The Political Representation of Immigrants and Minorities: Voters, parties and parliaments in liberal democracies 9780415492720, 9780203843604

256 103 3MB

English Pages [305] Year 2011

Report DMCA / Copyright


Polecaj historie

The Political Representation of Immigrants and Minorities: Voters, parties and parliaments in liberal democracies
 9780415492720, 9780203843604

Table of contents :
Book Cover
Series editor’s preface
1 Ethnic diversity, political participation and representation: A theoretical framework
Part I: Immigrants and members of visible minorities as voters: Turnout and party choice
2 Voter turnout among immigrants and visible minorities in comparative perspective
3 Party choices among immigrants and visible minorities in comparative perspective
Part II: Immigrants and members of visible minorities as candidates for elective office
4 New citizens – new candidates? Candidate selection and the mobilization of immigrant voters in German elections
5 Minority representation in Norway: Success at the local level; failure at the national level
6 Ethnic inclusion or exclusion in representation? Local candidate selection in Sweden
7 Yes they can: An experimental approach to the eligibility of ethnic minority candidates in France
Part III: Immigrants and members of visible minorities as legislators
8 Minority representation in the US Congress
9 Patterns of substantive representation among visible minority MPs: Evidence from Canada’s House of Commons
10 Presence and behaviour: Black and minority ethnic MPs in the British House of Commons
11 Migrants as parliamentary actors in Germany
12 Epilogue: Towards a strategic model of minority participation and representation

Citation preview

The Political Representation of Immigrants and Minorities

In 2005, almost 700,000 immigrants acquired the citizenship of a member state of the European Union; over 600,000 became US citizens; nearly 100,000 became Australians and approximately 200,000 Canadians, and 2005 was not an exceptional year. During the past decades, many advanced liberal democracies have become more ethnically diverse societies. This book breaks new ground in the analysis of the political representation of immigrants and visible minorities both theoretically and empirically. It examines the upward trend in migrant and minority representation and demonstrates that there remain crucial differences across liberal democracies in the timing of these developments; in channels of access for minority representatives, in the policy focus and outcomes of minority representation; in the nature of the connections between minority representatives and minority communities, and in the nature of their relationships with constituents at large. Part I analyses immigrants and visible minorities as voters, who must be the starting point of any analysis of political representation. Part II deals with the stage of candidate selection within political parties, a crucial and under-­ researched stage in the process of political representation. Part III deals with immigrants and members of visible minorities, once elected to parliament and includes analyses of the Canadian Parliament, the German Bundestag, MPs in the United Kingdom and Members of the United States Congress. The book will be of interest to students and scholars of migration and ethnicity studies and political science, especially those with an interest in political representation, democratic institutions, voting behaviour, party organisation, legislative behaviour and comparative politics. Karen Bird is Associate Professor in the Department of Political Science, McMaster University, Canada. Thomas Saalfeld is Professor of Political Science at the University of Bamberg, Germany. Andreas M. Wüst is Research Fellow at the Mannheim Center for European Social Research, University of Mannheim, Germany.

Routledge/ECPR studies in European political science Edited by Thomas Poguntke Ruhr University Bochum, Germany on behalf of the European Consortium for Political Research

The Routledge/ECPR Studies in European Political Science series is published in association with the European Consortium for Political Research – the leading organization concerned with the growth and development of political science in Europe. The series presents high-­quality edited volumes on topics at the leading edge of current interest in political science and related fields, with contributions from European scholars and others who have presented work at ECPR workshops or research groups.   1 Regionalist Parties in Western Europe Edited by Lieven de Winter and Huri Türsan

  6 Social Capital and European Democracy Edited by Jan van Deth, Marco Maraffi, Ken Newton and Paul Whiteley

  2 Comparing Party System Change Edited by Jan-­Erik Lane and Paul Pennings

  7 Party Elites in Divided Societies Edited by Kurt Richard Luther and Kris Deschouwer

  3 Political Theory and European Union Edited by Albert Weale and Michael Nentwich   4 Politics of Sexuality Edited by Terrell Carver and Véronique Mottier   5 Autonomous Policy Making by International Organizations Edited by Bob Reinalda and Bertjan Verbeek

  8 Citizenship and Welfare State Reform in Europe Edited by Jet Bussemaker   9 Democratic Governance and New Technology Technologically mediated innovations in political practice in Western Europe Edited by Ivan Horrocks, Jens Hoff and Pieter Tops 10 Democracy without Borders Transnationalisation and conditionality in new democracies Edited by Jean Grugel

11 Cultural Theory as Political Science Edited by Michael Thompson, Gunnar Grendstad and Per Selle 12 The Transformation of Governance in the European Union Edited by Beate Kohler-­Koch and Rainer Eising 13 Parliamentary Party Groups in European Democracies Political parties behind closed doors Edited by Knut Heidar and Ruud Koole 14 Survival of the European Welfare State Edited by Stein Kuhnle 15 Private Organisations in Global Politics Edited by Karsten Ronit and Volker Schneider 16 Federalism and Political Performance Edited by Ute Wachendorfer-­Schmidt 17 Democratic Innovation Deliberation, representation and association Edited by Michael Saward

20 Estimating the Policy Position of Political Actors Edited by Michael Laver 21 Democracy and Political Change in the ‘Third World’ Edited by Jeff Haynes 22 Politicians, Bureaucrats and Administrative Reform Edited by B. Guy Peters and Jon Pierre 23 Social Capital and Participation in Everyday Life Edited by Paul Dekker and Eric M. Uslaner 24 Development and Democracy What do we know and how? Edited by Ole Elgström and Goran Hyden 25 Do Political Campaigns Matter? Campaign effects in elections and referendums Edited by David M. Farrell and Rüdiger Schmitt-­Beck 26 Political Journalism New challenges, new practices Edited by Raymond Kuhn and Erik Neveu 27 Economic Voting Edited by Han Dorussen and Michaell Taylor

18 Public Opinion and the International Use of Force Edited by Philip Everts and Pierangelo Isernia

28 Organized Crime and the Challenge to Democracy Edited by Felia Allum and Renate Siebert

19 Religion and Mass Electoral Behaviour in Europe Edited by David Broughton and Hans-­Martien ten Napel

29 Understanding the European Union’s External Relations Edited by Michèle Knodt and Sebastiaan Princen

30 Social Democratic Party Policies in Contemporary Europe Edited by Giuliano Bonoli and Martin Powell 31 Decision Making Within International Organisations Edited by Bob Reinalda and Bertjan Verbeek 32 Comparative Biomedical Policy Governing assisted reproductive technologies Edited by Ivar Bleiklie, Malcolm L. Goggin and Christine Rothmayr 33 Electronic Democracy Mobilisation, organisation and participation via new ICTs Edited by Rachel K. Gibson, Andrea Römmele and Stephen J. Ward 34 Liberal Democracy and Environmentalism The end of environmentalism? Edited by Marcel Wissenburg and Yoram Levy 35 Political Theory and the European Constitution Edited by Lynn Dobson and Andreas Follesdal 36 Politics and the European Commission Actors, interdependence, legitimacy Edited by Andy Smith 37 Metropolitan Governance Capacity, democracy and the dynamics of place Edited by Hubert Heinelt and Daniel Kübler

38 Democracy and the Role of Associations Political, organizational and social contexts Edited by Sigrid Roßteutscher 39 The Territorial Politics of Welfare Edited by Nicola McEwen and Luis Moreno 40 Health Governance in Europe Issues, challenges and theories Edited by Monika Steffen 41 Republicanism in Theory and Practice Edited by Iseult Honohan and Jeremy Jennings 42 Mass Media and Political Communication in New Democracies Edited by Katrin Voltmer 43 Delegation in Contemporary Democracies Edited by Dietmar Braun and Fabrizio Gilardi 44 Governance and Democracy Comparing national, European and international experiences Edited by Yannis Papadopoulos and Arthur Benz 45 The European Union’s Roles in International Politics Concepts and analysis Edited by Ole Elgström and Michael Smith 46 Policy-­making Processes and the European Constitution A comparative study of member states and accession countries Edited by Thomas König and Simon Hug

47 Democratic Politics and Party Competition Edited by Judith Bara and Albert Weale

56 Intra-­Party Politics and Coalition Governments Edited by Daniela Giannetti and Kenneth Benoit

48 Participatory Democracy and Political Participation Can participatory engineering bring citizens back in? Edited by Thomas Zittel and Dieter Fuchs

57 Political Parties and Partisanship Social identity and individual attitudes Edited by John Bartle and Paolo Belucci

49 Civil Societies and Social Movements Potentials and problems Edited by Derrick Purdue

58 The Future of Political Community Edited by Gideon Baker and Jens Bartelson

50 Resources, Governance and Civil Conflict Edited by Magnus Öberg and Kaare Strøm

59 The Discursive Politics of Gender Equality Stretching, bending and policy making Edited by Emanuela Lombardo, Petra Meier and Mieke Verloo

51 Transnational Private Governance and its Limits Edited by Jean-­Christophe Graz and Andreas Nölke 52 International Organizations and Implementation Enforcers, managers, authorities? Edited by Jutta Joachim, Bob Reinalda and Bertjan Verbeek

60 Another Europe Conceptions and practices of democracy in the European social forums Edited by Donatella Della Porta

53 New Parties in Government Edited by Kris Deschouwer

61 European and North American Policy Change Drivers and dynamics Edited by Giliberto Capano and Michael Howlett

54 In Pursuit of Sustainable Development New governance practices at the sub-­national level in Europe Edited by Susan Baker and Katarina Eckerberg

62 Referendums and Representative Democracy Responsiveness, accountability and deliberation Edited by Maija Setälä and Theo Schiller

55 Governments, NGOs and Anti-­ Corruption The new integrity warriors Edited by Luís de Sousa, Barry Hindess and Peter Larmour

63 Education in Political Science Discovering a neglected field Edited by Anja P. Jakobi, Kerstin Martens and Klaus Dieter Wolf

64 Religion and Politics in Europe, the Middle East and North Africa Edited by Jeffrey Haynes 65 New Directions in Federalism Studies Edited by Jan Erk and Wilfried Swenden 66 Public Policy and the Media The interplay of mass communication and political decision making Edited by Sigrid Koch-­Baumgarten and Katrin Voltmer 67 Changing Government Relations in Europe From localism to intergovernmentalism Edited by Michael J. Goldsmith and Edward C. Page

68 Political Discussion in Modern Democracies A comparative perspective Edited by Michael R. Wolf, Laura Morales and Ken’ichi Ikeda 69 Dominant Political Parties and Democracy Concepts, measures, cases and comparisons Edited by Matthjis Bogaards and Françoise Boucek 70 The Political Representation of Immigrants and Minorities Voters, parties and parliaments in liberal democracies Edited by Karen Bird, Thomas Saalfeld and Andreas M. Wüst

Also available from Routledge in association with the ECPR: Sex Equality Policy in Western Europe, Edited by Frances Gardiner; Democracy and Green Political Thought, Edited by Brian Doherty and Marius de Geus; The New Politics of Unemployment, Edited by Hugh Compston; Citizenship, Democracy and Justice in the New Europe, Edited by Percy B. Lehning and Albert Weale; Private Groups and Public Life, Edited by Jan W. van Deth; The Political Context of Collective Action, Edited by Ricca Edmondson; Theories of Secession, Edited by Percy Lehning; Regionalism Across the North/South Divide, Edited by Jean Grugel and Wil Hout.

The Political Representation of Immigrants and Minorities

Voters, parties and parliaments in liberal democracies Edited by Karen Bird, Thomas Saalfeld and Andreas M. Wüst

First published 2011 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Avenue, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group, an informa business This edition published in the Taylor & Francis e-Library, 2010. To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to © 2011 Karen Bird, Thomas Saalfeld and Andreas M. Wüst for selection and editorial matter; individual contributors, their contribution All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN 0-203-84360-6 Master e-book ISBN

ISBN13: 978-0-415-49272-0 (hbk) ISBN13: 978-0-203-84360-4 (ebk)


List of figures List of tables List of contributors Series editor’s preface Preface Acknowledgements

  1 Ethnic diversity, political participation and representation: a theoretical framework

xi xii xiv xvi xviii xx


K aren B ird , T H omas S aalfeld and A ndreas M . W Ü st

Part I

Immigrants and members of visible minorities as voters: turnout and party choice


  2 Voter turnout among immigrants and visible minorities in comparative perspective


  3 Party choices among immigrants and visible minorities in comparative perspective


Part II

Immigrants and members of visible minorities as candidates for elective office


  4 New citizens – new candidates? Candidate selection and the mobilization of immigrant voters in German elections


S ara C laro da F onseca

x   Contents   5 Minority representation in Norway: success at the local level; failure at the national level


J o H annes B erg H and T or B j Ø rklund

  6 Ethnic inclusion or exclusion in representation? Local candidate selection in Sweden


M aritta S oininen

  7 Yes they can: an experimental approach to the eligibility of ethnic minority candidates in France


S ylvain B rouard and V incent T iberj

Part III

Immigrants and members of visible minorities as legislators


  8 Minority representation in the US Congress


J ason P . C asellas and D avid L . L eal

  9 Patterns of substantive representation among visible minority MPs: evidence from Canada’s House of Commons


K aren B ird

10 Presence and behaviour: black and minority ethnic MPs in the British House of Commons


T H omas S aalfeld and K alliopi K yriakopoulou

11 Migrants as parliamentary actors in Germany


A ndreas M . W Ü st

12 Epilogue: towards a strategic model of minority participation and representation


T H omas S aalfeld , A ndreas M . W Ü st and K aren B ird





Political opportunities and strategies for ethnic minority representation 2.1 Turnout rates of German and EU citizens in local elections since 1995 4.1 The process of legislative recruitment 4.2a Non-­German candidate nomination, Bundestag elections 1990–2005, by political party 4.2b Ethnic German candidate nomination, Bundestag elections 1990–2005, by political party 7.1 Impact of ethnic prejudice on the average probability of judging candidate as responsive 7.2 Impact of ethnic prejudice on the average probability of voting for the hypothetical candidate 8.1 Minority representation in the U.S. House, 1957–2007 10.1 Hierarchical cluster analysis of minority-­related questions for written answer by policy area for 30 UK MPs 11.1 Parliamentarians with a migratory background in Germany’s Bundestag, in state parliaments and of the European Parliament members elected in Germany (N) 11.2 Share of primary committee memberships of MPs with a migratory background compared to all MPs (national, regional and EU level combined) 12.1 Sequential evaluation game between constituency party selectors, voters and MPs in parliamentary democracies

13 48 113 122 173 177 187 244 244 254 258 268


2.1 2.2 2.3 2.4 2.5 2.6 2.7 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 4.1 4.2

Reported voter registration in France, 2005 Turnout of migrant voters in two Dutch cities, 1994–2006 Potential and actual electoral impact of EU citizens and non-­EU citizens for the 2006 local elections in Brussels’ municipalities Voter turnout in Swedish national and municipal elections among minority citizens, non-­citizens and majority population Minority participation in the 2007 Norwegian local elections, weighted by background variables, and compared to the general level of turnout Electoral participation by citizenship and country of origin in local elections in Denmark 2001 and in Norway 2003 Political participation of immigrant groups in Spain, national, regional and local elections Impact of candidate ethnicity and party on voter support in an experiment Vote choice by educational degree, occupation and religious belief Party choice of the three main groups of migrant voters in the Netherlands, 1994–2006 Net effects of socio-­demographic variables on voting behaviour in Schaarbeek (multinomial logistic regression) Party preferences among migration-­related groups in Sweden, 2006 Voting among non-­Western immigrants in the 2007 Norwegian local elections, by continent of origin, compared to election results Party preferences of immigrants and their Descendantsa by country of origin Vote intention in different immigrant groups in Spain (if they had the right to vote) Candidate nomination by the political parties, Bundestag elections 1990–2005, by ethnicity Percentage immigrant candidate nomination by the political parties, German national elections 1990–2005, by ethnicity and type of nomination

29 34 38 40 45 46 55 68 72 77 81 83 87 90 97 121 123

Tables   xiii 5.1

5.2 5.3 5.4 7.1 7.2 7.3 8.1 8.2 9.1 9.2 9.3 9.4 10.1 10.2 11.1 11.2

Numbers of candidates for Norwegian municipal councils with non-­Western backgrounds given perfect representation, compared to the actual numbers of candidates from this group in 2007 Numbers of representatives with non-­Western backgrounds in Norwegian municipal councils given perfect representation, compared to actual representation of this group in 2007 Candidates for seats in Norwegian municipal councils, in four types of ballot placements Percentage of voters who cast a preferential vote, by ethnic background and municipality Responses to hypothetical candidate experiment Determinants of perceived responsiveness of the candidate (logistic regression) Determinants of candidate’s electoral viability (logistic regression) OLS models of House of Representatives Nominate and ACU scores OLS models of Senate Nominate and ACU scores Mentions of ethnic related issues by category of MP Nature of ethnic-­related topics discussed Ethnic related issues mentioned by party affiliation Ethnic and women’s issues, by candidate ethnicity and gender Defying the Party Whip in the 1997, 2001 and 2005 Parliaments (May 1997 to April 2007) Poisson Regression Model for the number of minority-­related parliamentary questions for 30 Members of Parliament (2005–2008) Absolute numbers and shares of MPs with a migratory background on different political levels in Germany (as of February 2008) Questions of MPs with a migratory background in the 16th German Bundestag, November 2005 until March 2008

132 134 141 142 170 172 176 195–6 197–8 218 219 222 223 240 242 255 260


Johannes Bergh is Senior Researcher at the Institute for Social Research Oslo. Karen Bird is Associate Professor of Political Science at McMaster University. Tor Bjørklund is Professor of Political Science at the University of Oslo. Sylvain Brouard is Senior Research Fellow FNSP at SPIRIT, Sciences Po Bordeaux, University of Bordeaux. Jason P. Casellas is Assistant Professor of Government at the University of Texas at Austin. Pascal Delwit is Professor of Political Science at ULB Brussels (Université Libre de Bruxelles). Claudia Diehl is Professor for Migration and Ethnicity at the University of Goettingen. Sara Claro da Fonseca is Researcher at the Social Science Research Centre (WZB) Berlin. Dirk Jacobs is Associate Professor of Sociology at ULB Brussels (Université Libre de Bruxelles). Marcelo Jenny is Assistant Professor of Political Science at the University of Vienna. Kalliopi Kyriakopoulou is Visiting Fellow at the UK International Study Centre of Queen’s University (Canada) and Visiting Lecturer at the School of Law, Lille Catholic University. David L. Leal is Associate Professor of Government at the University of Texas at Austin. Mónica Méndez Lago is Professor of Political Science at the University of Murcia. Laure Michon is PhD candidate at the Institute for Migration and Ethnic Studies at the University of Amsterdam.

Contributors   xv Flemming Mikkelsen is Senior Researcher at the Academy for Migration Studies in Denmark (AMID), University of Copenhagen. Santiago Pérez-Nievas is Lecturer in Political Science at the Universidad Autónoma Madrid. Thomas Saalfeld is Professor of Political Science at the University of Bamberg. Maritta Soininen is Associate Professor of Political Science at the University of Stockholm. Nazem Tahvilzadeh is PhD candidate at the School of Public Administration, University of Göteborg. Vincent Tiberj is Senior Researcher at the Centre for European Studies (CEE) at Sciences Po, Paris. Jean Tillie is Deputy Director of the Institute for Migration and Ethnic Studies and Professor of Electoral Politics at the University of Amsterdam. Lise Togeby († 2008) was Professor of Political Sociology at the University of Aarhus. Andreas M. Wüst is Post-­Doctoral Research Fellow of the Volkswagen Foundation at the MZES, University of Mannheim.

Series editor’s preface

There can be little doubt that immigration has substantially changed Western European and North American societies; but how much has it changed the politics of these countries? To be sure, there has been a considerable effect of immigration on the substance of politics. Debates over wearing headscarves, burkas or building minarets are but the most recent examples of the impact of immigration on the politics of the receiving countries. Yet we still know very little about the effects of migration on the core of these countries’ political process, namely elections, parties and parliaments. More precisely: while there has been considerable debate about how to deal with immigration and its societal repercussions ranging from citizenship to ghettoization, the degree to which immigrants and visible minorities have themselves become involved in these debates is largely unknown. Much has been written about anti-­immigration parties and their effect on the cleavage structures of European party systems but research on the involvement of migrants within established parties is only now beginning to yield results. These few remarks may suffice to demonstrate that this volume fills a considerable gap in our knowledge about a phenomenon whose impact on the quality of Western democracies can hardly be overestimated. After all, the share of people with a migratory background reaches double-­digit figures in several countries, and it is evident that the way these people become involved in the political process is of crucial importance. To be sure, nowhere are we dealing with a homogenous group but there are subgroups in many countries where the questions of collective identity and loyalty vis-­à-vis the receiving state are far from being settled. The editors have assembled a team of scholars who investigate how and to what degree immigrants and visible minorities have turned from objects of subjects of the democratic political process. What are the levels of turnout among those who belong to this group of citizens? Do they show specific party political preferences? What are the specific opportunity structures which determine their chances of getting nominated for a parliamentary mandate? And how do parties of different ideological families use the nomination of candidates with a migratory background strategically? These are but some of the questions addressed in this volume that begins with two collective written chapters on turnout and party choice among immigrants

Series editor’s preface   xvii and visible minorities providing rich comparative information. As indicated already, the book limits its focus on advanced liberal democracies, excluding post-­colonial and post-­communist societies as well as emerging democracies. There are good reasons for this, as otherwise the problems of comparability might have become too large. On the other hand, the countries of East-­Central Europe provide a rich if somewhat different range of open questions related to the political involvement of minorities which may well warrant a similar volume. As the dominant focus of this book is on the political opportunity structure and strategic choices made by individual and collective actors, it shifts our attention from the debate over the relative merits of different shades of multiculturalism to the inherent mechanisms of the political process. The chapters demonstrate that, as the editors write in their epilogue; minority representation ‘is structured by political institutions but is ultimately shaped by human agency (and strategy)’. They also show that descriptive representation is not only of symbolic value but has mobilizing effects in that it will pull others along and enhance the integration of minorities into the system. However, this can be a double-­edged sword for nominating political parties because electoral gains among migrant communities may be offset by losses among those who view immigration critically. Yet modern democracies can only gain if such strategic choices are made within the existing party systems. Thomas Poguntke, Series editor Bochum, February 2010


The idea and initial preparations for this book took shape at the ECPR Joint Sessions of Workshops held in Helsinki in May 2007. The large number of exceptionally high-­quality papers presented on that occasion provided the basis for extensive discussion and enriched understanding of issues related to immigrant and minority political representation. As editors, this provided us with an embarrassment of riches, but also the very difficult task of selecting papers for this volume. The final selection was based on the need to ensure a book of reason­ able coherence and normal length. Further chapters were commissioned to strengthen the book’s treatment of candidate selection (Bergh and Bjørklund address this stage in Norway, while Brouard and Tiberj examine France), and to extend its geographical reach (Casellas and Leal cover minority representation in the U.S. Congress). We are particularly grateful to these scholars for their willingness to collaborate in this project at a later stage. As a matter of record, the Helsinki papers that were not included in this volume are: • • • • • • • • •

Johannes Bergh, ‘Political Representation in Norway’ Kristina Boréus, ‘Making Natives “Us” and Immigrants “Them” – A Study of Discrimination in Election Campaigns in Sweden, Denmark and Austria’ Justin Gest, ‘The Bande A Part: How Globalization Has Affected Immigrant–Host Relations in Western Democracies’ Laure Michon, ‘Immigrants as Political Actors: The Use of Voting Rights by Immigrants in Amsterdam, 1986–2002’ Aurélie Audrey Morin, ‘The Ambivalent Role of Ethnicity: French Candidates of North-­African Origin in an Electoral Campaign’ Lise Togeby, ‘Immigrant Exceptionalism: Immigrant Turnout at Local Elections in Denmark’ Tamara Semenova, ‘Representation of the Indigenous Peoples of the North in National and International Organizations’ Maria Suojanen, ‘Minority Party Success in Heterogeneous European Countries’ Nazem Tahvilzadeh, ‘Ethnic Minority Representation in the Swedish Local Public Administration – Do Ethnic Minority Administrators Perceive Themselves as Representatives of an Ethnic Group or Interest?’

