliamentary opposition.

Notes 1. Our examination focuses on the three national parties (and their predecessors) in each country. This excludes parties that compete only on a regional basis such as the Scottish National Party, Plaid Cymru, and the Bloc Québécois. 2. To be elected on the first ballot, a candidate required a majority of the weighted votes. If a second ballot was necessary, the top three candidates were included and a plurality of the weighted votes was required for victory.

References Alderman, K. (1999) ‘Revision of Leadership Election Procedures in the Conservative Party’, Parliamentary Affairs, 52(2), 260–74. Archer, K. and A. Whitehorn (1997) Political Activists: The NDP in Convention (Toronto: Oxford University Press). Brogan, B. (2003, May 2) ‘Tory’s Shot at IDS May Have Missed its Mark’, Daily Telegraph, 10. Carty, R. K. (1988) ’Campaigning in the Trenches: The Transformation of Constituency Politics’, in G. Perlin (ed.) Party Democracy in Canada: The Politics of National Party Conventions (Scarborough, ON: Prentice Hall), pp. 84–96. Carty, R. K. and D. E. Blake (1999) ‘The Adoption of Membership Votes for Choosing Party Leaders: The Experience of Canadian Parties’, Party Politics 5(2), 211–24. Carty, R. K. and W. Cross (2009) ‘Political Parties and the Practice of Brokerage Politics’ in J. C. Courtney and D. Smith (eds) The Handbook of Canadian Politics (New York: Oxford University Press). Courtney, J. (1973) The Selection of National Party Leaders in Canada (Hamden, CT.: Archon Books). Courtney, J. (1995) Do Conventions Matter? (Montreal: McGill-Queens University Press). Cross, W. (1996) ‘Direct Election of Provincial Party Leaders in Canada, 1985– 1995: The End of Leadership Convention?’, Canadian Journal of Political Science, 29(2), 295–315. Cross, W. (2004) Political Parties (Vancouver: UBC Press). Cross, W. and L. Young (2004) ‘The Contours of Political Party Membership in Canada’, Party Politics, 10(4), 427–44.

William Cross and John Crysler

193

Cross, W. and L. Young (2008) ‘Factors Influencing the Decision of the Young Politically Engaged to Join a Political Party: An Investigation of the Canadian Case’, Party Politics, 14(3), 345–69. Katz, R. S. (2001) ‘The Problem of Candidate Selection and Models of Party Democracy’, Party Politics, 7(3), 277–96. LeDuc, L. (2001) ‘Democratizing Party Leadership Selection’, Party Politics, 7(3), 323–41. Mair, P. (1994) ‘Party Organizations: From Civil Society to the State’, in R. S. Katz and P. Mair (eds) How Parties Organize: Change and Adaptation in Party Organizations in Western Democracies (London: Sage Publications), pp. 1–22. Michels, R. (1966) Political Parties: A Sociological Study of the Oligarchical Tendencies of Modern Democracy (New York: Collier Books). Poguntke, T. and P. Webb (eds) (2005) The Presidentialization of Politics: A Comparative Study of Modern Democracies (New York: Oxford University Press). Punnett, R. M. (1992) Selecting the Party Leader: Britain in Comparative Perspective (New York: Harvester Wheatsheaf). Quinn, T. (2004) ‘Electing the Leader: The British Labour Party’s Electoral College’, British Journal of Politics and International Relations, 6, 333–52. Russell, M. (2005) Building New Labour: The Politics of Party Organisation (New York: Palgrave Macmillan). Sartori, G. (1976) Parties and Party Systems: A Framework for Analysis, vol. 1 (Cambridge: Cambridge University Press). Seyd, P. (1999) ‘New Parties/New Politics? A Case Study of the British Labour Party’, Party Politics, 5(3), 383–405. Sokoloff, H. (2003, September 22) ‘Turnout is Poor at Delegate Selection Meeting: Less than a Quarter of Liberal Party Members Cast Votes’, National Post, A5. Stark, L. P. (1996) Choosing a Leader: Party Leadership Contests in Britain from Macmillan to Blair (Basingstoke: Macmillan – now Palgrave Macmillan). Stewart, D. K. and R. K. Carty (2002) ‘Leadership Politics as Party Building: The Conservatives in 1998’ in W C. Cross (ed.) Political Parties, Representation and Electoral Democracy in Canada (Don Mills, ON: Oxford University Press), pp. 55–67. Whitaker, R. (1977) The Government Party: Organizing and Financing the Liberal Party of Canada: 1930–58 (Toronto: University of Toronto Press). Young, L. and W. Cross (2002) ‘The Rise of Plebiscitary Democracy in Canadian Political Parties’, Party Politics, 8(6), 673–99.

This page intentionally left blank

Part IV Beyond Elections and Parties: Innovations in Activating Citizens

This page intentionally left blank

9 Participation and the Good Citizen Jon H. Pammett

Although citizenship is a concept with a variety of meanings and dimensions, participation usually occupies a central position, since it provides a standard by which to evaluate the democratic nature of citizenship and also addresses directly the process by which citizenship is carried out (Giesen and Eder, 2001). The recognition of individuals as constituent members of a political community carries with it not only a set of legal and social rights (Marshall, 1950) but also the right to fully take part in some way in the public life of that community. In many models of democracy, participation is not only a right, but an obligation (MacPherson, 1977; Barber, 1984; Held, 1996). Social capital theorists (Putnam, 2000; Norris, 2002) watch the nature and extent of public participation closely for clues to the overall health of the community. If nothing else, widespread public interest and participation in the public life of a community creates legitimacy for it. The citizenship supported by participation has traditionally been keyed to the boundaries of the nation-state. With the establishment of extra-national agreements, alliances, or unions of states, questions have been raised as to whether the idea of citizenship can be applied beyond national boundaries. These questions have been raised most directly in recent years with respect to Europe. In a legal sense, of course, citizenship beyond borders has already occurred, with the series of agreements and treaties that have expanded the European Economic Community to the European Community to the European Union (EU). But the elements of European citizenship based on public feelings of identity with Europe as a whole and public participation in specifically European political activities have been more problematic. Many of the more positive statements about the emergence of European citizenship focus on institutional and policy development on the continental level (Meehan, 1993; 197

198 Activating the Citizen

Weiner, 1998), leaving aside the dimensions of public identification and participation – hence, the ‘democratic deficit’ so often mentioned. The creation of an exclusive ‘European political identity’ would be a very stringent test for the existence of a European citizenship, which would legitimate the existence of the EU (Eder, 2001). Public opinion data show that very few people (7 per cent overall) consider themselves ‘European only’ when offered alternatives involving their own nationality solely or in combination with Europe (Beetham and Lord, 1998). Similarly, fewer than 20 per cent of the population of most European nations consider themselves ‘very attached’ to Europe (McLaren, 2006). In addition, low and/or declining voting rates in European Parliament elections suggest a lack of identification with Europe as a whole. There is a long list of deficiencies of European institutions in terms of opportunities for public engagement (Nentwich, 1998; Schmitter, 2001). A recent ‘democratic audit’ of the EU found that its institutions were ‘poorly linked to the citizen’ (Lord, 2004, p. 222). On a more optimistic note, despite the fact that ‘today’s Europe is more oriented towards economic and commercial values’, people with postmaterialist values identify more with Europe (as opposed to their own nation) than do materialists (Dell’Olio, 2005, pp. 114–15). Furthermore, surveys also reveal substantial pride in ‘being European’ and majorities in most countries trust people from other EU-member countries (though these measures were taken before the most recent round of EU expansion) (Scheuer, 1999). Schmitt and Thomassen (1999) suggest that ‘valence issues’ (those of ‘common concern’), such as reducing unemployment, fighting crime, and protecting the environment, can serve as pan-European issues for political parties and help to create a meaningful party system for Europe. Similarly, it has been noted that ‘the Left has attempted to conceptualize the main social problems as instrumental to the development of citizens’ consensus . . . [which] can only be reached if citizens see the EU as a unity fully capable of tackling social problems through a common response’ (Dell’Olio, 2005, p. 80). Canada, too, has seen its share of speculation about whether citizenship can transcend boundaries. Throughout the country’s history, these questions have involved the regional basis of Canadian identity, asking whether Quebeckers, or Albertans, or Maritimers, or even Ontarians, could think more in terms of the country as a whole than their specific area. More recently, the passage of the North American Free Trade Agreement has raised the possibility that Canadians might come to think more in terms of a North American identity than was previously the case. This treaty, and a North American democratic deficit that accompanies

Jon H. Pammett 199

it, may simultaneously erode confidence in the power of the national government, because it does not provide adequate avenues for citizen input on the continental level (Clarkson, 2000). While the constraints of continental policy making are not as rigid in North America as they are in Europe, the effects of extra-national political, social, and economic integration are important to explore. This chapter will utilize an international data set to explore two subjects: the qualities associated with ‘the good citizen’ and the extent and nature of political participation. While the questions are not explicitly oriented to participation in pan-European or North American activities, or to qualities associated with extra-national citizenship, they are of sufficient generality that some associations may be drawn. The data used here are from 2004–05 interviews done in the 37 countries that comprised the membership of the International Social Survey Programme (ISSP) at that time. More information about the ISSP surveys may be found at http://www.issp.org. The structure of the chapter is straightforward. I will situate the discussion of citizenship by examining measures of the ‘Good Citizen’ in the ISSP data, and then identify the underlying structure of public opinion on the obligations of citizenship in Canada, the old EU countries, and the new EU members. Then, I will turn my attention to participation, considering the connections between different kinds of political action. Finally, I will look at the correlations between the different dimensions of good citizenship and the orientations to participation. In the conclusion, I will consider whether the findings may have applicability for the question of the development of European citizenship or extra-national citizenship more generally.

The good citizen The duties or obligations of the good citizen form the flip side of the rights that accrue to those who have achieved citizen status in a community. There have been a number of theoretical and empirical attempts to codify the elements of good citizenship. For example, Heater (1990) says, The good citizen has more than loyalty; he has a sense of responsibility to take positive and supportive actions. . . . The good citizen performs his legal duties, such as paying taxes or military service not for fear of the consequences for himself of avoidance but for fear of the consequences for his country. . . . The good world citizen treats

200 Activating the Citizen

the whole planet in the same way . . . recognis[ing] his obligation to serve in his own small way the world and its peoples. (p. 197) Denters, Gabriel, and Torcal (2007) identify three components from the international Citizenship, Involvement, and Democracy project: solidarity, critical and deliberative principles, and law-abidingness. Westheimer and Kahne (2004) sort conceptions of good citizenship into three types: the personally responsible citizen, the participatory citizen, and the justice-oriented citizen. Janoski’s (1998) model identifies four types of citizen obligations: support, caring, service, and protection. In his examination of citizenship in the United States, Dalton (2008) refers to two dimensions: duty and engagement. Despite the different terminology, there are broad similarities between many of these conceptions of citizenship. To some extent, the empirical studies of the duties of the good citizen are constrained by the choices of duties presented to the respondents. Thus, Denters et al., referred to in the previous paragraph, included in their test battery items dealing with the importance of forming one’s own opinion and being self-critical. These two items formed a distinct factor in their results. In the case of the ISSP questions on the obligations of citizenship, ten questions were used, but these were a subset of a larger battery, which could not be fully included because of space limitations. Nevertheless, the results provide a coherent set of citizenship duties, even if not necessarily complete. Table 9.1 gives the mean scores of importance (on a 7-point scale) for the ten statements about ‘what it takes to be a good citizen’. For our eventual purpose of considering the potential development of a supranational citizenship, it should be noted that this set of questions does not generally ask people to answer in the context of their own state. An exception is the question asking about the importance of helping those worse off in their country, which is accompanied by a question asking about those worse off in the rest of the world. The item about serving in the military also has national implications. Most of the others, however, deal with values and actions on a more general level. Items dealing with the law-abiding aspects of good citizenship lead the list in terms of public perceptions of importance. They are followed by the importance of voting in elections; indeed, this attitude is often used as a measure of ‘civic duty’ to explain trends in voting turnout. Despite evidence of declines in the belief about the civic duty of voting, it appears high here. Political consumerism, and social and political activity in associations are at the bottom, but these are still well above the halfway point on the scale.

Jon H. Pammett 201 Table 9.1:

Duties of the good citizen, pooled ISSP data

Always to obey laws and regulations Never to try to evade taxes Always to vote in elections To help people in [country] who are worse off than yourself To try to understand the reasoning of people with other opinions To keep watch on the actions of government To help people in the rest of the world who are worse off than yourself To serve in the military at a time of need To choose products for political, ethical or environmental reasons, even if they cost a bit more To be active in social or political associations

Mean (out of 7)

Standard deviation

6.2 6.0 5.8

1.2 1.4 1.7

5.7

1.4

5.7 5.4

1.4 1.6

5.0 4.6

1.8 2.2

4.4 4.2

1.9 1.9

The structure that lies behind the measures we have of good citizenship is presented in Table 9.2, in three different forms. First, Canada is shown, to establish the basic structure and point out the underlying dimensions of citizens’ conceptions of their most important duties. Second is that group of countries from the ISSP surveys that constitutes the ‘old EU’. Either original members or those that joined in the intermediate stages, they are generally the more established countries of western, northern, and southern Europe. This does not include all of the more established EU members – unfortunately, Italy’s ISSP membership was in flux at this point and no survey was conducted – but still gives us a substantial number of cases. Finally, the ‘new EU’ is shown, primarily eastern European countries, plus (southern) Cyprus. The factor analysis in Table 9.2a shows a clear and comprehensible division of Canadian opinion about what it takes to be a good citizen into three dimensions, labelled here as service, activity, and obedience. The service factor consists of four of the ten items, grouping the judgements that it is important to help people in Canada and in the rest of the world who are worse off than the respondent with the beliefs that it is important to choose products for political, ethical, or environmental reasons even if they cost a bit more, and that it is necessary to try to understand people with other opinions. It is instructive to note that the desire to help alleviate poverty at home and abroad is associated with conscious

202 Table 9.2:

The good citizen: factor structure

a) Canada Vote Pay taxes Obey laws Watch government Active in associations Understand others Political consumerism Help worse off (country) Help worse off (world) Serve in military FACTOR: b) old EU* Vote Pay taxes Obey laws Watch government Active in associations Understand others Political consumerism Help worse off (country) Help worse off (world) Serve in military FACTOR:

1

2

3

.02 .08 .08 .00 .27 .45 .53 .83 .85 .15 SERVICE

.58 .11 .09 .80 .71 .35 .47 .07 .01 .26 ACTIVITY

.37 .85 .88 .15 .04 .14 − .07 .17 .07 .33 OBEDIENCE

1

2

3

.06 .14 .11 .20 .42 .57 .67 .83 .84 .04 SERVICE

.74 .15 .17 .74 .63 .27 .30 .04 .04 .32 ACTIVITY

.21 .83 .85 .15 − .05 .08 − .04 .21 .14 .25 OBEDIENCE

*countries: Germany, United Kingdom, Austria, Ireland, the Netherlands, Sweden, Spain, France, Portugal, Denmark, Finland, Flanders (The ISSP survey for Belgium was conducted only in Flanders.)

c) New EU* Vote Pay taxes Obey laws Watch government Active in associations Understand others Political consumerism Help worse off (country) Help worse off (world) Serve in military FACTOR:

1

2

.30 .11 .14 .50 .61 .63 .63 .80 .80 .48 SERVICE/ACTIVITY

.61 .84 .83 .49 .26 .24 .12 .14 .07 .24 OBEDIENCE/VOTE

*countries: Hungary, the Czech Republic, Slovakia, Slovenia, Poland, Latvia, Cyprus Note: Shading identifies factor loadings above .4 Method: principal components, varimax rotation.

Jon H. Pammett 203

consumer choice and also with a belief in tolerance and understanding of those people who do not agree with them. The second factor in the making of the good Canadian citizen is one that incorporates various aspects of action and attentiveness. The good citizen is one who always votes in elections, is active in social or political associations, keeps watch on the actions of government, and makes an effort to choose products for ‘correct’ reasons even if they cost a little more. It will be noted that this ‘political consumerism’ item loads on both the service and activity factors. Inspection of the factor structure of opinions about good citizenship in the other ISSP countries considered here shows that ‘political consumerism’ is sometimes associated with either the service or activity factor and sometimes associated with both factors. Finally, there is the factor here called obedience, which captures beliefs that laws should be obeyed without exception and that people should not attempt to evade their taxes. This characteristic of the dutiful citizen – supportive of the legitimately promulgated rules and regulations of the state – is sometimes thought to be the primary responsibility of the citizen. This is a passive, rather than active, view of citizenship. Sometimes the factor of obedience includes the obligation to vote. For citizens, however important this dimension of law-abidingness may be, it does not emerge as the first factor underlying the structure of citizen duty attitudes, but rather the last. The pattern seen with regard to Canada is similar to the pooled data from twelve countries that have been members of the EU for over a decade, as shown in Table 9.2b. The service, activity, and obedience dimensions are essentially the same as those already seen. One exception to this pattern concerns the ‘political consumerism’ item, which in the old Europe is associated only with the service/helpfulness dimension, and not with activity per se. Another interesting item to note is that, as in Canada, the perceived obligation to serve in the military during a time of need is not associated with any of the other citizen duty items. In the new Europe, we have a less differentiated pattern of citizen duties, as shown in Table 9.2c. Here, the service and activity dimensions are essentially fused together, as six (or even seven if one counts the lower correlation of the obligation of military service) items correlate with the background dimension. The other factor is a slightly expanded obedience factor, with the obligation to always vote in elections combined with the necessity of paying taxes and obeying laws. In the new EU countries, most of which are in eastern Europe, there has been extensive regime change in the last decade and a half. In these countries, there is yet to

204 Activating the Citizen

emerge a distinct ‘service’ dimension of public opinion regarding citizen duty. An examination of correlates of these three orientations to citizen duty with age and education show that in all three country groupings (Canada, old EU, new EU), older citizens are more likely to rate the activity and obedience dimensions higher. These more traditional conceptions of citizen obligations are also rated as more important by those with lower education. When it comes to the service dimension, however, there are very slight tendencies for younger people to rate it as more important. Similarly, there is a small but positive correlation with years of education. So we may conservatively say that the desire for service cuts across the age and educational spectrum of the societies under observation and, if anything, is associated very weakly with those who are younger or more educated. With this demographic profile, service is likely to come more into focus as the years go by. It appears, from this initial examination of the dimensions of citizenship, that the service or helpfulness dimension is the one that is most likely to extend the horizons of citizen duty beyond the nation-state. Feelings of duty to aid in the alleviation of poverty, to play a part in making responsible consumer decisions, or to reduce burdens on the environment identify social purposes for action of a broader scope. I will later consider the kinds of participation that can be associated with the dimensions of good citizenship, with particular attention to the service dimension.

Patterns of participation As with the measures of good citizenship, a selection of types of participation was included in the ISSP surveys. Respondents were asked whether they had done these things in the past year and in the more distant past, as well as whether they might do them or would never do them. Considering the most frequently reported act, just over one-fifth of all the survey respondents in the countries we are analysing reported having signed a petition in the past year, and another quarter said they had done so in the more distant past. The next most popular activity involved donating money or raising funds ‘for a social or political activity’, a rather broad statement that might be expected to encompass quite a few possible actions. Measuring political consumerism, a question asking whether people had ‘boycotted, or deliberately bought, certain products for political, ethical, or environmental reasons’ brought agreement from about a quarter of the overall ISSP samples. Less cited were participation in

Jon H. Pammett 205

demonstrations and attendance at political meetings or rallies, though many more people reported doing these things in the more distant past than in the past year. A fairly small number had contacted or appeared in the media to express their views, and a very small number had joined an Internet political forum or discussion group. More important for our current purposes than the numbers of people who reported engaging in various types of political participation, however, is the interrelationship between them. Table 9.3 presents the correlations between the items available to us. In all cases, these correlations are positive, and in many cases they are of substantial strength. In addition, this table adds at the bottom the report as to whether the respondent voted in the most recent national election. These correlations are also positive, though at lower levels of association. The one exception is at the very bottom right of the table, where it is shown that there is no relationship at all between voting and joining an Internet discussion group. The top part of the correlation matrix, shaded in Table 9.3, displays the same relationships for that part of the total sample aged 25 and under. The overall pattern is very similar to the totals for the entire samples, displaying strong and positive connections between all types of political participation. As far as voting is concerned, there are once again positive and statistically significant relationships between voting and all other forms of participation with the exception of the Internet. There may indeed have been changes in the nature of political participation over time, and there may be some differences between the types of activities favoured by young and old (Norris, 2002), but the overall picture is quite consistent for all age groups; those who engage in one kind of participation are more likely to engage in another. These findings about the interconnected nature of political participation are not unique. In a study of non-voting done for Elections Canada in 2002, voting was found to relate to all other kinds of participation measured, as well as membership in a wide variety of social groups (Pammett and LeDuc, 2003). The authors of the Canadian Democratic Audit series book on citizens say, ‘One of the most striking findings to emerge from this audit is the extent to which different forms of political engagement tend to go together’ (Gidengil, Blais, Nevitte, and Nadeau, 2004, p. 172). In the European Science Foundation-funded study on citizenship, involvement, and democracy, the findings for thirteen European countries were unequivocally in the same direction. The implications for the study of political participation are very clear. As the authors of the above cited study put it,

206

Table 9.3: Correlations between political actions in pooled ISSP data (Tau b) Petitions Boycotts Demonstrations Meetings

Petitions Boycotts Demonstrations Meetings Contact politicians Donate/raise Contact media Internet Voted

.41 .44 .36 .30 .33 .30 .29 .23 .14

.36 .33 .36 .37 .34 .30 .12

Notes: *relationship not statistically significant at < .01. Shading identifies those aged 25 and under.

.34 .35 .48 .32 .28 .33 .31 .08

.25 .33 .40 .46 .34 .37 .33 .15

Contact Donate/ Contact politicians raise media .26 .35 .26 .46 .34 .50 .36 .14

.30 .36 .26 .33 .32 .34 .27 .14

.23 .29 .25 .36 .47 .30 .48 .08

Internet

Voted

.19 .26 .25 .36 .39 .24 .46

.14 .09 .07 .12 .09 .09 .07 .02*

.01*

Jon H. Pammett 207

This means that citizens in all 13 societies under study see no tradeoff in their choice of mode of participation. It is not the case, as is commonly held, that some citizens continue to use the traditional channels of participation, such as contacting officials and working for a political party, whereas others favour the less conventional modes such as protest or consumer participation. On the contrary, activists within one mode of activity tend to be activists within the others as well. (Teorell, Torcal, and Montero, 2007, p. 334)

It could be added that, whatever generation gaps might exist in political attitudes and behaviour, this is not one of them. The coherence of the underlying structure of participation is further reinforced by the fact that, when the pooled data from Canada and the EU are subjected to a factor analysis, the eight participation items in the ISSP data set form a single factor, which is shown in the second column of Table 9.4a. Actions ‘conventional’ or ‘less conventional’, ‘orthodox’ or ‘less orthodox’ (Dalton, 2006) are for many people part of the same dimension. Table 9.4b lists those countries of interest in the current context that have only one underlying participation factor. Although the pooled data produce a single factor within standard limits, the procedure can force out a second factor for examination purposes. This two-factor result is presented in the last two columns of Table 9.4a. It divides participatory actions into those associated with the politics of representation and media utilization (attending meetings or rallies, contacting public officials, contacting the media, using the Internet) and the politics of direct action (signing petitions, engaging in boycotts, attending demonstrations, donating or raising money). Table 9.4b lists those countries in which this pattern appears from the factor analysis on its own, that is, where two factors appear without being forced. In the concluding analysis in this paper, I will utilize these two participation factors for these countries when examining the connections of participation with conceptions of the good citizen. The empirical distinction between a preference for direct action and a reliance on representatives and media fits well with distinctions in democratic theory between representative and participative democracy. It is worth remembering, however, that they are both orientations towards political actions of some sort, as opposed to passive political orientations, held by those who trust elites to govern and have minimal political interest. These people, recently called ‘decisionists’ (Martin and van Deth, 2007), limit themselves to voting (at times).

208 Activating the Citizen Table 9.4: Factor structure of participation, components) a) Overall

Petition Boycott Demonstration Meeting Contact Donate/raise Media Internet

Single factor

.66 .69 .66 .71 .72 .64 .70 .59

pooled ISSP data (principal

Two forced factors Media/ representative .11 .19 .23 .47 .66 .32 .79 .77

Direct action .79 .74 .67 .53 .38 .57 .24 .10

Shading identifies factor loadings above .4 b) By country: Countries with single participation factor

Countries with two participation factors (unforced)

Germany United Kingdom Ireland Hungary Spain Portugal Slovenia Poland Bulgaria

Canada Austria Netherlands Sweden Czech Republic Latvia Slovakia France Cyprus Denmark Flanders Finland

Participation and good citizenship So far I have examined the underlying structure of both political participation and good citizenship. For the former, I found that for a subgroup of countries in Europe, and for Canada, there are two participation dimensions, direct action and using mediated or representative participation channels. For the latter, I found another subgroup of countries in which good citizenship can be considered to have three underlying dimensions, service, activity, and obedience. Considering the two sets of results, I can identify eleven countries where the prerequisites

Jon H. Pammett 209 Table 9.5: Participatory patterns* for dimensions of good citizenship (correlations of factor scores) Service

Activity

Obedience

With direct action

With media/representative

Usually low or negatively associated with both

Canada Austria Netherlands Sweden France Denmark Finland Flanders Latvia Czech Republic

.293 .268 .357 .278 .328 .377 .385 .251 .114 .07

Canada Denmark Flanders Finland France Sweden Latvia Czech Republic

With media/representative

With direct action

Latvia Slovakia Czech Republic

Netherlands Slovakia

.173 .114 .08

.296 .246 .264 .341 .212 .216 .192 .117

.119 .246

Note: Only statistically significant correlations < .01 reported. *For countries with two participatory factors.

exist to correlate the two sets of factors. These are Canada, Austria, the Netherlands, Sweden, France, Denmark, Finland, Flanders, Latvia, the Czech Republic, and Slovakia. Some of these countries are in the old EU and some in the new EU. Though no sample of countries can be considered representative, when considering the inter-country differences in culture and experience with EU participation, we may use this group of states to examine an underlying hypothesis that can shed some light on the possibilities of extra-national citizenship. The hypothesis of interest is this: The service dimension of good citizenship will correlate with a direct action approach to participation. In addition, it is hypothesized that the activity dimension of good citizenship, based as it is on the necessity of voting, keeping watch on government, and association membership, will be more likely to correlate with the other dimension of participation, which is founded on attention to media and participation through representatives. This latter approach is more likely to be concentrated within the boundaries

210 Activating the Citizen

of the nation-state, whereas the activities associated with direct action participation are broader in scope. With good citizenship, the service orientation is also likely to extend beyond the national boundaries in the case of one of its components, helping those worse off throughout the world. The participation patterns for the dimensions of good citizenship are displayed in Table 9.5. For the majority of the countries for which the variables are in the appropriate form, the hypothesis is upheld whereby the definition of good citizenship in terms of service to others is connected with a direct action approach to participation. This relationship is found in virtually all the countries under examination. Two of the three countries (Latvia and the Czech Republic) where the other relationship – that between a service orientation and a media/representative orientation to participation–exists also have one with service. That only leaves one country of the eleven (Slovakia) with such a primary connection of service with mediated participation.

Conclusion ‘Most people are apparently involved in the political sphere in ways consistent with their citizenship perspective’ (Theiss-Morse, 1993, p. 370). If feelings of commitment to a citizenship beyond the nation-state are to develop or be nurtured, they must be based on something other than a willingness to abide by laws or to evaluate the direct financial benefits/losses from services or taxes. They also need to have an overarching social purpose, in addition to the voting, attentiveness to government actions, and association memberships that may signal a reasonably active individual within the national context. This investigation has shown that the existence of a social purpose for citizen duty is associated with participation techniques involving direct action, measured here as petitions, boycotts, demonstrations, and fundraising. If direct action is the key to the expansion of citizen consciousness beyond the national level, what are the prospects for EU citizenship, in the context of public identification with the EU in this manner? The continued dependence of the EU on coordinated economic policies to justify its existence may engender obedient citizens but not engaged and committed citizens who identify with the community itself. The provision of voting opportunities for the European Parliament may tap into the activity dimension of citizenship, but unless these elections can address more general social issues, voting by itself will not advance the cause

Jon H. Pammett 211

of European citizenship. In fact, the concentration of European Parliament elections on national issues may inhibit it. The activity dimension of citizenship can also be linked to participation in organizations and through the media, but what is the activity for? The service orientation to citizenship calls for a purpose behind the activity, particularly rooted in larger causes, such as the alleviation of poverty at home and abroad and the creation of a better environment. The establishment of a citizen identity based on service is the main possibility for generating EU citizenship, but the public preference for direct action in achieving such goals would make a very uneasy fit with current or future EU institutions, unless these are of a highly democratic and participative nature. The challenge for the EU, as it is for North America, is to develop institutions that will accommodate the public desire to participate personally in political and social decisions. The literature is not bereft of suggestions for participative EU policies or institutions. Schmitter (2001) suggests a system of ‘Euro-sabbaticals’ whereby citizens are granted paid leave from their employment for periods of time in order to engage in a variety of extra-national activities of individual or social benefit. Nentwich (1998) proposes a Europe-wide communication network to provide information and connect activists, enhanced polling, widespread public policy consultations and public hearings, sponsorship of trans-European political parties, greater use of referenda and the implementation of a public initiative procedure for holding them, and even a recall procedure for European commissioners. One drawback to the serious consideration of these and other initiatives comes from the fact that the service orientation to citizenship and the desire for enhanced direct participation opportunities is not universal. The distinct identification of a service dimension to citizenship is found mostly within the older EU countries. The new EU fuses the service dimension with the activity one, possibly meaning that national institutions and mediated action through representatives will be relied on to achieve it. EU expansion may have blunted the potential for a broader citizenship perspective that transcends the legal and economic basis of the community. On the North American stage, the chances for implementing mechanisms for innovative citizen action seem highly dependent on the political complexion of future national governments, as there are no central institutions powerful enough to create them and the tradition of governments voluntarily giving up power to the people is not established. It may take further participation declines and an atmosphere of crisis before action can succeed.