Preface   xix We are delighted that a number of these workshop participants (Bergh, Michon, Togeby and Tahvilzadeh), as well as several other authors, have contributed to the collaborative chapters on immigrant and minority turnout and party choice. Finally, it is with tremendous sadness that we report the untimely passing of one of this volume’s contributors. A pioneer in the study of the political participation among immigrants in Europe, Lise Togeby passed away on 15 October 2008. Lise was an active participant in the Helsinki workshop, and this volume owes much to her incisive and stimulating comments. Already suffering from a terminal illness, Lise contributed fully to the collaborative chapter on voter turnout. We would like to dedicate this book to her memory.


This volume is the product of a workshop held at the ECPR Joint Sessions of Workshops in Helsinki in May 2007. We would therefore like to acknowledge the assistance and support of the ECPR in bringing us together and providing a forum for our deliberations. In particular we would like to thank Sandra Thompson, the Joint Sessions Coordinator, and Clare Dekker, the Administrative Director, for organizing the Joint Sessions with their usual efficiency. We would also wish to thank our local hosts at the University of Helsinki for providing fine facilities and ensuring that our stay was very productive. All workshop participants left Helsinki with fond memories of the University and the Finnish capital itself. We would like to acknowledge the support of Thomas Poguntke, the editor of the ECPR/Routledge series, for his encouragement and advice in preparing the original proposal. Two reviewers offered excellent comments and suggestions, which improved the volume significantly. We would also like to thank editorial staff at Routledge for guiding us through to final publication. We owe thanks to Jessica Merolli at McMaster University and to Constanze Schmitz of the MZES for assisting us with editing the book and preparing it for submission. Finally, we are indebted to the Social Sciences and Humanities Research Council of Canada International Opportunities Fund, and to the Volkswagen Foundation, whose financial support has helped to sustain our research collaboration. We, along with the contributing authors, remain responsible for any errors that the book may still contain. Karen Bird, Thomas Saalfeld and Andreas M. Wüst

1 Ethnic diversity, political participation and representation A theoretical framework Karen Bird, Thomas Saalfeld and Andreas M. Wüst Over the past two decades, states have grown increasingly ethnically diverse. This has been a result of mass migration, of complex shifts in cultural identification, and of changes to state borders. In the case of the EU, we have also seen the emergence of a new regional form of sovereignty that has created both new citizens and new minorities. To be clear, this recent increase in ethnic diversity is related not only to movement across borders. Most countries in Europe trace their diversification back several decades, to the period between 1945 and 1973, when developments of mass industrialization and decolonization brought growing numbers of (initially male) foreign workers, followed by their family members.1 In Britain, for example, the broadest wave of non-­white immigration arrived between 1951 and 1971 from commonwealth colonies in the Caribbean, the Indian subcontinent and Africa. In France, large-­scale post-­colonial immigration arrived in the 1960s and early 1970s, especially from Algeria, Morocco, Tunisa, as well as from former West African colonies of Senegal, Mali and Mauritania. In the Netherlands, the main flows came between 1945 and the early 1960s from the former Dutch East Indies (now Indonesia) and, from 1965 until its independence in 1975, from the Dutch Caribbean territory of Surinam. By 1973, most Western European countries had closed their doors to new migration, though family reunification continued. Scandinavian countries are somewhat distinctive, having experienced diversification largely through liberal policies of refugee acceptance in the 1980s and 1990s. For most countries in Europe, then, the very recent period of socio-­demographic change has been propelled not by migration directly, but rather reflects a second-­stage settlement and incorporation process. This process involves the development of community structures and ethnic consciousness, and the political emergence of second and third generations. As Castles and Miller note (1998: 79), ‘By the 1980s, colonial migrants and their descendants had become clearly visible social groups.’ The story is somewhat different in the former ‘settler societies’ such as Australia, Canada and the United States. Historically, immigration policies in these countries were linked to ‘nation-­building’ strategies, and, while policies were restrictive and often racially defined, it was usually understood and accepted that immigrants would come to stay. This meant that naturalization policies were

2   K. Bird et al. typically liberal, and that immigrants were incorporated rapidly into the national fabric. Diversity was thus present from the start in these countries, though, as in Europe, the scope of that diversity expanded dramatically in the post-­war period. In these countries, the shift towards large-­scale non-­white immigration was prompted by an era of civil rights advancements in the mid-­1960s.2 Canada notably eliminated race-­based distinctions in immigration policy between 1965 and 1967, and introduced a points system for the recruitment of skilled immigrants. This was followed by the introduction of a legislative framework for official multiculturalism in 1971. Australia had eliminated the last vestiges of its racist ‘White Australia’ policy by 1973, and introduced its own official multicultural policy in 1978. In addition, in the U.S., 1965 amendments to the Immigration and Nationality Act that removed the discriminatory national-­origins quota system were part and parcel of broad civil rights legislation during that period. These civil rights developments had two consequences for diversity in these countries. First, they led to rapid growth in non-­European sources of migration. For example, the ‘visible minority’ share of the population of Canada grew from 5 to over 16 per cent between 1981 and 2006. Second, they ushered in a new normative disposition towards diversity, encouraging groups to express and celebrate their cultural identity. This normative shift helped create a more welcoming environment for diverse newcomers, and also influenced the identity consciousness of those minorities whose ancestors had arrived many generations before. Part of the change we are witnessing today thus comes not from migration directly but from the growing salience or ‘politicization’ of ethnicity. By this we mean primarily the emergence of ethnic minorities as political actors – as community activists, voters, candidates and, increasingly, as elected representatives – who at least to some extent express their ethnicity through their political projects. We have noted, for example, a broad trajectory towards increasing representation of minorities in local and national government across many liberal democracies (Anwar 2001; Bergh and Bjorklund 2003; Bird 2005; Black and Hicks 2006; Donovan 2007; Messina 2007; Norris and Lovenduski 1995; Saggar 2000; Studlar and Welch 1992; Togeby 2008; Wüst 2006). The causes of this are as yet unclear, and there appear to be important differences across countries in the timing of these developments, in channels of access for minority citizens to become candidates, in the policy focus and outcomes of minority representation, and in the nature of connections between minority representatives and minority communities. Nevertheless, there appears to be something driving these changes. We believe that changing patterns of electoral turnout and party choice among ethnic minority voters may be a key underlying factor. At a very basic level, increases in ethnic minority representation may reflect growing political engagement of minority citizens, and increasing responsiveness to these potential voters among political parties and candidates. Yet the ‘politicization’ of ethnicity can also refer to intensifying public debates about the management of ethnic diversity. In this latter vein, we have seen rising concern about states’ capacity to incorporate pluralism and maintain

Diversity, participation and representation   3 social cohesion, such that all citizens feel a sense of trust and belonging and a willingness to engage in a system of mutual governance. Most notably, Robert Putnam (2007) has worried about the role of diversity in eroding civic ties and the bonds of trust that hold democratic societies together. In Europe, David Goodhart (2004) has lamented that multiculturalism has worn thin the sense of collective identity and reciprocity that undergirds the welfare state. And Samuel Huntington (2004: 32) has raised even broader concerns about deepening cultural divisions in the context of ‘the popularity in intellectual and political circles of the doctrines of multiculturalism and diversity’. Partly as a result of these debates, many countries in Europe are now retreating from multiculturalism. Indicative of this trend, the Council of Europe in its 2008 White Paper on Intercultural Dialogue noted that a major concern emerging from its consultation with member states was that ‘old approaches to the management of cultural diversity were no longer adequate’. In particular, it found a predominant belief among European states that ‘what had until recently been a preferred policy approach, conveyed in shorthand as ‘multiculturalism’, had been found inadequate’ (Council of Europe 2008: 5). We take these two developments – an apparent increase in minority participation and representation in democratic politics, but also growing concern about ‘failed’ policies of multiculturalism – as a puzzle that guides this book. To lay bare our hypothesis from the start, we suspect that political participation and political representation of minorities has less to do with formal policies of multiculturalism than is often assumed. While the effects of multicultural policy in terms of social and economic integration of minorities are now highly debated (e.g. Koopmans 2008; Kymlicka 1995, 1998), there has been a fair degree of scholarly consensus that multiculturalism has a positive impact on political integration. In an influential study based on paired comparisons of ethnic groups in Canada and the US, Irene Bloemraad argues that Canada’s multicultural policies produce higher levels of citizenship acquisition, political participation and electoral representation. She concludes: ‘the results seem clear: on average Canada has been doing better than the United States, in part because multiculturalism provides the symbolic and material resources needed to take out and exercise political membership’ (Bloemraad 2006: 236). The main thesis of her book is that the political incorporation of ethnic minorities and immigrants occurs through a process of ‘structured mobilization’, and that this is facilitated where governments provide symbolic and material support, notably through policies such as multiculturalism and newcomer settlement services. Bloemraad makes a fine case for this theory with respect to rates of naturalization and community organization. Yet when it comes to formal political participation and, most specifically, electoral representation, her findings seem doubtful. We do not dispute that multicultural policies may play some role. However, we suspect that this is but a small part in a complex system of factors, many of which have little or nothing to do with multiculturalism – for example, candidate nomination procedures, electoral rules, party systems, and campaign financing. Indeed, by Bloemraad’s own account, these latter factors appear at

4   K. Bird et al. least as important in explaining differences in minority representation between Canada and the United States (Bloemraad 2006: 202–14). Furthermore, it is possible that multicultural policies may have no effect at all on minority political representation in the absence of critical political structures, such as a competitive party system that serves to mobilize ethnic minority voters. This appears to be the case for municipal elections in Canada (Bird 2004). Equally, it may be possible to obtain significant minority electoral participation and representation even where state policies are highly antagonistic toward immigration and multiculturalism – as appears to be true in Denmark (Togeby 2008). Our purpose then is to explore closely, and in comparative fashion, the formal inclusion of immigrants and ethnic minorities in democratic politics. Specifically, we wish to understand the factors that explain differences in political participation and voter choice, and in the descriptive and substantive representation of immigrant and ethnic minorities both across and within states. We believe that this research agenda is an important one, especially in the current context of retreat from multiculturalism. For if it is the case that there are mechanisms other than multicultural policy that serve to enhance or reduce minority political incorporation, then it is crucial that we identify these and point to constructive adjustments that should be made.

Minority representation: from normative theory to political outcomes From the perspective of politics, attention has focused on how to ensure adequate participation of ethnic minorities in public life and political institutions, to make sure that their interests are included and represented in the legislative process. The notion of representation is particularly important because, in contemporary liberal democracies, political decisions ‘are made by elected representatives and implemented by appointed officials to whom the representatives delegate some of the tasks of governing. The representatives decide what citizens must and cannot do, and they coerce citizens to comply with their decisions’ (Przeworski et al. 1999: 1). If there is a growing segment of the population that neither participates electorally nor has access to elected office, there is an increasing risk that the interests of such persons are not part of the electoral preference aggregation mechanisms of modern liberal democracies. If this is the case, the process of democratic representation is incomplete and the legitimacy of public policy is in doubt. Furthermore, where representatives appear to be ‘persistently at odds with the wishes of the represented without good reason in terms of their interest’ (Pitkin 1967: 209), there is considerable potential for political alienation and civic unrest. Such a breakdown in the process of democratic representation appears to have been one factor in the large-­scale destructive rioting by second- and third-­generation descendants of immigrants that shook France in November 2005 (Bowen 2006). It is difficult to underestimate the importance of fair inclusion of ethnic diversity within the representative process; but what is ‘fair inclusion’ in

Diversity, participation and representation   5 representative democracy? This question has significant implications for the normative basis, legitimacy and institutional performance of modern democracy. It also has consequences for political behaviour both within the majority population and among immigrants and ethnic minorities. The normative arguments addressing the need for enhanced ‘descriptive representation’ in parliamentary assemblies are based on the premise that the constant and systematic exclusion of a group from elected office tends to signal that group’s exclusion from full membership in the political community. Remedial measures to enhance group representation within the legislature are thought to be important for at least three distinctive reasons. First, group representation is of important symbolic value, signalling to both dominant and marginalized groups that its perspectives matter and rightfully belong in public debate. Representation may also provide an important point of access for marginalized groups, providing them with a less intimidating channel through which to engage with elected representatives, and to let their preferences be known on a quotidian basis, outside of periodic elections. Finally, descriptive representation is thought to be valuable insofar as it facilitates the introduction of new perspectives and a broader range of reasons to democratic debate. Despite these arguments, the premise of descriptive or symbolic representation raises thorny questions as to whether elected representatives of marginalized groups really can or will speak for a community of interest. Groups are clearly not homogenous entities in which members all think identically. Moreover, the very focus on the fact that a representative shares particular traits with his or her constituency may overshadow the important matter of one’s policy views. Hannah Pitkin (1967: 209–10) most notably has underscored that active representation in the sense of ‘action for’ and ‘being responsive toward’ constituents catches the deeper meaning of representation, and that this may be undermined by the focus on descriptive or mirror representation. In addition, there is a danger that some mechanisms to enhance the political representation of minority groups may have the secondary and untoward effect of depressing turnout and undermining the accountability of elected members to citizens. Anne Phillips (1995) has challenged Pitkin’s critical views on the descriptive and symbolic dimensions of political representation. Phillips disagrees with Pitkin’s assertion that making representation even partially dependent on personal or group characteristics inevitably undermines the basis for political accountability. She makes a strong and ambitious argument for a ‘politics of presence’, where improvements in the descriptive representation of excluded groups improve representation, if they are accompanied by sustained conditions for more deliberative and consultative processes. These supplementary conditions alter the balance between citizens and representatives and shift democracy toward a more participatory model (Phillips 1995: 187–91). Phillips argues, along with Jane Mansbridge (1999) and Melissa Williams (1998), that the parliamentary presence of groups such as women and ethnic minorities enhances the representativeness of an assembly both symbolically and substantively.

6   K. Bird et al. For our purposes, these normative debates are useful insofar as they tend to map on to distinctive citizenship regimes across states. For example, the French republican regime formally rejects the notion of group identity and descriptive representation. The role of a French legislator is not to represent the distinctive identity of any group or geographical constituency. Rather it is the member’s role to represent the nation in its collective interest. On the other hand, the notion of descriptive representation is relatively consistent with liberal pluralist regimes in the United States and Canada. Commentators have also pointed out that federal systems operate on a principle of territorial or geographic identity that is little different from ethnic identity (Kymlicka 1998). This volume considers differences in the descriptive representation of immigrants and ethnic minorities across distinctive citizenship regimes. However, we also push beyond this question, to assess whether the nature of representation varies depending on concrete mechanisms of the political structure. Specifically, we ask whether beyond and perhaps even in contrast to the normative dimensions of a citizenship regime, there are structural dynamics and incentives that affect both the statistical side of descriptive representation (the number of minorities elected) and the substantive side (what a representative does, and who he or she speaks for). One of the key concerns shared by many contributors to this volume is the question of whether the descriptive presence or absence of minority groups in parliaments and other forums of democratic representation actually makes a difference in policy terms. After all, their interests may be represented perfectly well by responsive majority legislators. How effective are the seemingly individualistic and ‘colour-­blind’ mechanisms of preference aggregation, delegation and accountability in representing the interests of immigrants and ethnic minorities? Reviewing the existing literature on ethnic minority representation across mature democracies, it seems fair to conclude that there are still broad gaps in our knowledge on this question. Melissa Williams (1998) found that decision-­ making in the US Congress was affected by the presence of minorities. There is some evidence that African-­American Members of Congress tend to be more responsive than other Members of Congress to demands from African-­American citizens (Fenno 2003; Gay 2002, 2007; Lublin 1997). In Australia, too, minority MPs were found to be more responsive to minority groups, but only on the political Left (Zappalà 1997). Nixon (1998), by contrast, did not find any confirmation for such claims in a study of the United Kingdom’s House of Commons. Even where the representational behaviour and policy positions of elected ethnic minorities are not significantly different from the behaviour of ethnic majority members, there may still be a significant effect on the attitudes and political integration of minority citizens, especially with regard to civic trust, efficacy and participation. If a stronger presence of minority representatives leads to the increased political integration of minority groups, this might in turn produce diffuse changes in policy outcomes, as well as improvements in system legitimacy. A number of studies in the US confirm that members of minority ethnic groups tend to express more trust in parliaments, politicians and policies if they are represented by MPs belonging to their own ethnic group (Abney and

Diversity, participation and representation   7 Hutchinson 1981; Gay 2002). In one of the few cross-­national studies of these effects, Banducci and colleagues (2004) have shown that there are positive ‘civic empowerment effects’ both among blacks in the US and Maori in New Zealand who are represented by a member of their own ethnic group, as compared to those with non-­minority representatives. Yet the precise nature of these effects differs across countries, likely due to the particular mechanisms employed to enhance descriptive representation – that is, minority-­majority districts in the US, and separate Maori electorates in New Zealand. Again, our argument is that there has not been nearly enough focus on the complex features of political systems. Examining these features across countries, with special attention given to the implications for the political incorporation of ethnic minority citizens, should help us to better address these questions.

Conceptual issues and the scope of comparison At the moment we know little about the processes of ethnic minority participation in public life, and how these vary across states. Part of the puzzle is that countries count ‘minorities’ differently (Kertzer and Arel 2002). In many countries throughout Europe, objective distinctions are made based on a person’s place of birth, their parents’ and their grandparents’ place of birth. In France, there is an official rejection of the concept of ‘ethnic minorities’, though this has come under increasing criticism (Amiraux and Simon 2006). At the other extreme, countries such as the US and Canada have evolved complex measures of ethnic and racial self-­identification. For example, in Canada, the measure of ‘ethnic origin’ refers to the ethnic or cultural origins of the respondent’s ancestors. In the census, respondents are told ‘an ancestor is usually more distant than a grandparent’ and are asked to specify, in an open-­ended format, as many ethnic origins as applicable. Ethnic origin responses are thus a reflection of each respondent’s perception of their ethnic ancestry, and, consequently, the measurement of ethnicity is affected by changes in the social environment in which the question is asked and changes in the respondent’s understanding or views about the topic (Statistics Canada 2006). More significant still are the differing characteristics of people who might be classified as indigenous, racialized, national minorities, or newly arrived. Each group has different political interests and faces different structural obstacles (or indeed opportunities) with respect to political participation and representation. These include, to list just a few, different degrees of ‘ethnic’ consciousness and capacity for collective mobilization, variations in access to citizenship and voting rights, different degrees of social integration, levels of poverty, official language skills, and so on. Despite these differences, ethnic minorities usually want to play a full part in the political life of their country. There are some exceptions. For example, the Amish in North America are reluctant to integrate socially and politically. In addition, some members of indigenous communities and national minorities may reject political representation within existing institutions of a state that they view as imperialistic and coercive. But even those

8   K. Bird et al. groups that seek a degree of political autonomy or self-­government in some areas also wish to have a say in issues of shared concern.3 This range of difference among ‘ethnic minorities’ and in the arrangements that might be made to ensure fair representation in public life brings us to a central methodological problem of comparative politics, one that we find necessary to address at the outset. The problem concerns the appropriate scope of comparison. As Donald Horowitz (1985: 16) cautions in his seminal work on ethnically divided societies, ‘the problem of comparability in cross-­national studies of ethnic conflict is likely to be substantial. . . . The class of cases needs careful delimitation, for the simple reason that comparison requires comparability.’ We heed this warning, in large part. We have chosen to limit our comparison in two respects. First, we look exclusively at advanced liberal democracies, and not at emerging democracies, post-­colonial or post-­communist societies, where the political exclusion of ethnic minorities tends to be part of a more fundamental problem rooted in the absence of state consolidation and democratic institutions. We recognize that critical scholars raise important questions about the commitment to democratic principles and to racial equality in countries that are ostensibly ‘advanced democracies’ (e.g. Day 2000; Delgado and Stafancic 2000). Nevertheless, we believe that differences in state capacity, state–society relations, economic development, and the entrenchment and/or implementation of human rights and core democratic values sets advanced liberal democracies apart from other societies with respect to minority participation and representation in politics. Second, we exclude from our study national minorities and indigenous groups. These groups have tended to articulate different kinds of remedies for historic exclusion. Namely, they seek institutional mechanisms of self-­ government. This is a conceptually distinctive form of political representation. As Will Kymlicka (1995: 142–4) explains, self-­government entails a transfer of power from one government body to another, and consequently reduced influence for the national minority on some issues in the former legislative arena.4 This vastly complicates any empirical comparison of ethnic minority representation. Kymlicka has made a strong normative argument for a ‘targeted’ or group-­ differentiated approach to ethnic minority rights, in which indigenous, national and immigrant-­origin minorities are treated distinctively (2007: 78–9). While we acknowledge some quibbles with Kymlicka’s strict conceptual distinction among types of minorities (Brock 2005, Packer 2005), we do agree that for the purposes of conducting sound comparative work it is more helpful to make these distinctions. Overall, by limiting the scope of comparison in these ways, we hope to avoid the risks contemplated by Horowtiz (1985: 16–17) of ‘including like and unlike cases, entangling the analysis in unnecessary complexity, and in the end probably diluting the power of the explanations that emerge’. We seek comparability, and have therefore asked each of the contributors to this volume to address the analytic model that we present below. At the same time, we have been wary of overly strict similarity in our country coverage. Many of the existing works on minority political integration are subject to this

Diversity, participation and representation   9 problem. These include single case studies that provide detailed analysis of the processes of minority representation in one country, but little leverage in understanding the role of a range of political-­institutional factors including electoral rules or parliamentary forms. There are also a number of narrowly comparative studies, usually limited to a pair of countries with similar migration experiences and ethnic histories. This approach tends to limit the field of comparison to a relatively small circle of countries, usually drawn from Western Europe. While such comparisons reveal a great deal about political system characteristics and their effects, we learn much less about the patterns of political integration across different groups of ethnic minorities. The failure to include more distinctive countries within this comparative framework has thus constrained our ability to identify and specify the effects of a broad range of contextual variables. Our intent then is to occupy a middle ground in terms of comparative scope. We have delimited the comparison in certain ways, but we have also deliberately broadened it beyond the most common ‘similar systems’ comparisons. Notably, we have included cases across Europe and North America. In so doing, we have opened up the conceptual framework to include actors who are typically described as ‘immigrants’ or of ‘immigrant-­origin’, as well as ‘ethnic’ and ‘visible’ minorities whose identity may be largely, or even entirely, unrelated to the migration process. In this respect, we heed the advice of Adcock and Collier (2001: 535), who argue that the effect of allowing the empirical domain to which a concept is applied to vary is to take a ‘productive step toward establishing equivalence among diverse contexts’. We believe that there is much room for conceptual and theoretical improvement in our understanding of minority representation across diverse contexts. Casellas and Leal (Chapter 8) point to some of the conceptual complexities of ethnic identity in the case of Latinos in the United States. They suggest that ‘the category is a somewhat artificial construct’ that exists only in the US. Latino identity has been shaped partly by migration and partly by territorial annexation, but it has also been influenced by the distinctive paradigm of racial politics in the US. We take this cue as a reminder that ethnicity is socially and politically constructed in different ways across different countries. From our perspective, ethnic minority political participation and representation play a central part in this process of identity construction. However, as we emphasize, this takes place within a particular politico-­structural context.