212 Activating the Citizen

References Barber, B. (1984) Strong Democracy (Berkeley: University of California Press). Beetham, D. and C. Lord (1998) ‘Legitimacy and the European Union’ in A. Weale and M. Nentwich (eds) Political Theory and the European Union (London: Routledge), pp. 15–33. Clarkson, S. (2000) ‘Do Deficits Imply Surpluses? Toward a Democratic Audit of North America’ in M. Th. Greven and L. W. Pauly (eds) Democracy Beyond the State: The European Dilemma and the Emerging Global Order (Toronto, ON: University of Toronto Press). Dalton, R. J. (2006) Citizen Politics, 4th edn (Washington, DC: CQ Press). Dalton, R. J. (2008) The Good Citizen: How a Younger Generation Is Reshaping American Politics (Washington, DC: CQ Press). Dell’Olio, F. (2005) The Europeanization of Citizenship (Aldershot: Ashgate). Denters, B., O. Gabriel, and M. Torcal (2007) ‘Norms of Good Citizenship’ in J. W. van Deth, J. R. Montero, and A. Westholm (eds) Citizenship and Involvement in European Democracies (Abingdon: Routledge), pp. 88–108. Eder, K. (2001) ‘Integration through Culture? The Paradox of the Search for a European Identity’ in K. Eder and B. Giesen (eds) European Citizenship between National Legacies and Postnational Projects (Oxford: Oxford University Press), pp. 222–44. Giesen, B. and K. Eder (2001) ‘Introduction: European Citizenship’ in K. Eder and B. Giesen (eds) European Citizenship between National Legacies and Postnational Projects (Oxford: Oxford University Press), pp. 1–16. Heater, D. (1990) Citizenship: The Civic Ideal in World History, Politics and Education (London: Longman). Held, D. (1996) Models of Democracy (Stanford, CA: Stanford University Press). Gidengil, E., A. Blais, N. Nevitte, and R. Nadeau (2004) Citizens (Vancouver: University of British Columbia Press). Janoski, T. (1998) Citizenship and Civil Society: A Framework of Rights and Obligations in Liberal, Traditional and Social Democratic Regimes (Cambridge: Cambridge University Press). Lord, C. (2004) A Democratic Audit of the European Union (Basingstoke: Palgrave Macmillan). MacPherson, C. B. (1977) The Life and Times of Liberal Democracy (London: Oxford University Press). Martin, I. and J. W. van Deth (2007) ‘Political Involvement’ in J. W. van Deth, J. R. Montero, and A. Westholm (eds) Citizenship and Involvement in European Democracies (Abingdon: Routledge), pp. 303–33. Marshall, T. H. (1950) Citizenship and Social Class and Other Essays (Cambridge: Cambridge University Press). McLaren, L. M. (2006) Identity, Interests and Attitudes to European Integration (Basingstoke: Palgrave Macmillan). Meehan, E. (1993) Citizenship and the European Community (London: Sage). Nentwich, M. (1998) ‘Opportunity Structures for Citizens’ Participation: The Case of the European Union’ in A. Weale and M. Nentwich (eds) Political Theory and the European Union (London: Routledge), pp. 125–40. Norris, P. (2002) Democratic Phoenix (Cambridge: Cambridge University Press).

Jon H. Pammett 213 Pammett, J. H. and L. LeDuc (2003) Explaining the Turnout Decline in Canadian Federal Elections (Ottawa, ON: Elections Canada). Putnam, R. (2000) Bowling Alone (New York: Simon and Shuster). Scheuer, A. (1999) ‘Political Community?’ in H. Schmitt and J. Thomassen (eds) Political Representation and Legitimacy in the European Union (Oxford: Oxford University Press), pp. 25–46. Schmitt, H. and J. Thomassen (1999) ‘Distinctiveness and Cohesion of Parties’ in H. Schmitt and J. Thomassen (eds) Political Representation and Legitimacy in the European Union (Oxford: Oxford University Press), pp. 111–28. Schmitter, P. (2001) ‘The Scope of Citizenship’ in K. Eder and B. Giesen (eds) European Citizenship between National Legacies and Postnational Projects (Oxford: Oxford University Press), pp. 86–121. Teorell, J., M. Torcal, and J. R. Montero (2007) ‘Political Participation: Mapping the Terrain’ in J. W. van Deth, J. R. Montero, and A. Westholm, Citizenship and Involvement in European Democracies (Abingdon: Routledge), pp. 334–57. Theiss-Morse, E. (1993) ‘Conceptualizations of Good Citizenship and Political Participation’, Political Behavior, 15(4), 355–80. Weiner, A. (1998) ‘European’ Citizenship Practice: Building Institutions of a Non-State (Boulder, CO: Westview). Westheimer, J. and J. Kahne (2004) ‘Educating the “Good” Citizen: Political Choices and Pedagogical Goals’, PS: Political Science and Politics, XXXVII(2), 241–7.

10 Institutionalizing Participation through Citizens’ Assemblies Jonathan Rose

The question of why citizens ought to participate in politics is one that is as old as the study of politics itself. As far back as Aristotle, participation was seen as a vital part of the res publica, as a way for citizens to take part in activities that were important for the collective good. The ancient Greeks understood the importance of forms of participation such as the council of five hundred – a randomly selected deliberative body chosen for a fixed period of time and charged with oversight of the assembly. For them, robust participation was not seen exclusively through voting but also through direct involvement in democratic life. The seed of this idea has spread throughout the ages and germinated in Italian city-states, medieval Italian communes, kibbutz, and communes. The advent of representative institutions, the nation-state, and capitalism led to a conception of democracy that favoured a clash of interests. Individuals were assumed to be self-interested and democracy was viewed as the result of competing demands made on the political system. Mansbridge (1980) calls this ‘adversary democracy’. It is at odds with an older version that privileged reciprocity and consensus, which she calls ‘unitary democracy’. Studies of interests within parliament, factionalism, debates within political parties, and consociationalism – to name a few – all assume adversary democracy. The unitary model, on the other hand, more often examines town hall meetings, deliberative forums, and small-scale democratic institutions. Each of these democratic models is based on different assumptions about the nature, calibre, and quantity of participation needed in a democracy, as well as the locus of participation. The political parties and elections that are examined in the earlier parts of this book are key institutions of adversary democracy. The decline of public interest and participation in these institutions may indicate that people are becoming 214

Jonathan Rose

215

more receptive to the unitary model, which calls for quite different, and far more hands-on, forms of participation. Rousseau, whose ‘general will’ was a unitary concept, wrote of democracies where debate in face-to-face settings could reveal a common interest (Mansbridge, 1980). We know from myriad studies how citizens behave in adversary democracies. But we know far less about citizen behaviour, attitudes, and participation in unitary democracy. The older Athenian model of participatory democracy has been revived in the last few decades. According to Snider (2007), its modern variant can be traced to ‘the Yale school of democratic reform’. It began with Dahl’s (1970) After the Revolution, which argued for an institutional arrangement of random citizens to serve as ‘advisory councils’. Later, in Democracy and its Critics, Dahl (1989) elaborated on this arrangement by calling it a ‘minipopolus’, whose ‘task would be to deliberate, for a year perhaps, on an issue and then to announce its choices’ (p. 340). The disciples of this school include Barber and Gastil. Fishkin is arguably the best known of the Yale school. His 1991 work on deliberative polling is one attempt to learn how citizens, initially uninformed about a policy issue, can learn through collective deliberation. While Fishkin’s work is important in examining the impact of citizen education over a short period of time, little work has been done on how citizens learn and participate in deliberative exercises that take place over a longer period. This chapter examines a type of deliberative body that meets over a long period of time and is charged with making policy recommendations that have considerable weight – the citizens’ assembly (CA). A CA takes a group of random citizens of diverse ages, ethnic backgrounds, and socioeconomic status and after an intensive education programme followed by a public consultation and deliberation phase has them make a policy recommendation. Using this definition, there have been only three CAs, two in Canada (in British Columbia [BC] and Ontario) and one in Europe (the Netherlands). All three examined the issue of electoral reform – BC in 2003–04, the Netherlands in 2006, and Ontario in 2006–07. In the two Canadian cases, the results were put to referendums that ultimately failed. In the Dutch case, the government did not act on the CA’s recommendation. For our purpose here, the subject matter of a CA is less significant than its defining characteristic – the quality of its public judgement and deliberation – the focus of this chapter. Public judgement and deliberation are terms in common currency but mean something particular in this case. According to Yankelovich (1991), public judgement is a form of public opinion that exhibits ‘more thoughtfulness, weighing of alternatives, more engagement with the

216 Activating the Citizen

issue, more taking into account a wide variety of factors than ordinary public opinion as measured in opinion polls and more emphasis on the normative, valuing, ethical side of questions than on the factual, informational side’ (p. 5). Deliberation involves the reasoned and enlightened formation of policy preferences, based on a discussion of shared interests. Advocates such as Barber (1984) believe that transformative talk that goes beyond self-interested perspectives lies at the heart of deliberative democracy. Citizens involved in such exercises must eschew choice as simply the aggregation of individual preferences, seeing choice instead as a result of common interests based on the decision rule of consensus. With its emphasis on equality, common good, mutual respect, and organic relationships, the roots of deliberation can be found in friendship rather than competition. This echoes Aristotle’s maxim that ‘friendship [philia] appears to hold city-states together’ (cited in Mansbridge 1980, p. 9). The deliberative approach assumes that participation ought to be non-adversarial, like friendship. While much has been written on the concept of deliberative democracy, ‘there has been far less attention paid to actual efforts to develop forums where citizens can and will talk to one another about public issues’ (Button and Mattso, 1999, p. 610). We simply do not have good data on the calibre and nature of decisions made in bodies where citizens meet to make policy recommendations. CAs are one such venue and, aside from work by Warren and Pearce (2008), there is very little research done on them. This chapter attempts to add to that body of literature.

Why use a citizens’ assembly? Many might question whether a CA constituted of ordinary citizens would be any better than the representative assemblies that we elect. The answers are both theoretical and practical. Because CAs represent an instance of policy making by deliberation, experience with them offers insights into curing the ills of most legislative bodies. The qualities that characterize CAs are the very ones lacking in legislatures. CAs are marked by equality and they are not divided by gender, party affiliation, age, or experience. They have legitimacy because the participants, at least collectively, enter policy making with open minds and full transparency, making decisions only after consultation with fellow citizens. Individual interests are subsumed in the search for the common good. Perhaps most crucially, decision rules are not majority based, which is an arithmetic rule about individual votes; rather, they are consensus based, which sees value in using conversation to reach a common goal. There are also

Jonathan Rose

217

practical arguments in favour of creating CAs. When politicians have a conflict of interest, CAs are an obvious solution. By putting the decision in the hands of citizens, the appearance of bias is removed. On decisions where there is no right answer or when policy makers are divided, CAs might result in policy that is seen as legitimate and defensible. Assemblies also have an important participation spillover. Not only do the participants learn about the trade-offs required in policy making but the public they consult also learns. While it is true that governments engage in consultation exercises routinely, the consultations carried out by CAs are more likely to be seen as authentic and without a pre-set political agenda since they are divorced from partisan interests or political ideology. Creating a CA does not necessarily guarantee these attributes. These are a result of several prerequisites. Assemblies must have time to learn, consult, and deliberate. They must have independence from government and must be transparent. To fulfil these conditions, assemblies must be well resourced in order to ensure diversity in the material considered. Unlike citizens’ juries, participants do not choose between two or more competing options. Assemblies do not choose which expert is most persuasive but rather learn through a neutral presentation of arguments from a variety of perspectives and, most crucially, from themselves. A further criterion is that assemblies should deliberate on issues that are value driven and meaningful rather than on those that are exclusively scientific/empirical. Reliance on a CA makes several assumptions about citizen participation. First, the raison d’être for a CA is that good policy can be made by a group of citizens who are given the appropriate learning tools and deliberative space in which to listen, hear alternative views, and not be invested in any outcome. Second, to be successful, a CA must be diverse in its make-up. This is so that the body is not ‘captured’ by any ideological or socioeconomic position. Surowiecki (2004) suggests that good deliberation is dependent in part on wide sources of knowledge and background. This is important in ensuring that one faction does not dominate but also enriches the calibre of discussion. It appears that both Canadian CAs and the Dutch version met these criteria, although through different methods. In the case of the BC CA, participant selection was made by the Assembly Secretariat, the administrative body that planned, organized, and ran the meetings. In the case of Ontario, it was made by Elections Ontario, the government agency that oversees elections in the province. In both cases, gender balance was achieved. In the Ontario case, a regulation mandated that there be

218 Activating the Citizen Table 10.1: Age distribution of Ontario and BC assembly members and population

18–24 25–39 40–54 55–70 71+ Total

Ontario CA

Ontario population

BC CA

BC population

11% 22% 31% 24% 12% 100% (103 members)

12% 29% 30% 17% 12% 100%

7% 24% 35% 28% 6% 100% (160 members)

11% 26% 32% 17% 14% 100%

Table 10.2: Age distribution of Dutch civic forum Civic forum

The Netherlands

18–30 31–40 41–50 51–60 61–70 71+

17.6% 18.2% 25.8% 21.8% 13% 3.6%

16 22 22 17 13 7

Total

100% (143 members)

100%

at least one Aboriginal member, whereas in BC, an amendment to the terms of reference added two Aboriginal members. Over 120,000 initial letters were sent from Election Ontario’s Register of Electors asking prospective assembly members if they were interested in participating. Of those who received the letter, 7033 responded affirmatively. From this pool, 1253 were invited to one of the 29 selection meetings held across the province where 103 members and two alternates from each electoral district were chosen by random draw. As Table 10.1 shows, in terms of demographics, the assembly members came close to mirroring the population of Ontario. In other ways, too, assembly members were diverse. Collectively, they spoke over 28 languages and 66 of them were born in Ontario, 11 were from other provinces, and 27 were originally from outside Canada. In BC, Elections BC provided the Assembly Secretariat with 200 names – equally divided by gender – per electoral district. Of this pool of 15,800 names, 1441 were invited to attend a selection meeting in their

Jonathan Rose

219

electoral district. This process yielded a very close approximation to the demographics of BC as Table 10.1 shows. The process was slightly different in the Netherlands but the effect of mirroring the population was the same, as Table 10.2 shows. In the Netherlands, 50,000 letters were sent out based on a national database called the Municipal Persons Records Database. From the 4000 responses, 2107 were invited to attend one of the nine selection meetings.

How do citizens’ assembly members learn? Diversity is obviously not just about place of origin. Assembly members represented a broad range of learning styles and educational backgrounds as well, which raised pedagogical challenges. Some members were students enrolled in university- or college-level courses. Many more were recent graduates. A few members had little or no formal education, and many others had not attended school for several years or many decades. Some members were comfortable in a traditional school setting and had no difficulty reading advanced materials. Others preferred learning by listening, talking, or doing. A small group of members learned quickly and easily while a minority initially found learning about electoral systems difficult or the lack of an obvious answer frustrating. Despite this diversity, a majority of the members appear to have learned a great deal by being exposed to a variety of learning techniques and tools. Surveys administered at the beginning and end of the learning phase showed that over that period, the level of the participants’ comfort with the materials increased. In the Ontario case, on a scale of 0 (not informed) to 10 (very informed), members’ self-reported confidence about electoral systems increased from 4.31 before the learning phase to 7.68 after the learning phase.1 In BC and the Netherlands, the data are similar. Citizens’ own perceived capacity to understand the material is not sufficient to claim that an exercise in deliberation was successful. We need to be sure that citizens, with their broad range of educational abilities, were competent in making an informed decision. As far back as Plato, there has been a current of literature questioning the capacity of citizens to fulfil their democratic duty. Early twentieth-century democratic elitists such as Schumpeter (1942) did not believe citizens had the requisite knowledge or interest to play a meaningful democratic role. Lippmann (1925) wrote that the ‘private citizen today has come to feel rather like a deaf spectator in the back row, who ought to keep his mind on the mystery off there, but cannot quite manage to keep awake’ (p. 13). Others, such as Berelson, Lazarsfeld, and McPhee (1954) and more recently Neuman

220 Activating the Citizen

(1986), Converse (1990), and Zaller (1992), have been equally sceptical about citizens’ capacity. So we have good reasons to think that a democratic experiment such as a CA is beyond the capacity of most citizens and doomed to fail. However, evidence suggests that this was not the case in any of the three assemblies. Thompson (2008) has argued that competence in choosing an electoral system can be measured by five criteria. First, participants should be interested in the project. Second, they need to be able to formulate standards or goals that a system should fulfil. Third, they need to understand basic characteristics, that is, they need sound knowledge. Fourth, citizen-experts should be able to understand the consequences of their goals; in other words, they need to be able to link goals with systems. Finally, they should consider the views of other citizens when making their decision. Evidence indicates that members of all three CAs fulfilled Thompson’s criteria. Regarding the first one, according to several indicators, members exhibited high levels of interest. In Ontario, attendance at weekly meetings averaged 98 per cent over the 12 weekend meetings in the learning and deliberation phases. In BC, ‘over the eleven-month course of the Assembly, only one of 161 members withdrew and attendance was close to perfect’ (British Columbia Citizens’ Assembly on Electoral Reform, 2004, p. xiii). In the Dutch CA, only 6 of 143 withdrew over the process. Members were also asked to complete weekly surveys that assessed everything from their knowledge to comfort and interest in the process. In the BC and Ontario cases, one of the questions asked members to assess their degree of enthusiasm for the following week’s session. In both cases, the responses indicated a very high degree of engagement. In Ontario, for example, in response to ‘I am looking forward to the next weekend session’, where 1 indicated ‘strongly disagree’ and 5 indicated ‘strongly agree’, the range over the 12 weeks was between 4.6 and 4.84 (Institute on Governance, 2007) and in BC, it was between 4.86 in the learning phase and 4.87 in deliberation. The data for the Dutch CA showed a similar commitment to the project. Thompson’s second criterion is that good deliberation is dependent on the formulation of standards or goals that the electoral system should fulfil. This is important because decision makers require some benchmark by which to weigh competing options. Since members were given principles by which to assess electoral systems, this was established before their meetings began. Moreover, members initially had a low knowledge of electoral systems, so the teaching of the material was done as a principles-led exercise. Farrell (2001) writes that electoral systems can

Jonathan Rose

221

be understood either as mechanics or as outcomes. Both approaches are important for members in making decisions about how electoral systems work and what they produce. The mechanics-led approach was more dominant in the deliberation phase as members had to design a system from scratch. Understanding the outcomes by assessing principles was one of the pillars in the learning phase. When electoral systems were discussed in plenary sessions, design elements were introduced, but much of the teaching and discussion involved members thinking about how well and in what ways various electoral systems might meet the criteria they were given as well as those added by the group. For example, in discussing principles of representation, members were taught about how different systems privilege different kinds of representation, such as how singlemember plurality privileges territorial representation and proportional systems privilege identity-based representation. They were thus able to conclude that electoral systems involve trade-offs of competing principles and that no one system is ‘the best’. Their principles-led discussion meant that they were able to think about how each system is related to both mechanics and outcomes. Establishing standards is not only important in ensuring sound policy but is also essential in providing a focal point around which competing options can be assessed and discussed. These standards help ensure that personal or ideological preoccupations are avoided in favour of a commonly shared set of principles. For CAs with a large decision-making body (158 in BC, 103 in Ontario, and 143 in the Netherlands), a clear set of standards is vital so that options are assessed using the same set of measurements. They also have an equalizing effect on members, giving everyone the same set of conversational tools. Thompson’s third criterion asks whether members understand the basics of electoral systems. Participation must be predicated on having sound knowledge. We know that citizens’ knowledge of politics, let alone the specialized field of electoral studies, is low. Surveys of the members showed that their knowledge about the fundamentals of electoral systems increased during the learning phase. In both Canadian cases, members were asked four questions about ‘political facts’; specifically, they were requested to identify countries around the world that use particular electoral systems. The education and deliberation process resulted in a four-fold increase in knowledge in the BC and Ontario assemblies and a corresponding increase in self-assessments of knowledge. In the learning phase of the CA, members were quickly introduced to the four elements of electoral systems: ballot structure, district

222 Activating the Citizen

magnitude, formula, and size of legislature. Each family of electoral system was explained not only in terms of these elements but also in terms of how well it fulfilled the criteria the CA had established. Discussion from the recorded plenary sessions indicates that members had a high level of engagement with the material and a strong facility with using the technical language of electoral systems. This ability to understand and discuss the intricacies of electoral systems may be in part a result of the different ways the characteristics of systems were presented to members. Traditional classroom methods (reading, note taking, and listening) were supplemented by materials for visual learners and for verbal learners and by materials that had a range of complexity. In Ontario, simulations also figured prominently early on in the planning of the assembly’s learning phase. For example, at their first meeting, members were encouraged to vote in an election simulation to determine the choice of snacks served at breaks. Using three kinds of ballots (single-member plurality, alternative vote, and list proportional representation), members were given a tangible lesson in how preferences and choices are shaped by ballot structure. The norm of practice for CAs has been to ensure that learning materials are presented in different venues, such as plenary sessions for the introduction of broad concepts, small ‘tutorial style’ groups where indepth discussion can occur, and semi-structured learning time at the hotel in the evenings after the formal sessions. As Table 10.3 indicates, the Ontario CA listed the learning phase as the single most important element contributing to the success of the process. Not only was this ranked the highest, but the standard deviation among responses was the lowest. The data are similar for the BC assembly. Plenary lectures by visitors and staff were deemed the most important for the success of the assembly, followed by discussion group sessions and personal study (British Columbia Citizens’ Assembly on Electoral Reform, 2004). In the two Canadian cases, a robust educational component, which Karpowitz and Mansbridge (2005) identify as an essential element of deliberation, allowed participants to ‘continually and consciously update their understandings of common and conflicting interests as the process [evolved]’ (p. 238). Members of the CAs started the process with a knowledge of electoral systems that was like ‘scattered croutons floating in undifferentiated cognitive soup’ (Abelson, cited in Luskin, 1987, p. 860) but were transformed by their education and deliberation into citizen-experts. Thompson’s (2008) fourth criterion of competence is the ability ‘to connect the goals with the systems’ (p. 26). This ability to give a rational

Jonathan Rose

223

Table 10.3: Ontario CA members’ views of what contributed to the success of the assembly

The learning phase The academic director The chair The deliberation phase The citizens’ assembly members The interaction among members The secretariat staff The facilitators The consultation phase Conversations with family, friends, and/or other people in your riding

N

Minimum

Maximum

Mean

Standard deviation

96 97 97 97 96

4 4 3 1 5

7 7 7 7 7

6.64 6.62 6.61 6.58 6.56

0.651 0.756 0.785 0.801 0.662

94

3

7

6.47

0.799

94 96 97 95

2 4 2 1

7 7 7 7

6.44 6.39 5.95 5.48

0.979 0.910 1.054 1.428

Asked at the end of the deliberation phase, the question was as follows: ‘How would you rate the following aspects in terms of their importance in contributing to the success of the assembly? (Please circle your answer for each question where 1 means not important and 7 means extremely important).’

account to the public is essential in order to create legitimacy for any deliberative body’s decision (Chambers, 2003). While on the face of it, this seems like a reasonable criterion of success, making it an element of competence for a CA may be asking more of CA members than we ask of experts working in the field. In a survey of political scientists whose research is focused on electoral systems, Bowler and Farrell (2006) found that experts agreed that mixed-member proportional (MMP) representation was the most highly regarded electoral system but could not explain why. As Bowler and Farrell say, ‘[While] the preference for MMP among of our sample of experts is unmistakable, the reasons for that preference, and for preferring MMP over list systems or STV [single transferable vote], are not nearly so clear cut’ (2006, p. 455). The process of connecting goals with systems must take sufficient account of the inherently contingent nature of electoral systems. Are they ‘embedded institutions’ whose effects are dependent on the context in which they are employed, as Bowler and Grofman (2000) argue? Do we even have, as Bowler, Farrell, and Pettitt (2005) suggest, ‘commonly

224 Activating the Citizen

accepted measures of key concepts’ (p. 16)? What is the purpose of an electoral system? Should an electoral system provide the best representation (such as a diversity of parties) or should system designers focus on governability issues (such as stability or effective parliaments)? Should one even assume that these two sets of concerns are in opposition? Perhaps we ought to acknowledge that we cannot answer which principles best connect goals to electoral systems and admit that ‘it depends’, as Katz (1997) argues. When we compare how assembly members reasoned, we find that there is a strong connection between the goals that each assembly thought were important and the system each designed. In other words, the system they preferred was closely correlated with what they thought a system should accomplish. Choosing among nine goals related to the outcomes and inputs of electoral systems, both Canadian assemblies chose proportionality, greater voter choice, and strong local representation. In the BC case, this led members to recommend the single transferable vote system. In Ontario, it led them to recommend MMP. In the Netherlands, members valued their system of proportional representation and the quality of representation offered by that system and wanted a system that was simple and understandable. These criteria led them to recommend only minor modifications to their electoral system. Yankelovich (1991) says that citizens who engage in these kinds of deliberative exercises have to choose principles that are ‘simple enough to be intelligible to the public. Yet, at the same time, they could not be so simplistic as to be irrelevant and meaningless’ (p. 152). The process that led each CA to recommend an electoral system was the core of the deliberative exercise. Deliberation began with members evaluating what they had heard in their public consultation meetings, what they believed were the most important elements of an electoral system, and how they might accommodate each other’s interests. Position taking was avoided as it would lead to either a defensive or offensive dialogue. Instead, articulation of interests was seen as the way in which to accommodate diversity. It is important to recall that at this point, both Canadian assemblies had spent considerable time together in a learning phase (of over six weekends) and a public consultation phase, as well as informally on an electronic bulletin board and website. This had created a significant storehouse of good will among members such that civility was the dominant principle of interaction. Gadamer argues that in such philosophical dialogue, ‘one does not seek to score a point but to strengthen the other’s argument’ (cited in Yankelovich, 1991, p. 212). This was exactly the mode of interaction that characterized the CAs.

Jonathan Rose

225

Members had to decide which underlying principles of electoral systems were most important. The principles had to 1) determine which alternative system they would work up, 2) determine how the design elements fit together within each system, and 3) be used to compare their preferred alternative system to the present system. Of course, each principle involves a trade-off with other principles. For example, in an MMP system, the principles of proportionality and local representation might be seen in opposition to one another. The greater the degree of local representation, the less proportionality. In the Ontario CA’s discussion of the structure of the list tier, members had two options: region list tiers or a province-wide list tier. Region list tiers would ensure that each geographic area of the province has some representation on the list tier (one of the objectives) but may compromise proportionality (another objective).2 In the BC CA, this trade-off required balancing a smaller geographic electoral district, which could compromise proportionality, or increasing proportionality but sacrificing local representation. In the Dutch case, the emphasis on proportionality and simplicity led members to make minor modifications to their very proportional system. The members there recommended allowing voters to choose either to vote for the party of their choice or to cast a preference vote for the candidate of their choice. They also recommended a very technical and minor change to the electoral formula to transform it from a highest averages model to a largest remainder. The system recommendations of all three assemblies were driven by and evidently a product of choices about design objectives. What is noteworthy about the decisions arrived at by the assemblies is that their models achieved their priority objectives without additional features that may have made the models more complex than necessary. As one member said, quoting Einstein, ‘Make everything as simple as possible but not simpler.’

How do assembly members consult the public? While the four previous criteria about competence relate to the capacity of a CA (interest, ability to formulate goals, knowledge of electoral systems, and connection between goals and models), Thompson’s fifth criterion has to do with consultation with other citizens. On this element, we also find that CAs have been successful as deliberative and competent decision-making bodies. On both a theoretical and practical level, this is one of the most important elements of a CA. At a theoretical level, Gutmann and Thompson

226 Activating the Citizen

(2004) tell us that ‘to justify imposing their will on you, your fellow citizens must give reasons that are comprehensible to you. If you seek to impose your will on them, you owe them nothing less’ (p. 4). At a practical level, Catt and Murphy (2003) see public consultation deriving ‘from an understanding of the policy-making process as a forum for weighing competing preferences and priorities rather than as a procedure for uncovering hidden and incontestable truths. Thus consultation aims to improve the policy process by increasing the information or the range of perspectives available to decision makers’ (p. 408). Broadly stated, public consultation can serve several purposes. It can strengthen relationships between government and citizens (Barber, 1984), thereby increasing the legitimacy of government; it can have a teaching function, where the public has an opportunity to examine policy options (Arnstein, 1969); and it can help produce better policy (Irvin and Stansbury, 2004). When consultation is assessed, both the process of consultation and the outcome should be considered. Moreover, a consultation process needs to offer advantages to those consulting as well as to those being consulted (cf. Irvin and Stansbury, 2004). While the consultations of the three assemblies performed all of these functions, what is unique about them is that they were deliberative bodies comprised of citizens as well as consultative bodies consulting citizens. In this respect, a CA is different from other consultative bodies, where typically experts consult citizens. Though the participatory budgeting exercises and management councils of Brazil are good examples of citizens consulting citizens, they have more constraints on membership and policy recommendations than CAs (see Coelho, Pozzoni, and Montoya, 2005). What was the purpose of the consultation phase? The nominal purpose was to seek additional information from members of the public and stakeholders about the specific policy issue, that is, electoral reform. More broadly however, the purpose of this phase was to create legitimacy for the assembly in the minds of the public. Substantively, the public consultation phase was about hearing from those who wanted change. In all three cases, oral presentations and written submissions were largely about improvements needed to the electoral system. In Ontario, 90 per cent of them argued for some kind of change. In BC, 76 per cent of the written submissions argued for change while only 2 per cent made a case for the status quo (British Columbia Citizens’ Assembly on Electoral Reform, 2004). While the impact of this overwhelming endorsement of change on the CA is hard to measure, in Ontario at least CA members claimed they had learned a lot from the public and were

Jonathan Rose

227

impressed by the presentations (Ontario Citizens’ Assembly on Electoral Reform, 2007). In a survey given to Ontario CA members after consultation, 87.4 per cent said they had found written submissions from the public very informative or somewhat informative, while 95.7 per cent found the public meetings they attended informative or very informative. Table 10.3 provides some contradictory evidence, however, since CA members also identified the public consultations as the second least important to the success of the process (though the range from the most important to the least important is small). So members found the consultation sessions informative but not important to their decision. It is difficult to know whether this is because the consultations reinforced the ideas members held at the time or because members were unsure of their role in the consultations. Usually in public consultations, those consulting have some expertise and those who are consulted are non-experts. The CAs created an opposite dynamic; assembly members were in the unusual situation of being citizens who believed they had little expertise, consulting the public whom they believed had, at least in some cases, significant expertise. As one member in Ontario said, ‘I am a member of the Citizens’ Assembly, but I’m also a member of the public. I am an ordinary person’ (cited in Hannigan, 2007, p. 91). Another said about those who appeared at the public meetings, ‘A lot of these people have been thinking about these views for years, and as ordinary citizens it’s nice [for us] to get different views’ (cited in Hannigan, 2007, p. 98). In other cases, there was an opposite dynamic, namely a disjunction between the calibre of talk heard at the consultation meetings and the quality of CA members’ own internal deliberations. The set of skills that CA members had developed was often not reflected in the public meetings when citizens spoke to advocate a position without having understood the complexity of competing options. It is not surprising, therefore, that like other public hearings, those of the assembly ‘routinely failed to resemble even a crude form of deliberation’ (Gastil and Black, 2008, p. 24). As vehicles of participation, CAs are caught in an interesting paradox. On the one hand, they are imbued with great authority, independence, and power. On the other hand, they are comprised of citizens who have developed expertise but in many cases are still deferential to the ‘expert’ citizens they hear. Unlike politicians who understand the symbolic and real importance of public consultation, members were often not clear on the competing roles that these public meetings had. Gastil and Black (2008) describe these competing functions as expert, official, and public perspectives, which need to be balanced during deliberations. Members

228 Activating the Citizen

had developed a certain expertise, but as citizens they represented public perspectives. These two roles may have been tough to reconcile. This is not to say that this form of participation was not useful to the CAs. Furthermore, the process likely benefited the public whose views were sought. Of the public who attended the meetings, 45 per cent completed a survey administered by the Ontario CA Secretariat, numbers similar to BC. In Ontario, over 80 per cent of respondents agreed or strongly agreed that they had learned more about the assembly after attending a meeting. This suggests that one of the benefits of the consultation process was to raise public awareness of the project. Some scholars claim that public consultation hearings ‘are mere democratic rituals that provide a false sense of legitimacy’ (Adams, 2004, p. 44), since presentations are more pro forma than substantive and the public is not given a chance to raise questions or participate meaningfully. This interpretation seems to contradict the views of members of the public who participated in all three assemblies. The Ontario case is illustrative: Over 90 per cent agreed or strongly agreed that ‘the presenters were given enough time to present and answer questions’ while 79.7 per cent agreed or strongly agreed that ‘there was a chance for members of the public to raise questions’. Perhaps the strongest warrant for the consultation phase was to make the process as well as the content accessible. Guttman and Thompson (2004) argue that deliberative democracy demands accessibility in both process and content; in particular, information must be presented in ways understandable to the audience. The structure of discourse in the consultation hearings made what could be technical issues into issues driven by values. Complex language was explained in as simple a manner as possible in an attempt to avoid one of the major limitations of deliberation (see Yankelovich, 1991).

The importance of time, resources, and support One of the most significant limitations on the work of the CAs was the very compressed time frame in which members had to learn, consult, and decide. In Ontario, there was no break between any of the three phases and, in fact, consultation began during the last week of the learning phase. In the BC case, there was a summer break between consultation and deliberation, allowing members time to absorb the material they had been exposed to. The Ontario CA’s independent evaluator’s report indicated that more time was needed, especially in the deliberation phase but also in the learning phase (Institute on Governance, 2007). The issue of

Jonathan Rose

229

time constraint was commented upon by members themselves throughout the 12 weekends of these phases. In the Netherlands, the time frame was even more compressed, with three weekends devoted to learning and four devoted to deliberation. In order to teach such a diverse group, it is imperative to provide educational materials that meet very different needs. Active learning was one of key foundations of the learning phase, with members engaged in simulations and learning by doing. Detailed planning of small group sessions began in advance of each weekend. Facilitator notes were sent to small group facilitators for feedback and discussed with them on Friday before each meeting. After the weekend, there was a debriefing session with facilitators to ensure that the techniques used in the small groups worked. In addition to debates and simulations, some of the other techniques used are described in the learning literature as ‘buzz groups’, ‘think-pair-share’, and problem-based learning that teaches students ‘to develop the ability to think critically, analyse problems, find and use appropriate learning resources’ (Centre for Teaching and Learning, 2007). Many members did want traditional textbook resources and these were provided as beginner, intermediate, and advanced readings, ensuring that all members were given material of the appropriate level. The third and perhaps most critical element to a successful CA is to do two apparently contradictory things: to support the CA in its decision and deliberation but also to ensure its independence. According to Surowiecki (2004), the independence of a deliberative body is a crucial element for it to reach a sound decision. The process needed to be strongly supported by staff whose job was to ensure that members received the resources and tools necessary to help them with decision making without steering them in any particular direction. Members needed to seek their own sources of knowledge and be willing to share them with their colleagues. This was done through a ‘members-only’ web forum, which served as a place for members to post articles and web links and to engage in debates about issues they were working through. Members’ extra-curricular learning occurred at evening sessions at the hotel where ad hoc groups formed to discuss issues before the assembly met that weekend. Weekly surveys indicated that CA members felt well supported and that plenary presentations were neutral. The crucial measurement of neutrality was upheld with 93.3 per cent of members saying that ‘the presentation of the options of the Academic Director and research staff’ was very or somewhat unbiased. In the Netherlands, answering a yes/no question, 80 per cent agreed that the staff did not have a preference. All of this suggests that citizens have the capacity to

230 Activating the Citizen

learn and that institutions can be established that aid in that endeavour without ‘leading’ citizens to a particular outcome.

Conclusion Political participation can be defined by adversarial activities that place a premium on individual interest aggregation. Much of political science is devoted to examining these activities. Another older form of participation stresses the importance of talk that is framed around consensus. According to this view, citizens talking to citizens about politics is the essence of democracy. This chapter has explored three case studies of this form of democratic participation. The creation of a civic space where equality, fairness, and an open mind are the goals is the hallmark of this kind of participatory democracy. CAs are an excellent – and far too rare – example of this type of robust participation. They are unique because they represent a transformation of citizens into citizen-experts through an in-depth educational programme and remind us about the capacity of citizens to make sound policy. Their democratic procedures and transparency provide a rare instance of participatory democracy that is not grounded in adversarial relationships but shared values. While none of the three assemblies so far has resulted in electoral reform, this does not impugn the process of the assemblies or the model of participation. Farr (1993) reminds us that ‘more light might come of our theoretical and empirical research if we look not to public opinion or to elites but to actual discussions held between citizens’ (p. 378). Perhaps this is the real value of citizens’ assemblies.

Notes 1. This was in response to the question, ‘How informed about electoral systems do you feel?’ 2. The Ontario CA chose a province-wide list tier knowing that proportionality would be compromised by regional lists because they necessarily have a smaller district magnitude than a province-wide list.

References Adams, B. (2004) ‘Public Meetings and the Democratic Process’, Public Administration Review, 64(1), 43–54. Arnstein, S. (1969) ‘A Ladder of Citizen Participation’, Journal of American Institute of Planners, 35, 216–24.

Jonathan Rose

231

Barber, B. (1984) Strong Democracy: Participatory Democracy for a New Age (Berkeley, CA and Los Angeles: University of California Press). Berelson, B. R., P. F. Lazarsfeld, and W. N. McPhee (1954) Voting: A Study of Opinion Formation in a Presidential Campaign (Chicago: University of Chicago Press). Bowler, S. and D. Farrell (2006) ‘We Know Which One We Prefer but We Don’t Really Know Why: The Curious Case of Mixed Member Electoral Systems’, British Journal of Politics and International Relations, 8(3), 445–60. Bowler, S., D. Farrell, and R. Pettitt (2005) ‘Expert Opinion on Electoral Systems: So Which Electoral System Is “Best”?’, Journal of Elections, Public Opinion and Parties, 15(1), 3–19. Bowler, S. and B. Grofman (2000) Elections in Australia, Ireland, and Malta under the Single Transferable Vote: Reflections on an Embedded Institution (Ann Arbor, MI: University of Michigan Press). British Columbia Citizens’ Assembly on Electoral Reform (2004) Making Every Vote Count: The Case for Electoral Reform in British Columbia, http://www.citizensassembly.bc.ca/resources/TechReport(full).pdf Button, M. and K. Mattso (1999) ‘Deliberative Democracy in Practice: Challenges and Prospects for Civic Deliberation’, Polity, 31(4), 609–37. Catt, H. and M. Murphy (2003) ‘What Voice for the People? Categorising Methods of Public Consultation’, Australian Journal of Political Science, 38(3), 407–21. Centre for Teaching and Learning (2007) ‘Good Practice: Problem-Based Learning’ Queen’s University, http://www.queensu.ca/ctl/goodpractice/problem/index. html Chambers, S. (2003) ‘Deliberative Democratic Theory’, Annual Review of Political Science, 6, 307–26. Coelho, V. S., B. Pozzoni, and M. C. Montoya (2005) ‘Participation and Public Policies in Brazil’ in J. Gastil and P. Levine (eds) The Deliberative Democracy Handbook: Strategies for Effective Civic Engagement in the 21st Century (San Francisco: Jossey-Bass), pp. 174–85. Converse, P. E. (1990) ‘Popular Representation and the Distribution of Information’ in J. A. Ferejohn and J. H. Kuklinski (eds) Information and Democratic Processes (Urbana, IL: University of Illinois Press), pp. 369–87. Dahl, R. A. (1970) After the Revolution: Authority in a Good Society (New Haven, CT: Yale University Press). Dahl, R. A. (1989) Democracy and Its Critics (New Haven, CT: Yale University Press). Farr, J. (1993) ‘Framing Democratic Discussion’ in G. E. Marcus and R. Hanson (eds) Reconsidering the Democratic Public (University Park, PA: Penn State Press). Farrell, D. (2001) Electoral Systems: A Comparative Introduction (New York: Palgrave – now Palgrave Macmillan). Fishkin, J. S. (1991) Democracy and Deliberation: New Directions for Democratic Reform (New Haven, CT: Yale University Press). Gastil, J. and L. W. Black (2008) ‘Public Deliberation as the Organizing Principle of Political Communication Research’, Journal of Public Deliberation, 4(1), 1–47. Gutmann, A. and D. Thompson (2004) Why Deliberative Democracy? (Princeton, NJ: Princeton University Press). Hannigan, L. (2007) ‘The Effectiveness of Public Consultation: A Case Study of the Ontario Citizens’ Assembly on Electoral Reform.’ Unpublished BA honours thesis, Queen’s University, Kingston, ON, Canada.

232 Activating the Citizen Institute on Governance (2007) Citizen Deliberative Decision-Making: Evaluation of the Ontario Citizens’ Assembly on Electoral Reform (Ottawa: Institute on Governance). Irvin, R. and J. Stansbury (2004) ‘Citizen Participation in Decision Making: Is It Worth the Effort’, Public Administration Review, 64(1), 55–65. Karpowitz, C. and J. Mansbridge (2005) ‘Disagreement and Consensus: The Importance of Dynamic Updating in Public Deliberation’ in J. Gastil and P. Levine (eds) The Deliberative Democracy Handbook: Strategies for Effective Civic Engagement in the 21st Century (San Francisco: Jossey-Bass). Katz, R. S. (1997) Democracy and Elections (New York: Oxford University Press). Lippmann, W. (1925) The Phantom Public (New York: Harcourt Brace Jovanovich). Luskin, R. (1987) ‘Measuring Political Sophistication’, American Journal of Political Science, 31(4), 856–99. Mansbridge, J. J. (1980) Beyond Adversary Democracy (New York: Basic Books). Neuman, W. R. (1986) The Paradox of Mass Politics: Knowledge and Opinion in the American Electorate (Cambridge, MA: Harvard University Press). Ontario Citizens’ Assembly on Electoral Reform (2007) Public Consultation Reports, http://www. citizensassembly.gov.on.ca/ assets /Citizens ’% 20Assembly % 20 Consultation%20Reports.pdf Schumpeter, J. (1942) Capitalism, Socialism and Democracy (New York: Harper). Snider, J. H. (2007) ‘From Dahl to O’Leary: 36 Years of the “Yale School of Democratic Reform” ’, Journal of Public Deliberation, 3(1), 1–7. Surowiecki, J. (2004) Wisdom of Crowds: Why the Many are Smarter than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations (New York: Random House). Thompson, D. (2008) ‘Who Should Govern Who Governs: The Role of Citizens in Reforming the Electoral System’ in M. Warren and H. Pearce (eds) Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press), pp. 20–49. Warren, M. and H. Pearce (eds) (2008) Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press). Yankelovich, D. (1991) Coming to Public Judgement: Making Democracy Work in a Complex World (Syracuse, NY: Syracuse University Press). Zaller, J. R. (1992) The Nature and Origins of Mass Opinion (New York: Cambridge University Press).

11 Citizen Involvement in Constitutional Politics: European and Canadian Experiences Lawrence LeDuc

The declines in electoral participation and political party membership discussed in the preceding chapters of this book have led to new thinking about ways in which citizens might be engaged or reengaged in political processes. This has been particularly important with respect to constitutional politics, since the legitimacy of a constitution depends on its acceptance by the citizens. In modern democratic societies, proposals for constitutional change often involve the direct engagement of citizens in an elaborate process of consultation and decision. Typically, such a process consists of a protracted period of negotiation among elites followed by submission of proposals to the people for approval in a referendum. Referendums have been seen by some democratic theorists as one means of redressing the decline in participation that has widened the ‘democratic deficit’ in both North America and Europe. They have also been seen as a means of legitimizing constitutional change in a way that ordinary electoral processes or institutions cannot. This chapter will examine the possibilities and pitfalls of using referendums to engage citizens in the constitutional process. In some countries, such as Ireland and Australia, constitutional amendments must be submitted to a referendum vote. In others, such as France or Denmark, a referendum is one option that may be followed.1 In Canada, Britain, and a number of other countries, the decision to submit proposed constitutional changes to a referendum is a purely political one, since direct public consultation is not a part of any formal legal process. Yet increasingly, public approval in some form has become an essential element of the constitutional regime. Following the signing of the Treaty Establishing a Constitution for Europe in Rome on 29 October 2004, the 25 European Union (EU) member nations each had to decide on an appropriate process for ratification of the document. 233

234 Activating the Citizen

While Ireland was required by law to submit the treaty to a referendum, in other EU countries the referendum was either one of several routes that might have been employed or an entirely optional matter. In reality however, holding a referendum quickly became a political necessity in several of the member states. The recent history of contention over EU issues in Denmark all but assured that a referendum would take place in that country. In Britain, Tony Blair abruptly reversed his government’s earlier position and announced that Britain would hold a referendum on the treaty. Jacques Chirac decided that a referendum would be held in France under Article 11 of the constitution. It becomes more difficult to make a case for preventing one’s own citizens from having their say in a referendum when others are doing so. In addition, political parties are often divided on European issues, and holding a referendum can be one way of managing these internal party divisions. Then there is the matter of precedent. Once a referendum on an important European issue has taken place, it becomes more difficult to argue against holding one on the next important issue that arises. Referendums thus become a kind of quasi-constitutional norm for dealing with certain categories of issues – for example, institutional changes, international treaties, or major European issues – even in the absence of any formal legal requirement that such issues be handled in this way. Ultimately, four EU member countries – Spain, France, the Netherlands, and Luxembourg – held referendums on the treaty, and it became clear that if the process of ratification had proceeded positively, several others would also have done so – probably Portugal, Poland, and the Czech Republic, in addition to Britain and Denmark. The 1992 Canadian constitutional referendum likewise came about as a result of a series of political decisions, rather than because of any legal requirement that it take place. In Canada, constitutional negotiations have been conducted on various occasions between the federal government and the provinces. The Meech Lake Accord, a package of proposals negotiated in this way, was submitted to the provincial legislatures for ratification in 1987. But that agreement expired in 1990 when two provinces – Manitoba and Newfoundland – failed to ratify it within the established three-year time limit. A subsequent attempt to reach a constitutional settlement found success in the Charlottetown Accord, an agreement between the federal government and the provinces reached in 1992. That agreement, like the Meech Lake document, might have simply been put to parliament and the provincial legislatures for ratification. But Quebec was already committed by law to hold its own referendum, and several other provinces had also signalled their intention to

Lawrence LeDuc

235

follow this route. Thus, the federal government decided to submit the Charlottetown proposals to a national vote. Constitutional settlements such as the Charlottetown Accord and the Treaty Establishing a Constitution for Europe were the culmination of efforts spread over many years, involving lengthy and complex negotiations. Yet the final outcomes were determined in referendum campaigns lasting only a few weeks. In both instances, the political elites who had secured the agreements became convinced that it would not be a difficult matter to persuade voters to support proposals that seemed perfectly reasonable to them and that enjoyed seemingly broad support across many diverse groups – support that had often been carefully and painstakingly cultivated over a substantial period. The long and complex process of negotiating the Charlottetown Accord had involved extensive consultation and compromise between the federal government, the provinces, and many other groups and organizations over the two-year period following the death of the Meech Lake agreement. The attempt to craft a new European constitutional document to replace the many separate treaties underpinning the enlarged EU began with the Laeken Declaration in December 2001, which created the European Convention on the Future of Europe. In the end however, the intricate work of the convention, which took place over a period of nearly three years, was abruptly derailed by the referendums in France and the Netherlands. What these examples have in common is that many voters decisively said NO to ambitious proposals for institutional change that appeared initially to enjoy wide public support. Yet there is more to the story than merely the political miscalculation of elites. When placed in a broader comparative context, it will be seen that these cases share a pattern that is disturbingly commonplace in referendum campaigns dealing with certain types of issues. While democratic values may dictate a view that ‘the voters are always right’, it appears increasingly to be the case that institutions that are intended to provide solutions to difficult problems may just as readily act to block them. Once solutions are blocked, it is not an easy matter to simply resume the debate. In the Canadian and European experiences, the result has been not a genuine resolution of the issue presented to the voters, but rather the onset of a protracted period of constitutional and political impasse. Referendums can be a way of legitimizing a political decision that might otherwise be a source of continuing conflict. But once the first steps down this path are taken, there is often no way out of the impasse created except through another referendum. Thus, in Canada and Europe today, it is clear that a referendum did not provide a ‘final’ answer to the constitutional question. In each of these instances,

236 Activating the Citizen

the constitutional debate continues, albeit with the question of future referendum(s) occupying a central part of the discussion.

Opinion formation and change in referendum campaigns Zaller’s (1992) model of opinion formation is particularly well suited to the study of public opinion and voting behaviour in referendums. As Zaller argues, any process of opinion formation proceeds from an interaction of information and predisposition. When strongly held predispositions are reinforced by a campaign, factors such as party identification or ideological orientation typically play a crucial role. But when parties are internally divided, ideological alignments are unclear, or an issue is new and unfamiliar, voters might be expected to draw much of their information from the campaign discourse. Constitutional referendums in particular tend to conform to this pattern, with elite-driven projects often being decisively rejected once the voters learn more about them. In such circumstances, the degree of change in opinion over the course of even a short campaign is often large, because there is little in the way of stable social or political attitudes that might anchor opinions on the issue. Predispositions on constitutional issues, in other words, tend to be very weak. By contrast, where an issue is a familiar one in the political arena, or where parties take clearly competing positions, the voting decision is easier and tends to be made earlier in the campaign. Here, opinions are firmer and less subject to rapid change or sudden reversal. Voters will often have strong cues based on partisanship or ideology and be receptive to arguments presented by familiar and trusted political leaders. In such a campaign, much of the attention is directed toward wavering or ‘undecided’ voters, in the knowledge that a swing of only a few percentage points might make the crucial difference. A good example is the 1995 Quebec sovereignty referendum; the Parti Québécois government knew that it could count on the nearly universal support of the partisan voters who had brought it to power, but it needed the votes of others in order to secure a majority for its sovereignty proposal. The Danish (2000) and Swedish (2003) referendums on the Euro also provide good examples of this pattern; the governments believed that they could win the referendums by persuading a relatively small number of undecided voters to support the YES side. Factors such as party identification, ideology, the linkage of the referendum issue to particular groups, or its identification with established political actors, provide operational examples of Zaller’s predispositions.

Lawrence LeDuc

237

Opinions about certain types of issues will change less readily than others, and perhaps even less readily than attitudes towards individual politicians, groups, or political parties. For some voters, opinions on Quebec sovereignty or on European integration might reflect strongly held fundamental beliefs about the nation or a sense of political community. For others however, such attitudes might be less the product of deeply held beliefs than a more transient opinion based on the persuasive arguments of an advertising campaign, apprehensions about the state of the economy, or judgements about the relative credibility of those delivering the message. Thus, a referendum that involves a cleavage or ideological issue and/or political parties taking well-known and predictably opposite positions ought to hold the least potential for opinion change over the course of a short campaign. One that involves a new or previously undiscussed issue or parties lining up in a non-traditional manner is likely to produce greater volatility. In the absence of strong predispositions, information becomes more critical to the voting decision. Generally, such information comes from the campaign itself, and more specifically from the actors who are part of that campaign. But a campaign is never merely about ‘informing’ voters – it is about winning. Thus, while there may (or may not) be some neutral sources of information available to voters over the course of any referendum campaign, much of the information that voters receive comes directly or indirectly from campaign actors who have a stake in the outcome. We should not be surprised therefore to find that, where predispositions are weak, such sources of information can be quite powerful. In constitutional referendums particularly, the NO side often possesses a distinct advantage. How unbalanced the field actually is depends on the nature of the issue at stake, the political context of the campaign, and the resourcefulness and ingenuity of the campaign actors in transmitting their versions of the message to the voters.

The 2005 European constitutional treaty referendums Zaller’s model suggests that the most volatile referendum campaigns are likely to be those with little partisan, issue, or ideological basis on which voters can more readily form opinions. The European constitution referendums fit this model quite well. Up until the signing of the Treaty Establishing a Constitution for Europe, debate on it had been conducted almost entirely by elites, and public awareness of the details of the treaty was very low. In a survey taken two months after the signing of the treaty, about a third of all Europeans reported that they had never heard of it.

238 Activating the Citizen

The vast majority of those who had heard of the treaty indicated that they knew little about its contents.2 This is not surprising. The convention that drafted the treaty was in most respects ‘business as usual’ regarding the arcane workings of the EU, despite some of the more extravagant rhetoric associated with its creation (on the background to the convention and the Laeken declaration, see Magnette and Nicolaidis, 2004). Headed by former French president Valèry Giscard d’Estaing, the convention was not, in the words of one analyst, ‘made up of idealistic actors promoting a public and deliberative style of constitution-making in the name of legitimacy and popular sovereignty’ (cited in Magnette and Nicolaidis, 2004, p. 383). The treaty that it eventually produced, with its hundreds of clauses written in typically opaque language, was not the type of document likely to be readily accessible to the public. Thus, it was not until well after the treaty had been signed and attention began to turn to the ratification process in each country that public awareness of the convention and its work began to slowly increase. Evidence from previous European referendum campaigns, such as the ones that took place in Ireland, France, and Denmark on the Maastricht Treaty in 1992, tells us that the outcome of a referendum taking place in such a setting is highly uncertain. Voters need time and information to form opinions and are often influenced directly by the discourse of a campaign. On the other hand, the larger European issues, of which the debate on the constitutional treaty was a part, lurked continuously in the background as the convention completed its work. But attitudes towards Europe are likely not in themselves strong enough to predetermine the outcome of a referendum on an issue such as the constitution. Nor are they firmly anchored to partisanship or ideology in most instances, since divisions over European issues exist within the major parties in many EU countries. Context and timing are all-important, and ‘second order’ effects are often present (Franklin, van der Eijk, and Marsh, 1995; Garry, Marsh, and Sinnott, 2006). The chances of an unpopular prime minister or president, late in a government’s term, winning such a referendum are undoubtedly smaller than if the campaign were led by more credible or popular figures in a more positive setting. Domestic political factors may also play an important role in the opinion formation process, even when these may have had little to do with the European constitutional treaty per se. The Spanish referendum – the first of the four that took place after the signing of the constitutional treaty – was not subject to many of these extraneous effects. The campaign was led by a new government and

Lawrence LeDuc

239

80

70

60

50

40

Yes No

30

20 Dec

Jan

Feb

Mar

Apr

May

R*

Figure 11.1: Public opinion in the French referendum on the EU constitutional treaty, December 2004–May 2005 (%) R∗ = actual referendum result. Source: CSA/France Info; SOFRES/Le Figaro.

prime minister, who had been in office for less than a year. Both major political parties actively campaigned for the YES side and stressed the broad theme of affirming Spain’s place in Europe. The governing party was strongly united in support of the treaty, and there was no prior history in Spain of divisive debates over European issues. The early and positive nature of the campaign (‘First in Europe’) gave potential opponents of the constitution little time to organize and few issues around which to frame opposing arguments. It is therefore not surprising that, under these favourable conditions, the YES side rolled to an easy (77 per cent) victory. The low turnout (42 per cent) reflected the relative lack of engagement on the part of the electorate and the one-sided nature of the contest. In distinct contrast was the referendum in France, which took place three months later. While initial public opinion polls, conducted at about the same time as the Spanish referendum (see Figure 11.1), suggested that more than 60 per cent of French voters would endorse the treaty, opinion on the issue was not yet well formed in the minds of many voters, and the actual campaign in France had yet to begin. President Chirac’s announcement of the date of the referendum in March

240 Activating the Citizen

appeared to precipitate a significant drop in support for the treaty and revealed the potential for volatility that would be evident throughout the campaign. The unpopularity of the president would prove to be a liability for the YES campaign, of which he was the de facto leader. As was the case in the Maastricht referendum of 1992, the campaign effectively pitted the mainstream political parties, Chirac’s Union for a Popular Movement and the Socialist Party, against more marginal forces such as the National Front and the Communists. This configuration guaranteed that, as with Maastricht, the treaty would be attacked from both left and right, but on different grounds. Splits within the two major parties also damaged the prospects for ratification of the treaty by providing conflicting cues to large groups of voters. The Socialist Party had voted in an internal party referendum to support the constitution, but some of its leading figures nevertheless campaigned on the NO side.3 Likewise, a faction of the Gaullists refused to follow Chirac’s leadership in support of the treaty and actively campaigned for the NO side. While the strategy that lay behind Chirac’s decision to hold a referendum rather than allow the treaty to be ratified by a joint sitting of the National Assembly and Senate may have been clear at the time that it was taken, the outcome later became uncertain because of the introduction into the campaign of new personalities, issues, and arguments. A campaign can introduce a new set of variables, and these are sometimes capable of producing substantial shifts in sentiment, particularly when the factors anchoring opinions on an issue are weak to begin with. Among those that appear to have been a factor in the outcome of the French referendum are the already mentioned unpopularity of the president and prime minister, the state of the French economy, and the strategic calculations of future potential presidential candidates. The fact that a YES vote seemed, at least in some minds, to constitute a vote for Chirac may have given some potential YES voters pause (Marthaler, 2005). Also influential were sets of issues only partially related to the constitutional debates, such as the implications for France of the 2004 EU enlargement and the possible future prospect of Turkish membership. The relatively low levels of support in France for Turkish membership in the EU made this an issue that was easily injected into the campaign by the far right parties, even though ratification of the constitution in itself held little direct connection to this question. The intensity of the campaign and the uncertainty of the outcome produced a substantial (69 per cent) turnout of French voters – higher than the previous parliamentary election and similar to that found in the 1992 referendum on Maastricht.4 The margin of victory for the NO side

Lawrence LeDuc

241

was slightly greater than had been predicted by the final opinion polls of the campaign. An exit poll analysis suggested that the split among Socialist voters was decisive in the result, with a narrow majority of this group supporting the NO side in spite of the fact that the party itself had officially endorsed the YES side (Marthaler, 2005). The partisan and ideological cues emanating from the left during the campaign portrayed the constitution as entrenching a ‘neo-liberal’ economic regime in Europe, thereby connecting it to a broad array of other social and economic issues (Hobolt, 2005; Ivaldi, 2006). The referendum in the Netherlands took place only two days after the French result, raising the strong possibility of a ‘domino effect’ (Jahn and Storsved, 1995). In comparison with France, the campaign in the Netherlands was surprisingly low key. But in the wake of the French result, there appeared little possibility of a YES victory. In the Netherlands, as in France, the constitution enjoyed the support of all the mainstream political parties, and opposition came mainly from the extremes on the left and right. But the unpopularity of the governing centre-right coalition that had called the referendum almost certainly introduced second-order effects, and the prior outcome of the French referendum substantially lowered the stakes for Dutch voters inclined to vote NO. Since the constitution was effectively dead as soon as one country had failed to ratify it, there appeared to be little at stake in the Dutch vote, making it easier for voters to use the referendum as a means to express dissatisfaction with the government or scepticism regarding other European issues. Dissatisfaction with the Euro may also have played a role (Aarts and van der Kolk, 2006). Available evidence indicates that voters of nearly all political parties were divided on the issue, but that support for the treaty in the Netherlands, as in France, faced somewhat greater problems on the left of the political spectrum. Turnout at 62 per cent was respectable, but low compared with national elections. While the outcome itself was not a surprise, the size of the NO vote (62 per cent) came as a shock in a country that had traditionally shown strong support for European institutions. Compared with the benchmark established by the November 2004 Eurobarometer, which found Dutch respondents to be among the most positively disposed in Europe towards the treaty, the decline in support over such a short period of time was indeed massive.5 The outcome of the referendums in France and the Netherlands effectively ended the process of ratification in other member countries and led to the cancellation of planned referendums in several others, notably Denmark and Britain. It therefore came as a surprise when Luxembourg

242 Activating the Citizen

opted to proceed with its referendum, which took place a little over a month later on 10 July. This decision was taken by the prime minister, Jean-Claude Junker, who made the issue a matter of confidence in his government. Indicating that he would resign if the constitution was not approved in the referendum, Junker placed his own considerable popularity firmly behind the treaty, thus explicitly introducing second-order effects into the campaign dynamic. This made the referendum in Luxembourg a quite different type of contest than in the other countries voting on the treaty. As in the other three countries, all of the major political parties endorsed the treaty, despite the fact that its fate was so closely tied to that of the government. While this strategy produced a clear victory for the YES side (57 per cent), it undoubtedly left voters confused about what they were voting on and made the outcome difficult to interpret. With compulsory voting in Luxembourg, turnout was not a factor, as it may well have been in the other three countries.

The 1992 Canadian constitutional referendum The origins of the 1992 referendum, the first such nationwide vote since 1942, lay in Canada’s long and complex struggle to repatriate the constitution from its British origins.6 Constitutional reform took on a new sense of urgency in 1976 with the election of a separatist government in Quebec. Prime Minister Trudeau, having failed to secure an agreement on a new constitution with the provinces after extensive negotiations, finally acted to impose a settlement in 1982 without the concurrence of Quebec. However, the election of a new Conservative federal government under Brian Mulroney in 1984 and the defeat of the Parti Québécois government in Quebec in the 1985 provincial election opened the door once again to the possibility of new constitutional initiatives. The first of these, the Meech Lake Accord, expired in 1990 when two provinces—Manitoba and Newfoundland—failed to ratify it within the time limits established in the agreement. A subsequent attempt to reach a settlement that could ‘bring Quebec in’ resulted in August 1992 in an agreement between the federal government and the provinces known as the Charlottetown Accord. That agreement might have simply been put to parliament and the provincial legislatures for debate and ratification, but because Quebec was already committed to holding its own referendum no later than 26 October 1992, the federal government opted instead for a national referendum to be held at the same time. The referendum ballot provided only for a simple YES or NO on ‘the agreement of August 28th’. Because the decision to hold a referendum

Lawrence LeDuc

243

80

70

60

50

40

Yes No

30

20 909

916

926

1009

1016

1021

R*

Figure 11.2: Public opinion in the Canadian constitutional referendum, August–October 1992 (%) R∗ = actual referendum result. Source: Environics; Angus Reid; ComQuest.

was taken so quickly, little thought had been given to matters of organization, strategy, or even the wording of the question. As was the case with the European constitutional referendums, the initial readings of public opinion on the agreement were very favourable, in part because Canadians appeared relieved to at last see an end to the long-running constitutional wars. All ten provincial premiers representing three different political parties supported the agreement, and the leaders of the three main federal parties announced that they would campaign actively for a YES vote. As seen in Figure 11.2, the campaign began with a seemingly high level of public support for the agreement. The Parti Québécois almost immediately announced its intention to campaign for a NO vote, and other opposition soon began to surface. The Reform party, at that time a much smaller, largely western-based political movement, announced that it would oppose the agreement and campaign against it nationally. Its leader, Preston Manning, would become one of the principal campaigners in English Canada for the NO side. At about the same time, several prominent Quebec Liberals came out against the proposals, revealing a split in Premier Bourassa’s own party and indicating that the agreement was in more serious trouble in Quebec than initially thought. Within a few weeks, other opposition surfaced – from

244 Activating the Citizen

women’s groups, some constitutional lawyers, and finally former Prime Minister Pierre Trudeau. While it was perhaps not all that surprising that Trudeau, who had also been a critic of the Meech Lake proposals, would oppose the Charlottetown Accord, his views commanded wide attention in the media. Momentum suddenly seemed to shift away from the poorly organized and overconfident architects of the agreement (on the dynamic of the campaign, see Johnston, Blais, Gidengil, and Nevitte, 1996). The various groups and individuals supporting the NO side had little in common with each other; they ranged widely across the political spectrum and often held contradictory views on many other issues. But arrayed against them were the pillars of the Canadian establishment – business, government, academia, and seemingly much of the press and media. Because there was really no long-term basis of public opinion on many of the specific issues arising from the Charlottetown Accord, predispositions as defined by Zaller were low for most voters. A few would have been able to make up their minds on the basis of partisan cues or familiarity with one or more of the long-standing issues in the constitutional debates. A few issues, such as the distinct society provision or the equal Senate concept were already fairly well known, and might by themselves have shaped some voters’ views of the entire agreement. In addition, there were the cues provided by the parties and by political leaders such as Trudeau, Mulroney, Bourassa, and Manning – personalities about whom many voters had strong opinions. The NO side pulled ahead in the polls during the first week of October (see Figure 11.2) and stayed there throughout the remainder of the campaign. With nearly three weeks to go, the contest was effectively over, even though public opinion polls continued to show large numbers of undecided voters right up until the very end of the campaign. The outcome of the referendum was decisive. In the country as a whole, 55 per cent voted NO compared with 45 per cent YES. Six of the ten provinces voted NO, with strong majorities against the agreement in all of the western provinces and in Quebec. The Atlantic provinces, with the exception of Nova Scotia, supported the YES side by fairly substantial majorities, while in Ontario the result was a razor thin margin for the YES side. Since the leaders of all three of the main federal parties actively campaigned for a YES vote, many of the normal partisan cues that voters might be expected to respond to were largely absent from the referendum campaign.7 Prime Minister Mulroney’s popularity with the public was close to its nadir, and he would indicate his intention to resign from office only four months after the referendum defeat. But in the final

Lawrence LeDuc

245

analysis, the agreement failed because of the message and the messengers. A multivariate analysis suggests that some of the key provisions of the agreement, together with feelings about the political leaders, particularly the prime minister, sealed the agreement’s fate (LeDuc and Pammett, 1995). The outcome was thus determined by a mixture of firstand second-order effects. The 1992 constitutional referendum was an unusual event in Canadian politics. Although not legally required to hold a referendum, the architects of the Charlottetown agreement ultimately had little choice but to seek some kind of popular mandate for their project. They were initially confident, but the referendum unexpectedly provided Canadian voters with a rare opportunity to pass judgement on the nation’s entire political establishment, together with one of that establishment’s most cherished projects. For many, the referendum seemed to represent a battle of the people against the establishment. Seen in this light, it is perhaps more surprising that 45 per cent of Canadians ultimately voted for the Charlottetown agreement than that it went down to defeat.