Analytical perspectives Are there differences in the political behaviour (participation and party choice) of ethnic minorities compared to non-­minorities? What are the channels by which ethnic minorities become political candidates? What are their campaign strategies? Do they cast their appeal to minority voters, or more broadly? What do they do when elected, both in terms of parliamentary and extra-­parliamentary behaviour? And what is the dynamic situation with respect to these questions: are minority voters, candidates and representatives becoming more distinctive, or are their patterns of behaviour converging with those of non-­minorities?

10   K. Bird et al. While it is not possible to provide an exhaustive survey of explanations (see discussions in Freedman 2000; Ireland 1994), many older accounts have focused on ethnic, class-­based and institutional approaches to explaining ethnic minority participation. In recent years, work based on notions of social capital and political opportunity structures has become increasingly influential in the literature. Although these approaches are mainly concerned with the quality of political participation (rather than with the dynamics of democratic representation), they provide important insights into likely factors influencing the process of representation. Ethnic approaches to understanding minority political participation and vote choice take certain ethnic, religious or cultural characteristics as fundamental independent variables. Immigrants and members of ethnic minorities are thought to ‘organize and articulate their political interests along ethnic, racial, or religious lines. Each group’s own distinctive mode of political participation develops from group socialization processes and in response to discrimination and interaction with other groups’ (Ireland 2000: 235). The political participation of immigrants is thought to be shaped by networks of kinship associations, specific ‘political cultures’, culturally determined notions of legitimacy and authority, and the need for the maintenance of ethnic boundaries. Among the most influential work in this vein is that of Michael Dawson, who argues that the dominant role of race in black political behaviour in the US has its roots in a ‘black utility heuristic’. The heuristic ‘simply states that as long as African-­Americans’ life chances are powerfully shaped by race, it is efficient for individual African-­ Americans to use their perceptions of the interests of African-­Americans as a group as a proxy for their own interests’ (Dawson 1994: 61). The black utility heuristic thus accounts for African-­Americans’ continuously high support for the Democratic Party, despite the rapid expansion over the past three decades of the black middle class. Brouard and Tiberj (2005) have applied the racial utility heuristic to explain exceptionally high levels of support for the French Socialist Party among Maghrebin immigrants and their descendants. Tillie (1998) in his work on migrant voting in the Netherlands, Wüst (2002) in his study of Germany, and Togeby (1999) in her work on migrant groups in Demark, have also found that there are highly distinctive patterns of political participation and vote choice across ethnic groups within a single country or even city. Despite these findings, ethnic approaches to political behaviour are often criticized, mainly because they fail to explain variation in political participation within certain ethnic communities, and among members of the same ethnic groups in different societies of settlement (e.g. Freedman 2000; Garbaye 2005; Ireland 1994, 2000). Class-­based approaches focus on an ethnic group’s overall socio-­economic status as the key independent variable in explaining participation and mobilization. With their frequently low position in the labour markets of their countries of settlement, ethnicity and race have ‘become the modalities in which migrants experience class relations in Western Europe’ (Ireland 2000: 234). Political participation, vote choice and levels of representation among ethnic minorities are

Diversity, participation and representation   11 expected, from this perspective, to be similar to patterns among non-­minorities in a similar socio-­economic position. This approach has been used to explain the strong and consistent support which left-­wing and centre-­left parties have enjoyed among immigrant and visible minorities. It is also thought to account for the relatively low levels of interest in politics and electoral participation among immigrants (e.g. Diehl and Urbahn 1999). However, class-­based voting and the extent to which party systems are organized along class lines varies widely by country, and so it is not clear that this approach is universally applic­ able. Furthermore, the linkage between ethnicity and class will depend partly upon the immigration mix that a country achieves among skilled and unskilled workers, family dependants and asylees. For example, in countries that focus largely on the recruitment of skilled workers and business class immigrants, the belief in upward mobility may be very strong, and ethnicity may become effectively de-­linked from class relations. Recently, there has been much scholarly interest in the relationship between diversity and social capital. Going back to Alexis de Tocqueville’s Democracy in America, there is a broad perception that democratic institutions and processes are kept alive by a rich associational life (Putnam 1993), but there is growing concern about the effects of ethnic diversity and ethnic identity for social capital and civic participation. One hypothesis is that belonging to ethnic associations, even inherently undemocratic ones, allows individuals to learn civic behaviour as they ‘solve the dilemmas inherent in collective action’ (Fennema and Tillie 1999: 723). The alternative hypothesis is that ethnic diversity and ethnic forms of belonging produce social fragmentation, leading populations to ‘hunker down’, and eroding the generalized trust that is a necessary component of civic-­ mindedness (Berman 1997; Putnam 2007). Theorists have sought to address these contradictory hypotheses by distinguishing between exclusive or so-­called ‘bonding organizations’ such as extremist political or sectarian religious groups, and organizations with a broader social base. Bonding organizations tend to reinforce social differences and divisions, whereas ‘bridging associations’ bring together people from different social and ethnic backgrounds and help transcend social divisions. In a similar vein, some social movement theorists argue that strong social (including ethnic) ties are a crucial source of both capacity for collective mobilization and political identity (McAdam 1982; McAdam et al. 2001) and for confidence in political elites. Yet others emphasize the importance of weak ties as a precondition of elite integration and essential prerequisite for democracy (Huckfeldt and Sprague 1995; La Due and Huckfeldt 1998). The question of whether ethnic diversity erodes social capital and political participation is far from settled, and much more research remains to be done (APSA-­CP 2009). Certainly, one could argue that there is a striking mismatch between the rich associational life of minorities in some European countries and their under-­representation in elected bodies. One example is the impressive associative movement of Maghrebin-­origin citizens in France that began with the ‘marche de Beurs’ in 1983. Despite the evident skills of the movement leaders, and the significant size and demographic concentration of this potential

12   K. Bird et al. electorate, these ethnic minorities remained fundamentally excluded from municipal and national politics for at least 25 years (Garbaye 2005; Geisser 1997; Masclet 2003).5 Meanwhile, in Canada, ethnic communities appear to have been a crucial site for political mobilization and advocacy, as well as an important wellspring for political candidates (Bloemraad 2005, 2007). It remains unclear whether the political elites who might emerge from these ethnic networks are a crucial source of system support within ethnic communities. In some cases they may be viewed by their community as representing neither an ethnic constituency nor an agenda explicitly related to ethnic disadvantage, but rather as symbols and instruments of a process of co-­optation by political parties. In short, we harbour doubts that social capital and social movement theories alone can explain differences in the quantity and quality of minority representation. In contrast to the literature on social capital, institutional approaches focus predominantly on citizenship rules, immigration and anti-­discrimination policies in the countries of settlement that either constrain or enable political participation. It is therefore not surprising that they have become very influential in the literature on immigration in comparative politics. Rogers Brubaker (1992) argued that different models of nationhood and citizenship have a long-­ lasting impact on immigration policies and migrant incorporation. Therefore, opportunities for immigrants and minorities to participate and to get represented are limited by nation-­specific traditions (Bleich 2003; Freeman 2004). Recent changes of citizenship laws in countries such as Germany or the United Kingdom, as well as the emergence of new European paradigms in immigration and anti-­discrimination policy, challenge such historical approaches, which rely on versions of ‘path dependence’ as explanations of long-­standing, institutionalized restrictions for immigrants. A number of authors have questioned these institutional approaches by pointing to significant immigration and anti-­discrimination policy convergence, for example, between Germany and the UK (Green 2007), and between Britain and France (Geddes and Guiraudon 2004). Beyond the formal institutional structure, scholars sometimes include more informal constraints often referred to as ‘political opportunity structure’. Sidney Tarrow (1998: 19–20), one of the main proponents of the political opportunity approach in social movement research, defines political opportunities as ‘consistent – but not necessarily formal, permanent, or national – dimensions of the political struggle that encourage people to engage in contentious politics’. These factors, which are external to the resources of the respective group, should, Tarrow (1998: 20) suggests, be understood ‘as a set of clues for when contentious politics will emerge, setting in motion a chain of causation that may ultimately lead to sustained interaction with authorities and thence to social movements’. In the study of the political participation of immigrants, the concept has been applied more generally to all forms of political participation, referring to constraints determining the opportunities for, and costs of, participation. The framework developed in this chapter will borrow significantly from scholarship on political opportunity structures.

Diversity, participation and representation   13

Political opportunities and strategies The chapters in this book are organized around a framework for the study of minority political representation that integrates and extends the scope of the more structuralist approaches summarized above (e.g. Bird 2005). We draw strongly on the concept of political opportunity structures, denoting the degree of openness or accessibility of a given political system for movement initiators. On the other hand, we seek to overcome the restrictive nature of the political opportunity structure approach (which has been criticized for its state-­centric bias) by extending the focus to the strategies used by ethnic minority political actors. In this model (Figure 1.1), the political opportunity structure for ethnic minority participation and representation is the theoretical core. At the most general level, the model consists of (1) the collective identity and capacity for political mobilization of a given minority group, and (2) the party system’s and political Micro-context


Size and spatial concentration of ethnic group

Access to legal citizenship rights

Communication and leadership

Political opportunity structure Collective identity and capacity to mobilize

Length of settlement in community

Cultural rights of citizenship

Social and economic rights

Responsiveness of political system

Historical migration/racial regime

Institutions and social networks

Party competition

Cost of campaigns

Party selection rules

Electoral rules

Openness of recruitment

Meso-context Strategic calculus of visible minority candidates

Figure 1.1  Political opportunities and strategies for ethnic minority representation.

14   K. Bird et al. system’s responsiveness to such mobilization. The former depends on a number of macro- and micro-­level variables listed on the top right and left of Figure 1.1. The latter is largely a function of the meso-­level factors listed at the bottom of Figure 1.1. Our model also emphasizes that the nature of the opportunity structure shapes the strategic calculus of minority candidates in the representative process, and that these strategies in turn influence the political opportunity structure (especially the collective identity and capacity to mobilize). Long-­term causal influences on the political participation of immigrants and visible minorities can be condensed into three general factors: citizenship regimes, interest constellations, and institutions. A country’s citizenship regime includes its rules of access to citizenship (legal rights), and the cultural rights of citizenship (cultural assimilation or cultural pluralism). In countries where immigrants enjoy easy access to citizenship and voting rights, and where ethnic minorities are recognized as possessing a distinct culture and set of interests, they should be more likely to mobilize and achieve political representation as a group. Countries that are officially multicultural may even have formal measures in place to promote ethnic minority participation and representation in politics. This is the case, for example, in the province of Nova Scotia, Canada (for French-­speaking Acadians and blacks). Yet patterns of political mobilization and representation vary significantly for different groups, even when they are subject to the same formal rules and rights of citizenship. Therefore, the extent to which minorities enjoy equal social and economic rights are relevant contextual factors. Where ethnic minorities face formal and informal barriers, where larger portions of them are segmented within the labour force and the housing market, where they lack a promising educational degree, they are therefore less likely to have the resources necessary to achieve political representation. Finally, the historical migration and/or racial regime is of potential relevance for the degree of political integration. For example, in post-­colonial societies, where minorities were created through a historic sequence of typically violent territorial annexation and de-­colonization, punctuated by migration both to and from the colony, relations between the ethnic majority and the minorities tend to be fraught with power inequality, mutual distrust and hostility. In such cases, minorities may be subject to old colonial stereotypes, and viewed by the majority population and/or party leaders as less qualified to participate in the task of government. In guest-­worker societies, foreign workers and their descendants may be subject to similar patterns of discrimination and political exclusion. Ethnic groups may also differ widely in their political interest and engagement. These differences may spring from the degree of (dis)similarity between the political cultures of the sending and receiving societies, from a group’s length of settlement in a community, or from economic factors that affect a group’s political resources. The potential for political mobilization of an ethnic group is also related to its size and spatial density, the completeness of its social institutions, and resources such as communication networks, ethnic media and leadership (Howe 2007; McAdam 1982). For example, ethnic groups may do especially well in terms

Diversity, participation and representation   15 of representation if their spatial location corresponds with electoral boundaries, if they can be mobilized to vote as a bloc, and if they are located in a competitive constituency where they can deliver seats for one party at the expense of another. Collective mobilization of ethnic minorities is well and good, but its effects in terms of representation will be muted if the political system is not responsive to such action. The responsiveness of the political system to ethnic mobilization is determined by a number of factors. Countries with a more open and participatory democratic culture may be more likely to draw upon the leadership resources of ethnic communities, while those with more rigid political elites may be resistant to promoting political outsiders as candidates. Factors such as strong party competition, a high degree of legislative turnover and public funding for political campaigns may also make political systems more or less open to ethnic minority candidates. Of course, political parties play a crucial role in candidate recruitment, and thus in the social composition of parliamentary elites (Kittilson and Tate 2004; Norris and Lovenduski 1995). It is especially crucial to consider the strategic positions which parties assume vis-­à-vis one another, as they decide either to compete for the minority/immigrant vote, or to capitalize on potential xenophobic attitudes among non-­minority voters. In addition to exogenous factors affecting the propensity to elect visible minority representatives, we emphasize the role of individual candidates and their strategic calculus. Visible minority candidates may use ethnicity in a selective and entrepreneurial fashion. Indeed, impression management of a candidate’s ethnic identity may be a particularly important element of one’s political strategy. As conceptualized in Figure 1.1, candidates are expected to develop an electoral strategy based on the political opportunity structure within their country and/or local community. Especially if they choose to address themselves to the ethnic minority community, these candidates may in turn become a factor in the political identity and mobilization of those groups. However, ethnic candidacies are rarely pitched exclusively toward ethnic voters. One of the most interesting characteristics of successful visible minority candidates is the way in which they manage the multiple and sometimes contradictory demands of their political party, their own ethnic community and majority voters. Each of the studies in this volume draws on, or indeed re-­specifies, some of the components of our POS model. A number of contributors focus on the interaction between the structural and strategic influences, while others explore the explanatory potential of politico-­structural mechanisms that may be at work in a particular context.

Plan of the book This volume is organized into three main parts, examining immigrants and ethnic minorities as voters, candidates and political representatives, respectively. In Part I, immigrants and minorities are examined as voters, who must be the starting point of any analysis of political representation. Chapter 2 looks at levels of electoral participation among ethnic minorities across 11 countries. We asked

16   K. Bird et al. contributors to this chapter to provide ‘state-­of-the-­art’ knowledge concerning voter turnout among minorities compared to the non-­minority population in their country for elections at national and subnational levels, and to discuss the reasons for variations. Each of them has identified components from our political opportunities and strategies (POS) model that are particularly pertinent to explaining turnout in their case. Chapter 3 follows a similar structure in surveying the voting and party choices of immigrants and ethnic minorities, again bringing together and comparing evidence across countries. This is a particularly unique and important contribution of our volume, as we know of no other work that has provided a comparative perspective on minority voter participation and party choice across such a large number of countries. The information presented in these two chapters suggests a rich agenda for future cross-­country analysis in this area. Part II deals with the stage of candidate selection within political parties. Candidate nomination, often portrayed as the ‘secret garden of politics’ in liberal democracies (Gallagher and Marsh 1987), is a crucial stage in the process of political representation, but is largely under-­researched in the comparative study of ethnic minority representation. The selection process tends to be strategically complex, because its dynamics are shaped by electoral as well as organizational imperatives within parties. Four country-­based case studies are devoted to this topic. Sara Claro da Fonseca (Chapter 4) examines the extent to which candidates of immigrant origin have been able to mobilize the immigrant vote in German elections, and looks specifically at the potential for change in the electoral relationship between political parties and immigrants as a result of the liberalization of the German citizenship regime. Johannes Bergh and Tor Bjørklund (Chapter 5) focus on different electoral rules for local and national elections in Norway, and show how these impact upon the selection and placement – and consequently the electoral success – of ethnic minority candidates at these different levels. Maritta Soininen (Chapter 6) takes us even more deeply into the inner workings of the main party organizations in Sweden, documenting some of the mechanisms of ethnic exclusion that operate at the candidate nomination level. Sylvain Brouard and Vincent Tiberj (Chapter 7) apply an experimental methodology to assess whether French voters demonstrate prejudice against ethnic minority candidates and in favour of those with a more traditional social profile, or whether it is the mainstream party selectorates that act as the principal barrier to improved political representation for ethnic minorities in France. Finally, Part III examines the behaviour of immigrant and ethnic minorities once they have been elected to Parliament. All four contributions to this section break new ground in moving ‘beyond numbers’ to focus on the substantive nature of ethnic minority representation. Jason Casellas and David Leal (Chapter 8) examine the descriptive and substantive representation of Latinos and African Americans in the US Congress, testing whether representatives with larger minority constituencies are more likely to vote with minority interests in mind. Karen Bird (Chapter 9) assesses substantive representation of ethnic minority interests by looking at parliamentary speeches of visible minority and non-­ minority MPs in the Canadian House of Commons. Thomas Saalfeld and

Diversity, participation and representation   17 ­ alliopi Kyriakopoulou (Chapter 10) provide a novel analysis of both personal K website content and legislative behaviour of black and minority ethnic MPs compared to non-­minority MPs in the United Kingdom. Finally, Andreas Wüst (Chapter 11) takes stock of parliamentarians with a migratory background in German parliaments; he examines the overall composition and memberships in parliamentary committees and, for the Bundestag members, he also analyses their legislative behaviour. A brief conclusion draws together some of the main patterns observed across the book, and suggests possibilities for further comparative research in this area.

Notes 1 Of course some countries, including Britain, France and Germany, experienced large-­ scale labour immigration throughout the 1800s. However, migrants in this period came mainly from within Europe (e.g. from Ireland, Italy, Poland, the Austro-­ Hungarian Empire and Russia, and later from Spain and Portugal). Canada and the United States experienced large-­scale immigration from similar source countries somewhat later, with significant intake occurring in the late 1800s and early 1900s. Newcomers during this early period were certainly viewed as ethnically distinct, and were subject to a variety of racially discriminatory policies and practices. However, the scale of diversity became much broader after 1945, when countries reached beyond the European periphery to satisfy the needs of their rapidly expanding post-­ war economies. 2 Europe was not exempt from this emerging civil rights paradigm. Notably, British race policy was heavily influenced by these developments in the anglophone world (Bleich 2003). 3 In other words, self-­government is always a question of degree, determined to a great extent by the scope of the transfer of legislative competences from the larger entity to the (semi-)autonomous entity. 4 There are both territorial and non-­territorial forms of self-­government. Territorial self-­ government typically involves the devolution of authority over matters of culture, language, education and so on to a territorial parliament. Non-­territorial forms of self-­government allow an ethnic group autonomy in some area of jurisdiction, irrespective of the group’s size or historical geographic concentration. For example, a religious or linguistic minority might be granted autonomy in running their own schools throughout the state. We can see, from this example, that there remains some ‘conceptual fuzziness’ in our categorization of national/indigenous versus immigrant minorities. Namely, some immigrant origin ethnic groups may be viewed as enjoying a degree of self-­government, for example, in the running of religious schools. 5 Vincent Tiberj notes in Chapter 2.2 that there has been greater effort by some politicians to appeal to this electorate following the French race riots of 2005.

References Abney, Glenn F. and Hutchinson, John D. (1981) ‘Race, Representation and Trust: Changes in Attitudes after the Election of a Black Mayor’, Public Opinion Quarterly, 45(1): 91–101. Adcock, Robert and Collier, David (2001) ‘Measurement Validity: A Shared Standard for Qualitative and Quantitative Research’, American Political Science Review, 95(3): 529–46.

18   K. Bird et al. Amiraux, Valérie and Simon, Patrick (2006) ‘There Are No Minorities Here: Cultures of Scholarship and Public Debate on Immigrants and Integration in France’, International Journal of Comparative Sociology, 47: 191–215. Anwar, Muhammad (2001) ‘The Participation of Ethnic Minorities in British Politics’, Journal of Ethnic and Migration Studies, 27(3): 533–49. APSA-­CP (2009) ‘Ethnic Heterogeneity and Social Solidarity in Advanced Countries’, Symposium of the Organized Section in Comparative Politics of the American Political Science Association, 20(1): 8–21. Banducci, Susan A., Donvan, Todd and Karp, Jeffrey A. (2004) ‘Minority Representation, Empowerment and Participation’, Journal of Politics, 66(2): 534–56. Bergh, Johannes and Bjorklund, Tor (2003) ‘The Political Representation of Immigrants in Oslo and Copenhagen: A Study of Electoral Systems and Voting Behavior’, Comparative Social Research, 22: 101–21. Berman, Sheri (1997) ‘Civil Society and the Collapse of the Weimar Republic’, World Politics, 49(3): 401–29. Bird, Karen (2004) ‘Obstacles to Ethnic Minority Representation in Local Government in Canada’, Our Diverse Cities, 1, 182–6. —— (2005) ‘The Political Representation of Visible Minorities in Electoral Democracies: A Comparison of France, Denmark and Canada’, Nationalism and Ethnic Politics, 11(4): 425–65. Black, Jerome H. and Hicks, Bruce H. (2006) ‘Visible Minority Candidates in the 2004 Federal Election’, Canadian Parliamentary Review, summer: 26–31. Bleich, Erik (2003) Race Politics in Britain and France, Cambridge: Cambridge University Press. Bloemraad, Irene (2005) ‘The Limits of de Tocqueville: How Government Facilitates Organisational Capacity in Newcomer Communities’, Journal of Ethnic and Migration Studies, 31(5): 865–87. —— (2006) Becoming a Citizen: Incorporating Immigrants and Refugees in the United States and Canada, Berkeley: University of California Press. Bowen, John R. (2006) ‘France’s Revolt: Can the Republic Live Up to its Ideals?’, Boston Review, January/February: 29–32. Brock, Gillian (2005) ‘Can Kymlicka Help Us Mediate Cultural Claims?’, International Journal on Minority and Group Rights, 12: 269–96. Brouard, Sylvain and Tiberj, Vincent (2005) ‘Race, Class and Religion: The Political Alignments of the “French Muslims” ’, Working Paper, CEVIPOF, Sciences Po Paris. Brubaker, Rogers (1992) Citizenship and Nationhood in France and Germany, Cambridge, MA: Harvard University Press. Castles, Stephen and Miller, Mark J. (1998) The Age of Migration: International Population Movements in the Modern World, 2nd edn, New York: Guilford Press. Council of Europe (2008) White Paper on Intercultural Dialogue, Strasbourg: Council of Europe. Online: EN.pdf. Dawson, Michael (1994) Behind the Mule: Race and Class in African–American Politics, Princeton, NJ: Princeton University Press. Day, Richard J.F. (2000) Multiculturalism and the History of Canadian Diversity, Toronto: University of Toronto Press. Delgado, Richard and Stafancic, Jean (eds) (2000) Critical Race Theory: The Cutting Edge, Philadelphia, PA: Temple University Press.