Comparing cases The 1992 Canadian constitutional referendum and the 2005 French and Dutch referendums on the constitutional treaty display many similarities in terms of the patterns followed by the campaigns, the types of arguments put forward by the YES and NO sides, the relative weakness of predispositions in the form of strong social or partisan anchors of opinion, and most of all in their negative outcomes. These cases clearly demonstrate that those opposed to a proposal do not necessarily have to make a coherent case against it. Not uncommonly, it is enough merely to raise doubts about it in the minds of voters, question the motives of the proposers, play upon known fears, or attempt to link a proposal to other unpopular issues or personalities. While the referendum outcome may be of critical importance to elites, voters do not necessarily perceive the consequences of a NO vote in quite the same way. These characteristics are shared by many other referendums that deal with constitutional issues or with other matters that are not immediately familiar to the mass public (LeDuc 2002, 2003). The first Irish referendum on the Nice treaty in 2001 displayed a similar campaign dynamic, as did the 1992 referendums on Maastricht held in Ireland, France, and Denmark. Australian referendums dealing with constitutional issues display many similar patterns (Galligan, 2001). However, it is also useful to draw a contrast, using cases that display a very different pattern of

246 Activating the Citizen

80

70

60

50

40

Yes No

30

20 Mar

Apr

May

Jun

Jul

Aug

Sep

R*

Figure 11.3: Public opinion in the Danish referendum on the Euro, March–September 2000 (%) R∗ = actual referendum result. Source: Brrsen Politiken.

opinion formation and change than was found in the Canadian and European constitutional referendums. A distinct contrast is found in those referendums where predispositions are stronger, due to the presence of clear partisan or ideological correlates or greater familiarity on the part of voters with the issue(s) of the referendum (LeDuc and Svensson, 2002). The referendums on the Euro held in Denmark in 2000 (see Figure 11.3) and in Sweden in 2003 displayed markedly less volatility, as did the 1995 Quebec sovereignty referendum (see Figure 11.4) or the 1997 Scottish referendum on devolution. As is seen in the 2000 Danish case, in which the NO side also prevailed, the contest was much closer over the duration of much of the campaign, and there was considerably less volatility. In part, this is because the issue of the Euro was well known to the public and also because Euroscepticism has been a much stronger force in Danish politics in recent years. One would thus expect predispositions on an issue such as the Euro to be considerably stronger than on an issue such as the constitution. Nevertheless, there are similarities to the other more volatile cases. In the 2000 Danish referendum, all of the mainstream political parties supported the YES side.8 Also lined up in support of the Euro were the leaders of the major trade unions, nearly

Lawrence LeDuc

247

80

70

60

50

40

Yes No

30

20 Mar

Apr

May

Figure 11.4: Public opinion March–October 1995 (%) R∗ = actual referendum result.

Jun in

the

Jul

Aug

Quebec

Sep

Oct

sovereignty

R*

referendum,

Source: Léger and Léger.

all elements of the business community, and most of the nation’s newspapers. At the time that the date of the referendum was announced in March 2000, it appeared that the chances of success in such a setting were fairly good. But by mid-summer, the YES lead had evaporated. Polls in June showed the race at a dead heat, but the number of ‘undecided’ voters nevertheless remained high (15 to 20 per cent). The government’s principal strategy was to stress the potential economic dangers of continued Danish exclusion from the Eurozone – slower economic growth, a weaker currency, higher interest rates, and unemployment, etc. It also sought to separate the Euro question from larger European issues, arguing that the Euro was an economic issue that did not necessarily have any larger political implications. The Danish NO campaign in contrast was more diverse and wide ranging. Ranging from far left to far right, NO campaigners had various reasons for their opposition to the Euro. Some drew heavily on patriotic and nationalistic themes or portrayed EU institutions and practices as a threat to Danish democracy. The left parties that opposed the Euro tended to stress potential threats to the safeguards of the Danish welfare state. As in the Canadian

248 Activating the Citizen

constitutional referendum, a NO vote could also be construed as a rebuke to the political establishment. A very different partisan context is found in the 1995 Quebec sovereignty referendum (see Figure 11.4) where the ballot question was rooted in long-standing divisions reflected in the polarized structure of the Quebec party system. In the first referendum in 1980, the sovereignty issue in Quebec was still a new political phenomenon, and that campaign represented an important part of a long learning process for many voters. The Quebec government’s strategy of promoting sovereignty together with a continuing economic association with Canada at first appeared to be a winning political formula.9 But the referendum proposal nevertheless went down to a rather decisive defeat, in part because the NO side was able to effectively shift the terms of the debate over the course of the campaign, arguing instead for ‘renewed federalism’ as an alternative vision, effectively changing the subject of the discourse. The relative newness of these issues at that time and the abrupt shift of the campaign discourse meant that the 1980 decision was not a clear cut or easy one for many Quebec voters. By 1995 however, the positions of the federal and provincial political parties and their leaders were very different. Positions on the sovereignty issue by that time were well known and well entrenched. A Quebec electorate frustrated with the failed constitutional initiatives of the past 15 years (including the 1992 constitutional referendum) was much more prepared to listen to the arguments put forward by the YES side during the course of the 1995 campaign. There were simply fewer voters in 1995 who had not already made up their minds on an issue that had by that time become the defining cleavage of Quebec politics (Clarke and Kornberg, 1996). Further, the YES side in 1995 benefited from the campaign role played by the leader of the Bloc Québécois in the federal parliament, Lucien Bouchard, whose personal popularity in Quebec far exceeded that of the federal prime minister (Pammett and LeDuc, 2001). Bouchard, who effectively led the YES campaign, was the most popular politician in Quebec while the Liberal party, representing the federalist alternative, had become highly unpopular, along with its leaders at both the federal and provincial levels. As is seen in Figure 11.4, the outcome of the 1995 Quebec referendum was extremely close, but the NO side ultimately prevailed. With turnout of 94 per cent – far higher than is typical in federal and provincial elections in Quebec – virtually the entire electorate was mobilized. This referendum campaign was, for the YES side in particular, a classic uphill struggle, in which gradually winning over small numbers of undecided

Lawrence LeDuc

249

or wavering voters became the essential campaign tactic. Survey evidence shows that more than three-quarters of the voters had already made up their minds well in advance of the referendum or at the time that it was called. While some of the same campaign tactics observed in other cases can be found in the 1995 Quebec sovereignty referendum, these were much less effective in such a highly polarized setting. In Zaller’s terms, the relative weights of information and predisposition in this case were simply quite different.

Campaign effects in constitutional referendums Voters typically enter referendum campaigns on constitutional issues with relatively weak predispositions and low levels of information. Over the course of a short campaign they will acquire enough information to allow them to come to a voting decision, often quite late in the campaign. Public opinion polls often find as much as a quarter to a third of voters still undecided in the final week of many referendum campaigns. Lupia and McCubbins (1998) argue that voters will utilize ‘short cuts’ to help them find their positions on sometimes complex issues. On an issue such as the European constitution, short cuts may be essential since relatively few voters will immerse themselves in the details of such a complex legal document. But while short cuts may help the voter to come to a voting decision on an issue, they can also act to impede the process of deliberation, which demands greater attention and sensitivity to competing points of view. Some types of short cuts may actually encourage voters to close their minds to the political debate going on around them. Referendum campaigns dealing with constitutional questions often send mixed partisan and ideological signals to voters, thereby rendering the voting decision more difficult. While a political party may formally take a position on a referendum issue, internal divisions within a party often figure prominently in a campaign. This does not occur in cases where the YES and NO sides of the issue reflect the existing partisan alignment, such as the 1995 Quebec sovereignty referendum. But such cases are relatively rare. More common are those in which several of the mainstream parties attempt to occupy the same political space or in which parties find themselves internally divided. In these instances, voters who are accustomed to relying on party cues will be subjected to mixed signals. Ideology produces similar patterns. Some referendum issues may be ideological in character, but most of those dealing with constitutional matters are not. Thus, as is the case with partisanship, voters who are accustomed to receiving clear ideological signals will find

250 Activating the Citizen

it more difficult to form an opinion. This problem is compounded when a proposal is attacked simultaneously from both left and right, as was the case with the EU constitution referendums and the Swedish and Danish referendums on the Euro. The more widely a proposal is attacked from opposite directions, the more doubts will be raised in the minds of wavering voters over the course of a short campaign. Turnout tends to fluctuate more widely in referendums than it does in elections. Turnout was high in the 1992 Canadian constitutional referendum and moderately high in the 2005 French and Dutch referendums on the constitutional treaty. But there are also several cases in which turnouts were very low in comparison with the levels generally obtained in elections. The Spanish referendum on the EU constitution drew participation from only 42 per cent of the electorate – 35 percentage points lower than in the general election of the previous year. Turnouts in both Irish referendums on the Nice treaty were very low in comparison with national elections, and the low turnout of only 35 per cent of voters in the first referendum was widely blamed for the defeat of the treaty. Polls in the run-up to the referendum had shown a majority of the Irish public in support of the treaty. But the combination of low levels of information regarding its content, a lacklustre campaign, and widespread disinterest combined to defeat it. But turnout cannot be considered in isolation. It is affected by many of the other factors discussed here, including the timing of votes, party mobilization, levels of information, and the context of the campaign. The presence of various second-order effects in many referendum campaigns is also important in affecting the outcome. Often, these are negative, and NO campaigners move quickly to tie the referendum issue to an unpopular political leader or attempt to turn the entire campaign into a referendum on the government. But there are also instances in which such secondary effects can be positive. As noted earlier in this paper, the outcomes of the EU constitutional referendums in Spain and Luxembourg were different than those in France and the Netherlands, in part because of the directional differences in second-order effects. Nevertheless, the number of cases of constitutional referendums in which second-order effects worked to the advantage of the NO side appears far greater than the few instances where they may have had a clearly positive effect. Because so many referendum campaigns originate as projects undertaken and promoted by political elites, it is often an effective campaign strategy to attempt to turn the referendum into a battle of ‘the people’ against ‘the establishment’. Almost counter-intuitively, the broader the

Lawrence LeDuc

251

base of support for a ballot proposition, the easier it is to shift the campaign discourse in the direction of anti-establishment or anti-politics sentiment. When the main political parties, big business, organized labour, academics, and the media all line up in support of a proposal, it is not as difficult as it might seem to turn the public against it. A very similar set of circumstances occurred in the 2008 Irish referendum on the Lisbon Treaty, in which 53 per cent of Irish voters rejected a revised set of constitutional proposals that had been put together by European leaders in the wake of the results of the 2005 referendums.10 Constitutional referendums often deal with a large package of issues rather than a single question. The Treaty Establishing a Constitution for Europe, for example, contained hundreds of separate articles. NO campaigners could thus choose between stressing broad themes or being more selective in their critiques. Similarly, the 1992 Canadian constitutional agreement contained a number of specific provisions that were contentious among different groups of voters. While the YES campaign urged voters to accept the package as a carefully balanced set of compromises on these difficult issues, NO campaigners were quick to take it apart. As the campaign progressed, opposition to any one of the proposals was sufficient to turn some voters against the agreement.

Conclusion The most volatile referendum campaigns are those where there is little partisan, issue, or ideological basis on which to form an opinion. Lacking such information, voters need more time to come to a decision, and that decision becomes highly unpredictable. Such cases often involve elites taking strong positions at the beginning of the campaign, to which the public gradually begins to react. The 2005 referendums on the EU constitutional treaty, the 1992 Canadian referendum, and many other European referendums, such as those on the Maastricht and Nice treaties and the 2008 Irish referendum on the Lisbon Treaty, display such a pattern (On some of the other important European referendums cited here, see Gallagher and Uleri, 1996; Gilland, 2002; and Svensson, 2002). Surveys taken in the aftermath of a referendum campaign regularly show that ‘insufficient information’ is one of the most common complaints of citizens about the referendum process. Particularly in circumstances where large and complex packages of proposals are put to a vote, such as in the Canadian and European constitution referendums, information becomes critical. The challenge presented by the European constitution referendums in this regard was indeed formidable.

252 Activating the Citizen

The Treaty Establishing a Constitution for Europe ran to just under 500 pages of complex legal text. Even a summary of the document widely circulated during the campaigns ran to over 90 pages. In the case of the 1992 Canadian constitutional referendum, one of the government’s first mistakes was its inability to produce an accurate legal text of the agreement for the first several weeks of the campaign. This allowed opponents of the proposals to accuse the government of having a ‘hidden agenda’ or of attempting to mislead or misinform the public. While the text of the European constitutional treaty was available for inspection during the 2005 referendum campaigns, its length and complexity opened the door to similar claims. In some jurisdictions, this problem is addressed by having neutral authorities provide ‘balanced’ information. In Ireland for example, an independent referendum commission is charged with the responsibility of disseminating unbiased information to the public through its own publications and of promoting debate and discussion of the issue(s) over the course of the campaign. But this mechanism proved insufficient in the case of the first Nice referendum, and the relative lack of information concerning the treaty was one of the factors leading to the low turnout of voters that was widely blamed for its defeat (Gilland, 2002). Changes in the law regarding the function and powers of the commission were made following that referendum, and it mounted a more substantial public information campaign during the second referendum in 2002. While many other factors certainly came into play, the level and quality of information available to voters may well have been the difference accounting for the opposite results of the two Irish referendums on Nice. But information was not sufficient to save the Lisbon Treaty in the 2008 Irish referendum, and the YES side made little effort to inform voters, relying instead on tepid slogans such as, ‘Say YES to Europe’. This made it easy for NO campaigners to attack the treaty from a number of different directions by claiming that it would entrench a neo-liberal economic agenda, threaten Irish neutrality, and deprive Ireland of its proper weight in EU governance. The political advantage in referendum campaigns, particularly those dealing with unfamiliar constitutional questions, often seems to rest with the NO side. Those opposed to a proposal do not necessarily have to make a coherent case against it. Not uncommonly, it is enough merely to raise doubts about it in the minds of voters, question the motives of the proposers, play upon known fears, or attempt to link a proposal to other less popular issues or personalities. Even some referendum campaigns that have begun with great optimism have fallen victim to these tactics

Lawrence LeDuc

253

over the course of a short intense campaign. This raises serious questions about the quality of public ‘deliberation’ in referendum campaigns and perhaps also about the capacity of direct democratic devices to fully resolve certain types of political issues. The common arrangement under which constitutional issues are first debated by elites and then turned over to the citizens for decision in a short campaign may not be the best formula for conducting constitutional politics. Involvement of citizens earlier in the process, or a lengthier period of deliberation, may make it possible for citizens to be better informed and participate more fully in the debate. In the wake of the collapse of the constitutional treaty following the 2005 French and Dutch referendums, one might have expected some attempt to engage citizens more systematically in a deliberative process regarding EU governance. Instead, political leaders opted for a process that was even more elite driven and that sought to avoid the necessity of submitting any new agreement to a referendum. Had it not been for constitutional provisions that made a referendum mandatory in Ireland, this strategy might have succeeded, at least in the very short term. But the Irish NO to the Lisbon Treaty, like the French and Dutch NO to its predecessor, has simply acted to block any progress towards a resolution of the problem. The current constitutional impasse in Europe, as in Canada, has been created by a flawed process that is particularly unsuited to facilitating the adoption of a constitution in a democratic political culture. The unwillingness of the public to accept a process of constitutional reform dominated by elites is understandable in that the only way many people feel their voice will be heard is to vote NO on any proposal put forward in such a manner.

Notes 1. In France, the president may submit a proposal directly to the people under Article 11 of the constitution or to a joint sitting of the National Assembly and Senate. In the latter case, a 3/5 vote is required for approval. 2. Data are from a special Eurobarometer (#62) conducted in November 2004, a month after the signing of the treaty. See European Commission (2005a). 3. For example, Laurent Faubius, a former prime minister and potential presidential candidate. 4. Turnout in the first round of the 2002 National Assembly election in France was 64 per cent, whereas that in the 2004 European Parliament election was only 43 per cent. Turnout in the 1992 referendum on the Maastricht Treaty was 70 per cent. 5. Of the 25 EU members, support among Dutch respondents was the third highest, surpassed only by Italy and Belgium. The Special Eurobarometer

254 Activating the Citizen

6.

7. 8.

9.

10.

found 63 per cent of those surveyed in November 2004 expressing support for the treaty and only 11 per cent opposing it. The remainder (26 per cent) had no opinion at that time European Commission (2005a). Russell divides the modern history of Canadian constitutional politics into six distinct periods, beginning with the Fulton-Favreau proposal for a constitutional amendment formula put forward in 1964. See Russell (1993). See also Cook (ed.) (1994). Bloc Québécois and Reform party identifiers, however, voted overwhelmingly NO. See LeDuc and Pammett (1995). Opposed were two right-wing parties – the Danish Peoples’ Party and the Progress Party; one centre-right party – the Christian Peoples’ Party; and two left-wing parties – the Socialist Peoples’ Party and the Red–Green Alliance. See Marcussen and Zølner, 2001; Downs, 2001. The text of the 1980 Quebec referendum question was as follows: ‘The Government of Quebec has made public its proposal to negotiate a new agreement with the rest of Canada, based on the equality of nations. This agreement would enable Quebec to acquire the exclusive power to make its laws, levy its taxes, and establish relations abroad – in other words, sovereignty – and at the same time, to maintain with Canada an economic association including a common currency. No change in political status resulting from these negotiations will be effected without approval by the people through another referendum. On these terms, do you agree to give the Government of Quebec the mandate to negotiate the proposed agreement between Quebec and Canada?’ Ireland was the only EU member country to hold a referendum on the proposals contained in the Lisbon Treaty, which took place after 18 other countries had already ratified the agreement without resorting to a referendum.

References Aarts, K. and H. van der Kolk (2006) ‘Understanding the Dutch NO: The Euro, the East and the Elite’, PS: Political Science and Politics, 39, 243–6. Clarke, H. D. and A. Kornberg (1996) ‘Choosing Canada?: The 1995 Quebec Sovereignty Referendum’, PS: Political Science and Politics, 29, 676–82. Cook, C. (ed.) (1994) Constitutional Predicament: Canada after the Referendum of 1992 (Montreal: McGill-Queen’s University Press). DeBardeleben, J. and A. Hurrelmann (eds) (2007) Democratic Dilemmas of MultiLevel Governance: Accountability and Legitimacy in the European Union (Basingstoke: Palgrave Macmillan). Downs, W. (2001) ‘Denmark’s Referendum on the Euro’, West European Politics, 24, 222–6. European Commission (2005a) The Future Constitutional Treaty: First Results (Special Eurobarometer), http://www.europa.eu.int/comm/public_opinion/index_ en.htm European Commission (2005b) Ratification of the Treaty Establishing a Constitution for Europe, http://europa.eu.int/constitution/referendum_en.htm Franklin, M., C. van der Eijk, and M. Marsh (1995) ‘Referendum Outcomes and Trust in Government: Public Support for Europe in the Wake of Maastricht’, West European Politics, 18, 101–17.

Lawrence LeDuc

255

Gallagher, M. and P. Vincenzo Uleri (eds) (1996) The Referendum Experience in Europe (London: Macmillan). Galligan, B. (2001) ‘Amending Constitutions Through the Referendum Device’ in M. Mendelsohn and A. Parkin (eds) Referendum Democracy: Citizens, Elites and Deliberation in Referendum Campaigns, (Basingstoke: Palgrave – now Palgrave Macmillan) pp. 109–24. Garry, J., M. Marsh, and R. Sinnott (2006) ‘Second-order versus Issue-voting Effects in EU Referendums: Evidence from the Irish Nice Treaty Referendums’, European Union Politics, 6, 201–21. Gilland, K. (2002) ‘Ireland’s (First) Referendum on the Treaty of Nice’, Journal of Common Market Studies, 4, 527–35. Hobolt, S. Binzer (2005) ‘When Europe Matters: The Impact of Political Information on Voting in EU Referendums’, Journal of Elections, Public Opinion and Parties, 1, 85–109. Ivaldi, G. (2006) ‘Beyond France’s 2005 Referendum on the European Constitutional Treaty: Second Order Model, Anti-Establishment Attitudes and the End of the Alternative European Utopia’, West European Politics, 29, 47–69. Jahn, D. and A.-S. Storsved (1995) ‘Legitimacy Through Referendum: the Nearly Successful Domino Strategy of the EU Referendums in Austria, Finland, Sweden and Norway’, West European Politics, 18, 18–37. Johnston, R., A. Blais, E. Gidengil, and N. Nevitte (1996) The Challenge of Direct Democracy: the 1992 Canadian Referendum (Montreal: McGill-Queens University Press). LeDuc, L. (2002) ‘Referendums and Elections: How Do Campaigns Differ?’ in D. Farrell and R. Schmitt-Beck (eds) Do Political Campaigns Matter? Campaign Effects in Elections and Referendums (London: Routledge). LeDuc, L. (2003) The Politics of Direct Democracy: Referendums in Global Perspective (Toronto: Broadview Press). LeDuc, L. (2005) ‘Saving the Pound or Voting for Europe?: Expectations for Referendums on the Constitution and the Euro’, Journal of Elections, Parties and Public Opinion, 15, 169–96. LeDuc, L. (2006) ‘Referendums and Deliberative Democracy’. Paper presented to the World Congress of the International Political Science Association in Fukuoka, Japan. LeDuc, L. (2007) ‘Voting NO: The Negative Bias in Referendum Campaigns’. Paper presented to the European Consortium for Political Research Joint Sessions Workshops in Helsinki, Finland. LeDuc, L. and J. H. Pammett (1995) ‘Referendum Voting: Attitudes and Behaviour in the 1992 Constitutional Referendum’, Canadian Journal of Political Science, 28, 3–33. LeDuc, L. and P. Svensson (eds) (2002) Interests, Information and Voting in Referendums [special issue] European Journal of Political Research, 41(6). Lupia, A. and M. McCubbins (1998) The Democratic Dilemma: Can Citizens Learn What They Need To Know? (New York: Cambridge University Press). Magnette, P. and K. Nicolaidis (2004) ‘The European Convention: Bargaining in the Shadow of Rhetoric’, West European Politics, 27, 381–404. Marcussen, M. and M. Zølner (2001) ‘The Danish EMU Referendum 2000’, Government and Opposition, 36, 379–401.

256 Activating the Citizen Marthaler, S. (2005) ‘France’s Referendum on the EU Treaty, May 2005’, Representation, 41, 228–36. Pammett, J. H. and L. LeDuc (2001) ‘Sovereignty, Leadership and Voting in the Quebec Referendums’, Electoral Studies, 20, 265–80. Russell, P. (1993) Constitutional Odyssey, 2nd edn (Toronto: University of Toronto Press). Svensson, P. (2002) ‘Five Danish Referendums on the European Community and European Union: A Critical Assessment of the Franklin Thesis’, European Journal of Political Research, 41, 733–50. Zaller, J. R. (1992) The Nature and Origins of Mass Opinion (New York: Cambridge University Press).

12 Youth Engagement, Civic Education, and New Vehicles of Political Participation Eileen Saunders

There is no shortage of pundits who decry what is wrong with youth today, especially when it comes to the subject of political participation and civic engagement. They describe a generation that does not vote, does not follow the news, is uninformed about current political issues, and does not engage with public life – in short, a disaffected, disengaged, and politically cynical cohort. Youth are seen as a symptom of, and a contributor to, a decline in democracy much lamented in recent academic research (for example, Putnam, 2000). Many embrace what Osler and Starkey (2003) have labelled as a ‘deficit model, which defines young people as less good citizens’ (p. 245). The problematic phenomena identified by these critics are precisely those detailed in the earlier sections of this book, namely declining participation in elections and political parties; we know that it is predominantly young people who are less likely to engage in these activities. There are others who argue that youth engagement has not declined so much as it has transformed; if we look beyond the traditional markers of formal political participation and activity, it is argued, we find evidence of an interested and politically active generation (for example, Bennett, 2007). They argue that we need to rethink what we mean by political participation and look beyond the more formal spheres to find examples of youth engagement. In this approach, indicators include volunteering in communities, membership in interest groups, and participation in activities that, while outside the traditional channels of political participation, may nevertheless be broadly defined as political. How are we to understand, for example, the politics of consumption as a means by which many youth choose to engage politically? Does seeking out fairtrade, sweatshop-free products constitute a political act? In a cultural

257

258 Activating the Citizen

and social context where lifestyle politics come to the fore, how are we to make sense of a new repertoire of political activity? Whatever position one takes in this debate between what Bennett (2007) characterizes as the ‘engaged youth’ and ‘disengaged youth’ paradigms and Loader (2007) labels the ‘disaffected citizen’ and ‘cultural displacement’ perspectives, there is considerable evidence of a marked generational change in the attitudes and political behaviour of youth in their relationship to formal political processes and institutions. Elsewhere in this collection is evidence of the participation trends of youth in the European Union as well as Canada. How one interprets the longterm consequences of these trends for democratic processes depends on one’s assumptions about the character of citizenship in modern democracies, how citizenship is formed, and the essential competencies citizens require. Increasingly, questions about the political engagement of youth are caught up in and reflect debates about the contested character of citizenship. The challenge is to understand how and where youth come to define themselves as citizens and what motivates political interest and action among youth. In the discussion that follows, I begin by suggesting we need to look at our understanding of citizenship and of the skills necessary for competent citizens. Civic education, or citizenship education, has taken on increased importance in recent academic and policy discussions in most Western countries as a possible solution to the declining formal political participation of youth. What follows is an exploratory discussion that attempts to look at citizenship education in a broad sense. It examines two venues, schools and the media, as potentially valuable sites for the development and acquisition of civic competency, but also as sites facing challenges. In particular, new information and communication technologies are changing the landscape and nature of political information and political action. My argument is that the lesson for policy makers and those interested in engaging youth is that we need to rethink our traditional practices and venues of citizenship education.