Diversity, participation and representation   19 Diehl, Claudia and Urbahn, Julia, in cooperation with Hartmut Esser (1999) Die soziale und politische Partizipation von Zuwanderern in der Bundesrepublik Deutschland, 2nd edn, Bonn: Forschungsinstitut der Friedrich Ebert Stiftung. Donovan, Barbara (2007) ‘Minority Representation in Germany’, German Politics, 16(4): 455–80. Fennema, Meindart and Tillie, Jean (1999) ‘Political Participation and Political Trust in Amsterdam: Civic Communities and Ethnic Networks’, Journal of Ethnic and Migration Studies, 25(4): 703–26. Fenno, Richard F. (2003) Going Home: Black Representatives and Their Constituents, Chicago, IL: University of Chicago Press. Freedman, Amy L. (2000) Political Participation and Ethnic Minorities: Chinese Overseas in Malaysia, Indonesia, and the United States, London: Routledge. Freeman, Gary P. (2004) ‘Immigrant Incorporation in Western Democracies’, International Migration Review, 38(3): 945–69. Gallagher, Michael and Marsh, Michael (1987) Candidate Selection in Comparative Perspective: The Secret Garden of Politics, London: Sage. Garbaye, Romain (2005) Getting Into Local Power: The Politics of Ethnic Minorities in British and French Cities, Malden, MA: Blackwell. Gay, Claudine (2002) ‘Spirals of Trust? The Effect of Descriptive Representation on the Relationship between Citizens and their Government’, American Journal of Political Science, 46(4): 717–33. —— (2007) ‘Legislating Without Constraints: The Effect of Minority Districting on Legislators’ Responsiveness to Constituency Preferences’, Journal of Politics, 69(2): 442–56. Geddes, Andrew and Guiraudon, Virginie (2004) ‘Britain, France, and EU Anti-­ Discrimination Policy: The Emergence of an EU Paradigm’, West European Politics, 27(2): 334–53. Geisser, Vincent (1997) Ethnicité Républicaine. Les Elites d’Origine Maghrébine dans le Système Politique Français, Paris: Presses de Sciences Po. Goodhart, David (2004) ‘Too Diverse? Is Britain Becoming Too Diverse to Sustain the Mutual Obligations Behind a Good Society and the Welfare State?’, Prospect, 95 (February): 30–7. Green, Simon (2007) ‘Divergent Traditions, Converging Responses: Immigration and Integration Policy in the UK and Germany’, German Politics, 16(1): 95–115. Horowitz, Donald L. (1985) Ethnic Groups in Conflict, Berkeley: University of California Press. Howe, Paul (2007) ‘The Political Engagement of New Canadians: A Comparative Perspective’, in Keith Banting, Thomas J. Courchene and F. Leslie Seidle (eds) Belonging? Diversity, Recognition and Shared Citizenship in Canada, Montreal: Institute for Research on Public Policy. Huckfeldt, Robert and Sprague, John (1995) Citizens, Political, and Social Communication, Cambridge: Cambridge University Press. Huntington, Samuel P. (2004) Who Are We? The Challenges to America’s National Identity, New York: Simon & Schuster. Ireland, Patrick R. (1994) The Policy Challenge of Ethnic Diversity: Immigrant Politics in France and Switzerland, Cambridge, MA: Harvard University Press. Ireland, Patrick R. (2000) ‘Reaping What They Sow: Institutions and Immigrant Political Participation in Western Europe’, in Ruud Koopmans and Paul Statham (eds) Challenging Immigration and Ethnic Relations Politics, Oxford: Oxford University Press.

20   K. Bird et al. Kertzer, David I. and Arel, Dominique (eds) (2002) Census and Identity: The Politics of Race, Ethnicity and Language in National Censuses, Cambridge: Cambridge University Press. Kittilson, Miki C. and Tate, Katherine (2004) Political Parties, Minorities and Elected Office: Comparing Opportunities for Inclusion in the U.S. and Britain (Paper 04/06), Irvine: Center for the Study of Democracy. Koopmans, Ruud (2008) Tradeoffs between Equality and Difference: Immigrant Integration, Multiculturalism, and the Welfare State in Cross-­National Perspective, Discussion Paper, Berlin: Social Science Research Center Berlin (WZB). Online: www.wzb. eu/zkd/mit/pdf/dp_sp_iv_2008–701.pdf. Kymlicka, Will (1995) Multicultural Citizenship, Oxford: Oxford University Press. —— (1998) Finding Our Way: Rethinking Ethnocultural Relations in Canada, Toronto: Oxford University Press. —— (2007) Multicultural Odysseys: Navigating the New International Politics of Diversity, Oxford: Oxford University Press. Lublin, David (1997) The Paradox of Representation. Racial Gerrymandering and Minority Interests in Congress, Princeton, NJ: Princeton University Press. Mansbridge, Jane (1999) ‘Should Blacks Represent Blacks and Women Represent Women? A Contingent “Yes” ’, Journal of Politics, 61: 628–57. Masclet, Olivier (2003) La Gauche et les Cités: Enquête sur un Rendez-­vous Manqué, Paris: La Dispute. McAdam, Doug (1982) Political Process and the Development of Black Insurgency, 1930–1970, Chicago, IL: University of Chicago Press. McAdam, Doug, Tarrow, Sidney G. and Tilly, Charles (2001) Dynamics of Contention, New York: Cambridge University Press. Messina, Anthony M. (2007) The Logics and Politics of Post-­WWII Migration to Western Europe, Cambridge: Cambridge University Press. Nixon, Jaqi (1998) ‘The Role of Black and Asian MPs at Westminister’, in Shamit Saggar (ed.) Race and British Electoral Politics, London: UCL Press. Norris, Pippa and Lovenduski, Joni (1995) Political Recruitment: Gender, Race, and Class in the British Parliament, Cambridge: Cambridge University Press. Packer, John (2005) ‘Confronting the Contemporary Challenges of Europe’s Minorities’, Helsinki Monitor, 3: 227–31. Phillips, Anne (1995) The Politics of Presence, Oxford: Oxford University Press. Pitkin, Hanna F. (1967) The Concept of Representation, Berkeley: University of California Press. Przeworski, Adam, Stokes, Susan C. and Manin, Bernard (1999) Democracy, Accountability and Representation, Cambridge: Cambridge University Press. Putnam, Robert (1993) Making Democracy Work: Civic Traditions in Modern Italy, Princeton, NJ: Princeton University Press. —— (2007) ‘E pluribus Unum: Diversity and Community in the Twenty-­first Century’, Scandinavian Political Studies, 30(2): 137–74. Saggar, Shamit (2000) Race and Representation: Electoral Politics and Ethnic Pluralism in Britain. Manchester: Manchester University Press. Statistics Canada (2006) Ethnic Origin Reference Guide, Ottawa: Statistics Canada. Online:–recensement/2006/ref/rp–guides/ethnic–ethnique– eng.cfm. Studlar, Donley T. and Welch, Susan (1992) ‘Voting for Minority Candidates in Local British and American Elections’, in Anthony M. Messina, Luis R. Fraga, Laurie A.

Diversity, participation and representation   21 Rhodebeck and Frederick D. Wright (eds) Ethnic and Racial Minorities in Advanced Democracies, New York: Greenwood Press. Tarrow, Sidney (1998) Power in Movement: Social Movements and Contentious Politics, Cambridge: Cambridge University Press. Tillie, Jean (1998) ‘Explaining Migrant Voting Behaviour in the Netherlands: Combining the Electoral Research and Ethnic Studies Perspective’, Revue Européenne des Migrations Internationales, 14(2): 71–95. Togeby, Lise (1999) ‘Migrants at the Polls: An Analysis of Immigrant and Refugee Participation in Danish Local Elections’, Journal of Ethnic and Migration Studies, 25(4): 665–84. —— (2008) ‘The Political Representation of Ethnic Minorities: Denmark as a Deviant Case’, Party Politics, 14(3): 325–43. Williams, Melissa S. (1998) Voice, Trust and Memory: Marginalized Groups and the Failings of Liberal Representation. Princeton, NJ: Princeton University Press. Wüst, Andreas M. (2002) Wie wählen Neubürger? Politische Einstellungen und Wahlverhalten eingebürgerter Personen in Deutschland. Opladen: Leske und Budrich. —— (2006) ‘Wahlverhalten und politische Repräsentation von Migranten’, Der Bürger im Staat, 56(4): 228–34. Zappalà, Gianni (1997) ‘Is There an Ethnic Electorate Effect on Representation? Evidence from the 1993 Australian Candidate Study’, Parliament of Australia Research Paper, 2: 1997–8.

Part I

Immigrants and members of visible minorities as voters Turnout and party choice

2 Voter turnout among immigrants and visible minorities in comparative perspective

To what extent do immigrants and visible minorities participate in the electoral process? The question is crucial to the legitimacy of representative democracies. This chapter surveys research pertaining to voter turnout in 11 liberal democracies. Contributors compare voter turnout among immigrants and minorities to turnout in the majority population, reporting on available data from national, regional and local elections in their respective countries. They discuss the dominant explanations for differences between migrant/minority and non-­minority turnout. Subject to data availability, they also discuss turnout differences across migrant/minority groups, assessing the role of factors such as ethnic or national origin, citizenship status, length of settlement in the country of immigration, and socio-­demographic characteristics. Furthermore, each country’s specific political opportunity structure for participation is discussed, in which the role of parties as a screening and selection agent deserves most attention. As a whole, the chapter takes stock of existing scholarship in this field, and identifies gaps and problems, as well as the progress scholars have made.

2.1  Canada Karen Bird A classic settler society, Canada’s population has long been shaped by immigration. However, it is the post-­World War II pattern of migration to Canada that most distinguishes it from the European paradigm. Notably, all nationality and race-­based criteria were eliminated from immigrant selection policy during the mid- to late 1960s. Today, Canada continues to admit between 250,000 and 300,000 immigrants per year (approximately 1 per cent of the total population). While skilled immigrants comprise the largest portion of new arrivals, there are also significant numbers of sponsored family members and refugees. This makes for an exceptionally diverse population. Currently, one in five people living in Canada was born outside the country and fully three-­quarters of recent immigrants (those arriving within the past five years) belong to a visible minority group.1 In immigrant gateway cities such as Toronto and Vancouver, foreign-­ born residents constitute as much as 46 per cent, and visible minorities as much as 43 per cent of the population. With a liberal approach to naturalization – for

26   Voter turnout in comparative perspective example, the residence requirement is just three years and dual citizenship is allowed – most immigrants quickly acquire citizenship and with it, voting rights for elections at all levels of government.2 In addition, voter registration is automatic and any person who was missed in the enumeration process is simply required to provide proof of identity at the polls. Overall, then, there are few formal barriers to electoral participation facing new Canadians. Voter participation among immigrants and minorities has been an under-­ researched topic in Canada. We have only recently had access to sufficiently large datasets, permitting us to tease apart the effects on participation of factors such as nativity, ethnicity, socio-­demographic and migration and settlement-­ related variations. Most data are drawn from survey measures of self-­reported vote and these, for the most part, show minimal differences between foreign-­ born and native-­born Canadians.3 For example, using data from the 2002 Ethnic Diversity Study, Tossutti (2005, 2007) finds no significant difference in self-­ reported vote between immigrants and Canadian-­born citizens in federal elections (78.6 and 79.3 per cent, respectively) or in municipal elections (63.5 and 63.9 per cent, respectively, said they voted in the last election). Differences among immigrants are quite striking however, and may be related to such factors as length of settlement, visible minority status, age and socio-­economic status. Exploring these possibilities, White et al. (2006, 2008) find that the most important predictor of immigrant voter turnout is the amount of time they have lived in Canada. This outweighs other traditional social background factors such as age, income and education. They also find no support for the ‘resistance hypothesis’ that immigrants previously socialized in non-­democratic countries would be less likely to vote than immigrants from advanced industrial countries. If it is exposure to politics in their new country that is the prerequisite for voting, immigrants may gain this naturally over time, or they may make up for lost time by paying particular attention to information about politics. In a comparative study of immigrant political engagement in Canada and Britain, Paul Howe (2007) shows that while new immigrants to both countries begin with a significant political knowledge deficit, Canadian immigrants close that gap more quickly than those in Britain. Other studies have also contributed evidence that many immigrants, over time, achieve higher levels of interest and knowledge about Canadian politics than the native-­born Canadians (Bilodeau and Kanji 2006; Henderson 2005). These findings give rise to a mobilization perspective on immigrant political behaviour. In particular, what is the role that the ethnic community might play in providing contextual cues and opportunities that heighten the involvement of community members? The argument here is that rather than erode civic ties, strong and institutionally complete ethnic communities can serve to promote political involvement and a sense of attachment to the larger political unit (Kymlicka 1995; Lapp 1999; Qadeer and Kuman 2006). But this does not happen accidentally. Rather, political incorporation of immigrants depends on mechanisms that can link individuals and immigrant communities to the political system (Bloemraad 2003). A key factor here may be Election Canada’s very extensive voter outreach and

Voter turnout in comparative perspective   27 education initiatives in multicultural communities.4 However, there are also less formal linkage mechanisms that appear to facilitate newcomer integration into the political process. One is the presence of a vigorous ethnic media providing coverage of Canadian politics that is absent in the ‘homeland’ sources of news to which immigrants would otherwise turn (Black and Leithner 1988; Howe 2007). Another important linkage mechanism is provided by Canadian political parties. Of particular interest here are the practices of the relatively autonomous local associations of the national parties. Especially at the candidate selection stage, riding associations have traditionally used their very open party membership rules to reach out and include ethnic minorities (Carty 1991; Dhillon 2005; Stasiulis and Abu-­Laban 1991). National party organizations also engage in ethnic mobilization. While this is a long-­standing practice, recent changes in the national party system as well as growing competitiveness over key urban ridings have made mobilization of the ethnic vote an increasingly key strategy in federal politics (Hyder 2005a, 2005b). The mobilization effort of the new Conservative Party of Canada has been especially well developed (LeBlanc 2007; Rana 2007). The party’s ‘ethnic outreach’ strategy is coordinated through the Prime Minister’s Office and is overseen by the current Secretary of State for Multiculturalism. The strategy targets key ridings and ethnic communities that are viewed as ‘CPC accessible’ and lends assistance to local candidates and MPs by ensuring ministerial attendance at major ethnic events in their ridings, mailing holiday greetings to ethnic constituents, and providing communications and policy support. The stated goal of the strategy is to ‘replace the Liberals as the primary voice of new Canadians and ethnic minorities’ (Conservative Party of Canada 2007). As the party in government, the Conservatives have also made a point of passing immigrant- and minority-­friendly policy. These strategies appear to have a positive effect on the electoral participation of immigrants and visible minorities at the national level. We know much less about mechanisms for political incorporation at the municipal level, where candidates generally do not run under party labels. In addition to these quite positive findings on the political incorporation of immigrants and minorities, some troubling trends have also emerged. Recent immigrants have fared less well in the labour market, despite having higher levels of education and training than their predecessors. In addition, there are signs of growing race-­based patterns of poverty and perceived discrimination (Reitz and Banerjee 2007). Visible minority immigrants appear to have significantly lower levels of social trust than non-­minority immigrants (Soroka et al. 2007). There is also some evidence that lower turnout persists among second-­ generation visible minorities. While there may be signs of lower turnout among some segments of the visible minority population, there is certainly no pervasive racial gap in voting. Canada is far from the United States in this respect.5 However, there is concern that we are moving in that direction. There remain many gaps in our understanding. Despite the concentration of recent immigrants and minorities in large cities, we know little about their participation in municipal politics. We have not yet had a sufficient minority

28   Voter turnout in comparative perspective over-­sample in the Canadian Election Studies, leaving researchers to rely either on relatively outdated pooled samples drawn across several elections, or on specialized minority surveys that are not focused specifically on electoral behaviour, or on exploratory studies of single communities. New data from a series of recent federal elections (2004, 2006, 2008) will provide an important opportunity to explore the correlates of voting behaviour among distinctive ethnocultural groups across geographically diverse settings. There remains much to be learned about the political participation of this increasingly important segment of the Canadian electorate.

2.2  France Vincent Tiberj French research has only recently paid attention to the political behaviour of visible minorities. Laggardness in this respect can be explained both by French traditions of social scientific research, and by the specificities of the Republican ideological paradigm. As measured by the population share, scope and variety of immigrants during certain periods, France is considered to be an old country of immigration – not unlike the United States (Noiriel 1992). Yet this historical context was widely disregarded until recently owing to the Republican colour-­ blindness principle. Nevertheless, the social reality is clear: racial discrimination is widespread (Meurs et al. 2006; Silberman and Fournier 2006). Despite this ideological closure towards visible minorities, the political opportunity structure for minorities in France is surprisingly favourable. Among foreigners, only EU citizens enjoy the right to vote. However, access to full citizenship is relatively easy. Until recently the only conditions were length of residence and mastery of the French language. Yet both conditions are quite easy to achieve, since most of France’s post-­World War II non-­European immigration originated from former French colonies.6 Furthermore, the ius soli principle prevails for the second generation born in France. On the other hand, political parties have largely disregarded voter mobilization and electoral registration campaigns among visible minorities, mostly because ethnicity was not seen as a valuable ‘card’. This disregard among the mainstream parties is due both to the prevailing Republican ideology, and to pressure and rising competition from the extreme-­ right. There appears to have been an exception to this pattern following the suburbs crisis of 2005. Civic associations have specifically tried to mobilize suburban youth, and many politicians have also sought support from ethnic networks at the local level. The electoral return of this strategy has yet to be assessed. Recently, we have witnessed a growing body of research addressing the politics of ethnic diversity, and in particular the political participation of minorities.7 However, we still lack systematic surveys and are forced to rely on national samples that are insensitive to local context. To the extent that they study ethnic diversity, scholars have focused mostly on a narrow part of the potential population of interest. Foreigners are left aside, since they are ineligible to vote even in local elections. The attention is largely on non-­European groups, particularly

Voter turnout in comparative perspective   29 those of Maghrebin and African origin who account for the large majority of visible minorities. Globally, almost 20 per cent of the French electorate has at least one foreign grandparent. Ethnic minorities of European descent make up the lion’s share (9.5 per cent) of the eligible voter population, while the visible minorities of interest account for just 3.2 per cent.8 The relatively small sample size in national surveys has prevented detailed analyses by national origin.9 Within France’s non-­compulsory voting system, turnout among visible minorities appears to be generally lower than turnout in the general population. Based on the French permanent demographic sample data, Rahsaan Maxwell (2010) estimates a 30 per cent rate of voter turnout among Maghrebin-­French in the 2004 European Parliament elections, and 56 per cent turnout in the second round of the 2004 regional elections. In comparison, the rate of participation among native metropolitans reached 46 and 70 per cent, respectively. Yet Maxwell found that after controlling for age, level of education, occupation and employment, Maghrebin origin itself has no significant impact on the probability for turnout. We have replicated Maxwell’s study using data from the 2007 post-­electoral survey. This presidential election was particularly mobilizing, whatever the social characteristics of the individuals,10 but the results are the same. Whatever their age, education, occupation and employment status, the Maghrebin- and African-­French do not differ from the rest of the population, or from European-­origin citizens.11 In France, then, it appears that ethnicity cannot be added to the list of ‘usual suspects’ that tend to distort the social composition of the voting public. Table 2.1  Reported voter registration in France, 2005 (%) Maghrebin- and African-French population

General population

Difference (percentage points)

Age groups 18 to 24 years 25 to 34 years 35 to 49 years 50 to 64 years 65 years and older

70 78 76 92 –

  85   84   94   97 100

–16   –6 –17   –5 –

Level of education No or primary education Secondary Baccalaureate University education

76 72 78 81

  96   92   95   92

–20 –19 –17 –12

Political interest A lot Some Hardly any Not at all

86 84 73 66

  98   95   93   84

–12 –11 –20 –18


77 (N = 1,003)

  93 (N = 1,006)

30   Voter turnout in comparative perspective These results refer specifically to the registered voter population, and may not apply to the broader eligible voting population.12 In the 2005 RAPFI survey (described in Chapter 7), only 77 per cent of Maghrebin- and African-­French indicated that they were registered on the electoral rolls, compared to 93 per cent among the general population. Although a declarative claim (some may be automatically enrolled), this gap nevertheless demonstrates these minorities’ relative unawareness regarding their electoral status. This registration gap persists even after controls for age, education and level of political interest. In fact, we see that Maghrebin- and African-­French are an overall more educated population13 and are almost as interested in politics as the general population (47 per cent compared to 53 per cent). Even though registration is highly correlated with age, education and political interest, we see that group differences in registration levels persist among youngsters, and among the most highly interested and highly educated. Traditional theories such as social domination or political sophistication do not help here. Our survey shows that Maghrebin- and African-­French have slightly more confidence and trust in politics than the general population, and that they would be more opposed to the suppression of political institutions such as parties or the Parliament. In addition, these ethnic minorities refuse less often to take sides in politics (25 per cent versus 38 per cent in the general population). These findings clearly do not portray minorities as ‘alienated’ citizens (Lane 1972); instead we might consider them to be ‘incomplete’ citizens. Political socialization may be a key explanation for the incompleteness of voter registration among these groups. As pointed out by Sophie Duchesne (2007), voting habits are strongly stimulated by regular interaction and discussion of politics with parents. Sharing with parents the symbolic act of voting seems to be particularly important. Two of our findings support the validity of the intergenerational socialization hypothesis. First, when members of an ethnic minority cannot state the political affiliation of their father (a sign of a low-­ politicized environment), the level of registration is 11 points lower than in a politicized household. Second, among the second generation, mixed French/non-­ French origin corresponds to an eight-­point increase in registration. This suggests a major consequence of not granting voting rights to foreigners: their children do not take over the habit of registering and consequently play a diminished part in the electoral process, even if they are full citizens. Voter turnout among visible minorities was long a terra incognita in France. However, French social scientists as well as politicians have begun to seriously explore this matter. With respect to their role as voters, French citizens of minority ethnic background have been incomplete citizens; yet the 2007 presidential election may be a turning point. Never before did the suburbs mobilize so strongly against a presidential candidate as they did against Nicolas Sarkozy,14 and never before was the issue of ethnic diversity so present in national debate. Whether this diversity at the polls will be transitory, or the beginning of a long-­ term process of political empowerment, we do not yet know. As well, little is known of differences among ethnic groups (such as Asian-­French), of period of

Voter turnout in comparative perspective   31 settlement effects, of the role of local networks, civic associations or ethnic minority candidates. We also know little about the political socialization of immigrants prior to naturalization, whether with respect to interest in French politics, or other political activities such as associational and trade union membership. Nevertheless, ethnic mainstreaming in survey research is now accepted in French political science, so it may be only a matter of time before we find answers to these and to related questions.

2.3  United Kingdom Thomas Saalfeld At least two fundamental elements of the UK’s approach to citizenship and immigration provided relatively benign conditions for the electoral participation of post-­war immigrants. First, the initial waves of post-­war immigration to the United Kingdom from the 1950s to the early 1960s consisted overwhelmingly of persons from Commonwealth member states. Most of these persons were granted citizenship rights upon arrival. This offered them at least the formal opportunity to vote at all levels of the political system from the start (Anwar 2001: 533). Second, British race relations legislation between the 1960s and 1990s focused strongly on the ‘adaption of local public service delivery to specific ethnic minority requirements and anti-­discriminatory policies’ (Garbaye 2005: 33) and provided strong incentives for minority groups to become involved in matters of immediate importance to their local communities. This, in turn, created a favourable backdrop for local mobilization and the growth of political minority elites ‘from below’. Nevertheless, turnout levels among visible minorities have historically been lower than for the non-­minority population, across all main tiers of government. For the 2005 general (national) election, Sanders and colleagues (2005: 13, 35) found that members of minority ethnic groups were significantly less likely to vote than whites (turnout levels were 56 per cent and 68 per cent, respectively). This gap between white citizens and those from visible minorities is replicated in local elections, albeit at much lower levels of involvement. Richards and Marshall (2003: 6) report that the turnout of white voters in the 2002 local government elections in England and Wales was 33 per cent compared to around 24 per cent among black and minority ethnic voters. Turnout levels in regional elections tend to be situated between general and local elections. In a representative survey of the population of Northern Ireland, for example, less than half of the minority respondents (48 per cent) said they voted in the November 2003 (regional) Assembly election. This compared with a self-­reported turnout of 64 per cent among the general population (MMMA and Omi 2005: 4). Although the sample of minority voters in surveys is usually not large enough to draw strong inferences about individual ethnic groups, there has been considerable cumulative evidence of stable differences over decades. Anwar notes (2001: 536–7) that ‘the Asian turn-­out on average is higher than that of non-­Asians and also that there are differences between Indians, Pakistanis and

32   Voter turnout in comparative perspective Bangladeshis’. This is also borne out by recent studies. In his analysis of the 1997 general election, Saggar (2000: 103) found that turnout in the 1997 general election among citizens of Indian origin was higher (83 per cent) than the turnout of any other group, including whites (79 per cent). The lowest level of turnout occurred among citizens of black African origin (65 per cent). Similar results were found for the 2001 election (Cutts et al. 2007; Purdam and Fieldhouse 2002: 35). One of the possible explanations for such inter-­group variations may be found in recent studies informed by social capital theory, which confirm that contextual factors such as the size and density of the minority population in a constituency are associated positively with the turnout among such voter groups (for a study of Muslim voters in Britain in the 2001 general election see Fieldhouse and Cutts 2008a). A special survey conducted among black and minority ethnic adults during the 2005 general election campaign and shortly after the election allows some insights into the socio-­demographic factors influencing turnout among minority voters. The study found that overall among members of black and minority ethnic groups, self-­reported turnout was lowest among respondents in the age groups 18–24 and 25–34 (48 per cent in both cases). It increased to approximately 63 per cent among people aged between 35 and 54, and to 79 per cent among voters aged 55 and over. The differences in claimed turnout between those born in the UK or elsewhere, between males or females, and between different social classes were either insignificant or relatively small compared to the strong variation between age cohorts (Electoral Commission 2005). The Electoral Commission’s study of minority voter engagement in the 2001 general election (Purdam and Fieldhouse 2002: 30–44) shows that variations in turnout levels within the minority communities are influenced by similar factors as those operating in the population overall: turnout was higher among those BME citizens who considered voting a ‘civic duty’ and lower among those who deemed it an ‘inconvenience’. Ecological factors such as constituency marginality and hence perceived impact of the individual vote were clearly related to self­reported turnout among minority voters as well as among white voters. Some differences were related to ‘generic’ factors stemming from the socio-­ demographic composition of minority ethnic groups. Generic factors that tend to depress BME turnout rates include ‘the younger age profile of these communities, the higher levels of social and economic deprivation experienced among these groups, and the fact that they predominantly live in urban areas where turnout levels tend to be lower than average’ (Purdam and Fieldhouse 2002: 7). Yet there are some community-­specific problems: For citizens from visible minority backgrounds, the process of registration on the electoral roll is still one of the key obstacles in exercising their right to vote. Although registration of minority ethnic citizens has increased significantly since the 1960s, the levels among people of African heritage still tend to be lower than those in the rest of the population. Saggar (1998) reports a nearly 13 per cent non-­registration rate among African-­origin citizens in 1997. Variations in registration levels among some groups seem to vary with the groups’ size and density in a constituency.

Voter turnout in comparative perspective   33 For example, Fieldhouse and Cutts (2008b) find that Muslim registration rates are higher in predominantly Muslim areas. In addition, Purdam and Fieldhouse (2002: 7) believe that the ‘lack of representation in high-­profile public positions’ may create disincentives for minority ethnic citizens to vote (for a similar conjecture see Anwar 2001: 539). A recent survey of black voters shows that they would be more likely to vote, if there were more black people in politics (Richards and Marshall 2003: 7). Responding to findings of this type, the political parties have begun to compete more vigorously for the visible minority vote: Although the Labour Party is still by far the most successful party among minority ethnic voters, Conservatives and Liberal Democrats have made some inroads, especially among Asian and Muslim voters (Anwar 2001: 538–9; Purdam and Fieldhouse 2002: 23). Despite some remarkably thorough studies, especially a whole array of reports commissioned by the Electoral Commission since 2002, there remain many open questions; for example, small sample size and clear evidence of non-­ response bias and misreporting current challenges to estimating the extent of abstentionism, and to evaluating differences in turnout levels among different BME groups (Purdam and Fieldhouse 2002). Despite these reservations, there has been relatively consistent and cumulative evidence of increased and relatively high turnout among citizens of Indian heritage, whereas turnout levels among persons from black African and Caribbean backgrounds are still much lower. However, there is no real agreement in the literature as to whether these apparent differences are ‘cultural’ or whether they are at least partly related to generic socio-­demographic factors such as age, gender, education or social class. The UK has seen a number of recent measures intended to increase public engagement with the electoral process (e.g. better information on the registration process, citizenship classes for recent immigrants, wider use of postal voting). There have also been broader constitutional reforms, such as devolution and the introduction of proportional representation at the level of regional assemblies. We do not yet know what effect these changes may have on the electoral participation of immigrants and visible minorities.

2.4  The Netherlands Laure Michon and Jean Tillie The Netherlands granted municipal voting rights to foreigners in 1985, making it one of the earliest European countries – after Ireland and the Scandinavian countries – to enfranchise foreigners at the local level. It is moreover relatively easy to participate in a Dutch municipal election: there is no obligation to register in order to be able to vote. After five years of legal residence, non-­national inhabitants automatically receive their voting pass at home (EU citizens have local and European voting rights after only six months of residence). Typically, immigrants themselves did not mobilize for this political right. Rather, the Dutch political elite felt that it was necessary to enfranchise foreigners at the local level (Jacobs 1998), and in subsequent years a lot has been done to include candidates

34   Voter turnout in comparative perspective with a foreign background on party lists. Almost 10 per cent of Dutch MPs are now of foreign descent (Michon and Tillie 2003b: 146) and in the main cities, the share of elected politicians is proportionate to the share of immigrants in the population.15 Since local enfranchisement in 1985, turnout of immigrants (and party choice – see Chapter 3) has been studied regularly, but mainly at the municipal level. Due to different definitions of the research population and a focus on specific cities,16 research results are not always comparable. The data gathered before 1994 have been especially difficult to compare (see Tillie 1994: 5). In the studies of municipal elections in 1994 (Tillie 1994), 1998 (Tillie 2000), 2002 (Michon and Tillie 2003a; Rhee 2002a) and 2006 (Heelsum and Tillie 2006), ethnic origin is determined based on one’s country of birth and the country of birth of both parents, irrespective of a person’s nationality.17 It is important to note that no question has been asked about the nationality of the respondents since 1994, and therefore, the distinction between nationals and non-­nationals cannot, unfortunately, be made. From these last four surveys of municipal voter turnout (1994–2006) it appears that, in general, immigrants participate less than the native Dutch population. However, Table 2.2 – summarizing data from these elections in the two largest Dutch cities – shows that there are important differences among cities, among elections and among ethnic groups. Turnout among migrants tends to be much higher in Rotterdam than in Amsterdam, and while there has been a steady increase over time in Rotterdam for all ethnic groups, the figures in Amsterdam fluctuate quite widely. Participation is consistently higher among people of Turkish origin than among those of Moroccan or Surinamese background. In 2006, turnout among Turkish voters was equal to the overall turnout in both cities. Research on migrant turnout at the national level is rare18 and leads to inconclusive results, owing to different approaches and methods.19 The lack of research attention is due to the fact that only Dutch citizens can participate in national elections. However, two-­thirds of Moroccans and Turks in the Netherlands hold dual citizenship (see CBS 2007). In the 2006 parliamentary Table 2.2  Turnout of migrant voters in two Dutch cities, 1994–2006 (%) Country of origin of voter

Turkey Morocco Surinam/Dutch Antilles Overall turnout









2002a 2006

67 49 30 56.8

39 23 21 45.7

30 22 26 47.8

51 37 26 50.8

28 23 24 56.9

42 33 25 48.4

54 40 31/19b 55

Sources: Tillie 2000; Rhee 2002a; Michon and Tillie 2003a; Heelsum and Tillie 2006. Notes a Source: Rhee (2002a), with a different research method. b These two groups were distinguished only in the Rotterdam data for 2002.

56 58 41 57.8

Voter turnout in comparative perspective   35 elections, 1.2 million persons originating from the main immigrant countries – approximately 10 per cent of the total electorate – had the right to vote (Dekker 2006). Due to the research method used in exit polling since 1994 (counting people coming out of a polling-­station),20 it has not been possible to differentiate turnout in terms of gender, age, nativity and period of settlement, education or other socio-­demographic characteristics. In a study of turnout at the municipal election in Rotterdam in 2002, Rhee found that although turnout among migrants is always lower than for the native population, the age distribution of those who cast their vote is the same for the native Dutch and migrants (Rhee 2002a: 13). Moreover, it turned out that for the native Dutch and Turks, turnout among men and women was equal, while for other groups (Surinamese and Moroccans) men were more likely to vote than women (Rhee 2002a: 14).21 The data show no difference in turnout between the first and second generation of migrants (Rhee 2002a: 15). Turnout explanations have focused largely on differences among ethnic groups with reference to the role of social capital within ethnic networks (see the ‘micro-­context’ side of Figure 1.1). Building on the work of Robert Putnam (1993), Fennema and Tillie have measured the civic involvement of ethnic communities in Amsterdam. Their network analysis of organizational density and interlocking directorates in the boards of ethnic organizations shows that the Turkish community in Amsterdam is the most civic, followed by the Moroccan, Surinamese and Antillean communities (Fennema and Tillie 1999: 721). This rank order corroborates the order for voter turnout among these groups. Fennema and Tillie conclude that level of civic involvement is a key factor explaining differences in voter turnout among ethnic groups. In their words: ‘the more an ethnic group is engaged in the own community’s affairs, the more it participates in local politics’ (Fennema and Tillie 1999: 721). Other explanations for turnout have not yet been assessed. We do not know whether the presence of ethnic minority candidates on party lists would enhance turnout. Why turnout increases or decreases over time is also understudied. We know that while migrant turnout decreased in Amsterdam in 2002, it has increased and continues to increase in Rotterdam. In line with the civic community argument, it is interesting to note that ethnic organizations have been supported less (via funding, for example) in Amsterdam than they have been in Rotterdam prior to 2002. In Rotterdam, the participation of immigrants was fostered through targeted campaigns in the languages of the ethnic groups, and meetings were organized through the migrant associations. In terms of the POS framework presented in Chapter 1, it may be that different local policies produce variation at the level of social networks, and that this in turn explains differences in turnout. However, the municipal turnout figures for 2006 blur these elements of analysis. Although there has been no policy change in Amsterdam, turnout has increased again; simultaneously, we have seen the introduction of more harsh policies towards migrants in Rotterdam between 2002 and 2006, yet migrant turnout has nevertheless continued to rise in that city.

36   Voter turnout in comparative perspective Another issue requiring attention in the Netherlands concerns the difference between nationals and non-­nationals. Research in other countries has privileged nationality as a crucial explanatory criterion for turnout, and a comparative agenda should urge Dutch researchers to consider this variable as well. We wonder, for example, about the impact of nationality on party choice, or on turnout at the local level. More studies on the correlates of immigrant participation at the national level would substantially complement our knowledge about migrant political participation in the Netherlands.

2.5  Belgium Dirk Jacobs All adult citizens of the state of Belgium are obliged to vote, and risk a fine if they do not. Unsurprisingly, turnout rates exceed 90 per cent. While there has been no study on this question, we have no reason to believe that Belgians with an immigrant background would have a different turnout rate compared to other Belgians. Of course, not all immigrants have obtained Belgian citizenship. Foreign residents are eligible to vote at the local level, but not at the regional or federal levels. EU citizens have enjoyed local voting rights and eligibility to stand as a candidate since 2000. Those not holding European citizenship are eligible to vote in local elections if they have resided in the country for over five years. They cannot stand as a candidate, however. The 1994 European directive on EU enfranchisement, implementing the Maastricht Treaty, prohibits member states from forcing EU citizens from other countries to exercise their voting rights in local elections. This presents a problem for Belgium, which has compulsory voting for state citizens. The Belgian policy response has been to require EU citizens to register as voters if they wish to exercise their right to vote in accordance with the Maastricht Treaty. To do so, they must send a written demand to the municipal administration at least two months prior to the election date. Once registered as voters, EU citizens are then legally obliged to vote, just as are all adult Belgians. When non-­EU citizens were enfranchised in 2006, a similar rule was applied to them. The consequence is that voter registration is not compulsory for a foreigner, but once registered, a person must vote or otherwise risk a fine. In this context, rates of voter registration among resident non-­Belgian EU citizens and non-­EU citizens can tell us a great deal about the electoral participation of these groups. We focus here on foreigner voter registration for the October 2006 local elections.22 Across Belgium, 20.9 per cent of non-­Belgian EU citizens were registered as voters for these elections. However the rate of EU citizen voter registration varied widely across the three regions of the country. The lowest rate is found in the Brussels Capital Region, where just 13.7 per cent of non-­Belgian EU citizens were registered. This is a higher rate of participation for this group than in the local elections of October 2000 (9.6 per cent), but can hardly be called a success. EU citizen voter registration was significantly higher in Flanders (16.9 per cent) and in Wallonia (28.5 per cent). Turning to non-­EU

Voter turnout in comparative perspective   37 citizens, we find that just 15.7 per cent were registered as voters across Belgium as a whole. Again we see regional variation, with a low of 12.6 per cent of non-­EU citizens registered as voters in Flanders, 15.7 per cent in the Brussels Capital Region, and rising to 21.3 per cent in Wallonia. Notably, the registration rate of non-­EU citizens in Brussels (15.7 per cent) was higher than the registration rate of EU citizens in that region. In Flanders and Wallonia, however (and in Belgium overall), the registration rate of non-­EU citizens was lower than the registration rate of EU citizens. Let us now look in more detail at the registration rates in the 19 municipalities of the Brussels Capital Region for the October 2006 elections. Suppose that all potential foreign voters had registered to vote. The first column of Table 2.3 indicates the electoral impact which EU voters would have in such a scenario, while the second column indicates the potential electoral impact of non-­EU voters. As one can see, this potential electoral impact is quite impressive. In a municipality such as Anderlecht, EU voters would constitute no less than 14.8 per cent of the electorate, while non-­EU voters would constitute 6.0 per cent of the electorate. The potential electoral impact of EU voters ranges from a substantial 8.5 per cent of the total electorate in Ganshoren, to a staggering 32.3 per cent in Saint-­Gilles. The potential electoral impact of non-­EU voters is smaller but still important, ranging from 1.7 per cent in Woluwe-­Saint-Pierre to 13.2 per cent in Saint-­Josse. Taken together, all foreign residents account for 39.1 per cent of the electorate in Saint-­Gilles. Given the relatively low registration rates of both EU citizens and non-­EU citizens, this potential electoral impact fails to materialize however. Columns three and four of Table 2.3 indicate the actual share of the overall electorate taken by EU citizens and non-­EU citizens respectively (for a more detailed analysis consult Jacobs et al. 2009). Compared to the potential impact, the actual impact is modest. For EU citizens there is a non-­negligible impact, ranging from 1.5 per cent in Evere up to 8.3 per cent in Saint-­Gilles. If EU citizens were to vote en masse, they would have a decisive impact on electoral outcomes in several municipalities. It is important to note that relatively few EU citizens choose to stand as candidates: in 2006 there were 175 EU candidates among a total of 3,529 candidates (5.0 per cent), only three of whom were elected. The picture is also less impressive when we turn to the actual impact of the non-­EU electorate. Their share of actual voters is highest in Schaerbeek, where they comprise 2.0 per cent of the electorate. In most other municipalities non-­EU voters make up less than 1 per cent of the total electorate. Of course, within the EU and non-­EU groups, there are differences across nationalities. Registration rates can vary significantly from one nationality to another. For example, immigrants from Sub-­Saharan Africa have much higher rates of electoral participation than other groups. Notably, 29 per cent of Congolese, 35 per cent of Camerounese and 39 per cent of Rwandese immigrants register as voters, compared to only 10.7 per cent of Moroccans and 15 per cent of Turks. Nevertheless, the actual share of these individual groups among the overall electorate remains very small. We can hence conclude that

14.8 15.2 9.2 18.6 28.6 10.5 19.8 8.5 30.8 8.9 12.3 11.2 32.3 14.6 15.9 19.9 11.9 21.3 22.2

6.0 2.0 2.5 7.4 4.1 3.7 4.9 2.4 5.9 3.8 5.8 9.2 6.8 13.2 9.9 2.5 1.7 2.3 1.7

Potential electoral impact of non-EU voters

Source: Ministry of the Interior, Department of Elections, 2006, treatment ULB.

Anderlecht Auderghem Berchem-Ste-Agathe Bruxelles Etterbeek Evere Forest Ganshoren Ixelles Jette Koekelberg Molenbeek-Saint-Jean Saint-Gilles Saint-Josse Schaerbeek Uccle Watermael-Boitsfort Woluwe-St-Lambert Woluwe-St-Pierre

Potential electoral impact of EU voters 2.6 3.1 1.5 3.2 4.3 1.5 3.3 2.1 4.5 1.6 2.4 1.5 8.3 2.2 2.6 3.5 3.2 3.4 5.3

Actual electoral impact of EU voters 0.9 0.4 0.4 1.8 1.0 0.9 0.9 0.5 1.3 1.0 1.3 1.5 2.0 1.2 2.0 0.4 0.4 0.5 0.3

Actual electoral impact of non-EU voters

Table 2.3  Potential and actual electoral impact of EU citizens and non-EU citizens for the 2006 local elections in Brussels’ municipalities (%)

Voter turnout in comparative perspective   39 enfranchisement of non-­nationals cannot have had a tremendous impact on electoral results in the Brussels local elections. This should not, however, lead to the conclusion that the ethnic minority vote would be of no importance in Brussels. Due to the liberal naturalization legislation, large numbers of non-­EU-origin immigrants have obtained Belgian citizenship (in contrast to EU citizens who tend not to apply for Belgian citizenship). Given that voting is compulsory in Belgium, almost all of them participate in the polls. Immigrant-­origin citizens comprise tens of thousands of voters of various ethnic backgrounds, and no mainstream party in the Brussels Capital Region can afford to totally neglect this new electoral group if it wants to win elections. This finding cannot, however, be extrapolated to the rest of Belgium, where the presence of ethnic minorities is much more modest. We currently have no data that would allow us to examine differences in registration rates due to socio-­demographic factors (gender, age, education), or to migration-­related variables (length of stay, level of acculturation). Nor has there yet been any analysis of the effects on registration rates of factors related to the political opportunity structure (such as having ethnic minority candidates on the lists), or of factors related to social capital (for instance, the density of ethnic associational networks). Such research is possible, and is planned for the future.

2.6  Sweden Nazem Tahvilzadeh In 2006 approximately 13 per cent of the Swedish population was foreign-­born, with more than half of these migrants being naturalized Swedish citizens. Certainly in terms of the Swedish citizenship regime and electoral rules, the political opportunity structure is quite conducive to migrants’ participation in the political process. Since the 1970s, Swedish immigration and integration policies have been characterized by cultural pluralism and the explicit goal of political incorporation. Neo-­corporatist policy has also contributed to a highly open democratic structure, in which the state facilitates the mobilization of ethnic groups through supporting their organizations (Borevi 2002). These organizations are intended to be the main channel of dialogue between migrant groups and the public sector (Aytar 2007). In addition, since 1975 foreign citizens with a minimum of three years’ residence in Sweden have the right to participate in municipal and county-­level elections. Despite these favourable conditions, Table 2.4 shows that voter turnout is consistently lower among migrant groups compared to non-­migrants. Turnout among foreigners with voting rights was 60 per cent in municipal elections in 1976, compared to 91.7 per cent for the electorate as a whole, and overall non-­ citizens constituted just 3.5 per cent of the voters in these elections (Hammar 1979). Since 1976, voter turnout among foreigners has declined considerably, reaching just 36.9 per cent in the elections of 2006. Across municipal elections from 1976 to 2006, the difference in turnout between non-­citizens and the total electorate has ranged from 30.4 to 44.4 points. Voter turnout in national elections




90.4 60.0 –30.4

Municipal elections

Total* Non-citizens Difference

89.6 52.0 –37.6



91.4    –


87.8 48.0 –39.8



89.9    –


84.0 43.0 –41.0



86.0 71.7


84.3 41.0 –43.3



86.7 76.4


84.4 40.0 –44.4



86.8 77.3


78.6 34.5 –44.1



82.1 67.3


77.9 35.1 –42.8



80.1 66.8


79.4 36.9 –42.5



82.0 68.0


Note * Percentage turnout among citizens entitled to vote SCB (2008); for 1976 to 1994 partially rounded.

Sources: Bäck and Soininen 1996 and SCB (Statistics Sweden). There are no data for naturalized citizens in national elections before 1988, and no detailed and systematic data for naturalized citizens in local elections. Data are also not available for county (mid-tier) elections. The category ‘naturalized citizens’ includes individuals born in Sweden who acquired Swedish citizenship after birth.