Youth, citizenship, and civic education Various solutions have been proposed for declining political participation, from lowering the voting age to reforming electoral systems. Recently there has been a renewed focus on, and optimism about, the role of civic education. In the United Kingdom (UK), for example, a revitalized citizenship education strategy was introduced in 2002 as a ‘core and transformatory element’ of a mandated curriculum (Selwyn, 2007,

Eileen Saunders 259

p. 131). In Canada, O’Neill (2007) has called on policy makers to recognize that civic education ‘stands as an effective, if not the most effective, mechanism for addressing deficiencies in political knowledge and understanding of the public responsibilities associated with citizenship’ (p. 32). It is not surprising that civic education is very much on the public agenda, raising questions regarding how and where youth learn to be citizens, as well as what they learn about citizenship. As Gifford (2004) notes, ‘Citizenship is a highly contested concept and this is reflected in the tensions that arise over the objectives of citizenship education’ (p. 145). A starting point is to define the goal of citizenship education. At a basic level we might agree that the goal is to establish civic competency, defined broadly to include the knowledge, skills, and attitudes necessary for effective participation in democratic society. Youniss, Bales, ChristmasBest, Diversi, McLaughlin, and Silbereisen (2002), for example, describe it as ‘an understanding of how government functions, and the acquisition of behaviours that allow citizens to participate in government and permit individuals to meet, discuss, and collaborate to promote their interests within a framework of democratic principles’ (p. 124). Others see this as too narrow and argue that we need to think of the requisite cultural conditions for effective citizenship. Hermes (2006), for example, uses the notion of ‘cultural citizenship’ to focus on community building and the construction of shared identities as key to democratic citizenship, while Dahlgren (2000) proposes the concept of ‘civic culture’ to describe ‘those social and cultural factors which constitute everyday preconditions for all democratic participation’ (p. 336). Specifically, Dahlgren cites four key elements of civic culture: ‘relevant knowledge and competencies’ (roughly equivalent to civic competency), ‘loyalty to democratic values and procedures’, ‘practices, routines, traditions’ (which can range anywhere from electoral voting to practices regarding political discourse), and ‘identities as citizens’ (pp. 336–7). A debate quickly ensues, however, over the constituent elements of necessary knowledge, values, skills, and practices. The choices we make are very much tied to the model of citizenship we embrace. In assessing the introduction of a mandatory citizenship education curriculum in the UK following the release of the Crick Report (see Advisory Group on Citizenship, 1998), Gifford (2004) argues that it was centred on a project of nation building ‘that uses political membership of the state as the basis for social integration’ (p. 146). If preparation for national citizenship is the key goal, it is not surprising that the emphasis will be on the knowledge, skills, and practices necessary for effective integration

260 Activating the Citizen

into, and acquisition of, a national identity. In this view, the purpose of civic education is ‘to ensure that young people understand their present and future roles within the constitutional and legal framework of the state in which they live’ (Osler and Starkey, 2003, p. 244). The problems with this model are several. Its assumption of a homogeneous national cultural identity and its supposition of political identity bound to geographical boundaries are undermined by the realities of increased immigration, cultural pluralism, and processes of regional integration. As Gifford adds, this model is ‘primarily focused on individual capacities and the re-direction of morals and values’ (2004, p. 150). An alternative model, which does not privilege national identity, suggests that in an increasingly diverse and globalized world, with membership in multiple communities and regional integration structures such as the European Union, citizens need knowledge and skills that permit them to participate locally, nationally, regionally, and globally. Bennett (2000) notes that we need to ask whether ‘national citizenship [is] slowly giving way to or being supplemented by “issue tribes” and global citizen initiatives?’ (p. 309). Some suggest this shift requires an education for cosmopolitan citizenship (Osler and Starkey, 2003). Increasingly, youth develop ties to others not through the bond of a national identity but ‘because they may be members of a diasporic group, have a common faith or political agenda, or live in a particular neighbourhood’ (Osler and Starkey, 2003, p. 246). We need, therefore, to develop skills that allow youth to participate effectively within and across a range of communities. A related area of debate concerns the norms and practices of good citizenship we wish to encourage in youth. A growing body of literature describes a fundamental qualitative shift in the norms of citizenship and the differential distribution of these norms. Described as a shift from a ‘dutiful citizen’ to an ‘actualizing citizen’ model by Bennett (2003) and from ‘duty-based citizenship’ to ‘engaged citizenship’ by Dalton (2006), what is being depicted is the emergence of competing understandings of the practices of citizenship in modern democracies and of what is required to be a ‘good citizen’. More importantly, it suggests that youth in Western democracies, facing new contexts, think differently about the rights and responsibilities of citizenship than previous cohorts. Briefly stated, what Bennett (2003) has called the dutiful citizen model emphasizes the following citizenship practices: the ‘obligation to participate in government activities’; voting in elections as ‘the core democratic act’; the duty to inform oneself about political matters; and expression of one’s interests through formal organizations, such as political

Eileen Saunders 261

parties or civil society associations (p. 6). In contrast, the engaged citizen model described by Dalton (2006) emphasizes norms of participation in the ‘non-electoral domain’ and ‘general political activity’ beyond formal institutions and processes. Bennett (2007) further characterizes this view of citizenship, one he labels the actualizing citizen, as ‘favoring loosely networked activism to address issues that reflect personal values’ (p. 4). The available evidence points to a differential dispersal of these views across age cohorts, with youth more likely to embrace norms of active citizenship and older cohorts more likely to think of the good citizen in the terms of a ‘dutiful citizen’. Dalton (2006), for example, demonstrates a ‘strong positive relationship’ between age and the espousal of a dutiful citizen mode and ‘somewhat greater support for norms of engaged citizenship among younger cohorts’ (p. 5). The question is this: What kind of civic competency are young people currently acquiring? Are we preparing youth to be dutiful citizens, while they actually need and are seeking the skills and knowledge to be engaged citizens? Are we focused only on passing on the requisite political knowledge for participation in formal political institutions and processes while youth are looking for ways to participate on a broader scale? Are we focused on preparing youth to participate only in the context of their national boundaries, when they require ways to connect transnationally? With little agreement on what kind of citizenship we want, it is not surprising that there is also little consensus regarding the kind of citizenship education young people need. When we ask where youth acquire citizenship skills and practices, several venues can be identified: Schools, government, mass media, family, and peer group settings all contribute in different ways to the formation of citizenship identity and the acquisition of citizenship skills among youth. Not surprisingly, however, when we talk about civic education we generally think in terms of formal school-based programmes, now mandatory in many educational jurisdictions. Certainly, this is where the bulk of academic research is focused. The role of schools in the construction of good citizens has long been taken for granted and civic education is ‘one of the oldest topics in political theory’ (Galston, 2001, p. 217). What is less clear is whether it produces the intended outcomes.

Schools, citizenship, and civic education Do school-based civic education programmes make a difference? The fact of the matter is that despite several decades of research, ‘we still know relatively little about what knowledge, both qualitatively and

262 Activating the Citizen

quantitatively, is desirable for an informed and active citizenry’ (Dudley and Gitelson, 2003, p. 266). In part this is a reflection of the lack of consensus on our models of democratic citizenship, and partly it reflects a debate over what the best indicators are for measuring civic competency. More importantly, there are huge gaps in our understanding of what civic competencies are the best predictors of formal political behaviour such as voting. There is conflicting evidence, for example, about whether civic education programmes can improve levels of political knowledge. In their analysis of data from the 1988 National Association of Education Progress Civics Assessment in the United States (US), Neimi and Junn (1998) argued that civic education classes can increase political knowledge by 4 per cent or more if current events and a broad range of issues are included in classroom discussion. Critics have disputed this claim, noting methodological problems with Neimi and Junn’s study and pointing out there is little evidence to suggest the increase in knowledge will persist over time (Greene, cited in Galston, 2004). There is also no agreement on whether an improvement in political knowledge, even if it can be demonstrated empirically, will have an impact on political behaviour such as voting. Often what is found are changes in attitudes toward voting, for example, rather than voting behaviour itself (Torney-Purta and Amadeo, 2003). There is also no consistent demonstration of higher levels of political knowledge or willingness to participate in established democracies compared to newly democratic countries. In their 28-country survey, Torney-Purta, Lehmann, Oswald, and Schulz (2001) found that post-communist countries such as Poland and the Czech Republic demonstrated levels of civic knowledge higher than the international mean, while established democracies such as Belgium and Portugal demonstrated levels below the international mean. Despite the lack of definitive proof of impact, the public consensus around the importance of school-based civic education appears strong, even among youth themselves. A Canadian Policy Research Networks workshop with Canadian youth in March 2007 found that participants thought civic education should ‘start earlier, be part of all courses and linked to real issues that are important to youth and their communities’ (Canadian Policy Research Networks, 2007, p. 2). The problem is that approaches to civic education vary significantly, ranging from highly centralized, compulsory programmes that span several age levels to decentralized, optional programmes with lax or undefined standards. In Canada, for example, education is a provincial jurisdiction and not all provinces have a mandatory course. This is much the same in Germany

Eileen Saunders 263

where civic education is subject to the guidelines of each federal state, resulting in no standardization of goals or content (Richter, 2006). In France, where national curricular standards are prescribed, a new compulsory course, ‘éducation civique, juridique et sociale’, was introduced in 1999 for students in their upper years of secondary schooling. Slovenia, on the other hand, which also has a national curriculum, has mandatory courses at both the primary and secondary levels (Kuhn, 2006). When civic education programmes are present, we find alternative visions of the ideal model. Ross (2004), for example, observes that citizenship education curricula can be placed on a continuum ranging from those that espouse a ‘citizenship transmission model’ to those that embrace an ‘informed social criticism model’. Dubnick (2003), on the other hand, suggests an underlying tension between a traditionalist philosophy, focused on moral education, and a developmentalist philosophy more concerned with teaching young people ‘how to think’ rather than ‘what to think’. There are also disagreements about pedagogical styles and practices, whether civic education is better taught as a single subject or across a range of courses, and the role of participation in extracurricular activities in the school or community. At a very broad level we can see there are several different elements involved in establishing civic competency. It is not feasible here to chart the relative distribution of these elements across the vast range of citizenship education initiatives. The different national contexts of education and the lack of consensus on what constitutes citizenship education prevent generalization. Rather I want to describe the main pedagogical goals in current practice, while recognizing that a particular civic education programme might combine some or all of the elements in different ways.

Character education Early civic education in North American schools reflected a traditional moral paradigm where the goal was ‘to instil in children those shared values that are central to the moral community of citizens, and in the process to inculcate those habits of character associated with being a good citizen’ (Dubnick, 2003, p. 253). While the most conservative versions of this have declined in significance, this element of citizenship education is still front and centre in many contexts, including western Europe. It assumes that there are predetermined norms and values necessary for democratic participation, which should be part of any civic education curriculum. In the UK, for example, the 1998 Crick Report, which

264 Activating the Citizen

established the foundation for the introduction of a renewed national citizenship education curriculum, identified ‘social and moral responsibility’ as the first element of citizenship education (cited in Gifford, 2004, p. 146). As the descriptor indicates, it is about building character and instilling the values deemed necessary in a democratic society. Knowledge transmission In order to participate effectively in political institutions and processes, youth need to be informed. Aside from instrumental knowledge, such as how elections function or how government works, youth need to be able to evaluate information in order to make reasoned political choices. The content is ‘based on the belief that certain factual information is important to the practice of good citizenship’ (Ross, 2004, p. 250). In practice this often translates into knowledge about legislatures. In Canada, for example, despite provincial variation in curricular guidelines, almost all ‘characterize civic knowledge as primarily procedural and legislative in content’ (Llewellyn, Cook, Westheimer, Girón, and Suurtamm, 2007, p. 28). Community service In the past decade in many jurisdictions, attention has focused on community service as a key component of citizenship education. Based on the assumption of learning by doing, community service programmes link citizen competency to participation in one’s community. As Westheimer and Kahne (2004) note, an attempt to organize civic education programmes around this goal assumes that social progress depends on the fact that ‘citizens must actively participate and take leadership positions within established systems and community structures’ (p. 27). Two different approaches to the incorporation of community service include mandated volunteering time and service learning. Ontario schools, for example, require 40 hours of volunteer time per individual student as a high school graduation requirement. Service learning programmes, on the other hand, tend to be course-based programmes that link volunteer activities to the curriculum. This option appears less often in European schools, where the tradition of service learning is more recent and less established than in North America. Critical citizen skills This approach focuses more on equipping students with the skills to engage in social critique. Its purpose is to provide ‘students opportunities for an examination, critique, and revision of past traditions, existing

Eileen Saunders 265

social practices, and modes of problem solving’ (Ross, 2004, p. 250). In France, for example, the mandatory civic education course is less geared towards the transmission of particular political knowledge and more towards ‘conceptual reflection’ on different aspects of citizenship in France and current events (Kuhn, 2006). Ideally, Westheimer and Kahne (2004) note, programmes of this type (what they describe as ‘justice-oriented’) are focused more on social change and ‘seek to prepare students to improve society by critically analysing and addressing social issues and injustices’ (p. 4). In practice, it is unclear whether students are learning the skills necessary to effectively participate in social change. In a recent study in Ontario schools, Llewellyn et al. (2007) concluded that youth were not acquiring ‘skills for political action’ (p. 25). In the UK, Selwyn (2007) notes, the focus is more on ‘delivery of information’ rather than the ‘lived experiences of citizenship-as-practice’ (p. 133). As noted earlier, we do not know definitively whether citizenship education will lead to formal political participation or whether some kinds of programmes are better predictors of behaviours, such as voting, than others. In the context of our earlier discussion of national and cosmopolitan citizenship and engaged and dutiful citizens, the implication is that educational policy makers need to design programmes that reflect generational shifts in political identity formation and political activity. More critically, they need to find pathways for youth to make connections between the cause/issue campaigns they espouse and the formal political channels for exercising democratic citizenship. This means giving students traditional information resources and skills as a foundation for political understanding, but it also means finding ‘ways of establishing the relevance of politics and of connecting the “micro-politics” of personal experience with the “macro-politics” of the public sphere’ (Buckingham, 2000, p. 221). We seem to be doing more of the former and not much of the latter. Ross (2004) suggests that in jurisdictions where the emphasis is on ‘standardized curriculum and strict accountability’, citizenship education that emphasizes ‘informed social criticism’ is under attack (p. 250). Then again, a focus on individual character as an element of citizenship education can sometimes be at odds with our understanding of democratic principles. As Westheimer and Kahne (2004) remind us, ‘A focus on loyalty and obedience . . . works against the kind of critical reflection and action many assume are essential in a democratic society’ (p. 242). Community service, when mandated as individual volunteer hours, has been criticized for doing little to develop an ethic of community engagement.

266 Activating the Citizen

An emphasis on individual responsibility and the capacity to improve or change one’s community deflects attention and skill development, critics argue, away from ‘the need for collective and often public sector initiatives’ (Westheimer and Kahne, 2004, p. 241). While we cannot state with any certainty that school-based civic education programmes lead to increased political participation, this is not a reason to dismiss their possible contribution in this area. I see them as a necessary but not sufficient condition for effective political participation. I further argue that if civic education is to have any lasting effect, it must address the means by which youth come into contact with political information in their everyday lives. In particular, we need to look at where youth develop a sense of political literacy and how this is changing in the context of new media use.

Youth, media, and informed citizenship News media have long been singled out as a key means of access to political information and civic knowledge. From media, we acquire the maps and scripts of citizenship; we learn the routines of democracy, the requisites of the good citizen, the values that bind us as a nation, and a range of citizenship practices. In much the same fashion as pundits have lamented the declining participation of youth in formal political processes, they have decried the turn of youth away from traditional sources of political information. Putnam (2000), for example, traces the decline in social capital partially to the decline in newspaper readership and the ‘connectedness’ it produces. This is in keeping with much political socialization literature, which argues for a correlation between exposure to newspapers and levels of political knowledge. Survey upon survey has documented the declining interest in news media among young people. Using data from the 2003 General Social Survey in Canada, Milan (2005) notes that only half of Canadians in their twenties follow the news daily compared to 89 per cent of seniors. Similar trends have been identified in the US and Europe. A 2006 Pew study on news consumption, for example, found that young people are far less likely to read newspapers on a regular basis; 24 per cent of participants ages 18 to 29 reported having read a paper the previous day, compared to 58 per cent of those ages 65 and older. Meanwhile in the UK, there has been a 25 per cent drop among youth who report watching television news (Schifferes, Ward, and Lusoli, 2007) Here, again, we see an implicit ‘deficit model’ of youth: ‘It is not democracy – or indeed news journalism, or the relationship between them – that is the problem, but people’s lack

Eileen Saunders 267

of interest in them. The resulting implication here is that the failure is that of young people themselves’ (Buckingham, 2000, p. 5). When we turn the question around and ask whether news media may actually hamper political engagement, we see a different scenario for citizenship. Lewis, Inthorn, and Wahl-Jorgensen (2005) argue that the manner in which citizens are portrayed in the news ‘helps to shape what it means to be a citizen in a democracy’ (p. 8). The evidence is not encouraging. In their extensive analysis of print and broadcast news in the UK and the US, Lewis et al. found that citizens were overwhelmingly represented as passive, rather than active, and as consumers, rather than citizens: ‘They do not make history, they react to it or simply experience it’ (p. 138). In a related vein, Barnhurst’s (1998) study of college-age youth in Spain and the US, describes a disconnect from traditional news media: ‘Much of what passes for political news on television and in the papers – because it simply doesn’t have meaning within the localities where they live – never reaches these young citizens. The news floats past them unanchored’ (p. 215). Buckingham (2000) found much the same disconnect among the young citizens he interviewed in the UK; they expressed frustration and cynicism with the format, content, manner of address, and relevance of news and a desire to be more informed. If news is conveyed in a manner that fails to connect with youth, and indeed that discourages active citizenship, it leads us to question whether the orientation of youth to political information in particular and civic culture more broadly is shifting. The question of what counts as news is especially important in regards to this cohort of young people, because ‘many of the previous boundaries have become problematic (e.g., journalism vs. popular culture, the personal and the political, citizen and consumer . . .)’ (Dahlgren, 2000, p. 337). One indicator of the blurring of lines between journalism and popular culture is the increasing trend, in North America at least, of comedy television as a source of political news for young people. A 2004 Pew study found that of all media sources of political news, comedy television gained the most young consumers between the 2000 and 2004 American elections, with over half of young people ages 18 to 24 reporting they had learned something about election news from these programmes (cited in Key Facts, 2004). Whether and how popular culture consumption might contribute to political participation is still unclear. The more important message, I believe, is the suggestion it offers about the orientation of youth to political information. It reminds us that youth political understanding is not constructed solely in relation to traditional ‘informational’ media but also in more broadly defined popular media venues. Moreover,

268 Activating the Citizen

popular media sites are emerging as important mobilizing agents in youth electoral turnout. Rock the Vote, originally a US-based effort to increase the turnout of young voters in the 2004 election by building on youth culture celebrities and venues, has been copied and extended around the globe. The question of the orientation of youth to politics and media becomes even more complex when we look at how their engagement with new media, in particular web-based media and mobile Internet technologies, affects their political participation. This is a hotly contested field with some commentators claiming transformative consequences and others dismissing this as naïve technological determinism. Without lapsing into a breathless optimism regarding new forms of electronic democracy, and with the available evidence suggesting mixed consequences, it is important to understand what forms of citizen participation new media allow and how youth are using these technologies to acquire civic competency and engage in political activity. Three areas appear critical for further investigation: information retrieval and sharing about politics, community building and networking, and political activism. Information and new media In one sense we can say new media simply offer a different venue for retrieving information about politics. In their research on Italian students’ Internet use, Calenda and Mosca (2007) confirmed that students used the Internet for retrieving political information. Moreover, their online use tended to reproduce their political participation offline. There is evidence to indicate that youth are more likely to turn to online news. In the 2005 UK election, for example, it was found that age was the key factor in determining who went online to follow election coverage (Schifferes et al., 2007) There is also evidence to suggest that youth are substituting traditional sources of information with the Internet. A 2004 study of online use among youth ages 15 to 24 years in ten different western European countries concluded that increased time spent online was accompanied by a reduction in time spent with other media such as television, radio, and newspapers. The distribution of total media use among the respondents was 24 per cent Internet, 31 per cent television, 27 per cent radio, 10 per cent newspapers, and 8 per cent magazines (Brown, 2004). While it is the case that adult cohorts are also going online in increasing numbers, the nature of their participation there seems to be different. O’Neill (2007) notes that Canadians ages 19 to 24 are much more likely than older cohorts to cite the Internet as a news source.

Eileen Saunders 269

This brings us back to the question of what counts as news or ‘relevant knowledge’. New media do more than simply deliver the news in a different venue; they provide alternative information for youth. The basic features of digital communication, when combined with new and innovative software that simplifies the creation and uploading of content, have transformed the information landscape available to youth. Whether in the form of blogs, interest group websites, text messaging, or video-sharing sites such as YouTube, new information sources are coming to play an unpredictable, and difficult to control, role in formal political processes. Moreover, the emergence of the citizen-as-journalist and the growth of alternative online news sites challenge traditional understandings of what counts as political information and who are the routine sources of that information. Indymedia.org, for example, first appeared in 1999 in the lead-up to the protests against the World Trade Organization meetings in Seattle with the purpose of reporting on issues and events traditional news did not cover; it is now a global network of over 135 news websites where volunteer contributors post news with local, regional, and international content (Chadwick, 2006). What is important here is that new media offer different voices and different views. In a study of young Swedish activists, Dahlgren and Olsson (2005) concluded that these youth saw the Internet as a way to ‘bypass traditional media and their views of the world in their search for various kinds of knowledge’ (p. 10). New media also offer a different format and ‘feel’. The interactive and often informal style of communication characterizing Internet exchanges ‘contrasts with the traditional, linear, hierarchical, logical, rule-governed conventions often used in official communications with youth’ (Livingstone, 2007, p. 166). It is no longer simply about retrieval. With the capacity for a user to be a producer and a consumer and with exchanges taking place in a less rule-bound, more interactive environment, we are beginning to see a more personalized orientation to political information. A 2004 survey, for example, found that young people are more attracted to online election information that is interactive and personalized, with chat rooms rated as the most popular online sites (cited in Key Facts, 2004). Community building and new media With the arrival of Web 2.0 as a platform for the expanded development of networked, web-based communities, we see a new potential for virtual spaces of interaction and discourse. Increasingly user-friendly software allows youth to create and link websites, forming extended networks

270 Activating the Citizen

around common interests. The significance of youth involvement in web-based communities for political participation is something that is little understood and only recently attracting serious attention. Researchers are pointing to the increased emergence and vitality of an online ‘youth civic culture’ composed of hundreds, if not thousands, of sites whose purpose is to ‘encourage and facilitate youth civic engagement by promoting thinking and dialogue about and participation in civic life, including partisan political participation, community involvement, philanthropy, and social activism’ (Montgomery and Gottlieb-Robles, 2006, p. 132). In a recent attempt to map and categorize these sites, over half of which were created by non-profit organizations, Montgomery and Gottlieb-Robles found that almost half of them incorporated participatory features and close to one-third invited participants to submit their own material, whether in the form of essays, opinion pieces, or artistic work. We must also consider the increasing presence of social networking sites, such as MySpace, Facebook, and Bebo, when examining the impact of new media. These sites have exploded in popularity in North America and Europe in recent years. The numbers are impressive when one considers they have largely appeared since 2004. Bebo, for example, registered more than 21 million users in its first year of operation, with 4 million in England alone (Sifrey, 2006). A recent survey found that 63 per cent of Canadians ages 18 to 34 have visited a social networking site, while approximately 55 per cent have registered their personal profile on at least one (Knight, 2007). Yet we are only beginning to explore their significance for political and civic engagement. Since its introduction in 2004, Facebook has primarily served as a social networking site, a place to connect with friends, post photos, and announce the next pub night. However, with over 60 million users, it also has increasingly served as a forum for youth to organize protests, form political issues groups, or run campaigns. These groups range across the political spectrum and across issues – from groups opposing gentrification in a Montreal neighbourhood to groups raising awareness about global poverty – but they tend to fly below the radar of what is considered political engagement. Political parties are only now starting to sit up and take notice. Increasingly, political leaders are reaching out to youth through these networks for recruitment purposes. We are also beginning to see the emergence of social networking sites built explicitly around politics. Essembly.com, for example, is a non-partisan social networking site where youth can find others with similar political interests, vote on posted ‘resolutions’, and engage in online discussion of political issues.

Eileen Saunders 271

What this points to is the role of the Internet not just as a tool f or political information but also as a space for political interaction and organization. Interestingly, a new trend is the merger of social networking and news information sites. In the spring of 2007, MySpace, which is owned by Rupert Murdoch’s News Corporation, launched a news aggregator service, MySpace News. In addition to searching news and blog sites from around the world, users can submit their own news items and vote on the top stories. Cyberprotests and cyberactivism Earlier, I noted the increasing interest of youth in engaged citizenship practices that extend beyond more formal institutions and processes. Here too, new media technologies facilitate new forms of political activism. From cell phone-organized protestors who gathered through text messaging to overthrow Philippines President Joseph Estrada in 2001 to vote-swapping websites that first appeared in the 2000 US election and quickly spread to the 2001 and 2005 UK general elections, we are seeing examples of new means of political activism (Chadwick, 2006). In some cases, such as the use of text messaging and Internet postings by youth in France during the 2005 street demonstrations, new media allow for fast and efficient mobilization for local offline political action. In other cases, mobilization is transnational in scope and centred on online political action. This can take, for example, the form of e-mail campaigns that target government or private sector actors, such as the massive e-mail campaign against Nike’s alleged use of sweatshop labour (Chadwick, 2006). The extent to which online political action can be distinguished from offline activities is open to debate, as is its longterm effectiveness. It is also unclear whether social networking sites will continue to expand as mechanisms for political recruitment and mobilization, but recent trends suggest they are growing in importance. As one example of the mobilization potential of these sites, more than 100,000 Californian high school students boycotted classes on 1, May 2006 to protest George Bush’s ‘anti-immigrant’ policies – and MySpace.com was the means by which many of them were mobilized to action (Melber, 2006). These online trends raise new research questions for scholars in this field. If we are to fully understand how to re-engage youth in formal political channels, we need first to understand how media – old and new – contribute to the formation of political understanding, political identity, and political action. We need to understand the changing citizenship practices that new media facilitate as well as the particular skills

272 Activating the Citizen

and competencies these media require for effective participation. And we also need to ask new questions about the relationship between news and political literacy. Rather than attempting to measure the effectiveness of news in communicating political information, we should be asking how it enables viewers to construct and define their relationship with the public sphere . . . How ultimately do they establish what it means to be a citizen? (Buckingham, 2000, p. 18)

Conclusion I began this chapter by arguing that if we want to find ways to engage youth politically, the challenge rests partly on how we define the objective of producing informed and active citizens. It also requires that we understand what motivates political interest and action among youth, how political participation is changing, and the consequences these have for formal political processes. Most importantly, we need to understand where youth learn to be citizens. There are several implications of the preceding discussion. First, we need to take account of the changing contexts in which youth seek out and consume political information as well as actively engage as producers of information. The challenge for policy makers is how to design and deliver political information in ways that attract and motivate youth. If they wish to make inroads in igniting the political interest and engagement of youth, this means paying more attention to the informal sites of popular media practices and the characteristic features of communication of those sites. Second, if we are to prepare young people effectively, we need to reconsider the role of school-based civic education. In light of the pervasiveness and patterns of media use, ‘the informal public pedagogies of popular (news and entertainment) media may be surpassing the formal public pedagogies of schooling and postsecondary education in terms of where and how we form citizens’ (Stack and Kelly, 2006, p. 9). This suggests that media education programmes designed to instil media literacy skills – for old and new media – need to be integral to any civic education programme. In addition to giving students the necessary political knowledge about formal institutions and processes, civic education must equip students with the skills to obtain knowledge and to organize, interpret, and prioritize it in an information-saturated society. It also means considering what digital citizenship skills are needed for youth to be effective

Eileen Saunders 273

participants and then giving them opportunities to practise and apply those skills, whether in their local communities around local issues or more globally in terms of transnational issues. Finally, we need to rethink deficit models in which youth figure as ‘less good’ citizens and focus on understanding how their norms and practices of citizenship are shifting. As Bennett (2007) has argued, the significance of these shifts ‘seems to point to changing the institutional and communication environments in which young people encounter politics rather than somehow fixing the attitudes of youth themselves’ (pp. 6–7). In other words, solving the ‘civic crisis’ may be a matter of adapting the system to a new informational environment rather than bringing youth back into the fold of a structure that in some senses is no longer relevant.

References Advisory Group on Citizenship (1998) Education for Citizenship and the Teaching of Democracy in Schools (London: Qualifications and Curriculum Authority). Barnhurst, K. G. (1998) ‘Politics in the Fine Meshes: Young Citizens, Power and Media’, Media, Culture & Society, 20, 201–18. Bennett, W. L. (2000) ‘Introduction: Communication and Civic Engagement in Comparative Perspective’, Political Communication, 17, 307–12. Bennett, W. L. (2003) Civic Learning in Changing Democracies. Working Paper #4: Center for Communication and Civic Engagement, University of Washington, http://depts.washington.edu/ccce/civiceducation/bennettciviclearning.pdf Bennett, W. L. (2007) Changing Citizenship in the Digital Age. Paper prepared for OECD/INDIRE Conference on Millennial Learners in Florence, Italy, http://www.oecd.org/dataoecd/0/8/38360794.pdf Brown, M. (2004) EIAA Media Consumption Study: Youth Online Results Europe 15–24 Age Group, http://www.unicef.org/voy/media/EIAA_media_ consumption_study. pdf Buckingham, D. (2000) The Making of Citizens: Young People, News and Politics (New York: Routledge). Calenda, D. and L. Mosca (2007) ‘The Political Use of the Internet’, Information, Communication and Society, 10(1), 29–47. Canadian Policy Research Networks (2007, Winter) ‘The Youth/Ballot Box Disconnect’, Network News, 36, http://www.cprn.org/documents/47408_en.pdf Chadwick, A. (2006). Internet Politics: States, Citizens and New Communication Technologies. (Oxford: Oxford University Press). Dahlgren, P. (2000) ‘The Internet and the Democratization of Civic Culture’, Political Communication, 17, 335–40. Dahlgren, P. and T. Olsson (2005) The Micro-Meshes of Civic Cultures: Political Engagement and Media Use. Paper presented at the Social Science Seminar at Karlstadt University, http://www.sam.kau.se/stv/ksspa/papers/dahlgren.pdf Dalton, R. J. (2006) Citizenship Norms and Political Participation in America: The Good News Is . . . the Bad News Is Wrong. CDACS Occasional Paper 2006-01.

274 Activating the Citizen Washington, DC: The Center for Democracy and Civil Society, Georgetown University. Dubnick, R. (2003) ‘Nurturing Civic Lives: Developmental Perspectives on Civic Education’, PS: Political Science and Politics XXXVI, 2, 253–5. Dudley, R. L. and A. R. Gitelson (2003) ‘Civic Education, Civic Engagement, and Youth Civic Development’, PS: Political Science and Politics XXXVI, 2, 263–7. Galston, W. A. (2001) ‘Political Knowledge, Political Engagement, and Civic Education’, Annual Review of Political Science, 4, 217–34. Galston, W. A. (2004, April) ‘Civic Education and Political Participation’, PS: Political Science & Politics, 263–6. Gifford, C. (2004) ‘National and Post-national Dimensions of Citizenship Education in the UK’, Citizenship Studies, 8(2), 145–58. Hermes, J. (2006) ‘Citizenship in the Age of the Internet’, European Journal of Communication, 21(3), 295–309. Key Facts (2004) Key Facts: Media, Youth, and Civic Engagement (The Henry J. Kaiser Family Foundation), http://www.kff.org/entmedia/upload/Media-Youthand-Civic-Engagement-Fact- Sheet.pdf Knight, K. (2007, 8 October) ‘Canadians Grow Social Networking Habits’, BizReport, http://www.bizreport.com/2007/10/canadians_grow_social_networking_ habits.html Kuhn, H.-W. (2006) European Approaches to Civic Education for the Promotion of Democracy. Paper presented at the German-American Conference DemocracyPromotion and International Cooperation in Denver, CO, available from http://www.civiced.org/index.php?page=calendar&eid=14423 Lewis, J., S. Inthorn, and K. Wahl-Jorgensen (2005) Citizens or Consumers? What the Media Tell Us about Political Participation (Maidenhead: Open University Press). Livingstone, S. (2007) ‘The Challenge of Engaging Youth Online’, European Journal of Communication, 22(2), 165–84. Llewellyn, K., S., Cook, J., Westheimer, L., Giron and K. Suurtamm, (2007) The State and Potential of Civic Learning in Canada. CPRN Research Report (Ottawa, ON: Canadian Policy Research Networks), http://www.cprn.org/ documents/48798_EN.pdf Loader, B. (ed.) (2007) Young Citizens in the Digital Age: Political Engagement, Young People and New Media (London: Routledge). Melber, A. (2006, 12 June) ‘MySpace, MyPolitics’, The Nation, http://www. thenation.com/doc/20060612/melber Milan, A. (2005) ‘Willing to Participate: Political Engagement of Young Adults’, Canadian Social Trends, 79, http://www.statcan.ca/english/freepub/11-008XIE/2005003/articles/8965.pdf Montgomery, K. and B. Gottlieb-Robles (2006) ‘Youth as e-Citizens: The Internet’s Contribution to Civic Engagement’ in D. Buckingham and R. Willet (eds) Digital Generations: Children, Young People, and New Media. (Mahwah, NJ: Lawrence Erlbaum Associates), pp. 131–47. Neimi, R. and J. Junn (1998) Civic Education: What Makes Students Learn (New Haven, CT: Yale University Press). O’Neill, B. (2007) Indifferent or Just Different? The Political and Civic Engagement of Young People in Canada. CPRN Research Report (Ottawa, ON: Canadian Policy Research Networks), http://www.cprn.org/documents/48504_EN.pdf

Eileen Saunders 275 Osler, A. and H. Starkey (2003) ‘Learning for Cosmopolitan Citizenship: Theoretical Debates and Young People’s Experiences’, Educational Review, 55(3), 243–54. Pew (2006) Pew Research Center for the People & the Press News Consumption and Believability Study, http://people-press.org/reports/pdf/282.pdf Putnam, R. (2000) Bowling Alone: The Collapse and Renewal of American Community (New York: Simon and Schuster). Richter, D. (2006) A Review of Research: What Do We Know About Civic Education in Classrooms in Germany? Paper presented at the German-American Conference Democracy-Promotion and International Cooperation in Denver, CO, available from http://www.civiced.org/index.php?mact=News,cntnt01,detail, 0&cntnt01articleid=83&&cntnt01returnid=338 Ross, E. W. (2004, April) ‘Negotiating the Politics of Citizenship Education’, PS: Political Science & Politics, 249–51. Schifferes, S., S. Ward, and W. Lusoli (2007) What’s The Story . . .? Online News Consumption in the 2005 UK Election. Unpublished manuscript, http://www.esri.salford.ac.uk/ESRCResearchproject/papers/schifferes_et_al_2007. pdf Selwyn, N. (2007) ‘Technology, Schools and Citizenship Education: A Fix Too Far?’ in B. Loader (ed.) Young Citizens in the Digital Age (London: Routledge), pp. 129–42. Sifrey, M. (2006, March 13) ‘Essembly.com: Finally, a Friendster for Politics’, Personal Democracy Forum, http://www.personaldemocracy.com/node/849 Stack, M. and D. Kelly (2006) ‘Popular Media, Education, and Resistance’, Canadian Journal of Education, 29(1), 5–25. Torney-Purta, J. and J.-A. Amadeo (2003, April) ‘A Cross-National Analysis of Political and Civic Involvement Among Adolescents’, PS: Political Science & Politics, 269–74. Torney-Purta, J., R. Lehmann, H. Oswald, and W. Schulz (2001) Citizenship and Education in Twenty-eight Countries: Civic Knowledge and Engagement at Age Fourteen (Amsterdam: International Association for the Evaluation of Educational Achievement). Westheimer, J. and J. Kahne (2004) ‘What Kind of Citizen? The Politics of Educating for Democracy’, American Educational Research Journal, 41(2), 237–69. Youniss, J., S. Bales, V. Christmas-Best, M. Diversi, M. McLaughlin, and R. Silbereisen (2002) ‘Youth Civic Engagement in the Twenty-First Century’, Journal of Research on Adolescence, 12(1), 121–48.