89.0 53.0 –36.0


90.7    –

91.8    –

Total* Naturalized citizens (1st and 2nd generation) Difference



National elections

Table 2.4 Voter turnout in Swedish national and municipal elections among minority citizens, non-citizens and majority population (%)

Voter turnout in comparative perspective   41 is also lower among naturalized citizens. In national elections from 1988 to 2006, the gap in voter turnout between naturalized citizens and the population overall ranged from 9.5 to 14.8 points. Unfortunately, we lack detailed and systematic data on voter turnout among naturalized citizens at the local level (as well as data on regional elections). Thus while the turnout data for ‘non-­natives’ at the local and national level are not strictly comparable (given that non-­citizens and naturalized citizens are distinctive groups), the pattern across these two levels is nonetheless similar: electoral participation rates of non-­natives are considerably lower than the population rates. That said, it is clear that the rate of participation among non-­citizens at the local level is especially low. Among naturalized Swedish citizens, there are clear variations in turnout depending on country of origin. In the national elections of 2002, the least inclined to vote were migrants from South America, Africa, Asia and non-­EU European countries. However, turnout among these four groups increased in the 2006 elections, surpassing even migrants from other EU countries. There are also clear variations in municipal turnout among foreigners, based on country of origin. Overall, South American groups tend to have the highest levels of voter turnout, followed by EU immigrants and North Americans. African, Asian and Non-­EU Europeans have the lowest rates of electoral participation (SCB 2002). While turnout across the entire electorate has decreased over time, the decline is more pronounced among some groups. Turnout levels among migrants from Greece and the former Yugoslavia fell by more than half in local elections from 1976 and 1994. One possible cause is the change in composition of these groups over time. Notably, the Balkan wars of the early 1990s contributed to a large influx of new migrants who may be less likely to participate in the political process than more settled immigrants. Notable also is the less steep decline in voter turnout among Western European and North American migrants. Migrants from Chile had the highest turnout, at 77 per cent in 1985, and remain on the relatively high end with a turnout rate of 59 per cent in 1994. In comparison, municipal turnout among foreigners overall during this period ranged from 48 to 40 per cent. One possible explanation for the very high rate of participation among Chileans is that this group comprises mainly political refugees, who tend to have a higher degree of knowledge and interest in politics (Bäck and Soininen 1996). Even among this group however, turnout has declined over time, reaching just 37 per cent in the most recent elections in 2006. Two important explanations for the differences in voter turnout are length of residence in Sweden and access to citizenship. According to Hammar (1979), the exceptionally high level of immigrant turnout in the ‘first immigrant election’ of 1976 was a result of strong political interest and union involvement among migrant groups. A multilingual information project from the electoral authorities also had a positive impact. Yet turnout was especially low among young migrants, among migrants with low levels of education, and among those not yet sure of permanent residence in Sweden (Hammar 1979: 144). Knowledge of the Swedish language and Swedish society, participation in civic activities, as

42   Voter turnout in comparative perspective well as contacts with channels of recruitment for political party activities, also help to explain differences in migrant voter turnout (Adman and Strömblad 2000). Bäck and Soininen (1998, 1999a, 1999b) have applied an ‘integration-­ culture-model’ to explain variations in turnout among different migrant groups. According to this model, migrants who belong to a hierarchical culture with strong group relations and an extensive normative system of rules and regulations have a strong potential to mobilize in elections. Alternatively, the fatalists, who have low group attachment but are still obedient to authority, have the lowest turnout rate. They also found that social integration in Swedish society – especially with regard to employment, organizational involvement, use of Swedish media and social interaction with native Swedes – was an important factor in explaining variations in migrant electoral participation. Behind these individual and cultural factors, there may be important structural factors. Notably, low participation in associational activities may be due to lack of human capital resources, which are in turn linked to high rates of unemployment and other class-­related factors (Adman and Strömblad 2000). However, recent analyses based on more thorough and reliable data at the municipal level show that turnout differences between migrants and non-­ migrants persist, even after controlling for gender, age, marital status, education and employment. These differences also hold true for second-­generation migrants (Öhrvall 2006: 77f.). These persistent minority-­related differences in electoral participation in Sweden are puzzling, as there are few formal barriers for electoral participation among naturalized migrants, and none of the colonial history of countries like France or the United Kingdom. Some research suggests that migrants’ experiences of discrimination may have a negative effect on turnout. Specifically, experiences of discrimination, marginalization, and nationalist political discourses that draw boundaries between a Swedish ‘Us’ and immigrant ‘Others’ have been shown to have a negative impact on turnout, especially among non-­ European and non-­Western migrant groups (Bäck and Soininen 2004; Dahlstedt and Hertzberg 2007). At the moment, our understanding of migrants’ capacity for political mobilization is underdeveloped. There has been a number of government-­sponsored information campaigns intended to increase migrant voter participation. However, the evaluation of one such effort in the 2002 elections suggests it had very limited impact (Bäck and Soininen 2004: IV). Political parties have also sought to mobilize migrant voters by nominating candidates from an immigrant background. These candidates have in turn tried to mobilize ethnic groups in the local elections. Knowledge of nominated candidates with the same ethnic background appears to have a positive effect on migrant voter participation (Bäck and Soininen 2004; Rodrigo Blomqvist 2000; 2005). This research also shows that that elected municipal politicians with an immigrant background do generally claim that they aspire to represent migrant groups as a whole (Rodrigo Blomqvist 2005). However, further research is needed to investigate patterns of collective identity among different migrant groups, and how these identities shape political action.

Voter turnout in comparative perspective   43 In conclusion, while we have some knowledge of migrant participation in Swedish elections, the picture is still fragmentary. Further studies should investigate the difference between first- and second-­generation migrants. There is also a need for better data at all levels of election. At the national level, we need data that would allow us to examine turnout differences by country of origin. At the local level, we need detailed and systematic data on participation among naturalized citizens. It would also be useful to have data on migrant turnout at the county level.

2.7  Norway Johannes Bergh The political opportunity structure for minority participation in Norway resembles that in other Scandinavian countries. The minority population23 is relatively small (5.2 per cent), but has grown rapidly, consisting mostly of first-­generation immigrants. There is a short history of active minority participation in Norwegian politics. The voting rights rules are that foreign nationals with at least three years’ legal residence have the right to vote in local elections, while voting in national parliamentary elections is restricted to (naturalized) citizens. The minority electorate is therefore larger in local elections than in parliamentary elections. The local level is thus the focus of most Norwegian research on minority participation, and most of the data to which we have access relate to those elections. What we do know about minority participation at the national level comes from a survey of the electoral rolls from 2005 conducted by Statistics Norway. Forty-­ nine per cent of eligible non-­Western minorities took part in that year’s parliamentary election. This may be compared to local elections by looking at turnout rates for naturalized citizens: 36 per cent of naturalized citizens with non-­ Western immigrant backgrounds voted in the local elections of 2003; 38 per cent did so in 2007. Turnout is always higher in national elections in Norway than in local elections, but these differences are larger among minority voters than among the electorate at large. Statistics Norway has collected extensive data on turnout of the minority population, especially so with respect to the last three local elections: 1999, 2003 and 2007. These data are drawn from the electoral rolls, and provide reliable information about turnout in the minority population.24 The most recent data from the local elections of 2007 show that turnout in the minority population was 35.6 per cent, compared to 61.2 per cent for the electorate at large. Minorities in Norway clearly have a lower level of electoral participation than the rest of the population. Furthermore, there is greater variation in minority participation from one election to the next than there is in the Norwegian electorate as a whole. Turnout declines sharply in the non-­Western immigrant electorate from 1999 (40.5 per cent) to 2003 (32.8 per cent), and is followed by a significant increase to 2007 (35.6 per cent).25 These changes are indicative of a volatile electorate. New immigrants arrive every year, leading to a rapid increase in the size of the minority electorate.

44   Voter turnout in comparative perspective There were almost twice as many such eligible voters in 2007 as there were in 1999. The changing turnout percentages from one election to the next are in part a reflection of major changes in the denominator in that equation (compositional effect). A further source of volatility in the minority electorate is the difference in turnout between groups of minorities. Sri Lankans, Pakistanis, and to a lesser extent Indians, have a fairly high turnout. Moroccans and Chinese are at the other end of the scale. These differences between individual groups are for the most part consistent over time, and similar patterns are evident in other countries (Bäck and Soininen 1998; Togeby 2003a). Evidence suggests that these differences can partly be explained by the demographics of each group. Those with a long history of residence in their adopted country and a close-­knit ethnic community tend to have higher rates of participation. There appears to be a correlation between the demographic concentration of an ethnic group and electoral participation (Bergh and Bjørklund 2003). Turnout is highest in municipalities with large minority populations. There is also more minority representation in these municipalities, and some active mobilization of minority voters by political parties. Perhaps turnout is affected by the presence of minority candidates, but it could also be that politically active minority communities beget minority politicians. As for party attempts to mobilize minority voters, these are limited to a few cases, with no discernible effect on turnout. Although there are notable differences in turnout among groups of minorities, the most striking differences are those among minority turnout on the one hand, and native turnout on the other. Minorities differ from the native population in a number of respects, beyond the fact that they have a minority background. Most of Norway’s minorities are immigrants. It seems reasonable to believe that recent arrivals have less interest in or knowledge of Norwegian politics, and thus lower turnout, than those who have resided in the country for most of their lives. There is a correlation between turnout and years of residence in our data.26 Electoral participation goes up when people have stayed in the country for a significant period of time. This partially explains the low turnout in the minority population. The electorate continues to expand with new arrivals, whose turnout is at a low level. In longitudinal cohort analyses, electoral participation rises from one election to the next. Another potential explanation for the differences in turnout between minorities and the majority population is variation in social background. The composition of the minority electorate differs from the general electorate on a number of characteristics that tend to correlate with turnout. Specifically, the minority electorate is younger than the rest of the population, and the average minority voter has a lower socio-­economic status (as measured by education and income) than others. The extent to which these factors explain low turnout in the minority electorate can be tested by weighting the turnout data. What would electoral participation in the minority electorate be if the age composition was identical to that of the entire Norwegian electorate? In addition, what if minorities had the same educational level or income level as the majority population?

Voter turnout in comparative perspective   45 Table 2.5 Minority participation in the 2007 Norwegian local elections, weighted by background variables, and compared to the general level of turnout (%) Immigrants of Asian, African and Latin-American background Not weighted

Weighted by age

Weighted by income level

Weighted by level of education

Weighted by age, income and education






Whole Norwegian electorate


Note N = 7,300 (immigrant data); the level of electoral participation for the whole electorate is the official number, published by Statistics Norway.

The second column in Table 2.5 shows the minority turnout given an age distribution that is identical to the electorate at large. Turnout goes up compared to the first non-­weighted column. A similar weighting by income level has a comparable effect. The same holds true for weighting by level of education. Finally, the data are weighted by all these characteristics simultaneously. Electoral participation rises to 46.8 if the composition of the minority electorate is adjusted to reflect the composition of the general electorate in terms of age, income and level of education. This suggests that the composition of the minority electorate with respect to these variables plays a large part in explaining the generally low turnout among minorities in Norway. Research on Norwegian minority turnout is nevertheless at an early stage. The extensive amount of available data should foster continued research in this area. The sources of variation in turnout between minority groups should be explored further. The reasons for the low levels of turnout among minorities are also an issue for continued research. Some of our findings suggest that minorities, given time, will get to the same levels of turnout as the majority population; other findings however, point to lasting differences.

2.8  Denmark Lise Togeby Only Danish citizens are eligible to vote in Danish national elections. However, since 1981, immigrants without Danish citizenship also have the right to vote in local elections after three years of uninterrupted legal residence. The turnout of immigrants – both Danish and foreign citizens – is well documented by way of a comprehensive register-­based study of the combined national and local election in 2001 (Elklit et al. 2005), supplemented by a smaller register-­based study from the local election in 1997 (Togeby 1999). The 2001 study covered about 130,000 immigrants of which two-­thirds come from non-­Western countries (Togeby 2007). Turnout varies with citizenship and country of origin. At the 2001 combined election, the turnout of ethnic Danish citizens was 87 per cent; for Danish citizens of non-­Western origin it was 68 per cent; and for foreign citizens of

46   Voter turnout in comparative perspective non-­Western origin turnout was 47 per cent. Even though voter turnout among non-­Western immigrants in Denmark is lower than among native-­born Danes, the turnout is higher than in most other countries. The difference in turnout between Danish and foreign citizens was larger in 2001 than is normally the case at local elections: the combined election day increased the turnout among Danish citizens and at the same time decreased the turnout among foreign citizens. In a ‘normal’ local election, such as the one in 1997, the turnout of specific immigrant groups (e.g. Turks in Aarhus) actually surpassed turnout among native Danes (Togeby 1999: 673). Even if we focus on immigrants from non-­Western countries there are major differences in turnout among different minorities. Among the large immigrant groups in Denmark the turnout in 2001 (naturalized and non-­citizens taken together) varied between immigrants from Sri Lanka with 69 per cent turnout, and immigrants from Lebanon with 39 per cent turnout. Among the high turnout groups we also find immigrants from Turkey, Poland and Iran. As Table 2.6 shows, differences in participation between Danish citizens of foreign origin and non-­citizens may be observed in all groups. The same pattern occurs in Norway. It is difficult to give well-­founded explanations for the differences between the different minorities, but the following factors appear to have an impact: the social and educational resources of the group’s individual members, the networks connecting the members of the group, the political experiences carried along from the country of origin, and the collective mobilization of the group in Denmark (Togeby 2003b). There are also variations among individual immigrants across ethnic groups based on socio-­demographic and migration-­related characteristics, and generally the socio-­demographic characteristics have a stronger impact on immigrants than on native-­born Danes (Togeby 2007). Turnout is higher among middle-­aged Table 2.6 Electoral participation by citizenship and country of origin in local elections in Denmark 2001 and in Norway 2003 (%) Country of origin

Denmark 2001 All



Sri Lanka Turkey Poland Iran Bosnia-Herzegovina Pakistan Vietnam Morocco Iraq Somalia

69 60 58 57 55 55 49 48 47 46

76 72 72 67 75 63 58 60 63 66

60 53 34 34 53 46 32 32 34 42

Turnout total


Source: Elklit et al. (2005: 140, 147).

Norway 2003 █ Citizens Non-citizens 57 42 39 28 41 40 30 28 27 33 59

39 24 25 23 20 40 – – 19 23

Voter turnout in comparative perspective   47 immigrants than among young or old immigrants, higher among well-­educated than less educated immigrants, higher among employed than non-­employed immigrants, and higher among immigrants who have lived for a long time in the local community, than among newcomers. Interesting enough, however, the differences in turnout between men and women are generally small – with a few exceptions. Among Lebanese, for instance, men have a 12 percentage point advantage over women. In addition, immigration-­related characteristics have an impact: length of stay in the country, naturalization and marriage to a Dane. Finally, in some ethnic groups, the second generation’s turnout is higher than the first generation’s, whereas in other groups it is not. The prime explanation for the relatively high turnout among immigrants in Denmark is the electoral system (Togeby 2008). The combination of a PR system with the possibility for preferential voting has created a political opportunity structure containing strong incentives for the collective mobilization of minority groups. It is easy for minority groups to get representation in the municipal councils, and immigrants are also relatively well represented in the councils. Accordingly, ethnic minority groups have strong incentives to mobilize in connection with local elections. Some ethnic groups do collectively mobilize in some municipalities, especially in situations where a countryman is nominated for election. One of the consequences of this mobilization is that the turnout of the same ethnic group varies considerably among municipalities. This pattern appears to be most pronounced among the Turks, who are also the largest immigrant group in Denmark. For instance, at the 2001 election 74 per cent of Turks in Aarhus municipality voted, compared to 49 per cent in the comparable municipality, Odense. Several Turks were on the lists in Aarhus and two were elected to the city council (Togeby 2008), whereas no Turks were nominated in Odense. However, only immigrants appearing on the big parties’ lists have a chance of being elected. There are only a few examples of candidates from specific immigrant lists getting a seat. This way, the electoral system pushes the political parties and the ethnic minority groups towards cooperation. The political parties need the immigrant vote to get more seats and the immigrants need a slot on the dominant parties’ lists to get elected. As far as we know, apart from placing immigrants on the party list, Danish parties have not done anything to mobilize immigrant voters. The collective mobilization appears to be the work of the immigrant groups themselves, facilitated by the opportunity structure embedded in the Danish electoral system. As it has been possible to base the analysis of immigrant turnout in Denmark on register data, our knowledge of immigrant participation is in many ways better founded than in most other countries. The biggest problem is that we have no continuous data collection. The data collection in connection with the 2001 combined local and national election was a one-­shot study. There has been no comparable data collection at later elections, and no further data collection is planned at the moment. Added to this, we have no information about the intra-­ party activities in connection with the nomination of immigrant candidates or the canvassing of the immigrant vote.

48   Voter turnout in comparative perspective

2.9  Germany Claudia Diehl and Andreas M. Wüst The political participation of migrants is a rather neglected topic in integration research, and information on their participation in elections is particularly limited. Non-­EU citizens cannot participate in elections in Germany at all, while EU citizens are entitled to vote in local elections and to register for European Parliament elections. For elections to the national parliament, the Bundestag, only naturalized migrants or their German-­born descendants are eligible to vote. In 2002, an estimate based on poll data suggested that barely 5 per cent of voters in the national election were naturalized citizens (Wüst 2004: 344). In 2005, Germany’s annual micro-­census for the first time not only differentiated between citizens and non-­citizens, but also between inhabitants with and without a migration background. Based on data for the first and second generation, it may be projected that a little under 9 per cent of eligible voters had a migration background – compared to about 20 per cent of the general population. This group not only comprises naturalized labour migrants and their descendants, but also naturalized asylees as well as ‘ethnic Germans’ and their children. Ethnic Germans immigrated in large numbers from Eastern Europe and from the former Soviet Union in the 1990s and acquired German citizenship prior to or upon arrival. Asylum seekers (e.g. from Afghanistan) have high naturalization rates because unlike other third-­country nationals, they do not have to renounce their former citizenship. Among labour migrants, Turkish immigrants are the largest 100

Germans EU citizens

80 68







69 64







40 27




22 18

B 95

B 99


B 01



B 06


HB HB HB 99 03 07


22 18



HH HH HH HH 97 01 04 08

S 99

S 04

Figure 2.1 Turnout rates of German and EU citizens in local elections since 1995 (sources: Election or Statistical Offices; B = Berlin, HB = Bremen, HH = Hamburg, S = Stuttgart).

Voter turnout in comparative perspective   49 group and also display the highest propensity to acquire a German passport. It is hard or even impossible to identify naturalized migrants, their offspring and ethnic Germans in existing data sources. Only about 7 per cent of EU migrants living in Germany registered to take part in the European Parliament elections in 2004. Due to a lack of data, one can only assume that most of them did actually participate in these elections. Statistics on EU migrants’ turnout in elections at the local level are available in a few municipalities that differentiate between native citizens and EU migrants. These reveal that EU migrants’ turnout rates are significantly lower than the rate among native Germans (Figure 2.1). Analyses of data from the city of Stuttgart (1999 and 2004) further suggest that migrants from Southern EU countries participate less in elections than those from Northern and Western Europe (Schwarz 2004). In contrast to the EU citizens, ethnic Germans have similar turnout rates as natives (Klinger 2001). In addition, according to the Politbarometer surveys of October 2001 to September 2002, self-­reported participation among naturalized citizens is just a few percentage points lower than among native Germans (Wüst 2003). The electoral turnout of naturalized citizens, it appears, barely differs from turnout among native Germans. Several studies have sought to identify factors that influence the degree and nature of migrants’ political participation. According to the Civic-­Voluntarism model, the central factors influencing political engagement are found in behavioural resources relevant to participation, such as time, money and civic skills, and in personal predispositions, such as political interest, personal belief in one’s own political efficacy and access to recruitment networks (Brady et al. 1995; Verba et al. 1995). This perspective assumes a positive correlation between available resources and an actor’s willingness to participate politically. Civic skills, in particular, are mediated by individual socio-­economic background (and by one’s non-­political affiliations), thereby increasing both one’s willingness and capacity to become politically active. On the macro level, the political opportunity structure affects migrants’ voting behaviour not only directly – by granting or not granting access to citizenship and participation – but also indirectly. It has been argued that long-­term political exclusion may lead to demobilization and ongoing homeland orientation of immigrants, beyond their mere inability to cast a ballot (Koopmans and Statham 2003). Participation opportunities also influence the likelihood and success of organizing ethnically based interests and the nomination of ethnic candidates. With regard to the factors just mentioned, the situation in Germany points towards low levels of political participation among migrants: First of all, this group has on average lower educational degrees and occupational positions than natives. This applies particularly to Turkish and Italian labour migrants. Second, their political interest is comparatively low. Besides, a large majority does not identify with a German party and those who do so have substantially weaker partisan identification than natives. Analyses for labour migrants show that this gap to natives remains substantial even after controlling for age and education (Diehl and Blohm 2001). Third, research suggests that many immigrants in

50   Voter turnout in comparative perspective Germany are preoccupied with homeland affairs. Ongoing political trouble in their countries of origin, most importantly in Turkey and the former Yugoslavia, has contributed to migrants’ focus on homeland matters (Diehl 2002; Koopmans and Statham 2003; Ögelman 2003). Fourth, the political opportunity structure in Germany lacks important incentives for minority participation in elections. In the past, Social Democrats and Greens have been more prone to defend the immigrants’ interests, but even these parties have only reluctantly attempted to mobilize the immigrant vote. The rise in numbers of minority candidates is a recent phenomenon, but these candidates often do not perceive themselves as representatives of group interests. Accordingly, the comparatively high turnout rates of naturalized migrants mentioned above reflect the selectivity of this subgroup. Analyses based on data from the Socio-­Economic Panel (SOEP) and the micro-­census have shown that migrants with higher levels of political interest, and who are more integrated economically, are more likely to be already naturalized or to apply for citizenship in the near future (Diehl and Blohm 2008). Again, this holds true particularly for immigrants from Turkey, who accounted for about 25 per cent of all naturalized citizens in 2006 (Statistisches Bundesamt 2007). Given the combination of micro- and macro-­level determinants just described and based on the experience in countries that have granted voting rights to immigrants, one can expect that turnout rates would remain low even if all migrants were granted voting rights in Germany today. Political integration is thus neither an immediate nor an automatic consequence of emerging opportunities for participation. Furthermore, it appears that formal political participation itself has little direct impact on the individual’s integration and prospects. This does not imply that opportunities for participation cannot be an important catalyst for the integration of migrants over the long run. The more minority members will become potential voters, the higher the pressure for German parties to attract respective votes – for example, by placing immigrants on their candidate lists and by addressing the specific needs and problems of this group. This, in turn, would affect the political opportunity structure for migrants’ political participation and increases the incentives to become involved in matters of the immigration country rather than of the ‘homeland’. With the rise in naturalizations since the late 1990s and the introduction of an ius soli component in Germany’s citizenship law in 1999, the political participation of new citizens has become an increasingly important topic for integration and electoral research. From 2018 on, when the ‘ius soli Germans’ will reach voting age, the topic’s importance will arise once again. The data available do not, however, permit extensive and systematic analysis of turnout among citizens with an immigration background. Identification of these individuals in regular surveys, and especially in exit polls, would be an important steps towards learning more about turnout, prospectively over time and across generations. To allow for more theory-­driven analyses of processes of political socialization among ethnic minorities, longitudinal data are needed, especially on those groups that are not captured by existing data sources such as Germany’s Socio-­Economic Panel.

Voter turnout in comparative perspective   51

2.10  Austria Marcelo Jenny Immigrant voting behaviour in Austria is a seriously under-­researched topic, despite the fact that the country has one of the larger shares of inhabitants of foreign origin among European states. Austria is a latecomer in this respect, only recently viewing immigrants as political agents, rather than merely as objects of political debate. There are several reasons for Austria’s laggardness on this matter. These include the high level of political polarization on many immigrant­related issues over the past 20 years, very limited opportunities for immigrants to be politically active up until the 1990s, and a recent awareness of the sizeable number of eligible voters with an immigrant background – stemming from a rise in naturalizations from the 1990s onward (Çynar and Waldrauch 2006). In the recent national election of September 2008 there were 400,000 eligible voters with a migrant background, an estimate provided by the National Bureau of Statistics. This amounts to 6.3 per cent of the total electorate. The immigrant population in Austria is concentrated in the cities in general, and in Vienna in particular. Vienna is the country’s largest city by far, with approximately one-­fifth of Austria’s total population of 8.3 million. Immigrants with a foreign passport and naturalized immigrants accounted for about a quarter of the city’s population in the census of 2001 (Bauböck and Perchinig 2006: 738). In 2008, the capital was the home of 39 per cent of the foreign population in Austria. Only one-­third of these were from the 26 Western and Eastern EU countries. The capital also harbours the most diverse and vibrant immigrant communities and organizations (Waldrauch and Sohler 2004). Austrian immigration policy has followed the ius sanguinis principle to citizenship with ten years of residence as the main time requirement for citizenship application. Dual citizenship is accepted in rare cases only. This exclusionary approach extends to the political sphere. The constitution gives the right to vote and to stand as a candidate in national and Länder elections of the federal polity to Austrian citizens only. Austria’s accession to the EU in 1995 extended eligibility for European parliamentary elections and Austrian municipal elections to citizens from all EU member countries. However, an exception applies to the capital city Vienna. The Vienna Municipal Assembly is simultaneously the Land diet. Therefore EU citizens are not allowed to vote in municipal elections in Vienna. They may participate only in elections to the city’s less important district assemblies (of which there are 23). In addition, foreigners can vote in elections to some corporatist bodies, such as the regional chamber of labour (Thienel 2007; Valchars 2007). In 2002 the Social Democratic Land/city majority government in Vienna, with support from the Greens, passed a law extending active and passive suffrage rights in the district elections to all foreign inhabitants with five years’ residence. However, the new regulations were never applied in practice. The Land’s Christian Democrats and the Freedom Party sent the law to the Constitutional Court, which repealed it in 2004 as being in conflict with the constitution.

52   Voter turnout in comparative perspective Ahead of that voting rights amendment an agency of the Viennese Land government commissioned a survey among 700 naturalized (N = 272) and non-­ naturalized (N = 426) immigrants. That survey is still the only available study providing some rudimentary data on the topic of immigrant turnout in elections (Jenny 2003; SORA 2002). The survey included a question on prior voting experience in Austria and a question on voting intention in the municipal election if provided with the right to vote – a topic of considerable public debate at the time of the study. The question on participation in the last municipal election of 2001 applied to naturalized immigrants only. Under half of them, 48 per cent, indicated that they indeed had voted in that election. This is 17 points below the official turnout rate for that election of 65 per cent. Among the reasons for the large difference of turnout among naturalized immigrants, compared to the general electorate, was a lack of accurate information about their voting rights. Notably, 10 per cent of naturalized immigrants surveyed answered that they had no right to vote, and another 8 per cent said they did not know whether or not they were entitled to vote. Due to the small number of cases, analysis by country of origin was restricted to the two quantitatively most important countries of origin – Turkey and the then still Federal Republic of Yugoslavia (Serbia and Montenegro) – while all other countries of origin were combined in a category of remainders. The participation rates among these three groups were 48, 42 and 51 per cent, respectively. Turnout was higher among men (53 per cent) than among women (45 per cent). All the numbers reported have to be read with care, as surveys on recalled voting are subject to producing higher than actual turnout rates (over-­reporting). Unfortunately, population data on recalled voting are not available for external validation of the survey. The second question on the intention to vote was presented to naturalized immigrants who had not yet participated in a municipal election, and to non-­nationals. Forty-­three per cent said they would ‘definitely’ vote, 27 per cent would ‘probably’ vote, 8 per cent said they would ‘rather not’, and 4 per cent said they ‘would definitely not’ participate in the election. Seventeen per cent did not know. There was a small gender gap: women were slightly less inclined to vote than men. There was also a more pronounced education gap: 56 per cent at the highest educational level said they would ‘definitely’ vote versus 38 per cent at the lowest educational level. As the number of cases is larger for this question, more details for each country are available. The distribution of responses among Turkish, Serbian and Montenegrine, Croatian and Bosnian immigrants were all very similar, while Polish immigrants stood out. They signalled a high propensity to vote in elections if they were given the right. A linear regression model on the intention to vote question among non-­ naturalized immigrants found that several factors had a positive effect on inclination to vote. These included the degree of satisfaction with life in Austria, the level of German-­language skills, a sense of political efficacy and prior experience of voting in the country of origin. Candidates of immigrant background had

Voter turnout in comparative perspective   53 only a small positive effect on the propensity to participate in a municipal election (SORA 2002: 77). Further data on immigrant turnout are available through the second round of the European Social Survey (Jowell et al. 2005), in which Austria participated. That study includes 112 respondents who were not born in Austria, but who had acquired Austrian citizenship. Carried out in spring 2005, the survey included a question on participation in the previous national election of October 2002. Subtracting non-­eligible youth and refusals, the data include just 98 respondents from 29 different countries. The self-­reported turnout rate for this group is 69 per cent, whereas the official turnout rate in the election stood at 84 per cent – about 15 points higher. The sample is too small and heterogeneous to examine differences by country of origin. However, on the positive side, the ESS data confirm the pattern observed in the Vienna study and support our expectations. That is, immigrant voter turnout in national elections in Austria is considerably lower than the overall turnout.

2.11  Spain Mónica Méndez Lago and Santiago Pérez-Nievas Immigration is a very recent and astonishingly rapid phenomenon in Spain (Bayona 2008; Cebolla and González Ferrez 2008). At just over five million people, the proportion of foreigners in 2008 (comprising 11 per cent of the total population) was eight times larger than the proportion in 1996 (1.4 per cent). The origins of immigration flows have also changed considerably throughout this period. While the largest proportion in 1996 came from the 15 EU member states and Morocco, by 2008 there had been a dramatic increase in immigrants from new EU member states such as Rumania and Bulgaria, and from Latin-­ American countries, particularly Ecuador and Colombia.27 If we take into account that some of the earlier waves of migrants have already naturalized (particularly Latin-­American citizens, who need only two years of legal residence before applying for naturalization), the population with an immigrant background probably reaches about 13 per cent of the total population. This gives Spain one of the largest immigrant-­origin populations in Western Europe. Only Spanish citizens are eligible to vote in Spanish national or regional elections. Non-­citizens from EU member states have been allowed to vote and stand as candidates in local elections since 1999, as a result of the implementation of the Maastricht Treaty. Interestingly, Norwegian immigrants enjoyed the right to vote in local elections somewhat earlier, on the basis of a 1990 treaty between Spain and Norway. Thus, in the 1999, 2003 and 2007 local elections, a considerable proportion of the foreign population remained outside the election process. However, in a remarkable policy shift at its last party congress in July 2008, the PSOE central government is now pursuing treaties of reciprocity – similar to that signed with Norway in 1990 – that would grant voting rights to immigrants. This is a necessary step since the 1978 Spanish Constitution only contemplates voting rights for foreigners in local elections on the basis of reciprocity. After a slight

54   Voter turnout in comparative perspective reform in 1992, the Constitution allows passive (that is, eligibility to stand as a candidate) as well as active suffrage (eligibility to vote) to foreigners in local elections, although again on the basis of reciprocity (Méndez 2007). By March 2009 the Spanish government had already signed treaties of reciprocity with Argentina, Colombia, Ecuador and Peru, and treaties with Bolivia, Chile, Paraguay, Uruguay and Venezuela were near completion. In all cases, migrants are required a minimum of five years’ legal residence in order to be able to exercise the right to vote. Unlike EU residents who need register only once, non-­EU residents must register for every election, a requirement that obviously increases the cost of participation. This enlargement of the franchise in local elections will, no doubt, contribute to balancing migrants’ political rights in relation to the autochthonous population. However, the constitutional requirement of reciprocity might also exacerbate inequities among different migrants groups, since it is clear that not all countries will agree to treaties.28 The requirement also stands as a serious barrier for migrants from countries without democratic elections – such as the Chinese. Finally, not all treaties include the same terms. For example, in the case of Peru, reciprocity applies to both active and passive suffrage, while for Colombia it only applies to active suffrage. Since only a minority of immigrants thus far has naturalized, and because voting rights for non-­EU nationals are still quite new, we have as yet little data on immigrant voting behaviour in local elections, and none nationwide. The first three rows of Table 2.7 show voter turnout rates in Madrid and Barcelona, of naturalized immigrants belonging to three of the largest migrant groups: Moroccans, Ecuadorians, and a mixed group of other Andean nationalities (Colombian, Peruvians and Bolivians).29 Electoral turnout is clearly lower for immigrants than for the native-­born population; however, there are notable differences across migrant groups. The mixed Andean group is consistently the highest, while the Ecuadorian group is generally the lowest. These differences are not easily explained by migration factors. Notably, the Moroccans are the group with the longest tradition in Spain (Cebolla and González Ferrer 2008), and yet they do not show the highest turnout. Furthermore, the two Latino groups are relatively similar in culture, and face lower barriers to participation (through easier access to nationality), yet have considerable differences in voter turnout. We lack information on other factors, such as organizational density, that might play some role here. While the generally higher levels of immigrant turnout in Madrid do correspond with a historical trend of higher overall turnout in that city, we do not know whether there may be unique characteristics of each immigrant group across the cities. Overall, the numbers of naturalized immigrants eligible to vote in these elections was very small, and thus, in aggregate terms, their electoral turnout has not been viewed as especially relevant up until now.30 In contrast, the electoral impact of EU citizens in recent local elections has been potentially much larger. After the last round of enlargement, with the incorporation of Rumanians and Bulgarians, some areas – for example, the coastal

80 (284) 84 (289) 75 (290) 81 (288) 81 (290) 87 (291)

Barcelona Madrid

Barcelona Madrid

Barcelona# Madrid#

% registered % turnout among those registered

Local 20031

Regional 20031

National 20041

Local 2007 (Arganda)

– –

49 (35)***O 53 (17)*

41 (37)*** 47 (19)*

29 (34)***O 47 (17)*


– –

29 (24)***O 41 (17)***

26 (31)***O 31 (16)***

30 (23)***O 31 (16)***


– –

66 (56)*M, E 66 (32)

49 (59)***E 55 (33)**

60 (50)*M, E 61 (31)

Other Andeans2

22 (99) 57 (53)***

– –

– –

– –


Notes 1 Figures for 2003 and 2004 elections are percentage turnout among specified populations. 2 Colombians, Peruvians and Bolivians. *** Difference compared to the autochthonous group is statistically significant at the 99 per cent level; ** Sig. at the 95 per cent level; * Sig. at the 90 per cent level. M, E, O: Differences across Moroccans, Ecuadorians and Other Andean groups are statistically significant at the 95 per cent level. #: Significant difference between the authochthonous population in the two cities at the 95 per cent level.

Source: Morales et al. (2009) for data on Madrid and Barcelona; our own data for the 2007 local elections in Arganda (Centro de Investigaciones Sociologicas at

– 79 (359)



Table 2.7  Political participation of immigrant groups in Spain, national, regional and local elections (%, N in parentheses)

56   Voter turnout in comparative perspective towns of Alicante or commuter towns in Madrid’s metropolitan area – have seen EU resident concentrations grow as high as 20 to 30 per cent, while interior provinces have seen the number of EU citizens grow to as much as 5 per cent of their total populations (Méndez 2007: 11–12, 22). In fact, the actual impact of these groups in terms of electoral turnout has been less significant than these figures would suggest. Unlike Spanish citizens, EU citizens are required to register in the electoral roll. Although registration is required only once, this nevertheless enhances the cost of participation. In the first local elections in which they were entitled to vote, in 1999, an estimated 24 per cent of EU citizens registered to do so. Even if all these individuals did vote, their level of turnout fell far below the overall electoral turnout of 65 per cent in those elections. In the most recent local elections in 2007, following the last round of EU enlargement, the registration rate among EU citizens was just 25 per cent. Registration was highest among immigrants from the richest EU countries (e.g. French, British or Germans), near average for those from middle-­income countries (e.g. Portuguese), and lowest among the poorest EU nationalities (e.g. Romanians or Bulgarians).31 Survey data collected in 2007 from the commuter town of Arganda (outside Madrid capital city) corroborate these findings. While Romanians immigrants exceed 20 per cent of the total population in this city, we found that only 22 per cent of them registered to vote in the local election (Table 2.7).32 Of those registered, only 57 per cent did actually vote in the election – a figure higher than the average across other immigrant groups, but well below the 79 per cent turnout among registered autochthonous in the same survey. As with other immigrants, turnout is surprisingly low considering the cost of registering in the first place. Given the data limitations, it is still difficult to discern and explain patterns of political mobilization among immigrant groups. We did, however, explore this matter among the Romanian immigrants in our survey. Looking at both voter registration and actual turnout, men tended to be more mobilized than women (9-point gap), frequent churchgoers more mobilized than non-­frequent churchgoers (12-point gap), and older individuals more mobilized than youth. Those who placed themselves on the right side of the political spectrum were also more likely to register than those on the left. Lastly, as might be expected, Romanians who have lived in Spain longer were more likely both to register and vote. Very few candidates of immigrant background have been placed in top positions in the party lists, and even fewer have been elected for local or regional office. Although Spanish parties do not stand out as having transparent and democratic procedures for candidate selection, it may be that immigration is still too recent, and the immigrant franchise still too limited, for established parties to yet reach out in inclusive fashion to these groups. We therefore do not know whether the lack of immigrant candidates is a factor hindering immigrant mobilization, or whether it instead reflects the prevailing weak political mobilization among these groups. It appears, in the case of local elections, that the self-­ registration requirement may be the greatest institutional disincentive for electoral participation. However, the data on electoral turnout among naturalized

Voter turnout in comparative perspective   57 immigrants show that turnout rates are also low where this obstacle is absent. We suspect that social and economic disadvantages linked to the migration and settlement experience are also at work, and hope that future research will focus on exploring these factors.

2.12  Conclusion Andreas M. Wüst In this chapter, social scientists with a focus on migrant participation have documented data sources and analyses pertaining to the voter turnout of immigrants and visible minorities for 11 countries. The main conclusions are that suitable data are scarce and analyses are often quite idiosyncratic. Data sources range from voter registers (Scandinavia), partially cumulated general surveys (e.g. Canada, UK, Germany), to special minority-­focused surveys (e.g. France, Netherlands, Austria, UK). In none of the countries are data available for all political levels. While some countries focus largely or exclusively on local elections, others look almost exclusively at the national level. This difference in the scale of analysis is clearly linked to state-­level variations in voter eligibility. The chapter also shows that there are hardly any common standards for identifying eligible voters with an immigrant or visible minority background (exceptions are Canada and the UK). Consequently, the conditions for comparative analyses across countries are modest. Despite the lack of cross-­national data, country-­specific definitions and analyses do reveal some general patterns. Citizens with a migratory background participate less frequently in elections. The observed participation gap (compared to the majority population) is larger among visible minorities and among foreign citizens (who are eligible to vote at lower political levels in some of the countries). As documented for Norway, France, Canada and the UK, higher socio-­ economic status narrows this participation gap, and analyses in various countries report a decrease in the turnout gap as the length of residence in the country of immigration increases. Social capital seems to also have positive effects on turnout. This may occur through integration into the broad civil society (Germany, Spain), or through ethnic communities (Netherlands). Collective mobilization by minority or migration groups can also be a positive factor influencing political participation – however, much depends on whether or not there are structures in place (e.g. voting rules, party outreach) that serve as linkage mechanisms between ethnic communities and the political process. Various analyses for Denmark provide evidence of such effects, and there are also hints of such effects in Canada, the UK and France. This short summary of participation patterns among immigrants and their descendants points to a number of promising research agendas within and across various countries. The main obstacle to such research remains the lack of data. Future research depends critically on the establishment of standard measures of  immigration background and minority status, and on the collection of large samples of data through surveys, exit polls and overtime panels. Further,

58   Voter turnout in comparative perspective cross-­country analyses require common group definitions reflecting either the more ‘subjective’ identity-­based approach to race and ethnicity, or the more ‘objective’ approach to immigration background. Finally, there is a need for further research on specific questions, such as the effect of immigrant generation, social capital or collective mobilization on turnout. As this chapter has begun to suggest, there is potentially great return in approaching this research through comparative cross-­country analyses. At the very least, we hope that this approach will be taken more frequently than in the past.

Notes   1 The concept of ‘visible minority’ was established in the 1970s under the terms of the federal government’s Employment Equity Act. It is defined in that legislation as ‘persons other than Aboriginal peoples, who are non-­white in colour or non-­Caucasian in race’. The concept is widely used in Canadian public discourse, has been measured in the census since 1981, and has been further entrenched in multiculturalism legislation, including the Charter of Rights and Freedoms.   2 In 2006, 85.1 per cent of eligible foreign-­born people were Canadian citizens. Non-­ citizens are ineligible to vote in local (municipal and school board), provincial or national elections.   3 The most useful data so far have been available through combining Canadian elections studies over multiple elections. Another important data source is the Ethnic Diversity Study (EDS), carried out in 2002 by Statistics Canada. The EDS focuses on ethnic identity, experiences of discrimination, social networks and civic participation. In terms of electoral behaviour, it asks respondents to recall whether they voted in the last federal, provincial and municipal elections.   4 For a useful review of these programmes in Canada and other countries, see Tossutti 2007.   5 There is an enduring effect of race on voter turnout in the US. Recent US Census reports show a persistent gap in voter turnout of about 10 per cent between non-­ Hispanic white and black citizens, and a gap of about 20 per cent between non-­ Hispanic whites and both Hispanic and Asian citizens.   6 A law introduced in 2003 adds a new condition: the ‘assimilation to the French community’. This condition was applied in 2008 in refusing the naturalization of a Muslim woman due to her ‘religious practices which contradict the republican values’.   7 Since 2005, electoral surveys have followed an ethnic mainstreaming approach: along with traditional socio-­demographics, some indicators of ethnic origin have been systematically integrated.   8 These figures are estimates based on the Baromètre Politique Français cumulative sample of 21,776 registered voters. These estimates do not take into account the black Caribbean population which can live both in the overseas territory and in mainland France. The overall overseas population accounts for 3.5 per cent of the registered voter population.   9 This problem of small sample size also applies to EU citizens in France. To our knowledge, no systematic survey has yet been conducted of the electoral participation of this group. 10 Only 16.2 per cent of the registered population abstained in 2007. This is the highest level of turnout on historical record. 11 In both rounds of this election, statistical tests of differences in turnout between these minorities and the other non-­minority groups never fell below 0.293 for each coefficient. 12 Until recently, voter registration in France was a voluntary process.

Voter turnout in comparative perspective   59 13 This is due to the fact that these populations are younger, on average, than the general population. Our study accentuates this trend because it samples exclusively from the second and third generations of the minority populations. 14 The overall registration rate increased by four percentage points in 2007. Yet the increase was as high as 8.5 per cent in heavily diverse areas such as Seine-­Saint-Denis where the ‘suburbs crisis’ began. 15 The effect of the integration of immigrants in the party system and in political arenas on the level of turnout of immigrants has never been studied, as far as we know. 16 For the 1986 elections, see Pennings (1987) and Buijs and Rath (1986); for the 1990 elections, see Rath (1990). 17 Using this definition, which was chosen by the Dutch Association of Municipalities (VNG), Rhee (2002a) used samples drawn from the administration of local voting bureaux combined with (anonymous) data from the population register, while all other studies rely on exit polls in selected voting bureaux. The number of registered (potential) voters for each ethnic group in these bureaux was provided by the municipality, and turnout estimates are based on observations and/or survey questions on the country of origin of voters leaving the polling station. 18 To our knowledge, turnout of immigrants has not been a subject of study for regional elections or European elections. 19 For example, an opinion poll in 2002 showed very high levels of turnout (between 76 and 94 per cent) among migrants compared to the overall population (79 per cent) (Source: However, the conclusion of another study, using a sample of the voting register, was that turnout among all ethnic minorities was somewhat lower than overall turnout (Rhee 2002b). Data on eligible migrant-­origin voters from the 2006 national elections showed a higher turnout nationally than in local elections that year, though turnout was still lower than in the electorate as a whole. These data were collected by a market research firm (Foquz 2006). 20 Poll takers from various ethnic groups were employed to observe voters leaving the polling stations, and to assess visually their ethnic origin. The proportion of each ethnic group among voters was then compared to municipal data, reporting the size of each ethnic group for the selected polling stations. 21 Regarding gender, Pennings (1986: 64) found similar results for the Dutch, Turks and Moroccans in his study of the 1986 district council elections in Amsterdam. However, Surinamese women participated more than Surinamese men in these elections. 22 Calculations are from data provided by the Ministry of Interior, Department of Elections (2006). 23 Minorities, in this definition, are people with immigrant backgrounds from Asia, Africa or Latin America. People who themselves emigrated from those parts of the world, or whose parents did so, are included here. 24 All eligible voters in Norway are registered in the electoral or census rolls. Those who vote and are not abstainers are marked by a cross. In addition, the electoral rolls contain information about such variables as country of origin, age, gender, education, income and years of residence. 25 Turnout for the immigrant population as a whole (including Western and non-­Western immigrants) was 43 per cent, 37 per cent and 38 per cent, in 1999, 2003 and 2007, respectively. 26 Pearson’s r = 0.14 (individual-­level data from 2007). 27 The greatest contributors to migration in Spain are Romania (14 per cent), Morocco (12 per cent), Ecuador (8 per cent), UK (7 per cent), Colombia and Bolivia (5 per cent each), Germany, Italy, Bulgaria and Argentina (3 per cent each), Portugal, China, France, Brazil and Peru (2 per cent each). Figures refer to percentages among total foreigner populations as of January 2008 (Source: National Instituto Nacional de Estadística INE,

60   Voter turnout in comparative perspective 28 Spanish invitations to open treaty negotiations with Morocco and Algeria, for example, have had no response yet. 29 For more details on data sources, refer to Chapter 3.11. 30 Only 10 to 20 per cent of each immigrant group sampled for these surveys were naturalized Spanish citizens (Morales et al. 2010). 31 Of course, citizens from the former EU-­15 have had a longer period of time for registry. Nevertheless, considering the registration rate in 1999 was 24 per cent, and that this increased to only 25 per cent after the last round of enlargement, we might safely conclude that registration rates are slower (and that electoral mobilization is weakest) among citizens from the new EU member states, compared to among the EU-­15 (Mendez 2007: 18). 32 It should be noted that Romanians had a very short period of time to register for these elections, beginning in January 2007 after Romania’s entry to the EU.