Bibliography Aarts, K. and H. van der Kolk (2006) ‘Understanding the Dutch NO: The Euro, the East and the Elite’, PS: Political Science and Politics, 39, 243–6. Abelson, J. and J. Eyles (2002) Public Participation and Citizen Governance in the Canadian Health System, Discussion Paper 7 (Ottawa, ON: Commission on the Future of Health Care in Canada). Achen, Christopher H (2002) ‘Toward a New Political Methodology: Microfoundations and ART;’ Annual Review of Political Science 5, 423–50. Adams, B. (2004) ‘Public Meetings and the Democratic Process’, Public Administration Review, 64(1), 43–54. Alderman, K. (1999) ‘Revision of Leadership Election Procedures in the Conservative Party’, Parliamentary Affairs, 52(2), 260–74. Almond, G. A. and S. Verba (1963) The Civic Culture: Political Attitudes and Democracy in Five Nations (Princeton, NJ: Princeton University Press). Andersen, J. and J. Goul Andersen (2003) ‘Klassernes forsvinden’ [The Disappearance of the Classes] in J. Goul Andersen and O. Borre (eds) Politisk forandring: Værdipolitik og nye skillelinjer ved folketingsvalget 2001 [Political Change: Value Politics and New Lines of Division in the Danish National Elections 2001] (Aarhus, Denmark: Systime Academic), pp. 207–22. Ansell, C. and J. Gingrich (2003) ‘Reforming the Administrative State’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 164–91. Appleton, A. M. (2000) ‘The France that Doesn’t Vote: Nonconsumption in the Electoral Market’ in M. S. Lewis-Beck (ed.) How France Votes (New York: Chatham House), pp. 206–26. Archer, K. and A. Whitehorn (1997) Political Activists: The NDP in Convention (Toronto: Oxford University Press). Arnstein, S. (1969) ‘A Ladder of Citizen Participation’, Journal of American Institute of Planners, 35, 216–24. Auers, D. (2005) ‘European Elections in Eight New EU Member States’, Electoral Studies, 24, 741–84. Baimbridge, M. D. (2005) ‘Euphoria to Apathy: EP Turnout in the New Member States’ in Juliet Lodge (ed.) The 2004 Elections to the European Parliament (Basingstoke: Palgrave Macmillan), pp. 45–56. Banducci, S. A., T. Donovan, and J. A. Karp (1999) ‘Proportional Representation and Attitudes about Politics: Results from New Zealand’, Electoral Studies, 18(4), 533–55. Barber, B. (1984) Strong Democracy: Participatory Democracy for a New Age (Berkeley, CA and Los Angeles: University of California Press). Bardi, L. (2002) ‘Italian Parties’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 46–76. Barnes, S. and M. Kaase (1979) Political Action (London: Sage). 276

Bibliography 277 Barnhurst, K. G. (1998) ‘Politics in the Fine Meshes: Young Citizens, Power and Media’, Media, Culture & Society, 20, 201–18. Beetham, D. and C. Lord (1998) ‘Legitimacy and the European Union’ in A. Weale and M. Nentwich (eds) Political Theory and the European Union (London: Routledge), pp. 15–33. Bellamy, R. and D. Castiglione (2000) ‘The Uses of Democracy: Further Reflections on the Democratic Deficit in Europe’ in E. O. Eriksen and J. E. Fossum (eds) Integration through Deliberation? On the Prospects for European Democracy (London: UCL Press), pp. 65–84. Bennett, W. L. (2000) ‘Introduction: Communication and Civic Engagement in Comparative Perspective’, Political Communication, 17, 307–12. Bennett, W. L. (2003) Civic Learning in Changing Democracies. Working Paper #4: Center for Communication and Civic Engagement, University of Washington, http://depts.washington.edu/ccce/civiceducation/bennettciviclearning.pdf Bennett, W. L. (2007) Changing Citizenship in the Digital Age. Paper prepared for OECD/INDIRE Conference on Millennial Learners in Florence, Italy, http://www.oecd.org/dataoecd/0/8/38360794.pdf Bennulf, M. and P. Hedberg (1999) ‘Utanför demokratin:Omdet minskade valdeltagandets sociale och politiska rötter’ [Outside of Democracy: On the Social and Political Roots of Declining Electoral Participation] in E. Amnå (ed.) Valdeltagande i förandring [Changes in Electoral Participation] Demokratiutredningens forskarvolym, XII (Stockholm: SOU), pp. 75–135. Benz, A. (2008) ‘The Treaty of Lisbon – A Step towards a More Effective and Democratic European Union’, EUCE Canadian Network Newsletter, 2(1), 1, 3, http:// www.euce-network.carleton.ca/newsletter/V2(1)-EUCENewsletter.pdf Berelson, B. R., P. F. Lazarsfeld, andW. N. McPhee (1954) Voting: A Study of Opinion Formation in a Presidential Campaign (Chicago: University of Chicago Press). Bernhagen, P. and M. Marsh (2007) ‘The Partisan Effects of Low Turnout: Analyzing Vote Abstention as a Missing Data Problem’, Electoral Studies, 26(3), 548–60. Bernhagen, P. and M. Marsh (2007) ‘Voting and Protesting: Explaining Citizen Participation in Old and New European Democracies’, Democratization, 14(1), 44–72. Biezen, I. van (2004) ‘Political Parties as Public Utilities’, Party Politics, 10, 701–22. Bjørklund, T. (1999) Et lokalvalg i perspektiv: Valget i 1995 i lys av sosiale og politiske endringer [A Local Election in Perspective: The 1995 Election in light of Social and Political Changes] (Oslo: Tano Aschehoug). Bjørklund, T. (2002) ‘Den synkende deltakelsen ved lokalvalg – noen årsaker og konsekvenser’ [The Declining Participation in Local Elections – Some Causes and Consequences] in B. Aardal (ed.) Valgdeltakelse og lokaldemokrati [Electoral Participation and Local Democracy] (Oslo: Kommuneforlaget), pp. 69–88. Blais, André (2000) To Vote Or Not To Vote? The Merits and Limits of Rational Choice (Pittsburgh: University of Pittsburgh Press). Blais, André (2006) ‘What Affects Turnout?’ Annual Review of Political Science, 9, 111–25. Blais, A. and K. Carty (1990) ‘Does Proportional Representation Foster Voter Turnout?’, European Journal of Political Research, 18, 167–81. Blais, A. and A. Dobrzynska (1998) ‘Turnout in Electoral Democracies’, European Journal of Political Research, 33, 239–61.

278 Bibliography Blais, A., E. Gidengil, N. Nevitte, and R. Nadeau (2004) ‘Where Does Turnout Decline Come From?’, European Journal of Political Research, 43(2), 221–36. Blais, André and Louis Massicotte (2002) ‘Electoral Systems’, in Lawrence LeDuc, Richard G. Niemi, and Pippa Norris (eds), Comparing Democracies 2: New Challenges in the Study of Elections and Voting (London: Sage). Blais, André, Louis Massicotte, and Agnieszka Dobrzynska (2003) Why is Turnout Higher in Some Countries than in Others? (Ottawa: Elections Canada). Blais, André and Daniel Rubenson (2007) ‘Turnout Decline: Generational Value Change or New Cohorts’ Response to Electoral Competition?’ Montreal: Typescript. Bowler, S. and D. Farrell (2006) ‘We Know Which One We Prefer but We Don’t Really Know Why: The Curious Case of Mixed Member Electoral Systems’, British Journal of Politics and International Relations, 8(3), 445–60. Bowler, S., D. Farrell, and R. Pettitt (2005) ‘Expert Opinion on Electoral Systems: So Which Electoral System Is “Best”?’, Journal of Elections, Public Opinion and Parties, 15(1), 3–19. Bowler, S. and B. Grofman (2000) Elections in Australia, Ireland, and Malta under the Single Transferable Vote: Reflections on an Embedded Institution (Ann Arbor, MI: University of Michigan Press). British Columbia Citizens’ Assembly on Electoral Reform (2004) Making Every Vote Count: The Case for Electoral Reform in British Columbia, http://www. citizensassembly.bc.ca/resources/TechReport(full).pdf Brogan, B. (2003, May 2) ‘Tory’s Shot at IDS May Have Missed its Mark’, Daily Telegraph, 10. Brown, M. (2004) EIAA Media Consumption Study: Youth Online Results Europe 15–24 Age Group, http://www.unicef.org/voy/media/EIAA_media_ consumption_study.pdf Brown, M. B. (2006) ‘Citizen Panels and the Concept of Representation’, Journal of Political Philosophy, 14, 203–25. Buckingham, D. (2000) The Making of Citizens: Young People, News and Politics (New York: Routledge). Button, M. and K. Mattso (1999) ‘Deliberative Democracy in Practice: Challenges and Prospects for Civic Deliberation’, Polity, 31(4), 609–37. Cain, B. E., R. J. Dalton, and S. E. Scarrow (eds) (2003) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press). Cain, B. E., P. Egan, and S. Fabbrini (2003) ‘Toward More Open Democracy: The Expansion of Freedom of Information Laws’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 115–39. Calenda, D. and L. Mosca (2007) ‘The Political Use of the Internet’, Information, Communication and Society, 10(1), 29–47. Canadian Policy Research Networks (2007,Winter) ‘The Youth/Ballot Box Disconnect’, Network News, 36, http://www.cprn.org/documents/47408_en.pdf Carty, R. K. (1988) ‘Campaigning in the Trenches: The Transformation of Constituency Politics’, in G. Perlin (ed.) Party Democracy in Canada: The Politics of National Party Conventions (Scarborough, ON: Prentice Hall), pp. 84–96.

Bibliography 279 Carty, R. K. (2002) ‘Canada’s Nineteenth-Century Cadre Parties at the Millennium’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 345–78. Carty, R. K. and D. E. Blake (1999) ‘The Adoption of Membership Votes for Choosing Party Leaders: The Experience of Canadian Parties’, Party Politics 5(2), 211–24. Carty, R. K. and W. Cross (2009) ‘Political Parties and the Practice of Brokerage Politics’ in J. C. Courtney and D. Smith (eds) The Handbook of Canadian Politics (New York: Oxford University Press). Castles, F. G. and P. Mair (1984) ‘Left-Right Political Scales: Some “Expert” Judgements’, European Journal of Political Research, 12, 73–88. Catt, H. and M. Murphy (2003) ‘What Voice for the People? Categorising Methods of Public Consultation’, Australian Journal of Political Science, 38(3), 407–21. Centre for Teaching and Learning (2007) ‘Good Practice: Problem-Based Learning’ Queen’s University, http://www.queensu.ca/ctl/goodpractice/problem/ index.html Chadwick, A. (2006) Internet Politics: States, Citizens and New Communication Technologies. (Oxford: Oxford University Press). Chambers, S. (2003) ‘Deliberative Democratic Theory’, Annual Review of Political Science, 6, 307–26. Christophorou, D. (2006) ‘Party Change and Development in Cyprus (1995–2005)’, South European Society and Politics, 11(3–4), 513–42. Cichowski, R. A. and A. Stone Sweet (2003) ‘Participation, Representative Democracy, and the Courts’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 192–220. Clarke, H. D. and A. Kornberg (1996) ‘Choosing Canada?: The 1995 Quebec Sovereignty Referendum’, PS: Political Science and Politics, 29, 676–82. Clarke, H. D. and M. C. Stewart (1998) ‘The Decline of Parties in the Minds of Citizens’, Annual Review of Political Science, 1, 357–8. Clarke, H. D., D. Saunders, M. C. Stewart, and P. Whiteley (2004) Political Choice in Britain (Oxford: Oxford University Press). Clarkson, S. (2000) ‘Do Deficits Imply Surpluses? Toward a Democratic Audit of North America’ in M. Th. Greven and L. W. Pauly (eds) Democracy Beyond the State: The European Dilemma and the Emerging Global Order (Toronto, ON: University of Toronto Press). Clarkson, S. (2002) Uncle Sam and Us: Globalization, Neoconservatism and the Canadian State (Toronto: University of Toronto Press). Coelho, V. S., B. Pozzoni, and M. C. Montoya (2005) ‘Participation and Public Policies in Brazil’ in J. Gastil and P. Levine (eds) The Deliberative Democracy Handbook: Strategies for Effective Civic Engagement in the 21st Century (San Francisco: Jossey-Bass), pp. 174–85. Coleman, J. (1990) Foundations of Social Theory (Cambridge, MA: Harvard University Press). Commission on Legislative Democracy (2004) Final Report and Recommendations, http://www.gnb.ca/0100/FinalReport-e.pdf Converse, P. E. (1990) ‘Popular Representation and the Distribution of Information’ in J. A. Ferejohn and J. H. Kuklinski (eds) Information and Democratic Processes (Urbana, IL: University of Illinois Press), pp. 369–87.

280 Bibliography Cook, C. (ed.) (1994) Constitutional Predicament: Canada after the Referendum of 1992 (Montreal: McGill-Queen’s University Press). Courtney, J. (1973) The Selection of National Party Leaders in Canada (Hamden, CT.: Archon Books). Courtney, J. (1995) Do Conventions Matter? (Montreal: McGill-Queens University Press). Cox, Gary W. (1999) ‘Electoral Rules and the Calculus of Mobilization’, Legislative Studies Quarterly, 24, 387–419. Crepaz, M. M. L. (1990) ‘The Impact of Party Polarization and Postmaterialism on Voter Turnout: A Comparative Study of 16 Industrial Democracies’, European Journal of Political Research, 18, 183–205. Crewe, I. (1981) ‘Electoral Participation’ 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), pp. 216–63. Cross, W. (1996) ‘Direct Election of Provincial Party Leaders in Canada, 1985–1995: The End of Leadership Convention?’, Canadian Journal of Political Science, 29(2), 295–315. Cross, W. (2004) Political Parties (Vancouver: UBC Press). Cross, W. and L. Young (2004) ‘The Contours of Political Party Membership in Canada’, Party Politics, 10(4), 427–44. Cross, W. and L. Young (2008) ‘Factors Influencing the Decision of the Young Politically Engaged to Join a Political Party: An Investigation of the Canadian Case’, Party Politics, 14(3), 345–69. Curtice, J. (1989) ‘The 1989 European Election: Protest or Green Tide?’, Electoral Studies, 8, 217–30. Cutler, F. and R. Johnston (2008), ‘Deliberation, Information, and Trust: The BC Citizens’ Assembly as Agenda-Setter’ in M. E.Warren and H. Pearse (eds) Designing Democratic Renewal: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press), pp. 166–91. Dahl, R. A. (1970) After the Revolution: Authority in a Good Society (New Haven, CT: Yale University Press). Dahl, R. A. (1989) Democracy and Its Critics (New Haven, CT: Yale University Press). Dahl, R. A. and E. R. Tufte (1973) Size and Democracy (Stanford, CA: Stanford University Press). Dahlgren, P. (2000) ‘The Internet and the Democratization of Civic Culture’, Political Communication, 17, 335–40. Dahlgren, P. and T. Olsson (2005) The Micro-Meshes of Civic Cultures: Political Engagement and Media Use. Paper presented at the Social Science Seminar at Karlstadt University, http://www.sam.kau.se/stv/ksspa/papers/dahlgren.pdf Dalton, R. J. (2004) Democratic Challenges, Democratic Choices: The Erosion of Political Support in the Advanced Industrial Democracies (Oxford: Oxford University Press). Dalton, R. J. (2006) Citizen Politics, 4th edn (Washington, DC: CQ Press). Dalton, R. J. (2006) Citizenship Norms and Political Participation in America: The Good News Is . . . the Bad News Is Wrong. CDACS Occasional Paper 2006–01. Washington, DC: The Center for Democracy and Civil Society, Georgetown University. Dalton, R. J. (2007) The Good Citizen: How a Younger Generation Is Reshaping American Politics (Washington, DC: Congressional Quarterly Press).

Bibliography 281 Dalton, R. J., B. E. Cain, and S. E. Scarrow (2003) ‘Democratic Publics and Democratic Institutions’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 250–75. Dalton, R. J. and M. Gray (2003) ‘Expanding the Electoral Marketplace’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 23–43. Dalton, R. J. and M. P. Wattenberg (2000) Parties without Partisans: Political Change in Advanced Industrial Societies (Oxford: Oxford University Press). DeBardeleben, J. and A. Hurrelmann (2007) ‘Introduction’ in J. DeBardeleben and A. Hurrelmann (eds) Democratic Dilemmas of Multilevel Governance: Legitimacy, Representation and Accountability in the European Union (Basingstoke: Palgrave Macmillan), pp. 1–14. Dell’Olio, F. (2005) The Europeanization of Citizenship (Aldershot: Ashgate). Denters, B., O. Gabriel, and M. Torcal (2007) ‘Norms of Good Citizenship’ in J. W. van Deth, J. R. Montero, and A.Westholm (eds) Citizenship and Involvement in European Democracies (Abingdon: Routledge), pp. 88–108. Denver, D. and G. Hands (1997) Modern Constituency Electioneering (London: Frank Cass). Department for Constitutional Affairs (2003) Constitutional Reform: Next Steps for the House of Lords, http://www.dca.gov.uk/consult/holref/index.htm Deschouwer, K. (2002) ‘The Colour People’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 151–80. Detterbeck, K. (2005) ‘Cartel Parties in Western Europe?’, Party Politics, 11(2), 173–91. Downs, Anthony (1957) An Economic Theory of Democracy. (New York: Harper). Downs, W. (2001) ‘Denmark’s Referendum on the Euro’, West European Politics, 24, 222–6. Dubnick, R. (2003) ‘Nurturing Civic Lives: Developmental Perspectives on Civic Education’, PS: Political Science and Politics XXXVI, 2, 253–5. Dudley, R. L. and A. R. Gitelson (2003) ‘Civic Education, Civic Engagement, and Youth Civic Development’, PS: Political Science and Politics XXXVI, 2, 263–7. Eder, K. (2001) ‘Integration through Culture? The Paradox of the Search for a European Identity’ in K. Eder and B. Giesen (eds) European Citizenship between National Legacies and Postnational Projects (Oxford: Oxford University Press), pp. 222–44. Eijk, C. van der, M. Franklin, and M. Marsh (1996) ‘What Voters Teach Us About Europe-Wide Elections: What Europe-Wide Elections Teach Us About Voters’, Electoral Studies, 15, 149–66. Elections Canada (2006) ‘Table 4: Voter Turnout for the 2006, 2004, 2000 and 1997 General Elections,’ Thirty-ninth General Election 2006: Official Voting Result, http://www.elections.ca/scripts/OVR2006/default.html Elections Canada (2008) Estimation of Voter Turnout by Age Group at the 39th Federal General Election, January 23, 2006 (Ottawa: Elections Canada). Elklit, J. (2005) ‘Denmark: Simplicity Embedded in Complexity (or is it the Other Way Round)?’ in M. Gallagher and P. Mitchell (eds) The Politics of Electoral Systems (Oxford: Oxford University Press), pp. 453–71.

282 Bibliography Elklit, J., B. Møller, P. Svensson, and L. Togeby (2005) Gensyn med Sofavælgerne: Valgdeltagelse i Danmark [The Non-Voters Revisited: Electoral Participation in Denmark] (Aarhus, Denmark: Aarhus University Press). Elklit, J., P. Svensson, and L. Togeby (2005) ‘Why Is Voter Turnout Not Declining in Denmark?’ Paper presented at the APSA Annual Meeting inWashington, DC. European Commission (2005) Eurobarometer 62: Public Opinion in the European Union, http://ec.europa.eu/public_opinion/archives/eb/eb62/eb_62_en.pdf European Commission (2005a) The Future Constitutional Treaty: First Results (Special Eurobarometer), http://www.europa.eu.int/comm/public_opinion/index_ en.htm European Commission (2005b) Ratification of the Treaty Establishing a Constitution for Europe, http://europa.eu.int/constitution/referendum_en.htm European Parliament (2004) European elections 10–13 june, http://www.europarl. europa.eu/elections2004/ep-election/sites/en/results1306/turnout_ep/index.html Farr, J. (1993) ‘Framing Democratic Discussion’ in G. E. Marcus and R. Hanson (eds) Reconsidering the Democratic Public (University Park, PA: Penn State Press). Farrell, D. (2001) Electoral Systems: A Comparative Introduction (New York: Palgrave – now Palgrave Macmillan). Fearon, J. D. (2003) ‘Ethnic and Cultural Diversity by Country’, Journal of Economic Growth, 8, 195–222. Fishkin, J. S. (1991) Democracy and Deliberation: New Directions for Democratic Reform (New Haven, CT: Yale University Press). Flickinger, R. S. and D. T. Studlar (2007) ‘One Europe, Many Electorates? Models of Turnout in European Parliament Elections After 2004’, Comparative Political Studies, 40(4), 383–404. Follesdal, A. and S. Hix (2006) ‘Why There Is a Democratic Deficit in the EU: A Response to Majone and Moravcsik’, Journal of Common Market Studies, 44, 533–62. Fornos, C., T. J. Power, and J. C. Garand (2004) ‘Explaining Voter Turnout in Latin America, 1980 to 2000’, Comparative Political Studies, 37(8), 909–40. Fowler, B. (2004) ‘Hungary: Unpacking the Permissive Consensus’, West European Politics, 27(4), 624–51. Franklin, Mark (1996) ‘Electoral Participation’, in Lawrence LeDuc, Richard G. Niemi and Pippa Norris (eds), Comparing Democracies: Elections and Voting in Global Perspective (California: Sage). Franklin, M. N. (2002) ‘The Dynamics of Electoral Participation’, in L. LeDuc, R. G. Niemi, and P. Norris (eds) Comparing Democracies 2: New Challenges in the Study of Elections (London and Thousand Oaks, CA: Sage), pp. 148–68. Franklin, M. (2004) Voter Turnout and the Dynamics of Electoral Competition in Established Democracies since 1945 (Cambridge: Cambridge University Press). Franklin, M. (2006) ‘You Want to Vote Where Everybody Knows Your Name: Targeting Adults in Need of Electoral Socialization’. Paper presented at 20thWorld Congress of the International Political Science Association in Fukuoka, Japan. Franklin, M., C. van der Eijk, and M. Marsh (1995) ‘Referendum Outcomes and Trust in Government: Public Support for Europe in the Wake of Maastricht’, West European Politics, 18, 101–17. Franklin, M. N., C. van der Eijk, D. Evans, M. Fotos,W. Hirczy de Mino, M. Marsh, and B. Wessels (2004) Voter Turnout and the Dynamics of Electoral Competition in Established Democracies since 1945 (Cambridge: Cambridge University Press).

Bibliography 283 Franklin, M. N., P. Lyons, and M. Marsh (2004) ‘The Generational Basis of Turnout Decline in Advanced Democracies’, Acta Politica, 39(2), 115–51. Fukuyama, F. (1995) Trust: The Social Virtues and the Creation of Prosperity (London: Hamish Hamilton). Fung, A. (2006a) ‘Democratizing the Policy Process’ in R. Goodin, M. Moran, and M. Rein (eds) The Oxford Handbook of Public Policy (Oxford: Oxford University Press), pp. 669–85. Fung, A. (2006b) ‘Varieties of Participation in Complex Governance’, Public Administration Review, 66, 66–75. Gallagher, M. and P. Vincenzo Uleri (eds) (1996) The Referendum Experience in Europe (London: Macmillan). Galligan, B. (2001) ‘Amending Constitutions Through the Referendum Device’ in M. Mendelsohn and A. Parkin (eds) Referendum Democracy: Citizens, Elites and Deliberation in Referendum Campaigns, (Basingstoke: Palgrave – now Palgrave Macmillan) pp. 109–24. Galston, W. A. (2001) ‘Political Knowledge, Political Engagement, and Civic Education’, Annual Review of Political Science, 4, 217–34. Galston, W. A. (2004, April) ‘Civic Education and Political Participation’, PS: Political Science & Politics, 263–6. Garry, J., M. Marsh, and R. Sinnott (2006) ‘Second-order versus Issue-voting Effects in EU Referendums: Evidence from the Irish Nice Treaty Referendums’, European Union Politics, 6, 201–21. Gastil, J. (2008) Political Communication and Deliberation (Los Angeles: Sage). Gastil, J. and P. Levine (eds) (2005) The Deliberative Democracy Handbook: Strategies for Effective Civic Engagement in the Twenty-first Century (San Francisco: JosseyBass). Geys, Benny (2006) ‘Explaining Voter Turnout: a Review of Aggregate-Level Research.’ Electoral Studies, 25: 637–63. Gibson, G. (2004) ‘Challenges in Senate Reform: Conflicts of Interest, Unintended Consequences, New Possibilities’, Public Policy Sources, 83, http:// www.fraserinstitute.ca/admin/books/files/ChallengesInSenateReform.pdf Gidengil, E., A. Blais, N. Nevitte, and R. Nadeau (2004) Citizens (Vancouver: University of British Columbia Press). Giesen, B. and K. Eder (2001) ‘Introduction: European Citizenship’ in K. Eder and B. Giesen (eds) European Citizenship between National Legacies and Postnational Projects (Oxford: Oxford University Press), pp. 1–16. Gifford, C. (2004) ‘National and Post-national Dimensions of Citizenship Education in the UK’, Citizenship Studies, 8(2), 145–58. Gilland, K. (2002) ‘Ireland’s (First) Referendum on the Treaty of Nice’, Journal of Common Market Studies, 4, 527–35. Goul Andersen, J. (2003a) ‘The General Election in Denmark, November 2001’, Electoral Studies, 22(1), 186–93. Goul Andersen, J. (2003b) ‘Partiernes image: De borgerlige bedst til at sikre velfælden’ [Party Images: The Right-of-centre Parties Best at Ensuring Welfare] in J. Goul Andersen and O. Borre (eds) Politisk forandring: Værdipolitik og nye skillelinjer ved folketingsvalget 2001 [Political Change: Value Politics and New Lines of Division in the 2001 National Elections] (Aarhus, Denmark: Systime Academic), pp. 151–70.

284 Bibliography Goul Andersen, J. (2004) Et ganske levende demokrati [A Perfectly Lively Democracy] (Aarhus, Denmark: Aarhus University Press). Goul Andersen, J. (2007) ‘Samfundskonflikter, partier i bevægelse og vælgere med omtanke: Rids af en generel teoretisk ramme’ [Conflicts in Society, Parties in Movement and Prudent Voters: The Outline of a General Theoretical Framework] in J. Goul Andersen et al. (eds) Det nye politiske landskab: Folketingsvalget 2005 i perspektiv [The New Political Landscape: The 2005 National Elections in Perspective] (Aarhus, Denmark: Academíca), pp. 11–55. Gratschew, M. (2001) ‘Compulsory Voting’, http://www.idea.int/vt/compulsory_ voting.cfm Gray, M. and M. Caul (2000) ‘Declining Voter Turnout in Advanced Industrial Democracies, 1950 to 1997: The Effects of Declining Group Mobilization’, Comparative Political Studies, 33(9), 1091–122. Green, D. P. and A. S. Gerber (2004) Get Out the Vote! (Washington, DC: Brookings). Green, J. C. (2002) ‘Still Functional After All These Years’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 310–44. Greenwood, J. (2007) ‘Organized Civil Society and Input Legitimacy in the EU’ in J. DeBardeleben and A. Hurrelmann (eds) Democratic Dilemmas of Multilevel Goveranance: Legitimacy, Representation and Accountability in the European Union (Basingstoke: Palgrave Macmillan), pp. 177–94. Grönlund, K. and M. Setälä (2004) ‘Low Electoral Turnout: An Indication of a Legitimacy Deficit?’ Paper presented at the ECPR workshop in Uppsala, Sweden. Gunther, R., J. R. Montero, and J. J. Linz (2002) Political Parties: Old Concepts and New Challenges (Oxford: Oxford University Press). Gutmann, A. and D. Thompson (2004) Why Deliberative Democracy? (Princeton, NJ: Princeton University Press). Haas, E. B. (1958) The Uniting of Europe (Stanford: Stanford University Press). Hajer, M. A. and H. Wagenaar (eds) (2003) Deliberative Policy Analysis: Understanding Governance in the Network Society (Cambridge: Cambridge University Press). Hannigan, L. (2007) The Effectiveness of Public Consultation: A Case Study of the Ontario Citizens’ Assembly on Electoral Reform. Unpublished BA honours thesis, Queen’s University, Kingston, ON, Canada. Heater, D. (1990) Citizenship: The Civic Ideal in World History, Politics and Education (London: Longman). Heath, A. and B. Taylor (1999) ‘New Sources of Abstention?’ in G. Evans and P. Norris (eds) Critical Elections: British Parties and Voters in Long-Term Perspective (London: Sage), pp. 164–80. Held, D. (1996) Models of Democracy (Stanford, CA: Stanford University Press). Henderson, K. (2004) ‘EU Accession and the New Slovak Consensus,’ West European Politics, 27(4), 652–70. Hermes, J. (2006) ‘Citizenship in the Age of the Internet’, European Journal of Communication, 21(3), 295–309. Heston, Alan, Robert Summers, and Bettina Aten (2006) Penn World Table Version 6.2. Center for International Comparisons of Production, Income and Prices at the University of Pennsylvania. Hibbing, J. R. and E. Theiss-Morse (2001) ‘Process Space and American Politics’, American Political Science Review, 95, 145–53.

Bibliography 285 Hibbing, J. and E. Theiss-Morse (2002) Stealth Democracy: Americans’ Beliefs about How Government Should Work (Cambridge: Cambridge University Press). Hirczy, W. (1995) ‘Explaining Near-Universal Turnout: The Case of Malta’, European Journal of Political Research, 27, 255–72. Hix, S. (2002) ‘Parliamentary Behavior with Two Principals: Preferences, Parties, and Voting in the European Parliament’, American Journal of Political Science, 46(3), 688–98. Hix, S. (2002) ‘Parties at the European Level’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 280–309. Hix, S. and C. Lord (1997) Political Parties in the European Union (New York: St. Martin’s Press – now Palgrave Macmillan). Hix, S., A. G. Noury, and D. Roland (2007) Democratic Politics in the European Parliament (Cambridge: Cambridge University Press). Hobolt, S. Binzer (2005) ‘When Europe Matters: The Impact of Political Information on Voting in EU Referendums’, Journal of Elections, Public Opinion and Parties, 1, 85–109. Holliday, I. (2002) ‘Spain’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press). Holmberg, S. (2000), Välja parti [Choose a Party] (Stockholm: Nordstedt Juridik). Hox, J. (2002) Multilevel Analysis: Techniques and Applications (London: Lawrence Erlbaum Associates). Huber, J. and R. Inglehart (1995) ‘Expert Interpretations of Party Space and Party Locations in 42 Societies’, Party Politics, 1(1), 73–111. IDEA (Institute for Democracy and Electoral Assistance) (2006) Engaging the Electorate: Initiatives to Encourage Voter Turnout from Around the World (Stockholm: IDEA). Inglehart, R. (1997) Modernization and Postmodernization: Cultural, Economic and Political Change in 43 Societies (Princeton, NJ: Princeton University Press). Inglehart, Ronald (2003) ‘How Solid is Mass Support for Democracy – And How Can We Measure It?’ PS: Political Science & Politics, 36: 51–7. Inglehart, R. and C. Welzel (2005) Modernization, Cultural Change, and Democracy: The Human Developmental Sequence (Cambridge: Cambridge University Press). International Institute for Democracy and Electoral Assistance (2002) ‘The International IDEA Database: Voter Turnout from 1945 to 2001’, in R. López Pintor and M. Gratschew (eds) Voter Turnout since 1945: A Global Report (Stockholm, Sweden: International IDEA), pp. 117–69. International Institute for Democracy and Electoral Assistance (2004) Parliamentary elections, http://www.idea.int/vt/parl.cfm Institute on Governance (2007) Citizen Deliberative Decision-Making: Evaluation of the Ontario Citizens’ Assembly on Electoral Reform, http://www.citizensassembly. gov.on.ca/assets/Citizen%20Deliberative%20Decision%20Making_English.pdf Irvin, R. and J. Stansbury (2004) ‘Citizen Participation in Decision Making: Is It Worth the Effort’, Public Administration Review, 64(1), 55–65. Irwin, G. (1995) ‘Second Order or Third Rate?: Issues in the Campaign for the Elections for the European Parliament 1994’, Electoral Studies 14(2), 183–99. Ivaldi, G. (2006) ‘Beyond France’s 2005 Referendum on the European Constitutional Treaty: Second Order Model, Anti-Establishment Attitudes and the End of the Alternative European Utopia’, West European Politics, 29, 47–69.

286 Bibliography Jackman, Robert W. (1987) ‘Political Institutions and Voter Turnout in Industrial Democracies,’ American Political Science Review, 81: 405–24. Jackman, R. and R. Miller (1995) ‘Voter Turnout in the Industrial Democracies During the 1980s’, Comparative Political Studies, 27, 467–92. Jahn, D. and A.-S. Storsved (1995) ‘Legitimacy Through Referendum: the Nearly Successful Domino Strategy of the EU Referendums in Austria, Finland, Sweden and Norway’, West European Politics, 18, 18–37. Janoski, T. (1998) Citizenship and Civil Society: A Framework of Rights and Obligations in Liberal, Traditional and Social Democratic Regimes (Cambridge: Cambridge University Press). Jeppesen, J. and P. Meyer (1964) Sofavælgerne: Valgdeltagelsen ved danske folketingsvalg [The Non-Voters: Electoral Participation in the Danish National Elections] (Aarhus, Denmark: Department of Political Science, University of Aarhus). Johnston, R., A. Blais, E. Gidengil, and N. Nevitte (1996) The Challenge of Direct Democracy: The 1992 Canadian Referendum (Montreal: McGill-Queen’s University Press). Johnston, R. and C. Pattie (1997) ‘Where’s the Difference? Decomposing the Impact of Local Election Campaigns in Great Britain’, Electoral Studies, 16, 165–74. Johnston, R., J. Scott Matthews, and A. Bittner (2007) ‘Turnout and the Party System in Canada, 1988–2004’, Electoral Studies, 26(4), 735–45. Karpowitz, C. and J. Mansbridge (2005) ‘Disagreement and Consensus: The Importance of Dynamic Updating in Public Deliberation’ in J. Gastil and P. Levine (eds) The Deliberative Democracy Handbook: Strategies for Effective Civic Engagement in the 21st Century (San Francisco: Jossey-Bass). Katz, R. S. (1997) Democracy and Elections (New York: Oxford University Press). Katz, R. S. (2001) ‘The Problem of Candidate Selection and Models of Party Democracy’, Party Politics, 7(3), 277–96. Katz, R., P. Mair, L. Bardi, L. Bille, K. Deschouwer, D. Farrell, et al. (1992) ‘The Membership of Political Parties in European Democracies, 1960–1990’, European Journal of Political Research, 22, 329–45. Key Facts (2004) Key Facts: Media, Youth, and Civic Engagement (The Henry J. Kaiser Family Foundation), http://www.kff.org/entmedia/upload/Media-YouthandCivic-Engagement-Fact-Sheet.pdf King, A. (1969) ‘Political Parties in Western Democracies. Some Sceptical Reflections’, Polity, 2, 111–41. Kirchner, A. (1985) The European Parliament: Performance and Prospects (Aldershot: Gower). Kittilson, M. C. and S. Scarrow (2003) ‘Political Parties and the Rhetoric and Realities of Democratization’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 59–80. Knapp, A. (2002) ‘France: Never a Golden Age’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 107–50. Knight, K. (2007, 8 October) ‘Canadians Grow Social Networking Habits’, BizReport, http://www.bizreport.com/2007/10/canadians_grow_social_networking_ habits.html

Bibliography 287 Koepke, J. R. and N. Ringe (2006) ‘The Second-order Election Model in an Enlarged Europe’, European Union Politics, 7(3), 321–46. Kostadinova, T. (2003) ‘Voter Turnout Dynamics in Post-Communist Europe’, European Journal of Political Research, 42, 741–59. Kristjánsson, S. (2004) ‘Iceland: Searching for Democracy along Three Dimensions of Citizen Control’, Scandinavian Political Studies, 27(2), 153–74. Kuhn, H.-W. (2006) European Approaches to Civic Education for the Promotion of Democracy. Paper presented at the German-American Conference DemocracyPromotion and International Cooperation in Denver, CO, available from http://www.civiced.org/index.php?page=calendar&eid=14423 LeDuc, L. (2001) ‘Democratizing Party Leadership Selection’, Party Politics, 7(3), 323–41. LeDuc, L. (2002) ‘Referendums and Elections: How Do Campaigns Differ?’ in D. Farrell and R. Schmitt-Beck (eds) Do Political Campaigns Matter? Campaign Effects in Elections and Referendums (London: Routledge). LeDuc, L. (2003) The Politics of Direct Democracy: Referendums in Global Perspective (Peterborough: Broadview). LeDuc, L. (2005) ‘Saving the Pound or Voting for Europe?: Expectations for Referendums on the Constitution and the Euro’, Journal of Elections, Parties and Public Opinion, 15, 169–96. LeDuc, L. (2006) ‘Referendums and Deliberative Democracy’. Paper presented to the World Congress of the International Political Science Association in Fukuoka, Japan. LeDuc, L. (2007) ‘Voting NO: The Negative Bias in Referendum Campaigns’. Paper presented to the European Consortium for Political Research Joint Sessions Workshops in Helsinki, Finland. LeDuc, L. (2007) ‘European Elections and Democratic Accountability: The 2004 Elections to the European Parliament’ in J. DeBardeleben and A. Hurrelmann (eds) Democratic Dilemmas of Multilevel Goveranance: Legitimacy, Representation and Accountability in the European Union (Basingstoke: Palgrave Macmillan), pp. 139–57. LeDuc, L. and J. H. Pammett (1995) ‘Referendum Voting: Attitudes and Behaviour in the 1992 Constitutional Referendum’, Canadian Journal of Political Science, 28, 3–33. LeDuc, L. and J. H. Pammett (2006) ‘Voter Turnout in 2006: More Than Just the Weather’ in J. H. Pammett and C. Dornan (eds) The Canadian Federal Election of 2006 (Toronto: Dundurn Press), pp. 304–26. LeDuc, L., J. H. Pammett, and H. Bastedo (2008) ‘The Problem of Young Voters: A Qualitative and Quantitative Analysis’. Paper presented at the annual Conference of the American Political Science Association in Boston. LeDuc, L. and P. Svensson (eds) (2002) Interests, Information and Voting in Referendums [special issue] European Journal of Political Research, 41(6). Lewis, J., S. Inthorn, and K.Wahl-Jorgensen (2005) Citizens or Consumers? What the Media Tell Us about Political Participation (Maidenhead: Open University Press). Lijphart, A. (1984) Democracies: Patterns of Majoritarian and Consensus Government in Twenty-One Countries (New Haven, CT and London: Yale University Press). Lijphart, A. (1997) ‘Unequal Participation: Democracy’s Unresolved Dilemma’, American Political Science Review, 91, 1–14.

288 Bibliography Lijphart, A. (1999) Patterns of Democracy: Government Forms and Performance in Thirty-six Countries (New Haven, CT and London: Yale University Press). Lijphart, A. (2000) ‘Turnout’ in R. Rose (ed.) International Encyclopedia of Elections (Washington DC: CQ Press), pp. 314–22. Lindblom, C. E. (2002) The Market System: What It Is, How It Works, and What To Make of It (New Haven, CT: Yale University Press). Lippmann, W. (1925) The Phantom Public (New York: Harcourt Brace Jovanovich). Livingstone, S. (2007) ‘The Challenge of Engaging Youth Online’, European Journal of Communication, 22(2), 165–84. Llewellyn, K., S., Cook, J., Westheimer, L., Giron and K. Suurtamm, (2007) The State and Potential of Civic Learning in Canada. CPRN Research Report (Ottawa, ON: Canadian Policy Research Networks), http://www.cprn.org/documents/ 48798_EN.pdf Loader, B. (ed.) (2007) Young Citizens in the Digital Age: Political Engagement, Young People and New Media (London: Routledge). Lodge, J. and V. Herman (1982) Direct Elections to the European Parliament: A Community Perspective (London and New Haven, NJ: Macmillan and Humanities Press). Lord, C. (2004) A Democratic Audit of the European Union (Basingstoke: Palgrave Macmillan). Lupia, A. and M. McCubbins (1998) The Democratic Dilemma: Can Citizens Learn What They Need To Know? (New York: Cambridge University Press). Luskin, R. (1987) ‘Measuring Political Sophistication’, American Journal of Political Science, 31(4), 856–99. Lutz, G. and M. Marsh (2007) ‘Introduction: Consequences of Low Turnout’, Electoral Studies, 26(3), 539–47. Lyons, William and Robert Alexander (2000) ‘A Tale of Two Electorates: Generational Replacement and the Decline of Voting in Presidential Elections,’ Journal of Politics, 62, 1014–34. MacPherson, C. B. (1977) The Life and Times of Liberal Democracy (London: Oxford University Press). McGann, A. J. and M. Latner (2006) ‘Consensus Without Veto-Players: Testing Theories of Consensual Democracy’, Center for the Study of Democracy, University of California, Irvine, CA, Paper 06-03, http://repositories.cdlib.org/csd/ 06–03 McKean, M. and E. Schiner (2000) ‘Japan’s New Electoral System: La Plus Ça Change . . .’, Electoral Studies, 19, 447–77. McLaren, L. M. (2006) Identity, Interests and Attitudes to European Integration (Basingstoke: Palgrave Macmillan). Magnette, P. and K. Nicolaidis (2004) ‘The European Convention: Bargaining in the Shadow of Rhetoric’, West European Politics, 27, 381–404. Mainwaring, S. and T. R. Scully (1995) ‘Introduction: Party Systems in Latin America’ in Mainwaring and Scully (eds) Building Democratic Institutions: Party Systems in Latin America (Stanford, CA: Stanford University Press), pp. 1–34. Mair, P. (1994) ‘Party Organizations: From Civil Society to the State’, in R. S. Katz and P. Mair (eds) How Parties Organize: Change and Adaptation in Party Organizations in Western Democracies (London: Sage Publications), pp. 1–22.

Bibliography 289 Mair, P. (2007) ‘Political Parties and Party Systems’ in P. Graziano and M. P. Vink (eds) Europeanization: New Research Agendas (Basingstoke: Palgrave Macmillan), pp. 154–66. Mair, P. and I. Van Biezen (2001) ‘Party Membership in Twenty European Democracies, 1980–2000’, Party Politics, 7(1), 5–21. Mansbridge, J. (1983) Beyond Adversary Democracy (Chicago: University of Chicago Press). Mansbridge, J. (2003) ‘Rethinking Representation’, American Political Science Review, 97, 515–28. Mansbridge, J. (2004) ‘Representation Revisited: Introduction to the Case against Electoral Accountability’, Democracy and Society, 2(1), 12–13. Marcussen, M. and M. Zølner (2001) ‘The Danish EMU Referendum 2000’, Government and Opposition, 36, 379–401. Marden, P. (2003) The Decline of Politics: Governance, Globalization and the Public Sphere (Aldershot, UK: Ashgate). Marquand, D. (1979) Parliament for Europe (London: Jonathan Cape). Marthaler, S. (2005) ‘France’s Referendum on the EU Treaty, May 2005’, Representation, 41, 228–36. Martin, I. and J. W. van Deth (2007) ‘Political Involvement’ in J. W. van Deth, J. R. Montero, and A. Westholm (eds) Citizenship and Involvement in European Democracies (Abingdon: Routledge), pp. 303–33. Martin, P. (2002) ‘The Democratic Deficit’, Policy Options, 24(1), 10–12. Marsh, M. (1998) ‘Testing the Second Order Election Model After Four European Elections’, British Journal of Political Science, 28, 591–607. Marshall, T. H. (1950) Citizenship and Social Class and Other Essays (Cambridge: Cambridge University Press). Meehan, E. (1993) Citizenship and the European Community (London: Sage). Melber, A. (2006, 12 June) ‘MySpace, MyPolitics’, The Nation, http://www. thenation.com/doc/20060612/melber Michels, R. (1915/1959) Political Parties (New York: Dover). Michels, R. (1996) Political Parties: A Sociological Study of the Oligarchical Tendencies of Modern Democracy (New York: Collier Books). Mikkel, E. (2006) ‘Patterns of Party Formation in Estonia: Consolidation Unaccomplished’ in S. Jungerstam-Mulders (ed.) Post-Communist EU Member States: Parties and Party Systems (Aldershot: Ashgate), pp. 23–49. Milan, A. (2005) ‘Willing to Participate: Political Engagement of Young Adults’, Canadian Social Trends, 79, http://www.statcan.ca/english/freepub/11-008-XIE/ 2005003/articles/8965.pdf Miller, Warren E. and J. Merrill Shanks (1996) The New American Voter (Cambridge, MA: Harvard University Press). Montgomery, K. and B. Gottlieb-Robles (2006) ‘Youth as e-Citizens: The Internet’s Contribution to Civic Engagement’ in D. Buckingham and R.Willet (eds) Digital Generations: Children, Young People, and New Media (Mahwah, NJ: Lawrence Erlbaum Associates), pp. 131–47. Moravcsik, A. (2002) ‘In Defence of the “Democratic Deficit”: Reassessing Legitimacy in the European Union’, Journal of Common Market Studies, 40, 603–24. Narud, H. M. and H. Valen (2007) Demokrati og ansvar: Politisk representation i et flerpartisystem [Democracy and Responsibility: Political Representation in a Multi-party System] (Oslo: N.W. Damm & Søn).

290 Bibliography Neimi, R. and J. Junn (1998) Civic Education: What Makes Students Learn (New Haven, CT: Yale University Press). Nentwich, M. (1998) ‘Opportunity Structures for Citizens’ Participation: The Case of the European Union’ in A. Weale and M. Nentwich (eds) Political theory and the European Union (London: Routledge), pp. 125–40. Neuman, W. R. (1986) The Paradox of Mass Politics: Knowledge and Opinion in the American Electorate (Cambridge, MA: Harvard University Press). Nevitte, N. (1996) The Decline of Deference: Canadian Value Change in Cross-National Perspective (Peterborough, ON: Broadview Press). Norris, P. (1997) ‘Representation and the Democratic Deficit’, European Journal of Political Research, 32, 273–82. Norris, P. (2002) Democratic Phoenix: Reinventing Political Activism (Cambridge: Cambridge University Press). Norris, P. (2004) Electoral Engineering: Voting Rules and Political Behavior (Cambridge: Cambridge University Press). Norris, P. (ed.) (1999) Critical Citizens: Global Support for Democratic Governance (Oxford: Oxford University Press). O’Neill, B. (2007) Indifferent or Just Different? The Political and Civic Engagement of Young People in Canada. CPRN Research Report (Ottawa, ON: Canadian Policy Research Networks), http://www.cprn.org/documents/48504_EN.pdf Ontario Citizens’ Assembly on Electoral Reform (2007) Public Consultation Reports, http://www.citizensassembly.gov.on.ca/assets/Citizens’ %20Assembly%20Consultation%20Reports.pdf Osler, A. and H. Starkey (2003) ‘Learning for Cosmopolitan Citizenship: Theoretical Debates and Young People’s Experiences’, Educational Review, 55(3), 243–54. Pammett, J. H. (forthcoming 2009) ‘Voting Turnout in a System of Multi-Level Governance’ in M. Kanji, A. Bilodeau, and T. Scotto (eds) Four Decades of Canadian Election Studies (Vancouver: University of British Columbia Press). Pammett, J. H. and L. LeDuc (2001) ‘Sovereignty, Leadership and Voting in the Quebec Referendums’, Electoral Studies, 20, 265–80. Pammett, J. H. and L. LeDuc (2003) Explaining the Turnout Decline in Canadian Federal Elections (Ottawa, ON: Elections Canada). Pammett, J. H. and L. LeDuc (2004) ‘Four Vicious Circles of Turnout: Competitiveness, Regionalism, Culture and Participation in Canada’. Paper presented to the Joint Sessions Workshops of the European Consortium for Political Research in Uppsala, Sweden. Parkinson, J. (2006) Deliberating in the Real World: Problems of Legitimacy in Deliberative Democracy (Oxford: Oxford University Press). Pattie, C., P. Seyd, and P. Whiteley (2004) Citizenship in Britain: Values, Participation and Democracy (Cambridge: Cambridge University Press). Pennings P. and R. Hazan (2001) ‘Democratizing Candidate Selection: Causes and Consequences’, Party Politics, 7, 267–75. Perrineau, P., G. Grünberg, and C. Ysmal (eds) (2002) Europe at the Polls: The European Elections of 1999 (London: Palgrave – now Palgrave Macmillan). Peters, B. G. (1997) ‘The Separation of Powers in Parliamentary Systems’ in K. von Mettenheim (ed.) Presidential Institutions and Democratic Politics: Comparing Regional and National Context (Baltimore, MD: The Johns Hopkins University Press), pp. 76–83.

Bibliography 291 Pew (2006) Pew Research Center for the People & the Press News Consumption and Believability Study, http://people-press.org/reports/pdf/282.pdf Pharr, S. J. and R. D. Putnam (2000) Disaffected Democracies: What’s Troubling the Trilateral Countries? (Princeton, NJ: Princeton University Press). Pitkin, H. (1967) The Concept of Representation (Berkeley: University of California Press). Poguntke, T. (2006) ‘Partiti e società nell’Europa Occidentale’ [Parties and Societies in Western Europe] in B. Luciano (ed.) Partiti e sistemi di partito [Parties and Party Systems] (Bologna, Italy: il Mulino), pp. 103–22. Poguntke, T. and P. Webb (eds) (2005) The Presidentialization of Politics: A Comparative Study of Modern Democracies (New York: Oxford University Press). Popkin, S. (1991) The Reasoning Voter: Communication and Persuasion in Presidential Campaigns (Chicago: University of Chicago Press). Powell, G. Bingham (1980) ‘Voting Turnout in Thirty Democracies’, in Richard Rose (ed.), Electoral Participation (Beverly Hills: Sage). Powell, G. Bingham (1982) Comparative Democracies: Participation, Stability and Violence (Cambridge, MA: Harvard University Press). Powell, G. Bingham Jr. (1986) ‘American Voter Turnout in Comparative Perspective’, American Political Science Review, 80, 17–43. Pridham, G. (2002) ‘EU Enlargement and Consolidating Democracy in PostCommunist States’, Journal of Common Market Studies, 40(3), 953–73. Punnett, R. M. (1992) Selecting the Party Leader: Britain in Comparative Perspective (New York: Harvester Wheatsheaf). Putnam, R. (1993) Making Democracy Work: Civic Traditions in Modern Italy (Princeton: Princeton University Press). Putnam, R. D. (1995) ‘Bowling Alone: America’s Declining Social Capital’, Journal of Democracy, 6(1), 65–78. Putnam, R. D. (2000) Bowling Alone: The Collapse and Revival of American Community (New York: Simon & Schuster). Quinn, T. (2004) ‘Electing the Leader: The British Labour Party’s Electoral College’, British Journal of Politics and International Relations, 6, 333–52. Raik, K. (2004) ‘EU Accession of Central and Eastern Europe: Democracy and Integration as Conflicting Logics’, East European Politics and Societies, 18, 567–94. Ramonaité, A. (2006) ‘The Development of the Lithuanian Party System: From Stability to Perturbation’ in S. Jungerstam-Mulders (ed.) Post-Communist EU Member States: Parties and Party Systems (Aldershot: Ashgate), pp. 69–90. Rapoport, R. B. and W. J. Stone (2006) Three’s a Crowd: The Dynamic of Third Parties, Ross Perot and Republican Resurgence (Ann Arbor: University of Michigan Press). Rattinger, H. and J. Krämer (1995) ‘Wahlnorm und Wahlbeteiligung in der Bundesrepublik Deutschland: Eine Kausalanalyse’ [Electoral Norm and Turnout in the German Federal Republic: A Causal Analysis], Politische Vierteljahrsschrift, 36(2), 267–85. Raudenbusch, S. W. and A. S. Bryk (2002) Heirachical Linear Models: Applications and Data Analysis Methods, 2nd edn. (Thousand Oaks, CA: Sage). Rehfeld, A. (2005) The Concept of Constituency: Political Representation, Democratic Legitimacy, and Institutional Design (Cambridge: Cambridge University Press). Reif, K. and H. Schmidt (1980) ‘Nine Second-order National Elections: A Conceptual Framework for the Analysis of European Election Results’, European Journal of Political Research, 8(1), 3–44.

292 Bibliography Rokkan, S. and A. Campbell (1960) ‘Citizen Participation in Political Life: A Comparison of Data from Norway and the United States of America’, International Social Science Journal, 12(1), 69–99. Rose, L. E. (2004) ‘Normer og roller – borgerpliktens endelikt?’ [Norms and Roles – The End of Civic Duty?] in B. Aardal (ed.) Valgdeltakelse og lokaldemokrati [Electoral Participation and Local Democracy] (Oslo: Kommuneforlaget), pp. 139–75. Rose, L. E. (2005) ‘Normer og stemmegivning: Både indre plikt og ytre press?’ [Norms and Voting: Internal Duty as well as External Pressure?] in J. Saglie and R. Bjørklund (eds) Lokalvalg og lokalt folkestyre [Local Government Elections and Local Democracy] (Oslo: Gyldendal Akademisk), pp. 228–51. Rose, R. (2004) Europe Expands, Turnout Falls: The Significance of the 2004 European Parliament Election (Stockholm: International Institute for Democracy and Electoral Assistance). Rosenblum, N. (2008) On the Side of the Angels: An Appreciation of Parties and Partisanship (Princeton, NJ: Princeton University Press). Rosenstone, S. J. and J. M. Hansen (1993) Mobilization, Participation, and Democracy in America (New York: Macmillan). Rubenson, Daniel, André Blais, Patrick Fournier, Elisabeth Gidengil, and Neil Nevitte (2007) ‘Does Low Turnout Matter? Evidence from the 2000 Canadian Election,’ Electoral Studies, 26: 589–97. Rubenstein, J. (2007) ‘Accountability in an Unequal World’, Journal of Politics, 69, 616–32. Russell, M. (2005) Building New Labour: The Politics of Party Organisation (New York: Palgrave Macmillan). Russell, P. (1993) Constitutional Odyssey, 2nd edn (Toronto: University of Toronto Press). Sartori, G. (1976) Parties and Party Systems: A Framework for Analysis, vol. 1. (Cambridge: Cambridge University Press). Scarrow, S. E. (1996) Parties and Their Members (Oxford: Oxford University Press). Scarrow, S. (2000) ‘Parties without Members? Party Organization in a Changing Electoral Environment’ in R. J. Dalton and M. P. Wattenberg (eds) Parties without Partisans: Political Change in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 79–101. Scarrow, S. E. (2002) ‘Party Decline in the Parties State? The Changing Environment of German Politics’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties In Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 77–106. Scarrow, S., P. Webb, and D. M. Farrell (2000) ‘From Social Integration to Electoral Contestation: The Changing Distribution of Power within Political Parties’ in R. J. Dalton and M. P. Wattenberg (eds) Parties without Partisans: Political Change in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 129–56. Scheuer, A. (1999) ‘Political Community?’ in H. Schmitt and J. Thomassen (eds) Political Representation and Legitimacy in the European Union (Oxford: Oxford University Press), pp. 25–46. Schifferes, S., S. Ward, and W. Lusoli (2007) What’s The Story . . .? Online News Consumption in the 2005 UK Election. Unpublished manuscript, http://www.esri.salford.ac.uk/ESRCResearchproject/papers/schifferes_ et_al_2007.pdf

Bibliography 293 Schmitt, H. and J. Thomassen (1999) ‘Distinctiveness and Cohesion of Parties’ in H. Schmitt and J. Thomassen (eds) Political Representation and Legitimacy in the European Union (Oxford: Oxford University Press), pp. 111–28. Schmitter, P. (2001) ‘The Scope of Citizenship’ in K. Eder and B. Giesen (eds) European Citizenship between National Legacies and Postnational Projects (Oxford: Oxford University Press), pp. 86–121. Schumpeter, J. (1942) Capitalism, Socialism and Democracy (New York: Harper). Selwyn, N. (2007) ‘Technology, Schools and Citizenship Education: A Fix Too Far?’ in B. Loader (ed.) Young Citizens in the Digital Age (London: Routledge), pp. 129–42. Semetko, H. A. (2006) ‘Parties in the Media Age’ in R. S. Katz and W. Crotty (eds) Handbook of Party Politics (London: Sage). Seyd, P. (1999) ‘New Parties/New Politics? A Case Study of the British Labour Party’, Party Politics, 5(3), 383–405. Seyd, P. and P. Whiteley (1992) Labour’s Grassroots: The Politics of Party Membership (Oxford: Oxford University Press). Siaroff, A. and J. Merer (2002) ‘Parliamentary Election Turnout in Europe since 1990’, Political Studies, 50(5), 916–27. Sifrey, M. (2006, March 13) ‘Essembly.com: Finally, a Friendster for Politics’, Personal Democracy Forum, http://www.personaldemocracy.com/node/849 Sokoloff, H. (2003, September 22) ‘Turnout is Poor at Delegate Selection Meeting: Less than a Quarter of Liberal Party Members Cast Votes’, National Post, A5. Smith, G. (2005) Beyond the Ballot: 57 Democratic Innovations from around theWorld (London: The Power Inquiry). Smith, G. (2009) Democratic Innovations: Designing Institutions for Citizen Participation (Cambridge: Cambridge University Press). Smith, J. (1996) ‘How European are European Elections?’, in J. Gaffney (ed.), Political Parties in the European Community (London: Routledge), pp. 275–90. Snider, J. H. (2007) ‘From Dahl to O’Leary: 36 Years of the “Yale School of Democratic Reform” ’, Journal of Public Deliberation, 3(1), 1–7. Sørensen, E. and J. Torfing (2003) ‘Network Politics, Political Capital, and Democracy’, International Journal of Public Administration, 26, 609–34. Special Committee on Reform of the House of Commons (1985) Report of the Special Committee on Reform of the House of Commons (Ottawa, ON: Government of Canada). Spirova, M. (2005) ‘Political Parties in Bulgaria: Organizational Trends in Comparative Perspective’, Party Politics, 11(5), 601–22. Stack, M. and D. Kelly (2006) ‘Popular Media, Education, and Resistance’, Canadian Journal of Education, 29(1), 5–25. Stark, L. P. (1996) Choosing a Leader: Party Leadership Contests in Britain from Macmillan to Blair (Basingstoke: Macmillan – now Palgrave Macmillan). Steinbrecher, M., S. Huber and H. Rattinger (2007) Turnout in Germany: Citizen Participation in State, Federal and European Elections since 1979 (Baden-Baden, Germany: Nomos). Steiner, J., A. Bachtiger, M. Sporndli, and M. R. Steenbergen (2004) Deliberative Politics in Action: Analysing Parliamentary Discourse (Cambridge: Cambridge University Press). Stewart, D. K. and R. K. Carty (2002) ‘Leadership Politics as Party Building: The Conservatives in 1998’ in W. C. Cross (ed.) Political Parties, Representation