References Adman, Per and Strömblad, Per (2000) Utanför demokratin. D. 3, Resurser för politisk integration, Norrköping: Integrationsverket. Anwar, Muhammad (1984) Ethnic Minorities and the 1983 General Election, London: Commission for Racial Equality. —— (2001) ‘The Participation of Ethnic Minorities in British Politics’, Journal of Ethnic and Migration Studies, 27(3): 533–49. Aytar, Osman (2007) Mångfaldens organisering: om integration, organisationer och interetniska relationer i Sverige, Stockholm: Acta Universitatis Stockholmiensis (AUS), Universitetsbiblioteket. Bäck, Henry and Soininen, Maritta (1996) Invandrarna, demokratin och samhället: om invandrarnas politiska deltagande i dagens Sverige, Göteborg: Förvaltningshögskolan. —— (1998) ‘Immigrants in the Political Process’, Scandinavian Political Studies, 21: 29–50. —— (1999a) ‘Invandrare som medborgare, väljare och politiker’, in Erik Amnå (ed.) Meborgarnas erfarenheter, SOU 1999: 113. —— (1999b) ‘Electoral Participation among Immigrants in Sweden: Integration, Culture and Participation’, in Steven Vertovec (ed.) Migration and Social Cohesion, Aldershot: Edward Elgar. —— (2004) Politisk annonsering eller nätverkande? Uppföljning och utvärdering av partiernas särskilda informationsinsatser till invandrarväljare vid 2002 års val, Göteborg: Förvaltningshögskolan. Bauböck, Rainer and Perchinig, Bernhard (2006) ‘Migrations- und Integrationspolitik’, in Herbert Dachs et al. (eds) Politik in Österreich: Das Handbuch, Vienna: Manz. Bayona, Jordi (2008) ‘La población de nacionalidad extranjera en España: Evolución y características demográficas’, Entelequía, 8: 118–37. Bergh, Johannes and Bjørklund, Tor (2003): ‘The Political Representation of Immigrants in Oslo and Copenhagen: A Study of Electoral Systems and Voting Behavior’, Comparative Social Research, 22: 101–21. Bilodeau, Antoine and Kanji, Mebs (2006) ‘Political Engagement among Immigrants in Four Anglo-­American Democracies’, Electoral Insight, 8(2): 43–8. Black, Jerome H. and Leithner, Christian (1988) ‘Immigrants and Political Involvement in Canada: The Role of the Ethnic Media’, Canadian Ethnic Studies, 20(1): 1–20. Bloemraad, Irene (2003) ‘Institutions, Ethnic Leaders and the Political Incorporation of Immigrants: A Comparison of Canada and the United States’, in Jeffery Reitz (ed.) Host

Voter turnout in comparative perspective   61 Societies and the Reception of Immigrants: Institutions, Markets and Policies, La Jolla, CA: Centre for Comparative Immigration Studies, University of California-­San Diego. Borevi, Karin (2002) Välfärdsstaten i det mångkulturella samhället, Uppsala: Acta Universitatis Upsaliensis, Univ.-bibl. distributör. Brady, Henry E., Verba, Sidney and Schlozman, Kay Lehman (1995) ‘Beyond SES: A Resource Model of Political Behaviour’, American Political Science Review, 89(2): 271–94. Buijs, Frank and Rath, Jan (1986) De stem van migranten en werklozen, De gemeenteraadsverkiezingen van 19 maart 1986 te Rotterdam, Leiden: Centrum voor Onderzoek van Maatschappelijke Tegenstellingen, Faculteit der Sociale Wetenschappen, Rijksuniversiteit te Leiden. Carty, R. Kenneth (1991) Canadian Political Parties in the Constituencies. Volume 23 of the Research Studies of the Royal Commission on Electoral Reform and Party Financing, Toronto: Dundurn Press. CBS (2007) Statline, Den Hagg/Heerlen: Centraal Bureau voor de Statistiek. Online: (accessed 22 February 2007). Cebolla, Héctor and González Ferrer, Amparo (2008) La inmigración en España, Madrid: Centro de Estudios Políticos y Constitucionales. Conservative Party of Canada (2007) ‘Conservative Party Conference 2007: Ethnic Outreach’, presented by Secretary of State, the Hon. Jason Kenney, at the Conservative Party Political Training Conference, 15–18 March. Cutts, David, Fieldhouse, Edward, Purdam, Kingsley, Steel, David and Tranmer, Mark (2007) ‘Voter turnout in British South Asian Communities at the 2001 General Election’, British Journal of Politics and International Relations, 9(3): 396–412. Çynar, Dilek and Waldrauch, Harald (2006) ‘Austria’, in Rainer Bauböck, Eva Ersbøll, Kees Groenendijk and Harald Waldrauch (eds) Acquisition and Loss of Nationality. Volume 2: Country Analyses, Amsterdam: Amsterdam University Press. Dahlstedt, Magnua and Hertzberg, Fredrik (2007) ‘Democracy the Swedish Way? The Exclusion of Immigrants in Swedish Politics’, Scandinavian Political Studies, 30: 175–203. Dekker, Lisette (2006) Vrouwen en allochtonen in de nieuw gekozen Tweede Kamer 2006, Amsterdam: IPP. Online: www.publiek-­ politiek/cijfers/tweede_kamer_2006 (accessed 16 April 2007). Dhillon, Sheila (2005) ‘Political Parties and Ethnic Participation: A Question of Access’, Canadian Issues, summer: 85–8. Diehl, Claudia (2002) Die Partizipation von Migranten in Deutschland, Rückzug oder Mobilisierung: Opladen. Diehl, Claudia and Blohm, Michael (2001) ‘Apathy, Adaptation, or Ethnic Mobilization? On the Political Attitudes of an Excluded Group’, Journal of Ethnic and Migration Studies 27(3): 401–20. —— (2008) ‘Die Entscheidung zur Einbürgerung: Optionen, Anreize und identifikative Aspekte’, in Frank Kalter (ed.) Migration und Integration, Wiesbaden: VS Verlag für Sozialwissenschaften. Duchesne, Sophie (2007) ‘Citoyenneté, nationalité et vote: Une association perturbée’, Pouvoirs, 120: 71–80. Electoral Commission (2005) Black and Minority Ethnic Survey, London: Electoral Commission. Online: ECBMEReportFINAL2_18810–13883__E__N__S__W__.pdf (last accessed 20 March 2009).

62   Voter turnout in comparative perspective Elklit, Jørgen, Møller, Birgit, Svensson, Palle and Togeby, Lise (2005) Gensyn med Sofavælgerne. Valgdeltagelse i Danmark, Århus: Aarhus Universitetsforlag. Fennema, Meindert and Tillie, Jean (1999) ‘Political Participation and Political Trust in Amsterdam: Civic Communities and Ethnic Networks’, Journal of Ethnic and Migration Studies, 25: 703–26. Fieldhouse, Edward and Cutts, David (2008a) ‘Diversity, Density and Turnout: The Effect of Neighbourhood Ethno-­Religious Composition on Voter Turnout in Britain’, Political Geography, 27(5): 530–48. —— (2008b) ‘Mobilisation or Marginalisation? Neighbourhood Effects on Muslim Electoral Registration in Britain in 2001’, Political Studies, 56(2): 333–54. Foquz etnomarketing (2006) Allochtonen bepalen door hoge opkomst 13 Kamerzetels, bijna allemaal gaan naar links, Nieuwegein: Foquz. Online:­allochtonen-tweedekamerverkiezing-­2006.html (accessed 14 February 2007). Garbaye, Romain (2005) Getting Into Local Power: The Politics of Ethnic Minorities in British and French Cities, Oxford: Blackwell. Hammar, Tomas (1979) Det första invandrarvalet, Stockholm, LiberFörlag. Heelsum, Anja van and Tillie, Jean (2006) Opkomst en partijvoorkeur van migranten bij de gemeenteraadsverkiezingen van 7 maart 2006, Amsterdam: IMES. Henderson, Ailsa (2005) ‘Ideal Citizens? Immigrant Voting Patterns in Canadian Elections’, Canadian Issues, summer: 57–60. Howe, Paul (2007) ‘The Political Engagement of New Canadians: A Comparative Perspective’, in Keith Banting, Thomas J. Courchene and F. Leslie Seidle (eds) Belonging? Diversity, Recognition and Shared Citizenship in Canada, Montreal: Institute for Research on Public Policy. Hyder, Goldy (2005a) ‘Real Challenge for Tories: Need to Bridge Rural—Urban Gap’, The Hill Times Online, 14 March. Online: hyder (accessed 18 May 2010). —— (2005b) ‘Gaining the Political Support of Minorities in Canada’, Canadian Issues (summer): 46–8. Jacobs, Dirk (1998) Nieuwkomers in de politiek. Het parlementaire debat omtrent kiesrecht voor vreemdelingen in Nederland en België (1970–1997), Gent: Academia Press. Jacobs, Dirk, Delwit, Pascal and Delmotte, Florence (2009) ‘Political Participation and Electoral Impact of EU-­citizens in the Brussels Capital Region: The October 2006 Local Elections’, in Roel De Groof (ed.) Brussels and Europe. The Position of Brussels in the World City Network – Interactions between the European Institutional Presence and the Brussels Capital Region, Brussels: Academic and Scientific Publishers. Jenny, Marcelo (2003) ‘Politikinteresse und Bereitschaft zur politischen Partizipation bei Migrantinnen in Wien’, Wiener Hefte zur Migration in Theorie und Praxis, 1: 127–40. Jowell, Roger and the Central Co-­ordinating Team (2005) European Social Survey 2004/2005: Technical Report, London: Centre for Comparative Social Surveys, City University. Klinger, Thorsten (2001) ‘Politische und gesellschaftliche Partizipation von Eingebürgerten in Hamburg’, Hamburg in Zahlen, 4: 132–5. Koopmans, Ruud and Statham, Paul (2003) ‘How National Citizenship Shapes Transnationalism: Migrant and Minority Claims Making in Germany, Great Britain and the Netherlands’, in Ewa Morawska and Christine Joppke (eds) Toward Assimilation and Citizenship in Liberal Nation-­States, London: Palgrave Macmillan.

Voter turnout in comparative perspective   63 Kymlicka, Will (1995) Multicultural Citizenship, New York: Oxford University Press. Lane Robert (1972) Political Man, New York: Free Press. Lapp, Miriam (1999) ‘Ethnic Group Leaders and the Mobilization of Voter Turnout: Evidence from Five Montreal Communities’, Canadian Ethnic Studies, 31(2): 17–42. LeBlanc, Daniel (2007) ‘Inside the Conservatives’ Election Plan: Tories Target Specific Ethnic Voters’, Globe & Mail, 16 October. Maxwell, Rahsaan (2010) ‘Political Participation in France among Non-­European Origin Migrants: Segregation or Integration?’, Journal of Ethnic and Migration Studies, 36(3): 425–43. Mendez, Mónica (2007) ‘The Participation of Non-­National EU Citizens in Spanish Local Elections’, Paper presented at the fourth ECPR General Conference in Pisa, 6–8 September. Meurs, Dominque, Paihle, Ariane and Simon, Patrick (2006) ‘Persistance des inégalités entre générations liées à l’immigration: l’accès à l’emploi des immigrés et de leurs descendants en France’, Population, 61(5–6): 763–801. Michon, Laure and Tillie, Jean (2003a) Amsterdamse polyfonie, opkomst en stemgedrag van allochtone Amsterdammers bij de gemeenteraads- en deelraadsverkiezingen van 6 maart 2002, Amsterdam: IMES. —— (2003b) ‘Politieke participatie van migranten in Nederland sinds 1986’, in Huib Pellikaan and Margo Trappenburg (eds) Politiek in de multiculturele samenleving, Amsterdam: Boom. MMMA Consultancy and Omi Consultancy (2005) Exploring and Measuring the Attitudes and Behaviour of Members of Minority Ethnic Communities in Northern Ireland to Participating in the Democratic Process Generally and with Specific Regard to Registering for and Voting at Elections, London: Electoral Commission. Online:–14951__N__.pdf (accessed 20 March 2009). Morales, Laura, Anduiza, Eva, Rodríguez, Elisa and San Martín, Josep (2010) ‘The Political Participation of Immigrants in Barcelona and Madrid: Voting Behaviour and Political Action’, in Amparo González Ferrer, David Moya and A. Viñas (eds) Aliens’ Political Participation in Europe, Barcelona: Fundació Pi i Sunyer. Noiriel, Gérard (1992) Le creuset français, Paris: Seuil. Ögelman, Nedim (2003) ‘Documenting and Explaining the Persistence of Homeland Politics among Germany’s Turks’, International Migration Review, 37(1): 163–93. Öhrvall, Richard (2006) ‘Invandrare och valdeltagande’, in Hanna Bäck and Mikael Gilljam (eds) Valets Mekanismer, Malmö: Liber. Pennings, Paul (1986) Migranten en de Amsterdamse deelraadsverkiezingen van 30 october 1985, Amsterdam: Afdeling Bestuursinformatie, Onderzoek en Statistiek & Subfaculteit der Algemene Politieke en Sociale Wetenschappen van de Universiteit van Amsterdam, Vakgroep Collectief Politiek Gedrag. —— (1987) Migrantenkiesrecht in Amsterdam. Een onderzoek naar de participatie en mobilisatie van etnische groepen bij de gemeenteraadsverkiezingen van 19 maart 1986, Amsterdam: Gemeente Amsterdam, Bestuursinformatie, Afdeling Onderzoek en Statistiek & Universiteit van Amsterdam, Subfaculteit Politicologie, Vakgroep Collectief Politiek Gedrag. Purdam, Kingsley and Fieldhouse, Edward (2002) Voter Engagement among Black and Minority Ethnic Communities, London: Electoral Commission. Online: alreport_11586–6190__E__N__S__W__.pdf (accessed 20 March 2009).

64   Voter turnout in comparative perspective Putnam, Robert (1993) Making Democracy Work: Civic Traditions in Modern Italy, Princetown, NJ: Princetown University Press. Qadeer, Mohammad and Kuman, Sandeep (2006) ‘Ethnic Enclaves and Social Cohesion’, Canadian Journal of Urban Research, 15(2) Supplement: 1–17. Rana, F. Abbas (2007) ‘Conservatives to Target Urban, Ethnic Ridings in Next Election’, The Hill Times, 22 January. Rath, Jan (1990) Kenterend tij. Migranten en de gemeenteraadsverkiezingen van 21 maart 1990 te Rotterdam, Utrecht: RUU, Vakgroep Culturele Antropologie. Reitz, Jeffrey G. and Banerjee, Rupa (2007) ‘Racial Inequality, Social Cohesion and Policy Issues in Canada’, in Keith Banting, Thomas J. Courchene and F. Leslie Seidle (eds) Belonging? Diversity, Recognition and Shared Citizenship in Canada, Montreal: Institute for Research on Public Policy. Rhee, Marn van (2002a) Analyse van de opkomst bij de gemeenteraadsverkiezingen in Rotterdam op 6 maart 2002, Rotterdam: COS. —— (2002b) Analyse opkomst Tweede Kamerverkiezing 2002, Rotterdam: COS. Richards, Laura and Marshall, Ben (2003) Political Engagement among Black and Minority Ethnic Communities: What we Know, What we Need to Know, London: Electoral Commission. Online: pdf_file/0011/16112/BMEresearchseminarpaper_11354–8831__E__N__S__W__.pdf (last accessed 20 March 2009). Rodrigo Blomqvist, Paula (2000) ‘Personvalets betydelse för valdeltagandet’, Integrationsverkets rapportserie, 2000: 17. —— (2005) Närvarons politik och det mångetniska, Sverige: om att ta plats i demokratin, Göteborg, Göteborgs universitet. Saggar, Shamit (1998) The General Election 1997: Ethnic Minorities and Electoral Politics, London: Commission for Racial Equality. —— (2000) Race and Representation: Electoral Politics and Ethnic Pluralism in Britain, Manchester: Manchester University Press. Sanders, David, Clarke, Harold, Stewart, Marianne and Whiteley, Paul (2005) The 2005 General Election in Great Britain: Report for the Electoral Commission, Colchester: University of Essex. Online: 0007/47257/BESreportforTheElectoralCommission_18771–13876__E__N__S__W__.pdf (last accessed 20 March 2009). SCB (2002) Electoral Participation of Foreign Citizens in the 2002 Municipal Council Elections. Statistiska meddelanden, ME 14 SM 0301, Stockholm: Statistics Sweden. Schwarz, Thomas (2004) ‘Wahlbeteiligung der Unionsbürger bei der Gemeinderatswahl 2004 in Stuttgart’, KomunIS 8: 203. Silberman, Roxanne and Fournier, Irène (2006) ‘Les secondes générations sur le marché du travail en France: une pénalité ethnique ancrée dans le temps’, Revue Française de Sociologie, 47(2): 243–92. SORA (2002) Erwartungen der Betroffenen an ein Ausländerwahlrecht: Ein Umfrage unter MigrantInnen in Wien, Vienna: Typoscript. Soroka, Stuart N., Johnston, Richard and Banting, Keith (2007) ‘Ties that Bind? Social Cohesion and Diversity in Canada’, in Keith Banting, Thomas J. Courchene and F. Leslie Seidle (eds) Belonging? Diversity, Recognition and Shared Citizenship in Canada, Montreal: Institute for Research on Public Policy. Stasiulis, Daiva K. and Abu-­Laban, Yasmeen (1991) ‘The House the Parties Built: (Re) Constructing Ethnic Representation in Canadian Politics’, in K. Megyery (ed.) Ethno-­ Cultural Groups and Visible Minorities in Canadian Politics. Volume 7 of the Research

Voter turnout in comparative perspective   65 Studies of the Royal Commission on Electoral Reform and Party Financing, Toronto: Dundurn Press. Statistisches Bundesamt (2007) Bevölkerung und Erwerbstätigkeit – Einbürgerungen, Topic Series 1, 2.1., Wiesbaden: Statistisches Bundesamt. Statistics Sweden (2008) Database 2006 and 2008, Stockholm: Statistics Sweden. Online: Thienel, Rudolf (2007) ‘Integration als rechtliche Querschnittmaterie’, in Heinz Fassmann (ed.) 2. Österreichischer Migrations- und Integrationsbericht, Klagenfurt: Drava Verlag. Tillie, Jean (1994) Kleurrijk kiezen, Opkomst en stemgedrag van migranten tijdens de gemeenteraadsverkiezingen van 2 maart 1994, Utrecht: NCB. —— (2000) De etnische stem, opkomst en stemgedrag van migranten tijdens gemeenteraadsverkiezingen, 1986–1998, Utrecht: Forum. Togeby, Lise (1999) ‘Migrants at the Polls. An Analysis of Immigrant and Refugee Participation in Danish Local Elections’, Journal of Ethnic and Migration Studies, 25(4): 665–84. —— (2003a) ‘Political Mobilization of Ethnic Minority Women’, in Grete Brochmann (ed.) The Multicultural Challenge, Amsterdam: Elsevier. —— (2003b) Fra fremmedarbejdere til etniske minoriteter, Århus: Aarhus Universitetsforlag. —— (2007) ‘Immigrant Exceptionalism? Immigrant Turnout in Local Elections in Denmark’, Paper presented at the ECPR joint sessions of workshops, Helsinki, May. —— (2008) ‘The Political Representation of Ethnic Minorities – Denmark as a Deviant Case’, Party Politics, 14: 325–43. Tossutti, Livianna (2005) ‘Electoral Turnout and Canada’s Changing Cultural Makeup’, Canadian Issues, summer: 53–6. —— (2007) ‘The Electoral Participation of Ethnocultural Communities’, in Elections Canada Working Paper series on Electoral Participation and Outreach Practices, Ottawa: Election Canada. Online: ethnocultural_e.pdf (accessed 22 July 2009). Valchars, Gerd (2007) ‘Wahlrechte von NichtstaatsbürgerInnen in Österreich’, in Heinz Fassmann (ed.) 2 Österreichischer Migrations- und Integrationsbericht, Klagenfurt: Drava Verlag. Verba, Sindey, Schlozman, Kay Lehman and Brady, Henry E. (1995) Voice and Equality: Civic Voluntarism in American Politics, Cambridge, MA: Harvard University Press. Waldrauch, Harald and Karin Sohler (2004) Migrantenorganisationen in der Großstadt: Entstehung, Strukturen und Aktivitäten am Beispiel Wien, Frankfurt am Main: Campus Verlag. White, Stephen, Nevitte, Neil, Blais, André, Everitt, Joanna, Fournier, Patrick and Gidengil, Elisabeth (2006) ‘Making Up for Lost Time: Immigrant Voter Turnout in Canada’, Electoral Insight, 8(2): 10–16. —— (2008) ‘The Political Resocialization of Immigrants: Resistance or Lifelong Learning?’, Political Research Quarterly, 61(2): 268–81. Wüst, Andreas M. (2003) ‘Das Wahlverhalten eingebürgerter Personen in Deutschland’, Aus Politik und Zeitgeschichte, B53: 29–38. —— (2004) ‘Naturalised Citizens as Voters: Behaviour and Impact’, German Politics, 13(2): 341–59.

3 Party choices among immigrants and visible minorities in comparative perspective

If immigrants and their descendants participate in elections they have the potential to change the outcomes of those contests and, consequently, the composition of government. They become subjects, and even agents of politics, rather than objects of policy. From the perspective of political sociology, the most interesting question is whether patterns of migrant and minority voter choice are different from patterns among the non-­migrant and non-­minority population. Where this is the case, immigrants and their descendants may have a lasting impact on party competition, and also potentially on policy issues such as immigration and citizenship, anti-­discrimination and racial equality, and international affairs. This chapter surveys research on party choice in 11 liberal democracies. The contributors report on available data for national, regional and local elections, focusing on the party choices of immigrants and minorities compared to the majority population. In their overviews, the authors also seek to address potential explanatory factors of migrant and minority voter choice according to the three dominant approaches for analysing electoral behaviour: socio-­demographic characteristics (Columbia School); long-­term attachments, and the role of issues and candidates (Michigan School); economic calculus and ideological proximity (Rational Choice). In addition, the specific party political context in each country is discussed.

3.1  Canada Karen Bird It is exceedingly difficult to provide any simple explanation for patterns of vote choice among immigrants and visible minorities in Canada. There are three main reasons for this. First, the population at the centre of this question is one of extraordinary heterogeneity in terms of ethnicity, national origin, religion, period of settlement and socio-­economic status. Second, Canadian politics is marked by highly distinctive patterns of voting across regions and official language groups, and these differences are, to a certain extent, superimposed across the voting patterns of migrant and visible minority Canadians. Finally, Canadian party politics has undergone a remarkable transformation over the past two decades. Canadians have witnessed during this time the emergence of two important regional protest

Party choices in comparative perspective   67 parties (the sovereignist Bloc Québecois, and the Reform Party in the west), followed by the ascendance of the new Conservative Party of Canada (CPC). As in previous cases of party transformation, this has ‘opened the system to new participants and forced the parties to invent new ways of doing Canadian politics’ (Carty et al. 2000: 5). The most important manifestation of this change for our purposes is the increasing competition between the two main ‘brokerage’ parties – the Liberals and the CPC – over the important immigrant vote. Traditionally, immigrants and visible minorities in Canada have been strong supporters of the centrist Liberal Party. Blais (2005) reports that the probability of voting Liberal in Ontario is 23 points higher when one is a visible minority (of Asian, Latin American or African origin), and 25 points higher when one is a visible minority immigrant, than when one is of British origin. These effects are politically significant, given that Ontario is the largest province in the country (thus electing the most seats to the House of Commons), and has the largest share of the country’s immigrant and visible minority population.1 Yet there are important regional variations in minority voting, due both to distinctive patterns of minority settlement, and to regional variations in the Canadian party system. For example, visible minorities in Atlantic Canada tend to support the left-­wing New Democratic Party. Many of the minorities in that region are descended from former slaves who came to Canada at the time of the American Revolution, ‘hence they are not immigrants and have different concerns and interests’ (Gerber 2006: 76). Regional differences are also apparent in Quebec, where the sovereignist Bloc Québecois is strongly rejected by immigrant and visible minority voters. There are few clear differences in federal parties’ policies towards immigration and multiculturalism, which raises the question why support for the Liberal Party in the immigrant and minority population has been historically so strong. The Liberals have been Canada’s most successful political party of the post-­war period, forming the government for all but 20 of the past 65 years, and so part of the reason may be related to incumbency. Blais (2005) finds no support for the hypothesis that immigrant support for the Liberal Party is stronger among those who entered the country under a Liberal administration. An alternative possibility is that immigrant voters behave as rational actors, backing the Liberals because they want their constituency MP to be close to the centre of power and able to dispense benefits to their community. Henderson (2005) finds no support for this hypothesis among immigrant voters, though it does factor into Canadian-­ born voting patterns. She finds that partisan attachment structures immigrant vote patterns to a large degree. All federal parties have increasingly moved immigration and ethnic minority issues to the centre of their policy agendas, and have adopted various strategies to court immigrant voters (e.g. Barber 2005; Hyder 2005b; Ibbitson 2005; LeBlanc 2007). However, it is the Conservative Party of Canada that has the most concerted strategy in this regard (see Chapter 2.1). The CPC has actively courted the ethnic vote in the belief that growing numbers of immigrants (especially skilled workers, business-­class investors and those from non-­Western

68   Party choices in comparative perspective societies) hold right-­of-centre views on issues such as tax relief or gay marriage. The Conservatives have also, in recent elections, nominated more visible minority candidates than other parties. The Conservative strategy may be working. For example, in the most recent election in 2008, the Liberals won 49 of the 80 ridings with visible minority populations greater than 20 per cent, down from 53 of such ridings in 2006. In comparison, the Conservatives won 18, up from just 12 in 2006, while the NDP held steady with 11. Because constituency-­level results do not allow disaggregation of minority versus non-­minority voter support, it is hard to tell whether the Conservatives are really winning more support from minority voters. To examine this question, I have applied an experimental design to assess the direct impact of candidate ethnicity on voter support (Bird 2009). The experiment (inspired by Brouard and Tiberj, Chapter 7) involved a large, ethnically diverse sample of university students who were presented with a single hypothetical candidate said to be running in the 2008 Canadian federal election. Unknown to the subjects, the candidate’s ethnicity and party identity were randomly varied. The study produced two major findings. First, the candidate’s ethnicity had a statistically significant impact on assessments made by visible minority subjects. Among visible minority respondents, 83.8 per cent of those presented with a visible minority candidate compared to 69.8 per cent of those presented with a non-­ minority candidate viewed that candidate as someone who ‘can understand my problems’ (p