294 Bibliography and Electoral Democracy in Canada (Don Mills, ON: Oxford University Press), pp. 55–67. Sundberg, J. (2002) ‘The Scandinavian Party Model at the Crossroads’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 181–216. Surowiecki, J. (2004) Wisdom of Crowds: Why the Many are Smarter than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations (New York: Random House). Svensson, P. (2002) ‘Five Danish Referendums on the European Community and European Union: A Critical Assessment of the Franklin Thesis’, European Journal of Political Research, 41, 733–50. Szczerbiak, A. (2004) ‘History Trumps Government Unpopularity: The June 2003 Polish EU Accession Referendum’, West European Politics, 27(4), 671–90. Szczerbiak, A. and P. Taggart (2004a) ‘The Politics of European Referendum Outcomes and Turnout: Two Models’, West European Politics, 27(4), 557–84. Szczerbiak, A. and P. Taggart (2004b) ‘Conclusion: Towards a Model of (European) Referendums’, West European Politics, 27(4), 749–77. Tarrow, S. (1994) Power in Movement, Social Movements, Collective Action and Politics (Cambridge: Cambridge University Press). Teorell, J., M. Torcal, and J. R. Montero (2007) ‘Political Participation: Mapping the Terrain’ in J. W. van Deth, J. R. Montero, and A. Westholm, Citizenship and Involvement in European Democracies (Abingdon: Routledge), pp. 334–57. Teorell, J. and A. Westholm (1999) ‘Att bestämma sig för att vara med och bestämma: Om varför vi röstar – allt mindre’ [To Decide to Participate in Deciding: On Why We Vote – Less and Less] in E. Amnå (ed.) Valdeltagande i förandring [Changes in Electoral Participation], Demokratiutredningens forskarvolym, XII (Stockholm: SOU), pp. 137–204. Theiss-Morse, E. (1993) ‘Conceptualizations of Good Citizenship and Political Participation’, Political Behavior, 15(4), 355–80. Thompson, D. (2008) ‘Who Should Govern Who Governs: The Role of Citizens in Reforming the Electoral System’ in M. Warren and H. Pearce (eds) Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press), pp. 20–49. Tillman, E. R. (2004) ‘The European Union at the Ballot Box?: European Integration and Voting Behavior in the New Member States’, Comparative Political Studies, 37(5), 590–610. Togeby, L. (2007) ‘Immigrant Exceptionalism: Immigrant Turnout in Local Elections in Denmark’. Paper presented at the ECPR workshops in Helsinki. Togeby, L. (2008) ‘The Political Representation of Ethnic Minorities. Denmark as a Deviant Case’, Party Politic, 14(3), 343–61. Trilateral Commission (1975) The Crisis of Democracy, Task Force Report #8 (New York: New York University Press). Urbinati, N. (2006) Representative Democracy: Principles and Genealogy (Chicago: University of Chicago Press). Urbinati, N. and M. E. Warren (2008) ‘The Concept of Representation in Contemporary Democratic Theory’, Annual Review of Political Science, 11, 387–412. Valmyndigheten [Swedish Election Authority] (2007) http://www.val.se/ valsystemet/rosta/personrosta/index.html

Bibliography 295 van der Brugge, J. and H. Voss (2007) ‘Årsagen til de socialistiske partiers tilbagegang i perioden 1990–2005’ [The Reason for the Decline of the Socialistic Parties in the Period 1990–2005] in J. Goul Andersen et al. (eds) Det nye politiske landskab: Folketingsvalget 2005 i perspektiv [The New Political Landscape: The 2005 National Elections in Perspective] (Aarhus, Denmark: Academíca), pp. 127–51. van Deth, J. (2000) ‘Interesting but Irrelevant: Social Capital and the Saliency of Politics in Western Europe’, European Journal of Social Research, 37, 115–47. van Deth, J. W., M. Maraffi, K. Newton, and P. F. Whiteley (eds) (1999) Social Capital and European Democracy (London: Routledge). van Deth, J.W., J. R. Montero, and A.Westholm (2006) Citizenship and Involvement in European Democracies: A Comparative Analysis (London: Routledge). Verba, S. and N. H. Nie (1972) Participation in America (Chicago: University of Chicago Press). Verba, S., N. Nie, and J. Kim (1978) Participation and Political Equality: A SevenNation Comparison (Cambridge: Cambridge University Press). Visser, J. (2006, January) ‘Union Membership Statistics in 24 Countries’, Monthly Labor Review, 38–49. Warren, M. E. (1996) ‘Deliberative Democracy and Authority’, American Political Science Review, 90, 46–60. Warren, M. E. (2000) ‘Democratic Theory and Trust’ in M. E. Warren (ed.) Democracy and Trust (Cambridge: Cambridge University Press), pp. 210–34. Warren, M. E. (2002) ‘What Can Democratic Participation Mean Today?’, Political Theory, 30(5), 678–702. Warren, M. E. (2003) ‘A Second Transformation of Democracy?’ in B. E. Cain, R. J. Dalton, and S. E. Scarrow (eds) Democracy Transformed?: Expanding Political Opportunities in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 223–49. Warren, M. E. (2008) ‘Citizen Representatives’ in M. Warren and H. Pearse (eds) Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press), pp. 50–69. Warren, M. E. and H. Pearse (eds) (2008) Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge: Cambridge University Press). Wass, H. (2007) ‘The Effects of Age, Generation and Period on Turnout in Finland 1975–2003’, Electoral Studies, 26(3), (September), 648–59. Webb, P. (2002) ‘Political Parties in Advanced Industrial Democracies’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press). Webb, P. (2002) ‘Political Parties in Britain: Secular Decline or Adaptive Resilience’ in P. Webb, D. Farrell, and I. Holliday (eds) Political Parties in Advanced Industrial Democracies (Oxford: Oxford University Press), pp. 16–45. Weiner, A. (1998) ‘European’ Citizenship Practice: Building Institutions of a Non-State (Boulder, CO: Westview). Weldon, S. (2006) ‘Downsize My Polity? The Impact of Size on Party Membership and Member Activism’, Party Politics, 12(4), 467–81. Westheimer, J. and J. Kahne (2004) ‘Educating the “Good” Citizen: Political Choices and Pedagogical Goals’, PS: Political Science and Politics, XXXVII(2), 241–7. Westheimer, J. and J. Kahne (2004) ‘What Kind of Citizen? The Politics of Educating for Democracy’, American Educational Research Journal, 41(2), 237–69.

296 Bibliography Whitaker, R. (1977) The Government Party: Organizing and Financing the Liberal Party of Canada: 1930–58 (Toronto: University of Toronto Press). Whiteley, P. F. (1999) ‘The Origins of Social Capital’ in J.W. Van Deth, M. Maraffi, K. Newton, and P. F. Whiteley (eds) Social Capital and European Democracy (London: Routledge), pp. 25–44. Whiteley, P. F. (2008) ‘What Makes a Good Citizen? Citizenship Across the Democratic World’ in (eds) A. Park, J. Curtice, K. Thomson, M. Phillips, M. C. Johnson, and E. Cleary British Social Attitudes, The 24th Report (London: Sage), pp. 173–201. Whiteley, P. and P. Seyd (2002) High Intensity Participation–The Dynamics of Party Activism in Britain (Ann Arbor: University of Michigan Press). Whiteley, P. F., P. Seyd, and J. Richardson (1994) True Blues: The Politics of Conservative Party Membership (Oxford: Oxford University Press). Wolfinger, Raymond E. and Steven J. Rosenstone (1980) Who Votes? (New Haven: Yale University Press). Yankelovich, D. (1991) Coming to Public Judgement: Making Democracy Work in a Complex World (Syracuse, NY: Syracuse University Press). Young, L. and W. Cross (2002) ‘The Rise of Plebiscitary Democracy in Canadian Political Parties’, Party Politics, 8(6), 673–99. Youniss, J., S. Bales, V. Christmas-Best, M. Diversi, M. McLaughlin, and R. Silbereisen (2002) ‘Youth Civic Engagement in the Twenty-First Century’, Journal of Research on Adolescence, 12(1), 121–48. Zaller, J. R. (1992) The Nature and Origins of Mass Opinion (New York: Cambridge University Press).

Index advocacy groups, see pressure groups Africa, 65 Amsterdam Treaty, see Treaty of Amsterdam assemblies, citizens’ consultation of public, 225–8 definition, 215 and democracy, 35–6, 230 education of members, 219–25 qualities, 216–17 resources, critical, 228–30 see also Canada; the Netherlands Australia, 174, 233, 245 voter turnout, see voter turnout Austria citizenship, extra-national, 209–10 demographics, 46 parties, political: competitiveness, 47; membership levels, 53; Socialist Party, 55 socioeconomic factors, 46 voter turnout, see voter turnout

Bulgaria, 46, 47 voter turnout, see voter turnout Bush, George, 271 Canada assemblies, citizens’, 35–6, 215, 217–30 citizenship, concepts and duties of, 198–9, 201–3, 204, 208–10, 211 demographics, 46 education, civic, 259, 262, 264, 265 electoral system, 23–4, 25–6, 31, 48, 163 media, 163, 266, 268, 270 parties, political: activity, 163, 165–6, 168; Bloc Québécois, 165, 176, 248; Canadian Alliance, 177, 183; competitiveness, 47; Conservative Party, 23, 175–6, 177, 182–3; identification with, 24; leadership power, 173–4; leadership selection, 168, 174–8, 185, 187–8, 191–2; Liberal Party (federal), 175–6, 177–8, 181–3, 184, 185; Liberal Party (Quebec), 243, 248; and media, 163; member activity, 168; membership levels, 133, 145; New Democratic Party (NDP), 177, 182, 183–4; Parti Québécois, 236, 242, 243; Progressive Conservative Party, 168, 176, 177, 185–7, 188; Reform Party, 165, 243 party system, 3, 5, 23–4, 42, 47, 151, 163, 165–6, 173, 174 political system, 2–3, 4, 5–7, 26–7, 30, 32, 49, 165–6, 168, 192, 234, 253 pressure groups, 163, 166 referendums, 22, 233: on the citizens’ assemblies’

Belgium education, civic, 262 electoral system, 48, 51, 107 parties, political: leadership selection, 174; and media, 162; membership levels, 53 party system, 46 political system, 49 socioeconomic factors, 46 union density, 54 voter turnout, see voter turnout Bennett, R.B., 175 Blair, Tony, 55, 180, 183, 234 Bloc Québécois (Canada), see Canada: parties, political Bouchard, Lucien, 248 Bourassa, Robert, 243, 244 Brazil, 226 Britain, see United Kingdom Brown, Gordon, 183 297

298 Index Canada – continued recommendations, 215; on the Charlottetown Accord in 1992, 234–6, 242–5, 250, 251, 252; on Quebec sovereignty in 1980, 248; on Quebec sovereignty in 1995, 236, 246, 248–9 socioeconomic factors, 46 voter turnout, see voter turnout youth, 7, 262, 266, 268, 270 Canadian Alliance, see Canada: parties, political Charlottetown Accord, 234–5, 242–5 Chirac, Jacques, 234, 239–40 citizens behaviours and characteristics of, 7, 20–2, 26, 33, 56, 133, 132–40, 145–6, 167, 204–7, 221, 236–7, 249 generations of, 7, 45, 66, 73, 89–90, 114, 124, 207 and trust, 1–2, 20–3, 19, 28, 33, 35–6, 50, 56, 57, 88, 95–6, 100–1, 115–16, 120, 124, 143, 145, 163, 198 citizenship and duty, 136–7, 139–40, 146, 197, 199–204, 210–11, 219–20, 230, 257–8, 260–1, 263–4 education, see education, civic extra-national, 197–9, 204, 209–11, 260, 265 models of, 259–61 and participation, 17, 20, 131, 139–40, 146, 173, 197, 204–7, 210–11, 214–16, 230, 233, 251–3, 257–8, 259, 261, 264–6 see also participation deficit Clark, Joe, 188 Clarke, Kenneth, 181 cognitive engagement model, 142, 145 Communist Party (France), see France: parties, political compulsory voting, see electoral system Conservative Party (Canada), see Canada: parties, political

Conservative Party (United Kingdom), see United Kingdom: parties, political The Council of the European Union, 154, 166; see also EU Council of Ministers Cyprus, 46, 48, 53 voter turnout, see voter turnout Czech Republic citizens and trust, 96 citizenship, extra-national, 209–10 demographics, 46 education, civic, 262 and the referendums on European Union constitutional change in 2005, 234 voter turnout, see voter turnout Danish People’s Party, see Denmark: parties, political Day, Stockwell, 192 democracy citizen views of, 56, 88 concepts, 17, 19–20, 21, 31, 61, 137, 197, 208, 214–16, 230, 265 history of: in the European Union, 50, 51; in Greece, 50; and party membership, 53; in Portugal, 50; in post-communist countries, 50, 51, 72, 122–3; in Spain, 50; in Switzerland, 50 importance of elections to, 21, 23, 27, 151–2 importance of parties to, 24, 33–4, 131, 138, 145–6, 156, 173, 188–91; see also parties, political and functions supplementary, 34–6: see also assemblies, citizens’; participation deficits and measures to address; parties, political and leadership selection; referendums see also democratic deficits; participation deficits democratic deficits, 1, 3, 17–29, 30–6, 42, 197–9, 233 see also democracy; participation deficits

Index Democratic Party (Italy), see Italy: parties, political Denmark citizens: and civic duty, 95, 99–100; and political interest, 93; and refugee and immigration issue, 94; and trust, 95–6, 99–100 citizenship, extra-national, 209–10 demographics, 46 electoral system, 87, 91 media, 162 parties, political: Danish People’s Party, 94–5; and media, 162; membership levels, 92–3; polarization of, 55, 93–4 party system, 46, 165 political system, 30, 49, 91 referendums, 233: on the euro in 2000, 236, 246–8, 250; on European Union constitutional change in 2005, 234, 241; on the Maastricht Treaty in 1992, 18, 238, 245 socioeconomic factors, 46 union density, 54 voter turnout, see voter turnout disproportionality, see electoral systems Douglas-Home, Alec, 179 Duncan Smith, Iain, 181 Eastern Europe, 72 education, civic goals of, 259–61 and media, 266–73 and schools, 261–6, 272 elections civic duty and, 95, 115, competitiveness, 54–5, 101, 120, 124 decisiveness of, 49–50, 63, 68–9, 70–1, 72, 112 to European Parliament, 4–5, 9, 151–2, 163; campaigning in, 47–8; and citizenship, extra-national, 210–11; timing of, 111; voting procedures, 96–7,

299

111; see also Europarties; voter turnout and European Parliament first-order, 5, 6, 86, 96, 99, 245 importance of, 6, 86, 96–7, 98, 115, 120 interest in, 93–4, mobilization and, 9–10, 67, 68, 84, 98–9, 101, 114, 124–5 second-order, 4, 24, 86, 99, 110–11, 115, 118, 122, 123, 160, 163, 238, 240, 241, 242, 245, 250 third-order, 97–8, 99 see also electoral systems; voter turnout electoral systems compulsory voting, 8–9, 48–9, 51, 63, 68, 69–70, 72, 85, 91, 107, 112, 113, 242 disproportionality, 48 mixed-member proportional, 31, 68, 223, 224, 225 nature of, 221, 223–4 proportional representation (PR), 24, 25, 27, 31, 48, 63, 68, 70, 72, 85–6, 87, 91, 112, 113, 114, 165, 221, 224 reform of, 31 representation, 23–7, 221 single-member plurality (SMP), 23, 25, 26, 31, 48, 182, 221 single transferable vote, 31, 113, 224 voter registration, 3, 85, 91, 114, 132 voting age, 10, 63, 68, 70, 72, 90, 112, 113, 114 Estonia citizen interest in the European Union, 123 demographics, 46 party system, 46, 47 referendum on accession in 2003, 120 voter turnout, see voter turnout Estrada, Joseph, 271 EU Council of Ministers, 154, 166

300 Index Europarties detachment from citizens, 156, 159–60 development, 151–4 as a distinctive party model, 155–6, 170 functions, 163–4, 166, 168–9, 170–1 participation in, 168–9 summary of, 169–71 European Coal and Steel Community, 152 European Commission, 18, 24, 25, 32, 50, 97, 166, 170 European Community, 151–2, 170 European Convention on the Future of Europe, 235 European Council, 25, 32, 97, 153, 169, 197 European Economic Community, 6, 197 European Parliament (EP) and citizenship, extra-national, 210–11 democratic deficit, 18, 32 elections, to, see elections Labour Party (United Kingdom) leadership selection and members of, 180 power of, 32, 50, 97, 115, 153, 159, 166 promotion of, 163–4 trust in, 100–1, 116 European Union (EU) citizenship: extra-national, 197–8, 210–11; and duty, 203–4, 210–11 democratic deficit, 18–20, 24, 25, 27, 32, 152–4, 198 electoral system, 113–14 member states, new, 108–9, 116–25, 152, 153, 201, 203–4, 209, 211, 240; see also post-communist countries party system, 47–8, 51, 152–4; see also Europarties permissive consensus, 18, 152 political system, 4–6, 27, 50 referendums: on accession in 2003, 120–3; on European Union

constitutional change in 2005, 29, 233–6, 237–42, 245, 250, 251, 253 trust in, 116 and unionization, 54 see also under individual institutions and treaties of the European Union Europeanization, 152–3 Flanders, 209–10 Fiji, 65 Finland citizens and trust, 96 citizenship, extra-national, 209–10 demographics, 46 parties, political: membership levels, 53 party system, 46 socioeconomic factors, 46, 53 union density, 54 voter turnout, see voter turnout first-order elections, see elections France demographics, 46 education, civic, 263, 265 electoral system, 5, 48, 113 media, 271 parties, political: Communist Party, 168, 240; competitiveness, 47; The Greens, 113; interest articulation and aggregation, 165; membership, 133, 168; National Front, 113, 165, 240; reform, 168; Socialist Party, 55, 168, 169, 240; Union for a Popular Movement, 240 political system, 3 referendums, 233: on European Union constitutional change in 2005, 234, 235, 239–41, 245, 250, 253; on the Maastricht Treaty in 1992, 18, 238, 240, 245 socioeconomic factors, 46 voter turnout, see voter turnout Free Democratic Party (Germany), see Germany: parties, political

Index

301

generations of citizens, see citizens Germany citizen trust, 96 demographics, 46 education, civic, 262–3 elections, 5 parties, political: Free Democratic Party, 168; and media, 162; member activity, 168; membership levels, 133; Social Democratic Party, 168 political system, 32 socioeconomic factors, 46 voter turnout, see voter turnout Weimar, 42 Giscard d’Estaing, Valèry, 238 Gladstone, William, 178 Greece citizens and trust, 56, 116 demographics, 46 electoral system, 48 and European Parliament election in 2004, 110 history, democratic, 50 parties, political: membership levels, 53 party system, 47 political system, 49 socioeconomic factors, 46 voter turnout, see voter turnout The Greens (France), see France: parties, political

International Social Survey Programme (ISSP), 8, 131, 132, 142, 199, 200, 201, 203, 204, 207 Internet, see media and new Ireland electoral system, 113 and the European Parliament election in 2004, 110 referendums, 233, 252: on European Union constitutional change, 234; on the Maastricht Treaty in 1992, 238; on the Treaty of Lisbon in 2008, 251, 252, 253; on the Treaty of Nice in 2001, 245, 250, 252; on the Treaty of Nice in 2002, 250, 252 voter turnout, see voter turnout Israel, 174 Italy demographics, 46 electoral system, 48 media, 268 parties, political: coalitions, 165; Democratic Party, 167; member activity, 167; membership levels, 53; movements of Second Republic, 160; Unione, 167 political system, 30 Second Republic, 160 socioeconomic factors, 46 voter turnout, see voter turnout and youth, 268

Howard, Michael, 181 Hungary parties, political: membership levels, 133 referendum on accession in 2003, 121 voter turnout, see voter turnout

Japan, 31 Junker, Jean-Claude, 242

Iceland parties, political: membership levels, 53 political system, 49 socioeconomic factors, 46, 53 voter turnout, see voter turnout interest groups, see pressure groups

Kennedy, Charles, 192 King, Mackenzie, 175 labour activity, 51, 54, 90–1, 86, 94, 164, 165, 177, 180, 182, 183–4, 191, 246 Labour Party (Norway), see Norway: parties, political Labour Party (United Kingdom), see United Kingdom: parties, political Laeken Declaration, 235 Latin America, 85

302 Index Latvia citizenship, extra-national, 209–10 demographics, 46 parties, political: membership levels, 133 party system, 46, 47 voter turnout, see voter turnout Liberal Democrats (United Kingdom), see United Kingdom: parties, political Liberal Party (Canada), see Canada: parties, political Liberal Party (United Kingdom), see United Kingdom: parties, political Lithuania, 46, 47, 120–1, 124 voter turnout, see voter turnout Luxembourg citizens and trust, 96, 107 electoral system, 48 media, 162 referendum on European Union constitutional change in 2005, 234, 241–2, 250 socioeconomic factors, 46 voter turnout, see voter turnout Maastricht Treaty, 18, 152, 163, 166, 238, 240, 245, 251 Mali, 65 Malta party system, 47, 51 parties, political: membership levels, 53 political system, 49, 51 socioeconomic factors, 46, 51, 53 voter turnout, see voter turnout Manning, Preston, 243, 244 McDonnell, Michael, 183 Meˇciar, Vladimír, 121 media mass, 87, 93, 99, 101 and mobilization, political, 10, 11, 93, 99, 101, 138, 142, 145 new, 10, 11, 140, 142, 162, 205, 207, 229, 268–72 news, 87, 99, 101, 207, 266–8

and participation, 205, 207, 229 and political parties, 140, 162–3, 165 and youth, see youth Meech Lake Accord, 234–5, 242, 244 Mexico, 6 mobilization, 8–11, 91, 86, 120–2, 166, 250, 267–8, 271 see also elections; parties, political Mulroney, Brian, 242, 244 National Front (France), see France: parties, political the Netherlands assembly, citizens’, 35, 215, 219–22, 224–30 citizens and trust, 96 citizenship, extra-national, 209–10 media, 162 parties, political: leadership selection, 174; and media, 162 party system, 46 referendum on European Union constitutional change in 2005, 234, 235, 241, 245, 250, 252, 253 voter turnout, see voter turnout New Democratic Party (NDP), see Canada: parties, political New Zealand, 31, 174 North America, 6, 66, 198–9, 211, 263, 267 North American Free Trade Agreement (NAFTA), 6, 198 Norway citizens: and civic duty, 95; and political interest, 93; and trust, 96 demographics, 46 electoral system, 91 media, 162 parties, political: Labour Party, 86; and media, 162; membership levels, 53, 92–3; and refugee and immigration policy, 94 party system, 46, 165 political system, 49, 91 socioeconomic factors, 46, 86 voter turnout, see voter turnout and youth, 89, 95

Index Oceania, 65–6 Orchard, David, 188 Parti Québécois, see Canada: parties, political participation deficits description of, 1 explanations for, 4–7, 7–8, 22–3, 45, 53–6, 145–6, 167, 251–3 impact of, 1–2, 7–8, 41–2, 146 measures to address, 3, 8–11, 22, 28–9, 34–6, 56–7, 188–91, 211, 233, 258, 273 and youth, 257, 266–7 see also parties, political and membership, explanations for; voter turnout and explanations for parties, political, 68 citizen views on, 1–2, 22, 33 and the European Union, see Europarties functions: policy influence, 138, 154; citizen integration, 154; citizen participation, 68, 138, 140, 145–6, 167–9; citizen and political elite linkage, 137–8, 146; civil society maintenance, 131; interest articulation and aggregation, 34, 164–6; organizational survival, 154–5; personnel recruitment, 138, 146, 173; vote structuring, 54–5, 99, 160–4 leadership selection, 167–8, 173–92 membership: characteristics, 41, 132–40, 145–6, 188, 191; definition, 132; explanations for, 1, 53–6, 143–4, 167–8, 173, 187, 214–15; levels, 1, 24, 41, 53, 92–3, 131, 145, 146, 167 and mobilization, 49, 51–7, 68, 92–5, 99, 123, 167 models of, 154–60: cartel party, 155, 156, 158–9, 162, 165, 169, 170; catch-all party, 155, 158, 160–1, 165; mass party, 154, 156–8, 160–1 presidentialization of, 173–4 reform of, 33–4, 57, 131–2, 146, 167–8, 173–4, 184–7

303

see also under individual countries party system, 42, 46–8, 51, 63, 70–1, 72, 112, 115 permissive consensus, see European Union Philippines, 271 Poland education, civic, 262 electoral system, 48 Kaczynski, ´ Lech, 116 parties, political: membership levels, 133 party system, 46, 47 referendum on accession in 2003, 120–1 and the referendums on European Union constitutional change in 2005, 234 socioeconomic factors, 46 voter turnout, see voter turnout Portugal education, civic, 262 history, democratic, 50 party system, 47 and the referendums on European Union constitutional change in 2005, 234 voter turnout, see voter turnout post-communist countries, 44, 46–7, 50, 51, 109, 116–23, 124–5, 133, 152, 203–4, 262 see also under individual countries; see also European Union and member states, new pressure groups, 24, 25, 29, 33, 34, 155, 158, 162, 163, 164–5, 173 Progressive Conservative Party (Canada), see Canada: parties, political proportional representation, see electoral systems public interest groups, see pressure groups referendums, 10–11, 29, 233–8, 249–53 see also under individual countries; see also European Union

304 Index Reform Party (Canada), see Canada: parties, political Reform Treaty, see Treaty of Lisbon representation, see electoral systems Romania, 46, 123 Rosebery, Lord, 178 Russia, 133 Scandanavia see Denmark; Norway; Sweden Scotland, 246 second-order elections, see elections Second Republic (Italy), see Italy Single European Act, 152 single-member plurality (SMP), see electoral systems Slovakia citizen interest in the European Union, 123 citizenship, extra-national, 209–10 demographics, 46 referendum on accession in 2003, 121 voter turnout, see voter turnout Slovenia education, civic, 263 party system, 46 referendum on accession in 2003, 120 voter turnout, see voter turnout social capital model, 142–3, 145, 197, 266 Social Democratic Party (Germany), see Germany: parties, political Social Democratic Party (United Kingdom), see United Kingdom: parties, political Socialist Party (Austria), see Austria: parties, political Socialist Party (France), see France: parties, political socioeconomic factors, 45–6, 53, 63, 66–8, 70, 72, 90–1, 99, 101, 133, 135, 145, 204 South America, 66 Spain citizens and trust, 56 demographics, 46

and European Parliament election in 2004, 110 history, democratic, 50 media, 267 parties, political: membership levels, 53; regionalist, 165 party system, 47 referendum on European Union constitutional change in 2005, 234, 238–9, 250 voter turnout, see voter turnout youth, 267 Steel, David, 179 Sweden citizens: and civic duty, 95; and political interest, 93; and trust, 96 citizenship, extra-national, 209–10 demographics, 46 electoral system, 91–2 and European Parliament election in 2004, 110 media, 162, 269 parties, political: and media, 162; membership levels, 92–3, 133; and refugee and immigration policy, 94 party system, 46, 165 political system, 91 referendum on the euro in 2003, 236, 246, 250 socioeconomic factors, 46 union density, 54 and youth, 89, 95, 269 voter turnout, see voter turnout Switzerland demographics, 46 electoral system, 48 history, democratic, 50, 51 party system, 67 political system, 49, 51 socioeconomic factors, 46 voter turnout, see voter turnout Thatcher, Margaret, 185 third-order elections, see elections trade unions, see labour activity Treaty of Amsterdam, 152 Treaty Establishing a Constitution for Europe, 233, 235, 237, 251, 252

Index Treaty Establishing the European Community, 153 Treaty of European Union, see Maastricht Treaty Treaty of Lisbon, 32, 166, 251, 252, 253 Treaty of Nice, 245, 250, 251, 252 Treaty of Rome, 151, 153 Trudeau, Pierre, 242, 244 trust, see citizens Turkey, 240 Union for a Popular Movement (France), see France: parties, political Unione (Italy), see Italy: parties, political unions, see labour activity United Kingdom citizens: and electoral reform, 165; and trust, 56, 96, 165 demographics, 46 education, civic, 258, 259, 263–4, 265 electoral system, 48 media, 162, 266, 267, 268, 270, 271 parties, political: Conservative Party, 55, 178–9, 180–1, 183, 185; identification with, 24; Labour Party, 55, 165, 178, 180, 182, 183–4, 185, 187; leadership selection, 167, 174–5, 178–81, 187–8, 191–2; Liberal Democrats, 179–80, 183, 185; Liberal Party, 179, 184; membership levels, 133; polarization of, 54–5; Social Democratic Party, 179 party system, 47, 54–5, 165 political system, 32, 49 referendums, 233, 234, 241 socioeconomic factors, 46 youth, 7, 266 voter turnout, see voter turnout United States citizens: characteristics, 20 citizenship, concepts of, 200 education, civic, 262 electoral system, 3–4, 33–4, 85, 132

305

media, 162, 266, 267, 271 parties, political: identification with, 24; leadership selection, 191; and media, 162; member definition, 132; membership levels, 133 and the North American Free Trade Agreement, 6 Rock the Vote, 268 socioeconomic factors, 86, 90 voter turnout, see voter turnout youth, 266, 267 Victoria, Queen, 178 voter registration, see electoral systems voter turnout Australia, 49 Austria, 44, 46, 47, 107 Belgium, 44, 46, 48, 49, 51, 107 Bulgaria, 46, 47 Canada, 5, 6–7, 22, 23, 26, 44, 45, 46, 47, 48, 49, 66, 73, 83, 85, 89, 248, 250 Cyprus, 44, 46, 48 Czech Republic, 46, 116–18, 124 Denmark, 44, 46, 49, 83, 85, 87, 88, 89–91, 92, 96–101, 107, 112, 114, 115 and electoral outcomes, 41–2 Estonia, 46, 47, 116–18, 120, 123, 124 Europe, 4, 5–6, 22, 26, 65, 67, 70, 71, 72–3 European Parliament, 4, 18, 22, 44, 47–8, 51, 95, 96–7, 106–25, 168–9, 198 explanations for, 4–7, 10, 44–51, 53–6, 63, 66–73, 85–101, 110–25, 138, 167, 168–9, 200, 205, 214–15, 252 Fiji, 65 Finland, 44, 45, 46, 83, 88, 89, 107–8 France, 44, 46, 47, 48, 165, 240, 250 Germany, 44, 46, 49, 89, 96 Greece, 44, 46, 47, 48, 49, 50, 107, 114, 116 Hungary, 44, 116, 120, 121

306 Index voter turnout – continued Iceland, 44, 46, 49 Ireland, 44, 250 Italy, 44, 46, 48, 114 Latvia, 44, 46, 47, 116, 120 Lithuania, 44, 46, 47, 116, 120–1, 124 Luxembourg, 44, 46, 48, 107 Mali, 65 Malta, 44, 46, 47, 49, 51, 65 the Netherlands, 44, 114, 124, 241, 250 Norway, 44, 46, 49, 83, 86, 88, 89, 90, 91, 93, 94, 95 Poland, 44, 46, 47, 48, 116–18, 120–1, 124 Portugal, 44, 47, 50, 107 post-communist countries, 44, 46–7, 50, 51, 109, 116–23, 124–5 referendum on accession in 2003, 120–3 referendum on European Union constitutional change in 2005, 239, 240, 241, 250 Romania, 46, 123

Slovakia, 46, 116, 121, 122, 123, 124 Slovenia, 46, 116, 120, 124 Spain, 44, 46, 47, 50, 107, 239, 250 Sweden, 44, 46, 83, 88, 89, 90, 91–92, 93, 94, 95, 107–8 Switzerland, 44, 46, 48, 49, 50, 51, 65, 67 the United Kingdom, 44, 46, 47, 48, 49, 55, 83, 89, 107, 114 the United States, 85, 86, 90 worldwide, 1, 65–6, 67, 70, 71, 72–3 youth, 7–8, 68, 83, 85, 89–90, 91, 94–55, 101, 114 voting age, see electoral systems

youth and citizenship, 7–8, 28, 204, 257–8, 259–61, 273 and education, civic, see education, civic and media, 266–72 political party membership, 94–5, 133, 173 voter turnout, see voter turnout